학술논문
Monday, December 4, 2006Poster Session III7:30 a.m. - 4:30 p.m
Document Type
Author abstract
Author
Source
Epilepsia. Oct, 2006, Vol. 47 Issue s4, p204, 86 p.
Subject
Language
English
ISSN
0013-9580
Abstract
To purchase or authenticate to the full-text of this article, please visit this link: http://dx.doi.org/10.1111/j.1528-1167.2006.00001_7.x Byline: Koichi Akaike (1), Shigeya Tanaka (2), Shin-Ichi Imamura (3), Motofumi Kasugai (1), Hideyuki Matsukubo (1), Hideshi Tojo (1), Akira Sano (1) Abstract: .sup.1 Koichi Akaike , .sup.2 Shigeya Tanaka , .sup.3 Shin-Ichi Imamura , .sup.1 Motofumi Kasugai , .sup.1 Hideyuki Matsukubo , .sup.1 Hideshi Tojo , and .sup.1 Akira Sano (Department of Psychiatry, Field of Social and Behavioral Medicine, Kagoshima University Graduate School of Medical and Dental, Kagoshima, Kagoshima, Japan ; Neurosurgery, Tanaka Neurosurgical Clinic, Kagoshima, Kagoshima, Japan ; and Department of Neurosurgery, Kagoshima University Graduate School of Medical and Dental, Kagoshima, Kagoshima, Japan) Rationale: Gabapentin (1-(aminomethyl) cyclohexane acetic acid) is used clinically to treat partial as well as generalized seizures. While various experimental studies have investigated the effects of gabapentin in animal models of epilepsy, the precise mechanism of action of gabapentin against limbic seizures remains unclear. We therefore studied the efficacy of gabapentin during limbic status epilepticus induced by microinjection of kainic acid (KA) into the amygdala. Methods: Fourteen male Wistar rats (250 to 300 g) were prepared for experiments by a stereotactic operation performed with intraperitoneal pentobarbital anesthesia (40 mg/kg). A stainless steel screw was placed in contact with the dura overlying the left sensorimotor cortex (LCx). An additional screw was placed in the frontal sinus as a reference electrode. For microinjection, a stainless steel cannula with an inner needle guide was inserted into the left basolateral nucleus of the amygdala (LA). Bipolar depth electrodes were placed in the LA and the left dorsal hippocampus (LH). Seven days after the operation, 1.0 microg of KA was injected into the LA (n = 8; KA group). In the control groups, 1.0 microl of PBS was injected into LA (n = 6). In the KA group, 120 minutes after KA injection, while rats were exhibiting limbic status epilepticus, gabapentin was administered intraperitoneally (i.p.) at a dose of 100 mg/kg. All rats were observed electrophysiologically and behaviorally over 7 days. The spike amplitude were measured in all records and values were expressed as persentage of controls. Results: On the EEG in KA-injected rats, multiple spike discharges initially appeared in the LA 10 to 20 min after KA injection, and rapidly propagated to the LdH. Sixty minutes after KA injection, the seizure involved the LCx and resulted in limbic status epilepticus. Behaviorally, the rats showed bilateral facial twitching, mastication, and salivation. Ten to twenty minutes after gabapentin administration, the seizures in the LCx were selectively suppressed (spike amplitude; 10% of controls). The seizures in LA and LH also reduced (spike amplitude; approximately 30-50% of controls) In addition, behavioral seizure manifestations disappeared. In controls, the severe seizures continued. Conclusions: These results suggested that gabapentin suppresses secondary generalization of limbic seizures by a direct effect on the cerebral cortex. (Supported by Pfizer.) Abstract: .sup.1 Gregory N. Barnes , .sup.1 Xiaoji Zhang , .sup.1 YanFang Li , .sup.2 George M. Smith , and .sup.3 Patricia E. Schauwecker (Neurology, Vanderbilt University School of Medicine, Nashville, TN ; Physiology, University of Kentucky College of Medicine, Lexington, KY ; and Cell and Neurobiology, University of Southern California School of Medicine, Los Angeles, CA) Rationale: Molecular determinants which modify activity synaptogenesis may form the core mechanistic components of cortical reorganization after brain injury. Semaphorin 3F (Sema 3F) signaling has been identified as a putative molecular determinant of activity dependent synaptogenesis in the mouse kainic acid model of epileptogenesis. The Sema 3F receptor is a multi-subunit complex composed of at least two proteins, Plexin A3, and Neuropilin 2 (NPN2). Methods: The activity dependent regulation and structure of the mouse NPN2 promoter is detailed here. Results: We have identified a 1.7 kb putative NPN2 promoter from the FVB/NJ (epileptogenic sensitive) and C57BL/6J (epileptogenic resistant strain) mouse strains. The distinct NPN2 promoters, in addition to common putative transcription factor binding sites, have six different single nucleotide polymorphisms (SNPs) that specify distinct putative transcription factor binding sites. The FVB/NJ NPN2 promoter has distinct activity dependent methylation sites, whereas these are absent in the C57BL/6J promoter. When the FVB/NJ nucleotide is substituted for C57BL/6J nucleotide, at least two sites, SNP site 2, and SNP site 6, reduce the activity of the C57BL/6J NPN2 promoter by 90%. These putative strain specific-enhancer sites may explain why NPN2 transcription is maintained in C57BL/6J hippocampal neurons but not FVB/NJ hippocampal neurons after kainic acid status epilepticus (KA-SE). In addition to regulation of the NPN2 transcription by glutamate receptors, we also report the NPN2 promoter may be regulated by reactive oxygen species. We have also determined the sequence of the human NPN2 promoter contains important SNPs distinguishing temporal lobe epilepsy patients from controls, which might be a biomarker or perhaps a causal factor in epilepsy. Promoter activity is an important determinant of NPN2 mRNA levels and thus sema 3F signaling; selective deletion of Sema 3F in the mature forebrain greatly facilitates epileptogenesis via PTZ kindling. Conclusions: This data provide preliminary support for the hypothesis that distinct alleles of synaptic plasticity genes may confer susceptibility for acquired epilepsy after brain injury. (Supported by NIH/NINDS, KSCHIRT, and PREP grant program.) Abstract: .sup.1 Lyndon F. Barnwell , .sup.2 Yajun Ren , .sup.2 Joaquin N. Lugo Jr. , and 1,2,3 Anne E. Anderson (Deptartment of Neurology, Baylor College of Medicine, Houston, TX ; Pediatrics ; and Division of Neuroscience) Rationale: Kv4.2 subunits are thought to compose hippocampal A-type K.sup.+ channels, which regulate dendritic excitability in hippocampus. ERK activation via PKA and PKC downregulates dendritic A-type K.sup.+ currents and is associated with an increase in post-synaptic excitability. Direct ERK phosphorylation of Kv4.2 channel proteins is a candidate mechanism for this effect. We previously demonstrated a dramatic increase in ERK activation and phosphorylation of Kv4.2 acutely following kainate seizures. In the studies presented here we characterize the cellular localization of ERK activation and Kv4.2 regulation in hippocampus during status epilepticus (SE). Methods: Seizures were induced in adult rats by kainate (KA) (15 mg/kg IP). Seizure activity was monitored and scored based on the Racine Scale. After 1-hr of SE animals were sacrificed for biochemical and molecular studies using whole cell homogenates, cellular membranes and synaptosomal preparations. Results: Studies using hippocampal membrane preparations showed an increase in ERK-phosphorylated Kv4.2 (p < 0.05) while total Kv4.2 remained unchanged acutely following kainate-induced SE. This increase in Kv4.2 phosphorylation with SE was accompanied by increased ERK activation (p < 0.05) without a change in total ERK in membrane preparations. In parallel, using hippocampal synaptosomal preparations, we found increased ERK-phosphorylated Kv4.2 (p < 0.05) and ERK activation (p < 0.05) following SE. Intriguingly, levels of total Kv4.2 were significantly decreased (p < 0.05) in the synaptosomal preparations following acute SE. During this same time period, total ERK remained unchanged in the synaptosomal preparations. Studies are underway to evaluate the time-course of these changes in total Kv4.2 regulation and ERK signaling following convulsant stimulation. Conclusions: Our results reveal an increase in ERK-phosphorylated Kv4.2 channels and a decrease in total Kv4.2 proteins in the synaptosomal cellular compartment. The net effect of these changes is predicted to lead to an overall increase in hippocampal excitability during status epilepticus. (Supported by: Funded by NINDS/NIH, Society for Neuroscience, Pediatric Partnership for Epilepsy Research.) Abstract: .sup.1 Simone K. Benassi , .sup.1 Jerome Engel Jr. , and .sup.1 Anatol Bragin (Neurology, David Geffen School of Medicine at UCLA, Los Angeles, CA; Neurology, Neurobiology, David Geffen School of Medicine at UCLA, Los Angeles, CA; and Neurology, David Geffen School of Medicine at UCLA, Los Angeles, CA) Rationale: Recently we have described the presence of pathological high frequency oscillations in the kainic acid rat model of epilepsy. These oscillations, which may be a marker of progressive epileptogenesis, appear to reflect changes in neuronal network excitability associated with an initial precipitating event that may lead to chronic epilepsy. The goal of these studies was to investigate (1) the existence of similar patterns in mice after pilocarpine (PILO) induced status epilepticus (SE) and (2) the neuronal substrates of such oscillations. Methods: C57B6/J adult mice (15-25g) were subjected to electrophysiological recordings under urethane anesthesia (1.5g/kg) between 1 and 3 weeks after PILO (320 mg/kg, s.c.) induced SE. Simultaneously, extracellular unit activity was recorded by a glass pipette (15-25M[OMEGA]) and local field potentials were recorded by a 20-[mu]m tungsten microelectrodes placed [less than or equal to]500[mu]m from the tip of the glass microelectrode. Juxtacellular labeling was performed according to the protocol described by Pinault (1996). After completion of electrophysiological experiments, brains were sliced and sections processed for peroxidase-based neurobiotin visualization used for morphological reconstructions. Cross-correlation analysis between unit activity and local field potentials and power spectral analysis were performed using Datapac (Mission Viejo, CA) software. Results: Similar to electrographic patterns observed in the kainic acid-treated rats, PILO-treated mice had interictal epileptiform events (IEE) with and without high frequency signals (100-500 Hz) and electrographic (EEG) seizures. Power spectral analysis of interictal local field potentials indicated the existence of activity with peaks at 2.0, 5.0-7.0 and 10-40 Hz. Two interneurons, one pyramidal and two granular cells were identified using juxtacellular labeling. One granular cell had no spontaneous activity, but discharged exclusively during epileptiform activity. During seizures, the discharge rate of the hilar interneuron increased to 0.59/sec from a baseline of 0.26/sec. The frequency of discharges of remaining cells (one CA1 interneuron, one pyramidal and one granular cells) was not affected either by the IEE or EEG seizures. Conclusions: During the early stage of epileptogenesis, the electrographic EEG patterns and spectral frequency characteristics of interictal and ictal discharges are similar between kainic-treated rats and PILO-treated mice. Simultaneous extracellular unit and field potential recording, combined with juxtacellular labeling, may identify neuronal mechanisms of progressive epileptogenesis in a mouse model of chronic epilepsy. (Supported by NIH grant NS 02808.) Abstract: .sup.1 Manjunatha B. Bhat , .sup.2 Stephen L. Yates , and .sup.1 Bridget V. Essley (Center for Anesthesiology Research, Cleveland Clinic, Cleveland, OH ; and Medical Affairs, UCB, Inc., Smyrna, GA) Rationale: Levetiracetam (LEV) is an antiepileptic drug indicated for the adjunctive treatment of partial onset seizures in adults and children age 4 and older, which exhibits good tolerance and no known drug interactions. In addition to its antiepileptic effects, LEV has also been shown to possess therapeutic potential for the treatment of neuropathic pain. While the molecular mechanisms of LEV's action are not fully known, it is known to interact with a specific binding site, synaptic vesicle protein 2A, which is a critical component in the neurotransmitter release process. Intracellular Ca.sup.2+ plays a major role in a variety of neuronal functions including excitability and neurotransmission. LEV has been shown to reduce Ca.sup.2+ release through IP.sub.3 receptors from the intracellular stores of endoplasmic reticulum (ER). In this study, we investigated the effects of LEV on Ca.sup.2+ release through ryanodine receptor (RyR), which is a Ca.sup.2+-regulated Ca.sup.2+ release channel in the ER, in sensory neurons of rat dorsal root ganglion (DRG). Methods: All procedures were approved by the Institutional Animal Care and Use Committee. Freshly dissociated lumbar DRG sensory neurons (L1-L6) from adult Sprague-Dawley rats were used in this study. Changes in intracellular Ca.sup.2+ were measured using the fluorescent Ca.sup.2+-sensitive dye, Fura-2. Caffeine (5 mM) was used to stimulate Ca.sup.2+ release through the RyR. LEV was used at a final concentration of 100 [mu]M. Results: In the presence of 2 mM extracellular Ca.sup.2+, stimulation of sensory neurons with caffeine elicited two types of responses. Small diameter, capsaicin-sensitive neurons showed a biphasic increase in intracellular Ca.sup.2+: an initial rapid rise (peak) due to release of Ca.sup.2+ from the ER, followed by a sustained Ca.sup.2+ rise due to Ca.sup.2+ entry through the capacitative Ca.sup.2+ entry (CCE) pathway. Large diameter neurons showed only a rapid but transient rise in intracellular Ca.sup.2+ due to ER Ca.sup.2+ release. Extracellular application of LEV did not elicit any changes in intracellular Ca.sup.2+ in either type of neurons. However, pretreatment of cells with LEV resulted in inhibition of caffeine-induced Ca.sup.2+ responses in both types of neurons. In small neurons, LEV reduced the peak Ca.sup.2+ response by 27% and the sustained Ca.sup.2+ response by 32%. In large neurons, LEV reduced the caffeine-induced Ca.sup.2+ response by 35%. These effects are similar to the inhibitory effects of the general anesthetic drug propofol, which also exhibits anti-epileptic activity. The mechanism(s) by which LEV exerts its effects on RyR-mediated Ca.sup.2+ signaling is not known, however, these effects are also mimicked by the activation of protein kinase C (PKC). Conclusions: Our results demonstrate that the antiepileptic drug LEV inhibits intracellular Ca.sup.2+ signaling mediated by RyR and CCE pathways in sensory neurons. The LEV-mediated regulation of Ca.sup.2+ signaling is similar to that mediated by PKC. (Supported by American Heart Association (National Center).) Abstract: .sup.1 Robert E. Blair , .sup.2 Katherine W. Falenski , .sup.2 Billy R. Martin , and 1,2 Robert J. DeLorenzo (Neurology, Virginia Commonwealth University School of Medicine, Richmond, VA ; and Pharmacology and Toxicology, Virginia Commonwealth University School of Medicine, Richmond, VA) Rationale: An imbalance between excitatory and inhibitory synaptic transmission is believed to be a primary mechanism underlying seizure discharge. Endocannabinoids in the brain have been shown to operate as retrograde modulators of inhibitory and excitatory neuronal transmission through activation of the presynaptic cannabinoid type-1 receptor (CB1). Studies from our laboratory have shown that exogenous cannabinoids infer CB1-dependent anticonvulsant activity in both in vivo and in vitro models of seizure and epilepsy (Blair et al., 2006, JPET, 317(3),1072-78; Wallace et al., 2003, JPET, 307(1), 129-37), and that the endocannabinoid system acts to tonically regulate epileptic seizure discharge through CB1 activation. Furthermore, we have observed a permanent redistribution of CB1 protein expression throughout the hippocampal formation in the rat pilocarpine-induced status epilepticus (SE) model of acquired epilepsy. This study was carried out to determine if the reorganization of the CB1 in the epileptic hippocampus acts to redirect the role of the endocannabinoid system towards regulating excitatory and inhibitory synaptic transmission. Methods: The rat pilocarpine-induced SE model of acquired epilepsy was utilized for this study. Epileptic and control animals were anesthetized and the brains were perfusion-fixed transcardially. Brain cryosections were made and processed for immunofluorescent co-localization analysis using antibodies specific for the CB1 and markers for excitatory (vesicular glutamate transporter 1: VGLUT1) and inhibitory (vesicular GABA transporter: VGAT) terminals. Results: When compared to control, CB1 immunostaining was increased in epileptic hippocampus in the stratum oriens and radiatum throughout the CA1-CA3 molecular regions that co-localized with VGLUT1-positive glutamatergic terminals. In the CA1-CA2 stratum pyramidale layer highly enriched with VGAT-positive GABAergic terminals, CB1 immunostaining was dramatically decreased in epileptic hippocampus as compared to controls. Conclusions: These findings demonstrate that pilocarpine-induced SE, with subsequent development of epilepsy, causes a redistribution of CB1 expression as indicated by both an increase at glutamatergic and decrease at GABAergic terminals. Such a reorganization of the CB1-dependent endocannabinoid pathway throughout the hippocampus would act to suppress excitatory and increase inhibitory synaptic transmission in the epileptic phenotype and may underlie the anticonvulsant properties of cannabinoids in this model. (Supported by RO1-NS23350, P50-NS25630, and NIDA (DA05274, DA07027,DA11322).) Abstract: .sup.1 Clayton B. Pulse , and .sup.1 Karin Borges (Pharmaceutical Sciences, Texas Tech University Health Science Center, Amarillo, TX) Rationale: The ketogenic diet is a high-fat, low-carbohydrate, low-protein diet that is anticonvulsant in drug-resistant epileptic children. Furthermore, it is anticonvulsive and neuroprotective in experimental brain injury in rodents, although its mechanism of action is still unknown. Our hypothesis is that the ketogenic diet leads to changes of brain metabolism and neurotransmitter levels reflected in the chemical composition of the extracellular fluid in the brain. Methods: Three week-old C3Heb/Fe mice (Jackson Labs) were fed the ketogenic diet (Harlan-Teklad TD.06233) or control diet (TD.06232) with matched amounts of minerals and vitamins ad lib for 2-4 weeks. The latency to flurothyl-induced seizures was measured after 3 weeks of feeding. We are using in vivo microdialysis to measure metabolic intermediates and neurotransmitters in the hippocampus of mice on the different diets. The CMA 600 microanalyser is used to measure glucose, lactate and glutamate. Results: Weight gain of both groups was similar, but mice on the ketogenic diet had significantly higher serum beta-hydroxybutyrate levels than control mice (2.7-fold, n = 10 mice each, p < 0.05). Latency to flurothyl-induced tonic seizures was significantly increased (10%) by the ketogenic diet (n = 10 mice each, p < 0.05). Preliminary microdialysis data suggests that the glucose/lactate ratio in the brain extracellular fluid is altered significantly in ketotic mice (p < 0.05 unpaired two-sided t-test). Mice on the control diet showed a glucose/lactate ratio of 1.5 [+ or -] 0.2 (mean [+ or -] standard deviation; n = 3), whereas ketogenic mice had a ratio of 3.4 [+ or -] 0.7 (n = 3). We are in the process to measure the absolute concentrations of glucose, lactate and glutamate in the brain extracellular fluid by microdialysis using the zero-flow method. Conclusions: In summary, our preliminary data indicate that brain energy metabolism is altered in mice on the ketogenic diet, which may contribute to protecting the brain from seizures and brain damage. (Supported by TTUHSC.) Abstract: .sup.1 Mark R. Bower , .sup.1 Timothy Y. Hui , and 1,2 Paul S. Buckmaster (Comparative Medicine, Stanford University, Stanford, CA ; and Neurology & Neurological Sciences, Stanford University, Stanford, CA) Rationale: Some Mongolian gerbils (Meriones unguiculatus) develop epilepsy 1-2 months after birth, but mechanisms are unknown. Seizures in these gerbils can be elicited by exposure to novel environments, providing control of seizure onset in an epilepsy model arising during natural development, rather than from injury or disease. It has been proposed that excessive inhibition of dentate interneurons could "disinhibit" dentate granule cells, thus predisposing these gerbils to seizures. To test this, we recorded from multiple, single neurons in dentate gyrus before and during seizures evoked by exposure to novel environments. If the hypothesis were correct, then action potential frequency would be expected to decrease in interneurons and increase in granule cells before seizure onset. Methods: Multiple, single neurons were recorded using tetrodes from the granule cell layer of dentate gyrus in 3, awake, freely-moving, adult gerbils. Recordings were obtained for 20 min ("baseline" interictal period), while gerbils rested in their home cage. Seizures (n = 1, 1 and 3, resp.) were initiated by moving gerbils from their home cage to a metal table top, which normally elicited a seizure within 1 min ("preictal" period). Seizures consisted of tonic-clonic convulsions. Seizure onset was identified offline as the earliest of five electrographic measures. Action potentials from individual neurons were identified by grouping ("cluster cutting") similar waveform parameters (e.g., peak amplitude) from data obtained during the baseline period. Neurons were classified as those with low firing rate (250 [mu]sec) that fired in bursts (putative granule cells; n = 9) or those with high firing rate (>5 Hz) and narrow spikewidth ( Results: No significant differences were observed between baseline and preictal action potential frequency in granule cells (2.7 [+ or -] 0.7 Hz baseline vs. 2.7 [+ or -] 1.0 Hz preictal, ANOVA, p = 0.958) or interneurons (10.6 [+ or -] 2.5 Hz vs. 11.9 [+ or -] 2.1 Hz, p = 0.665). Conclusions: The lack of significant changes in action potential frequency in either granule cells or interneurons during the preictal period suggests that seizures in Mongolian gerbils are not initiated by disinhibition in dentate gyrus. These preliminary results do not reject the possibility that disinhibition plays a role in seizure initiation in other brain structures. They do, however, show the utility of recording multiple, single neurons in vivo for testing hypotheses of ictal transition by identifying activity changes in specific neuron classes with respect to spontaneous seizure onset. (Supported by NIH/NINDS and the Eric W. Lothman fellowship from the Epilepsy Foundation.) Abstract: .sup.1 Amy L. Brewster , .sup.1 Roland A. Bender , .sup.2 Mira Kuisle , .sup.4 Didier Pinault , .sup.3 Ryuichi Shigemoto , .sup.2 Luthi Anita , and .sup.1 Tallie Z. Baram (Pediatrics & Anat/Neurobio, University of California, Irvine, CA ; Pharmacology & Neurobiology, Biozentrum, Basel, Switzerland ; Division of Cerebral Structure, National Institute of Physiological Sciences, Okazaki, Japan ; and Physiopathologie Clinique et Experimentaler de la Schizophrenie, Faculte de Medecine, Strasbourg, France) Rationale: (I.sub.h), the current carried by the HCN channels, contributes to neuronal membrane properties and network integration. Thalamic HCN channels participate in thalamocortical oscillations that are involved in sleep and have been implicated in absence epilepsies (Di Pasquale et al., 1997; Strauss et al., 2004; Budde et al., 2005). In both thalamus (absence) and hippocampus (febrile seizure-related limbic epilepsy), the epilepsies associated with HCN channel dysfunction commence during development. This suggests that disrupted developmental regulation of thalamic HCN channel expression may contribute to the process of initiation of these epilepsies. Therefore, we studied the developmental expression patterns of HCN channels in thalamus, and probed the role of these channels in an established model of absence epilepsy, the Genetic Absence Epilepsy Rat from Strasbourg (GAERS). Methods: We used in situ hybridization (quantitative and single-cell resolution) to investigate the thalamic expression of HCN channel isoforms 1, 2 and 4 in Sprague-Dawley rats at postnatal days (P) 2, 7, 11, 18 and 60, as well as in adult GAERS rats and age/strain matched controls. The developmental studies were augmented by immunocytochemistry. Results: (1) HCN1, 2 and 4 isoforms were expressed in distinct patterns in thalamus as early as P2. The HCN1 isoform resided primarily in thalamic nuclei that inter-connect with limbic regions (e.g., anterodorsal, lateral posterior). In contrast, HCN2 channels were robustly expressed in nuclei of the ventral group that relay somatosensory input to cortex (e.g., ventral posteromedial; VPm). HCN2, but not HCN4, was found in reticular nucleus (Rtn). In general, expression patterns in immature thalamus were similar to those in adults. (2) Levels of HCN2 and 4 isoforms did not distinguish GAERS rats from controls. However, HCN1 mRNA levels were strongly increased in the VPm (58%) and Rtn (28%). (3) Consistent with the increased contribution of HCN1 to the molecular make-up of thalamic I.sub.h in the GAERS, the h-current in these rats was much less sensitive to non-saturating levels of cAMP, influencing thalamocortical oscillations (Kuisle et al., 2006). Conclusions: (1) Nucleus-specific expression patterns of thalamic HCN isoforms are established early postnatally, consistent with a key role for these channels in fundamental functions of the thalamocortical network (2) Increased expression of HCN1 channels in GAERS rats is associated with I.sub.h dysfunction that may contribute to their epileptic phenotype. (Supported by NIH NS 47993 (ALB); NS 35439 (TZB).) Abstract: 1,2,3 Peter A. Abdelmalik , .sup.4 Patrick Shannon , .sup.1 Philip Liang , .sup.1 Marina Samoilova , 2,3 W. McIntyre Burnham , and 1,2 Peter L. Carlen (Department of Cellular and Molecular Biology, Toronto Western Research Institute, Toronto, ON, Canada ; University of Toronto Epilepsy Research Program, University of Toronto, Toronto, ON, Canada ; Department of Pharmacology, University of Toronto, Toronto, ON, Canada ; and Department of Pathology, Mount Sinai Hospital, Toronto, ON, Canada) Rationale: Severe hypoglycemia constitutes a medical emergency, manifested by seizures, coma and death. Methods: Using the isolated intact hippocampus of the immature mouse, we characterized hypoglycemic seizures in vitro during transient hypoglycemia, and their effects on synaptic transmission, neuronal morphology and hippocampal glycogen content. Results: Hypoglycemic seizures occurred less frequently with increasing extracellular glucose concentration. The irreversible loss of synaptic transmission, however, was only dependant on the occurrence of hypoglycemic seizure activity, irrespective of glucose concentration. Comparatively, no hypoglycemic seizures were generated in the isolated neocortical block during transient hypoglycemia, and the loss of synaptic transmission was reversible. Similar spontaneous seizure-like activity in the hippocampus was produced by the glycolysis inhibitor iodoacetate, and 500 [mu]M L-aspartate but not by hypoxia. Hypoglycemic seizure activity was abolished by NMDA and non-NMDA antagonists. The clinical anticonvulsant midazolam, but not phenytoin or valproate also abolished hypoglycemic seizures. In both cases, the prevention of hypoglycemic seizures led to the maintenance of synaptic transmission. Non-glycolytic, oxidative substrates attenuated hypoglycemic seizure activity, without supporting synaptic transmission, even in the presence of adenosine antagonist DPCPX. Methods used to abolish or significantly decrease hypoglycemic seizure activity led to the protection against glycogen depletion and a decrease in cell damage. Conclusions: These data suggest that suppressing neuronal hyperexcitability during hypoglycemia decreases subsequent neuronal damage. (Supported by: The authors would like to acknowledge the support of the Juvenile Diabetes Research Foundation and the Canadian Institutes of Health Research.) Abstract: .sup.1 S.S. Cash , .sup.1 N. Meng , .sup.1 C. Melinosky , .sup.2 I. Ulbert , .sup.3 O. Devinsky , .sup.3 W. Doyle , .sup.3 A. Mehta , .sup.3 C. Wang , .sup.4 E. Bromfield , .sup.5 J. Madsen , and .sup.6 E. Halgren (Neuro, Mass. General Hospital, Boston, MA ; Inst. for Psych., Hungarian Acad. of Sci., Budapest, Hungary ; New York University Medical Center, NY, NY ; Brigham & Womens Hosp, Boston, MA ; Childrens Hospital, Boston, MA ; and University of CA, San Diego, CA) Rationale: With limited available in vivo microphysiology, the neuronal mechanisms underlying the initiation and propagation of focal seizures in humans remain unclear. Yet this type of detailed information is important in the design of new therapeutic approaches to epilepsy. To more completely characterize the physiological evolution of seizures and how neuronal populations contribute to the development and spread of synchrony we recorded interictal and ictal activity in epileptic patients using intracortical linear microelectrode arrays, depth and subdural grid macroelectrodes. Methods: Microelectrode arrays measure potential gradient (PG), current source density (CSD) and multi-unit activity (MUA) from a restricted region of cortex [proportional to]300 [mu]m in diameter. These parameters, when combined, present a view of active microcircuitry during a given event. This information is complemented by macroelectrodes which provide contextual information about widespread cortical activity. We analyzed interictal discharges and 13 seizures from 7 patients who were undergoing intracranial investigation for intractable temporal or frontal lobe epilepsy of different etiologies including mesial temporal sclerosis, dysplasias, tumors and structural lesions. Results: Microelectrode recordings of some ictal events showed similarities to the PG, CSD and MUA profiles of interictal events. During most seizures, however, there was a progression in this profile suggesting changes in the cellular substrates underlying ictal spike generation. In particular, large amplitude burst discharges toward the seizure's end involved more robust participation of deeper cortical layers and greater population neuronal firing. Seizures also exhibited differing phase characteristics during their evolution: trains of ictal discharges recorded either from two laminar microelectrode arrays or from different points on the subdural grid showed phase locking during some epochs in the seizure but were phase variable at others. Likewise, interictal events often showed inconsistent onsets and spread patterns. Conclusions: These results lead us to speculate that network synchrony is achieved and then spreads by different modes during focal seizures in humans; epileptiform activity is generated not from a single cellular source but from different micro-circuitry networks and propagation relies on a variety of pathways. Furthermore, it is likely that both local and distributed processes interact to shape the overall spatio-temporal evolution and synchronization features of the seizure. This type of in vivo data may influence future designs of surgical procedures and brain stimulation paradigms. (Supported by NIH, AES, Grass Foundation.) Abstract: .sup.1 James W. Chen (Neurology, University of California, Los Angeles, Los Angeles, CA) Rationale: It has been established that optical intrinsic signal (OIS) imaging correlates well with seizure activity. In this study, we obtained high-spatial-resolution images of normalized oxy-hemoglobin (Hb), deoxy-Hb, total Hb, oxygen saturation and oxygen-consumed maps of acute neocortical seizures in rats by using spectral images recorded simultaneously at multiple wavelengths. Methods: In 17 Spraque-Dawley rats, surgery and OIS imaging were performed under general anesthesia in a thinned skull preparation. OIS images were obtained with two CCD cameras with combination of two optical filters among 852, 780, 610 and 568 nm. Electroencephalography (EEG) was recorded in real time with multiple depth electrodes or Michigan 16-channel electrodes in the somatosensory cortex. OIS spectral images were synchronously recorded with EEG. Acute seizures were induced by penicillin applied topically. Images of oxy-Hb, deoxy-Hb, total Hb, oxygen saturation, oxygen-consumed and scattering maps were obtained by performing the following sequential numerical steps:1) ratio analysis, 2) image transformation and co-registration, 3) pixel-by-pixel calculation of normalized oxy-Hb, deoxy-Hb, total Hb, oxygen saturation and oxygen consumed using the Beer-Lambert law, 4) reconstruct normalized images. Results: High-spatial-resolution normalized images of oxy-Hb, deoxy-Hb, total Hb, oxygen saturation and oxygen-consumed maps were reconstructed to study seizure induction and propagation. It was noted that before or at seizure onset, the concentration of deoxy-Hb transiently increases, in both parenchyma ([proportional to]0.03%) and venous space ([proportional to]0.06%). After that, it was quickly followed by increases of oxy-Hb in the parenchyma ([proportional to]0.03%) and venous space ([proportional to]0.06%). Conclusions: By obtaining OIS spectral images of multiple wavelengths simultaneously, normalized high-spatial-resolution maps of oxy-Hb, deoxy-Hb, total-Hb and oxygen saturation could be reconstructed. Based on these maps, changes of oxy-Hb, deoxy-Hb and total-Hb could be reviewed with information of the anatomical structures. These maps showed initial increase of deoxy-Hb transiently in both parenchyma and venous space before seizure onset. (Supported by NINDS K08 NS 42798.) Abstract: .sup.1 Kon Chu , .sup.1 Keun-Hwa Jung , .sup.1 Kyung-Il Park , .sup.1 Manho Kim , .sup.1 Jae-Kyu Roh , and .sup.1 Sang Kun Lee (Neural Stem Cell Laboratory, Stem Cell Research Center, Clinical Research Institute, Department of Neurology, Seoul National University Hospital, Neuroscience Research Institute of SNUMRC, Seoul National University, Seoul, Republic of Korea) Rationale: The acute brain insult, such as status epilepticus is followed by a latent period characterized by the functional or morphological changes of the hippocampus over several days to weeks, before the onset of spontaneous recurrent seizures. Although the significance of abnormal plasticity, such as cell genesis has been described in the hippocampus after seizures, little data so far ascertained the preventive strategies in epilepsy. We have tried to discover desirable drugs to rectify the abnormal plasticity for clinical use. Erythropoietin (EPO), recently known as cytoprotective and anti-inflammatory agent, and neurogenic modulator, may constitute an effective therapeutic modality in cases of epileptic insult. However, there is no study for the effects of exogenous EPO on the epileptogenic process. Methods: In this study, we examined the effects of EPO in the epileptic rats with lithium/pilocarpine-induced status epilepticus (SE), in terms of neuronal loss, inflammation, and cell proliferation . Results: The neuronal death, inflammation, and cell genesis was substantially linked together in more extensive areas, notably in entorhinal, perirhinal, piriform cortex, amygdala and parts of the thalamus as well as the hippocampus. The EPO receptor (EPOR) mRNA level was increased in the first week after SE, and decreased thereafter. EPOR was increased in the damaged areas such as hilus and CA1 subfield. During the latent period after pilocarpine-induced status epilepticus, EPO prevented neuronal death and microglia activation in these areas, and inhibited the generation of ectopic granule cells in the hilus and new glia in CA1. Conclusions: These findings suggest a potential therapeutic role for EPO in chronic epilepsy. Currently, we are trying to disclose the agents potentially conferring anti-epileptogenesis in a variety of epilepsy models, including febrile convulsion and neocortical epilepsy. (Supported by the Ministry of Health and Welfare (A060452), Republic of Korea. References: 1. Jung KH, Chu K, Kim M, Jeong SW, Song YM, Lee ST, Kim JY, Lee SK, Roh JK. Continuous cytosine-b-D-arabinofuranoside infusion reduces ectopic granule cells in adult rat hippocampus with attenuation of spontaneous recurrent seizures following pilocarpine-induced status epilepticus. Eur J Neurosci. 2004;19:3219-3226. 2. Jung KH, Chu K, Lee ST, Kim J, Sinn DI, Kim JM, Park DK, Lee JJ, Kim SU, Kim M, Lee SK, Roh JK. Cyclooxygenase-2 inhibitor, Celecoxib Inhibits the Altered Hippocampal Neurogenesis with Attenuation of Spontaneous Recurrent Seizures Following Pilocarpine-Induced Status Epilepticus. Neurobiol Dis. 2006 (In press).) Abstract: .sup.1 Ondrej Chudomel , 1,2 Aristea S. Galanopoulou , and 1,2,3 Solomon L. Moshe (Neurology, Albert Einstein College of Medicine, New York, NY ; Neuroscience, Albert Einstein College of Medicine, New York, NY ; and Pediatrics, Albert Einstein College of Medicine, New York, NY) Rationale: The substantia nigra pars reticulata (SNR) plays an important role in seizure control. There has been evidence that impairment of tonic GABA current may contribute to the development of generalized epilepsy in humans. We have found that the delta subunit, an important component of extrasynaptic GABA.sub.A receptors mediating tonic current, is present in the SNR and its expression significantly decreases with age in male and female rats. Thus, the objective of the current study was to determine whether tonic GABA current is present in GABAergic neurons and whether its magnitude changes with age. Furthermore, we studied alterations in phasic GABA inhibition as a function of age and sex. Methods: We studied two groups of Sprague-Dawley male and female rats at postnatal days (P) 10-15 and older than P22. Whole-cell patch clamp recordings were made from electrophysiologically identified nigral GABAergic cells in the presence of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and tetrodotoxin (TTX). Gabazine 500 nM was applied to inhibit GABA.sub.A receptor-mediated postsynaptic currents (mIPSCs). GABA.sub.A tonic current was revealed by bath-applied bicuculline 100 uM and was measured as the change in baseline current before and after bicuculline. IPSCs were analyzed before TTX and gabazine were applied. Results: The tonic GABA current was present in GABAergic neurons in younger rats, however it was significantly lower in older animals. On the contrary, frequency and amplitude of phasic GABA inhibition was higher after P22 compared with younger rats. We did not find significant differences between sexes in either parameter. Conclusions: The tonic GABA current can be revealed in GABAergic SNR neurons. Although there is no apparent difference between males and females, the magnitude of tonic GABA inhibition significantly decreases with age. These findings can be explained by the age-related decrease in delta subunit expression in SNR neurons and the absence of sex differences in its expression. On the other hand, phasic GABA inhibition prevails later in development probably reflecting more dense synaptic connections onto GABAergic neurons or increased expression of synaptic GABA.sub.A receptors. Given the importance of SNR GABAergic neurons in seizure control, these findings could facilitate treatment of epilepsy based on the maturational state of the brain. (Supported by NS 20253, NS045243 from NINDS and an RSRF grant.) Abstract: .sup.1 Robert J. Claycomb , .sup.1 Sandra J. Hewett , and .sup.1 James A. Hewett (Neuroscience, UConn Health Center, Farmington, CT) Rationale: IL-1 is a potent pro-inflammatory cytokine with neuromodulatory functions within the central nervous system. IL-1 signal transduction is mediated by the type I receptor (IL-1RI), which is expressed by neurons in the primary cell layers of the hippocampus. Using mice lacking IL-1R1, the hypothesis that endogenous IL-1 contributes to neuroplasticity associated with PTZ-induced kindling, a model of epileptogenesis, was tested herein. Methods: Homozygous male C57BL/6 mice (8-10 weeks) lacking a functional IL-1RI gene or wild-type controls were bred at the UCHC and acclimatized to handling stress for one week prior to use. A PTZ dose-response relationship was performed to determine the effect of IL-1RI deletion on acute convulsant activity of PTZ. Kindling was induced by a once daily administration of a subconvulsive dose of PTZ. All mice were monitored for 15 min following PTZ administration and scored for seizure severity using an established seizure severity scale (Stages 0-4). The latency (time to onset) of a generalized convulsive seizure (Stage 3-4) was also measured. Mice were considered kindled after exhibiting [greater than or equal to] 4 consecutive Stage 3-4 seizures. Results: PTZ induced kindling in wild-type C57BL/6 mice in a dose-dependent manner. After 18 days (i.e., 18 doses of PTZ), 40 and 57% of mice subjected to daily doses of 20 or 24 mg/kg PTZ, respectively, had become kindled. In stark contrast, IL-1RI null mice, treated in parallel, did not become sensitized to the convulsant properties of PTZ (0% kindled at either 20 or 24 mg/kg PTZ after 18 days). To determine if the inability of IL-1RI null mice to kindle was due to decreased sensitivity to PTZ, the dose-response relationship to acute PTZ-induced seizures was determined. Compared to controls, IL-1RI null mice were less sensitive to acute PTZ-induced seizures, as indicated by a rightward shift in the acute convulsive response between 20-58 mg/kg PTZ. For example, 36 mg/kg PTZ elicited a mean seizure score of 2.9 [+ or -] 0.3 in control mice but only 1.1 [+ or -] 0.1 in the IL-1RI null mice. Further, IL1-R1 null mice showed an increased latency to a seizure elicited by 50 mg/kg PTZ, a dose that resulted in generalized convulsions in both mouse strains. Time to onset was 138 [+ or -] 23 and 336 [+ or -] 60 sec for control and IL-1RI null mice, respectively. To determine if kindling could be achieved in IL1-R1 null mutants, these mice were administered once daily doses of 31 mg/kg PTZ. This dosing paradigm induced kindling in IL-1RI null mice that resembled kindling induced in control mice with 24 mg/kg PTZ. Conclusions: Results from this study indicate that IL-1RI null mice are less prone to PTZ-induced kindling. This is likely due to a reduction in the sensitivity of these mice to PTZ and not to an inability to kindle. These findings suggest that endogenous IL-1 facilitates PTZ-induced kindling by decreasing the PTZ seizure threshold. (Supported by Patterson Trust & Donaghue Medical Research Foundation.) Abstract: .sup.1 John R. Cressman , .sup.1 Jokubas Ziburkus , and 1,2 Steven J. Schiff (krasnow Institute, George Mason University, Fairfax, VA ; and Departments of Neurosurgery, Engineering Science and Mechanics, The Pennsylvania State University, University Park, PA) Rationale: Spontaneous, 4-AP induced seizures, as identified by paroxysmal electrical activity accompanied by a significant negative shift in extracellular voltage, occur after a significant period of high activity in the inhibitory cells. Seizures initiate as the inhibitory cells enter a transient state of depolarization block. The origin of this layer specific blockade of cellular spiking activity is unknown, but it shares physiologal similarity with hypoxic depolarization block. We present measurements with extremely high spatially resolutions utilizing nanoscale optical probes, PEBBLES (Barker et al., (1998) Anal. Chem) as well as measurements from optically active oxygen sensing whole-cell mircoelectrodes. Methods: Two probes were used in these experiments. In the first case we have developed a technique to use a combination of electrical current and pressure to drive small nanoscale oxygen sensitive probes, or PEBBLES, into the extracellular space of the brain slices. In addition we coat the tip of a microelectrode with a mixture of fluorescent dyes which permit the measurement of local tissue oxygen pressure simultaneous with extracellular or intracellular electrical potential. A ratiometric dye cocktail is made by dissolving the following components, PVC - Poly(vinyl chloride), DOS - Bis (2-ethylhexyl) sebacate, PtOEPKt - Pt(II) Octaethylporphine Ketone and BODIPY 577/618 in the solvent THF - tetrahydrofuran. Borosilicate glass micropipettes (3-7 MW) are dipped in the above cocktail, calibrated and inserted into tissue. Light from a metal hallide light passes through a 580 nm excitation filter, and is directed towards the sample using a 590 nm dichroic mirror. The fluorescence light is filtered alternatively at 620 nm and 760 nm to image the BODIPY and PtoEPKt-Pt(II) respectively. Results: Intracellular electrical measurements reveal interleaving activity between excitatory cells in the st. pyramidale and inhibitory cells in the st. oriens. We recently presented simultaneous measurement of extracellular electrical activity and quantitative oxygen measurement on the level of microdomains within neural tissue. We have further refined this technique to measure simultaneous whole-cell electrical recordings with local oxygen measurements. Optical measurements using both small nanoscale PEBBLES as well as optodes display an increase in intensity during the seizure which corresponds to a decrease in the local oxygen pressure. Conclusions: We utilized nanoscale probes, PEBBLES, to measure the oxygen pressure across different strata in the CA1 region of the rat hippocampus during 4-AP induced seizures allowing a detailed investigation of the role that oxygen plays on the level of microdomains of cortical layers or even near single cells in the formation and termination of seizures. (Supported by F32 NS051072-01A1, K02MH01493, R01MH500006.) Abstract: 1,3 S. Crowe , 1,3 K. Gale , and 1,2,3 A. Kondratyev (Pharmacology, Georgetown Univ, Wash, DC ; Pediatrics ; and Interdisc Program in Neurosci) Rationale: Continuous seizures or status epilepticus (SE) endanger vulnerable neurons, with prolonged durations causing cell death. Seizure-evoked oxidative damage, including DNA damage, can contribute to this injury (Crowe et al., AES, 2005), but the threshold duration of seizures to induce DNA damage is not known, nor do we know if DNA damage occurs after seizure durations not causing cell death. Because phosphorylation occurs at sites of DNA double-strand breaks (DSBs), DSBs can be quantified by measuring phosphorylated histone H2AX ([gamma]-H2AX). We examined [gamma]-H2AX in specific brain regions after seizure durations subthreshold for causing neuronal death. Methods: Four seizure conditions were studied: 1) acute repeated minimal electroshock seizures (ECS); 2) 30 min of kainic acid (KA)-evoked brief intermittent seizures (prior to the onset of SE); 3) 5 min of continuous seizure (SE) induced by KA; and 4) 120 min of KA-evoked SE in rats pre-exposed to a neuroprotective regimen of ECS for 7 days. Seizures induced by KA (12.5 mg/kg, ip) were terminated with diazepam (30 mg/kg, ip). ECS (200ms, 22-28mA) induced 5-10 sec limbic seizures. A single (acute) ECS session consisted of 3 ECS at 30 min intervals; chronic ECS consisted of daily sessions for 7d. Rats were perfused after seizure termination; brains were cryosectioned for [gamma]-H2AX immunohistochemistry. Results: Neither acute nor chronic ECS changed [gamma]-H2AX immediately or 24h after the last seizure. KA-evoked intermittent seizures (5-7 brief seizures in the first 30 min) caused significant increases in [gamma]-H2AX in neurons in CA1, CA3, and dentate gyrus (hilus and granule cells), amygdala, and entorhinal cortex. Even greater increases were seen following 5 min of SE, although in most areas (except for dentate granule cells) this was not as great as that obtained after 120 min of SE. With ECS preconditioning, the increase in [gamma]-H2AX induced by prolonged (120 min) SE was reduced by at least 80%; the residual increase in [gamma]-H2AX was less than that seen after 5 min SE in rats without preconditioning. Conclusions: Our observation that [gamma]-H2AX increased after seizures that did not cause significant cell death indicates that DSBs are an early sign of cell endangerment. Moreover, the transient increase in [gamma]-H2AX in death-resistant cells (dentate granule cells) suggests that rapid activation of repair processes may contribute to the resistance of these neurons to seizure-evoked injury. Because a single cluster of 5-7 seizures increased [gamma]-H2AX, whereas acute or chronic ECS did not, it appears that isolated seizures do not induce DSBs and that a minimum seizure density is needed to trigger DNA damage. Our data also suggest that 5 min of SE in seizure-naive animals causes substantial DNA damage in the absence of neuronal loss. Thus, early intervention to terminate SE may be necessary to minimize adverse neurological sequelae that can occur without neurodegeneration. (Supported by NIH F31NS461991, R01NS20576, K01MH02040, K07AG019165.) Abstract: .sup.1 Giulia Curia , .sup.1 Ruba Benini , and 1,2 Massimo Avoli (Neurology & Neurosurgery, and Physiology, Montreal Neurological Institute - McGill University, Montreal, QC, Canada ; and Human Physiology and Pharmacology, "La Sapienza" University, Roma, Italy) Rationale: Temporal lobe epilepsy (TLE) is the most common form of partial complex seizures in humans. For several years neuroscientists have focused on investigating network and intrinsic properties of the hippocampus proper, and its role in TLE. Recently however, they have been giving more importance to parahippocampal structures, including the subiculum. Subiculum is the closest structure to the hippocampus; it receives inputs from the hippocampus proper and projects to the entorhinal cortex. Methods: The brain was quickly removed from adult rats and cut into slices (450 [mu]m thickness) that initially included the hippocampus and adjacent enthorinal cortex. The subiculum was further isolated from these slices by cutting out the CA1, dentate gyrus, pre- and para-subiculum and enthorinal cortex. We performed extra- and intracellular recordings from isolated subiculum minislices obtained from both non-epileptic controls (NEC) and pilocarpine-treated rats, which represents an animal model of TLE. Results: In subiculum minislices, application of 4-aminopyridine (4-AP) induced spontaneous network activity in both NEC (interval of occurrence: 85.71 [+ or -] 10.56 s; duration: 0.78 [+ or -] 0.06 s; n = 15) and pilocarpine-treated (interval of occurrence: 73.35 [+ or -] 8.74 s; duration 0.64 [+ or -] 0.01 s; n = 7) animals. Intracellular recordings revealed that subicular cells at resting membrane potential exhibited three different types of spontaneous network activity. In the first group, field activity corresponded intracellulary to an initial hyperpolarization followed by a long-lasting depolarization (LLD). In the second group, the LLD was preceded by an initial depolarization leading to one or more action potentials. Finally, in the third group, an initial burst discharge was found to precede the LLD. In both NEC and pilocarpine-treated tissue, all three types of activity were contributed by GABA.sub.A receptors since the activity is markedly reduced by picrotoxin. Comparison of NEC and pilocarpine-treated tissue revealed a difference in the expression of these three types of activity. In NEC, 71.43% of the cells, demonstrated the first situation (n = 10/14); this percentage dropped to 57.14% in pilocarpine-treated animals (n = 4/7). The second situation occurred in 21.43% of NEC (n = 3/14) and in 14.28% of pilocarpine (n = 1/7). The third situation was present only in 7.14% of NEC (n = 1/14) but in 28.57% of pilocarpine-treated rodents (n = 2/7). Conclusions: These data demonstrate that 4-AP induces network-driven events in subicular minislices of both NEC and pilocarpine-treated rats. In addition, subicular cells from epileptic tissue are more likely to exhibit network-driven burst discharges as compared to those from NEC. (Supported by CIHR, FXRFC, CURE and FRSQ.) Abstract: .sup.1 Marit Dahl , and .sup.1 Peter Wolf (Danish Epilepsy Centre, Dianalund) Rationale: To discuss the purpose and methodology of comparative daily profiles of Lamotrigine (LTG). Methods: In Denmark, legislature since 2005 enforces, for all drugs including AEDs, delivery by the pharmacy of the cheapest preparation. For AEDs, the bioequivalence range has been set to 90 - 111% of the original. Bioequivalence is tested by comparison of a single dose of two preparations in a group of healthy volunteers. Frequent changes in price of 7 different generic Lamotrigine preparations expose patients to multiple preparation shifts. Some patients report problems following shifts, sometimes of a serious kind. In view of obtaining sufficiently detailed pharmacokinetic data including C.sup.max, C.sup.min, t.sup.max, C.sup.max/min and C.sup.x, we have started in selected cases to compare daily profiles of the preparations in question. There are no standard methods to establish such profiles. Ideally, blood sampling would have to take place at 30 - 60 min intervals after drug intake, followed by increasing intervals until the next dose. But compromises are necessary because blood sampling needs to be integrated into daily hospital routine. Another question is the adequate interval between the two comparative profiles. In addition, possible pharmacokinetic interactions with co-medications need to be considered. Results: We present preliminary findings with sampling intervals of three and four hours, and with intervals between profiles of one to three weeks. The three hour schedule appears to provide reasonably informative data, whereas a four hour schedule may fail to detect differences in t.sup.max and carries a risk of rather imprecise C.sup.max measurements from which other calculations depend. Deviations clearly exceeding the accepted bioequivalence range were observed under realistic clinical conditions, and could confirm and explain reports of both status epilepticus with lower bioavailability, and locomotor ataxia with the consequence of fall, skull fracture and intracranial bleeding with higher bioavailability. The deviations observed in the daily profiles were not necessarily apparent when only morning trough levels were compared. Conclusions: Comparative daily profiles provide differentiated information on the bioequivalence of different LTG preparations. In clinical conditions, deviations outside the accepted bioequivalence range were observed and could explain serious morbidity following preparation shift. For clinical purposes, a 3 hour sampling schedule seems to be adequate. Abstract: .sup.1 Celine M. Dube , .sup.1 Amy L. Brewster , .sup.1 Cristina Richichi , .sup.1 Elyzabeth Custodio , and .sup.1 Tallie Z. Baram (Anatomy & Neurobiology and Pediatrics, University of California at Irvine, Irvine, CA) Rationale: The current conducted by HCN channels (I.sub.h) contributes to neuronal and network properties of the hippocampal formation. HCN channels are expressed in diverse hippocampal interneuronal and principal cell populations. Their function depends on their subcellular distribution and the hyper-or depolarizing nature of synaptic input (Santoro and Baram, 2003). We previously showed that experimental prolonged febrile seizures (FS) in immature rats causes an enduring downregulation of HCN1 channel expression and leads to a hyperexcitable hippocampal network. Frank epilepsy develops in 35% of rats (Brewster et al., 2002, 2005; Dube et al., 2006). The generation of HCN1-deficient mice (HCN1-/-) provides a powerful tool to examine the potential causal role of reduced HCN1 channels in this hippocampal network dysfunction. These studies examined whether immature HCN1-/- mice are more sensitive to seizure provocation and prone to develop spontaneous recurrent seizures as adults. Concurrently, we examined the developmental expression and seizure-evoked regulation of the remaining hippocampal HCN isoforms in these mice. Methods: The threshold temperature required for the onset of a hyperthermic 'febrile' seizure was studied as a measure of hyperexcitability in HCN1-/- mice and wild-type controls. Experimental FS were induced on postnatal day 14 and animals monitored as adults for the emergence of spontaneous seizures using digital video hippocampal EEG recordings. Seizure events were determined according to both EEG (i.e., polyspikes or sharp-waves with amplitude > 200% of background lasting over 6 seconds) and behavioral criteria (i.e, freezing and 'limbic' automatisms). Hippocampal mRNA and protein expression of HCN channel isoforms 1-4 was quantified using in situ hybridization and Western blots. The cellular and subcellular distribution of HCN isoforms was analyzed via immunocytochemistry. Results: Threshold temperatures for experimental FS onset in HCN1-/- mice were significantly higher than those of controls: 41.9 [+ or -] 0.2 [degrees]C (n = 22) vs 40.7 [+ or -] 0.2[degrees]C (n = 24) p < 0.0001. These data indicate that HCN1-/- mice are less sensitive to seizure generation. EEGs were normal in all control mice and preliminary data have not yet revealed spontaneous seizures in HCN1-/- mice. Analyses of mRNA and protein expression are ongoing, but preliminary data support an absence of compensatory changes in HCN2 and HCN4 (Nolan et al., 2004). Conclusions: These experiments suggest that global absence of HCN1 channels from all cell types and cellular compartments yields a 'hypoexcitable' network. In addition, they suggest that HCN1 channels contribute to hippocampal network properties in a complex manner that is dependent on the cell type (e.g., interneurons vs principal cells), subcellular compartment and perhaps developmental expression programs. (Supported by NIH NS 47993 (ALB); NS 35439 (TZB).) Abstract: .sup.1 Sara J. Ernst , .sup.3 Lori Isom , .sup.2 Daniel L. Burgess , and 1,2 Jeffrey L. Noebels (Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX ; Department of Neurology, Baylor College of Medicine, Houston, TX ; and Department of Pharmacology, The University of Michigan, Ann Arbor, MI) Rationale: Loss of function mutations in voltage-gated sodium channels (VGSC) have been linked to epilepsy in humans, but the mechanisms underlying hyperexcitability remain unclear. One possibility for this unexpected seizure phenotype is a compensatory dysregulation of other sodium channel subunits expressed in brain. Methods: To evaluate this hypothesis, we examined two VGSC knockout mouse models with spontaneous seizures: one heterozygous for deletion of the Scn2a gene and the other homozygous for knockout of Scn1b. Quantitative RT-PCR was performed in triplicate on RNA from whole brain and selected brain regions from at least 4 pairs of mutant and wild-type littermate controls to compare transcript levels of all known brain VGSC subunit genes (Scn1a, 2a, 3a, 5a, 8a, and 1b-4b). Results: As expected, in each model we found that the disrupted gene had correspondingly decreased levels of mRNA expression ([proportional to]65% of wild-type in Scn2aKO/+ and 0% in Scn1bKO/KO mice). In addition, in both models, we found that the expression levels of other VGSC genes did not fall outside of the range of normal biological variation (among wildtype mice on the same strain background) with the exception of Scn5a. This TTX-insensitive VGSC was previously shown in our laboratory to be expressed at low levels in restricted limbic structures of the wild type brain. We found a consistent increase in Scn5a in the B1KO/+ and B1KO/KO mutant whole brain samples, as well as a highly variable increase in the Scn2a KO/+ forebrain samples compared to wildtype littermates. Conclusions: This result demonstrates that deficiency of VGSC alpha and beta subunit genes may lead to dysregulation of other pore forming subunit family members in the developing epileptic brain. These preliminary results in conjunction with cellular resolution analyses may help to define the network contributing to the seizure phenotype in these models. (Supported by an Epilepsy Foundation Pre-Doctoral Research Training Fellowship (SE), and the Wilson Medical Research Foundation (LI & JN).) Abstract: .sup.1 Michael J. Esser , .sup.2 Michael A. Galic , .sup.3 James G. Heida , .sup.1 Lorie D. Hamiwka , and .sup.2 Quentin J. Pittman (Pediatric Neurology, Alberta Children's Hospital, Calgary, AB, Canada ; Hotchkiss Brain Institute and Department of Physiology and Biophysics, University of Calgary, Calgary, AB, Canada ; and Department of Neurology, Albert Einstein College of Medicine, Bronx, NY) Rationale: Epilepsy ranks among the top three causes of neurological disability. In the pediatric population, seizure prevalence is estimated at 3-5%, the highest incidence of which is in the first year of life. This is a critical neurodevelopmental epoch and the long-term consequences of these neonatal seizures are not fully known. Thus we sought to determine whether maternal infection alone, or in combination with a neonatal febrile convulsion, affected subsequent seizure susceptibility as well as memory and learning in adults. Methods: Timed-pregnant Sprague-Dawley rats were administered either lipopolysaccharide (LPS; 100 [mu]g/kg, i.p.) or saline (sal; control) on gestational day 15 (G15), to mimic a gram-negative bacterial infection. On postnatal day 14 (P14), febrile convulsions (FC) were induced with LPS (200 [mu]g/kg) and kainic acid (KA; 1.75 mg/kg), which is known to evoke seizures in about 50% of rats at this age. Two months later, all groups were subjected to learning and memory tasks both before and after an adult seizure. One week prior to evocation of an adult seizure with KA (15 mg/kg) rats were subjected to a contextual fear conditioning paradigm. One week after the adult seizure, rats were then tested for learning and memory in a water maze paradigm. Adult seizures were attenuated within 30 minutes of seizure onset with parenteral diazepam (5 mg/kg) to mimic clinical practice. Results: Prenatal LPS exposure resulted in a trend towards increased neonatal seizure susceptibility at P14 compared to saline-treated controls. Further, a subset of animals with a history of prenatal LPS and FCs showed deficits in learned fear to contextual stimuli compared to cohorts of animals with no prenatal LPS and FC. Rats with a history of prenatal inflammation and postnatal FCs showed enhanced susceptibility to seizures as adults, compared to the control condition (no maternal treatment but suffered FCs). Results of the spatial learning water maze task revealed no statistically significant differences between any groups of rats regardless of prenatal treatment or FC outcome, when the KA provoked seizure were attenuated at 30 minutes. Conclusions: In this study, our findings suggest that prenatal LPS in combination with a neonatal FC alters adult seizure threshold as well as impairs memory performance to contextual fear stimuli. The lack of effect of adult seizure (in any group) on water maze testing, when seizures were attenuated, further emphasizes the importance of early seizure intervention. (Supported by Canadian Institutes of Health Research (CIHR), the Natural Sciences and Engineering Research Council of Canada (NSERC), and the Alberta Heritage Foundation for Medical Research (AHFMR).) Abstract: 1,2 Carl L. Faingold , and .sup.1 Srinivasan Tupal (Pharmacology, Southern Illinois University Sch. Med., Springfield, IL ; and Neurology, Southern Illinois University Sch. Med., Springfield, IL) Rationale: In response to acoustic stimulation genetically epilepsy-prone rats (GEPR-9s) exhibit tonic extension audiogenic seizures (AGS). However, with frequent repetitive AGS (audiogenic kindling) seizures increase in duration and an additional behavior, post-tonic clonus (PTC) is induced, which increases in incidence and duration as kindling proceeds. This AGS kindling-induced convulsive behavior change is long-lasting and is considered to be a model of epileptogenesis. Previous studies indicate that plastic changes in the synaptic projection to the lateral nucleus of amygdala (LAMG) are critical in the processes involved in production of PTC. Activation of NMDA receptors in LAMG can transiently cause the emergence of PTC in non-kindled GEPR-9s (NK-GEPR-9s) and conversely, inhibition of NMDA receptors by a NMDA receptor antagonist reversibly blocks PTC in kindled GEPR-9s (K-GEPR-9s). Previous studies of AGS kindling also observed increased expression of an immediate early gene, as indicated by c-fos changes in LAMG. A number of steps in the molecular cascade occur between NMDA receptor activation and c-fos expression, and cAMP is implicated in this cascade. The present study examined whether PTC incidence was altered by focal microinjection into LAMG of agents that alter adenylyl cyclase (AC) activity, which is the enzyme responsible for cAMP production. Methods: GEPR-9s (200-400 g) used in the present study all exhibited tonic hindlimb extension. Audiogenic kindling involved once daily presentation of an electric bell (122 dB SPL re: 0.0002 dyne/cm) for a maximum of 60 sec given on 14 consecutive days. Microinjection cannula guides (21G) were implanted stereotaxically above LAMG bilaterally under ketamine/xylazine (100/3 mg/kg) anesthesia, and all rats continued to exhibit tonic AGS one week after implantation. An AC activator forskolin (50-200 [mu]Mol/side) or an AC inhibitor (SQ22536, 0.5-1.0 nMol/side), was focally microinjected (at 0.5 [mu]l/min for 2 min) into LAMG bilaterally via an infusion cannulae (26G). The animals were then tested for AGS at 1, 12, 24 and subsequently every 24 hrs and then once a week. Results: Bilateral microinjection of forskolin caused the appearance of PTC in NK-GEPR-9s in response to presentation of the acoustic stimulus with an onset of 12 hr. However, AGS with PTC continued to be observed thereafter for a minimum of 5 weeks after the single microinjection. The AC inhibitor, SQ22536, transiently blocked PTC in K-GEPR-9s with an onset of 1 hr post microinjection. A complete blockade of seizures was observed at 12 hrs subsequent to which the animals gradually recovered to display a tonic hindlimb extension seizure followed by PTC at 120 hr post microinjection. Conclusions: These data indicate a potentially important role of AC activation in the LAMG in the molecular cascade subserving AGS kindling in GEPR-9s, a model of epileptogenesis. (Supported by SIUSM EAM grant.) Abstract: .sup.1 Leonardo C. Faria , and .sup.1 Lisa R. Merlin (Neurology and Physiology and Pharmacology, SUNY Downstate Medical Center, Brooklyn, NY) Rationale: Activation of metabotropic receptors in the CNS elicits long-lasting effects that may participate in epileptogenesis. Many such receptors are linked to phosphoinositide or phosphatidylcholine metabolism, and the resultant DAG production activates protein kinase C (PKC). We used a phorbol ester to activate PKC to determine its effect on picrotoxin-induced network activities in vitro. Methods: 400 [mu]m transverse guinea pig hippocampal slices were placed in an interface chamber and intracellular recordings were obtained from CA3 stratum pyramidale. Picrotoxin (50 [mu]M), a GABA.sub.A antagonist, was bath-applied to elicit baseline interictal activity (synchronized discharges < 500 ms long), and changes in burst length were monitored following the application of phorbol-12,13-dibutyrate (PDBu). Bursts were considered ictaform if they met the following three criteria: (1) unaffected by hyperpolarization, (2) the oscillations within each burst slowed in frequency toward the end of the burst, and (3) length of the burst exceeded 1 s, the presumed minimum length for a clinically-correlated experience to potentially occur if the discharge were expressed in vivo. Data are reported as means [+ or -] SE; significance was determined using the Student's t-test, with p < 0.05 deemed significant. Results: As phorbol esters activate second messengers, preincubation for at least 30 min is recommended. Yet within 40-60 min, bath application of 1 [mu]M PDBu typically increased the length of picrotoxin-induced interictal bursts to ictaform events 2-8 s long (mean peak burst length 5.6 [+ or -] 0.5 s; latency to produce 1 s ictal-length burst 48 [+ or -] 4 min; n = 7 out of 8 slices); this effect persisted for over 1 hr following washout of PDBu. When PDBu was applied in the presence of 10 [mu]M chelerythrine, a PKC inhibitor, the burst prolongation was either completely prevented (n = 4) or significantly suppressed (peak burst length 2.8 [+ or -] 0.7 s; time to ictal-length 68 [+ or -] 13 min; n = 6; p < 0.05). Bursts could be blocked by application of either 1 [mu]M tetrodotoxin (n = 3) or a combination of NMDA and AMPA receptor antagonists (50 [mu]M D-AP5 and 20 [mu]M CNQX, respectively; n = 2), suggesting the discharges were synchronized events; no intrinsic oscillatory activity was uncovered during these manipulations. NMDA antagonist alone could not prevent the induction of these bursts, but did reversibly shorten the length of expressed bursts, as did 25 [mu]M MPEP, an mGluR5 antagonist. While PDBu has been reported previously to enhance glutamate release, bath application of the glutamate reuptake inhibitor L-trans-2,4-PDC (100-200 [mu]M) was unable to simulate the response seen with PDBu (n = 2). Application of PDBu in the absence of picrotoxin also elicited burst prolongation, but a higher concentration was necessary to result in a similar peak burst length and time to ictal response (2-10 [mu]M; n = 4). Conclusions: Generalized PKC activation in the hippocampal slice has powerful excitatory effects on network activity with implications toward epileptogenesis. (Supported by NIH NS40387 (L.R.M.).) Abstract: .sup.1 Luisa P. Flores , .sup.2 Sebastian Ivens , 2,3 Alon Friedman , and 1,4 Daniela Kaufer (Helen Wills Neuroscience Institute, UC Berkeley, Berkeley, CA ; Institute of Neurophysiology, Charite University Medicine, Berlin, Germany ; Laboratory for Experimental Neurosurgery, Soroka University Medical Center and Zlotowski Center for Neuroscience, Ben-Gurion University of the Negev, Beer-Sheva, Israel ; and Department of Integrative Biology, UC Berkeley, Berkeley, CA) Rationale: Different types of cortical injury such as trauma, ischemia and infection are often accompanied by increased permeability of the blood-brain barrier (BBB). We have recently shown that perturbations in the blood-brain barrier (BBB) may lead to epileptiform activity in the neocortex. We also demonstrated that serum albumin acts as the epileptogenic factor. Our recent data suggest that the action of serum albumin is mediated via its interactions with transforming growth factor-beta (TGF-[beta]) receptors in astrocytes. Following albumin uptake into astrocytes we show that the glial inward rectifying potassium channel, Kir 4.1, was down regulated which results in reduced buffering of extracellular potassium. This, in turn, leads to activity-dependent accumulation of extracellular potassium, resulting in N-methyl-D-asparate (NMDA)-receptor-mediated neuronal hyperexcitability and epileptiform activity. Methods: In order to elucidate the cascade of transcriptional effects following albumin uptake and the consequent TGF-[beta] receptor activation we evaluated expression levels at different time points following induced permeation of BBB, exposure to albumin, or exposure to TGF-[beta]. Genome scale evaluations using DNA microarrays were carried out in conjunction with gene-specific quantitative RT-PCR of pre-selected genes. Results: All three treatments resulted in down regulation of both Kir 4.1 and the glial protein, glutamine synthetase as well as up regulation of the glial fibrillary acidic protein albeit with different time courses. All genes were normalized to [beta]-Actin which showed no change in expression across all treatments and time points. Analysis of the genome scale expression levels present transcriptional signatures common to all three treatments, revealing the molecular pathway underlying the process of epileptogenesis. Conclusions: Here, we highlight a novel cascade of events following common brain insults leading to brain dysfunction and eventually epilepsy, suggesting possible therapeutic targets. (Supported by the Mary Elizabeth Rennie Epilepsy Foundation.) Abstract: .sup.1 Corey Flynn , and .sup.1 G. Campbell Teskey (Department of Psychology, University of Calgary, Calgary, AB, Canada) Rationale: Epilepsy is characterized as a chronic brain state with a very low seizure threshold. In people that develop epilepsy, there must be changes in brain functioning that results in a lowered seizure threshold prior to the appearance of the first seizure. Currently there is no animal model of induced epilepsy that allows for the exploration of the neural basis of a low seizure threshold without the elicitation of seizures. For instance stimulation that results in kindling both lowers thresholds and elicits seizures. Likewise the models that develop spontaneous seizures result from a treatment that elicits seizures. In this study we employed repeated application of subthreshold electrical stimulation in an attempt to lower seizure thresholds without eliciting seizures. Methods: Expt 1: Rats were implanted with bipolar stimulating and recording electrodes in the corpus callosum and sensorimotor cortex respectively. Once daily stimulation consisted of biphasic rectangular waves (1 ms, 60 Hz, for 1 sec). A range of stimulation intensities (20, 30, 40, 50, and 500 uA) was used. Prior to, and following 20 stimulation sessions at a single intensity, the afterdischarge (seizure) threshold was determined. Expt 2: The goal of this experiment was to reduce seizure thresholds in the absence of ANY seizures. Because the measurement of afterdischarge threshold (ADT) necessitates one AD, the initial ADT measurement was excluded for Expt 2. The intensity groups chosen for Expt 2 were 20, 50, and 500 uA as well as two additional groups (150, 250 uA). Results: Repeated stimulation at 50 uA was most effective at reducing the ADT in sensorimotor neocortex. However, even at this intensity, it was difficult to achieve a complete absence of seizure activity. Unexpectedly the highest intensity stimulation groups (250, & 500 uA) showed repeated seizure activity but did not show a reduction of ADT. Conclusions: Low intensity electrical stimulation allowed for the reduction of seizure thresholds with few seizures, while high intensity electrical stimulation reliably elicits seizures without lowering seizure thresholds. These two stimulation protocols can be used to dissociate the neural basis of a low seizure threshold from the reorganizing effect of repeated seizures. (Supported by NSERC.) Abstract: .sup.1 Chris R. French (Neurology, Monash Medical Centre, Clayton, Victoria, Australia) Rationale: The rapidly activating voltage activated sodium current (Ina.sub.(t)) underlies the rapid depolarizing phase of the action potential in most neurons of the central and peripheral nervous system. This current has a rapid (sub-millisecond) activation phase, followed by by a slower (>10 ms) falling phase, termed inactivation. The aim of this study was to better characterise the inactivation process in a mammalian central neuron, and to observe the effects of chemical removal of inactivation on this current. Methods: Hippocampal CA1 neurons (n = 100) were acutely dissociated from female Wistar rats, 2-5 weeks of age, using a pronase-based protocol. Ina.sub.(t) was recorded under voltage clamp conditions with the whole-cell patch clamp technique using specially fabricated low-resistance patch electrodes at 20[degrees]C. With this method, the rising phase of the sodium current could be resolved. Results: Macroscopic inactivation of Ina.sub.(t) was evident at membrane potentials around -40 mV, had a bi-exponential decay (Fig. 1A), becoming more rapid with greater depolarization (fast and slow time constants of 0.5 and 2.5 ms respectively at 0 mV). The amplitude ratio of the fast and slow components approached a limiting value of 0.7, and the bi-exponential inactivation was remarkably well accounted for by the addition of a second open state in the kinetic description. The steady-state inactivation (h.sub.[alpha]) relation reached a maximum at -120 mV, with a half-inactivation voltage of -71 mV, complete inactivation at -40 mV and was best fitted by a double Boltzman distribution. Data from double pulse techniques suggested that the time course of macroscopic inactivation reflects the true interconversion rates between open and inactivated states. Similar experiments at subthreshold potentials revealed that transitions from activatable closed states to inactivated states were surprisingly long ([proportional to]100's of ms) which is likely to affect repetitive firing rates. Addition of the enzyme papaine (1 mg/ml) to the patch-electrode solution resulted a rapid increase in Ina.sub.(t) amplitude ([proportional to]100%) and gradual removal of inactivation, and a negative shift in activation (Figure 1 B,C). Time constants of activation and deactivation were largely unaffected by inactivation removal. Conclusions: These observations reveal major discrepancies between properties of rat CA1 INa.sub.(t) and conventional invertebrate models, and suggest that substantial alterations to the Hodgkin-Huxley model must be made to adequately model mammalian CNS excitability.[figure 1] Abstract: 1,2 Linda K. Friedman , and .sup.2 Robert Keesey (Neuroscience, New York College of Osteopathic Medicine, Old Westbury, NY ; and Neuroscience, Seton Hall University, South Orange, NJ) Rationale: Sustained increases of intracellular calcium ([Ca.sup.2+]i) following prolonged seizures or other neurological insults are thought to be responsible for neuronal cell death. However, elevated calcium in central neurons does not always lead to neurotoxicity. In fact, under certain circumstances calcium can even protect neurons from dying. Methods: The FURA-2AM dye indicator was used to measure intracellular Ca .sup.2+ after single (1xKA) or multiple (3xKA) episodes of KA seizures that were induced on postnatal (P) days P6, P9 and P13. Slices from control animals were exposed to middle doses of glutamate (30 [mu]M) and NMDA (30 [mu]M). Results: Basal Ca .sup.2+ measurements were very consistent and relatively low in control slices prepared from P6, P9, and P13 rat pups. At 5 hrs after a singe episode of status epilepticus at P6, P9 or P13, basal Ca.sup.2+ levels in the presence of TTX were elevated in an age-dependent fashion such that CA1/subiculum and DG had the largest concentrations of Ca.sup.2+, consistent with early studies (Hammon and Heinemann, 1988). After ratiometric calculations it was found that CA2 and CA3 neurons also increased their Ca.sup.2+ concentrations but to a lesser degree. Responses to NMDA were also significantly potentiated in CA1/subiculum and DG at the three perinatal ages examined. Calcium levels returned to baseline in the presence of the selective antagonist NMDA receptor APV. In contrast, after 3xKA, much smaller increases in both Ca.sup.2+ baseline and NMDA stimulated responses were observed 5 hrs after the third seizure relative to animals after 1xKA. This decrease in Ca2+ permeability coincides with a reduction in NMDA receptor protein expression in CA1/subiculum rather than the loss of GluR2 subunits that has been observed in adults. A full time course of Ca2+ baseline and agonist and antagonist responses at these early ages will be presented. Quantification of baseline and pharmacological blockade of NMDA receptors will be illustrated by pseudocolor images. Conclusions: Data show that a single episode of KA-induced status epilepticus increases hippocampal sensitivity to NMDA after seizures have subsided. However, multiple perinatal seizures reduce glutamatergic stimulated Ca.sup.2+ influx to induce long-lasting tolerance and neuroprotection. (Supported by NJ Neuroscience Institute and New York College of Osteopathic Medicine.) Abstract: 1,2 Aristea S. Galanopoulou , .sup.1 Ondrej Chudomel , .sup.1 James Heida , and 1,2,3 Solomon L. Moshe (Neurology, Albert Einstein College of Medicine, Bronx, NY ; Neuroscience, Albert Einstein College of Medicine, Bronx, NY ; and Pediatrics, Albert Einstein College of Medicine, Bronx, NY) Rationale: Neonatal prolonged seizures may cause long term functional changes which in certain cases may increase the risk for subsequent epilepsy. Using a model of 3 episodes of neonatal kainic acid induced status epilepticus (KA-SE), we have shown that the normal GABA.sub.A receptor mediated differentiation of the substantia nigra pars reticulata (SNR) and its role in seizure control are disrupted. Here we investigate whether the 3 episodes of neonatal KA-SE also lead to long term changes in GABA.sub.A receptor subunit expression, which could alter its sensitivity to GABAAergic drugs. Methods: Male Sprague-Dawley rats were subjected to 3 daily episodes of KA-SE at postnatal days (PN) 4, 5, and 6 by intraperitoneal injections of KA (PN4: 1.25 mg/kg; PN5: 1.5 mg/kg; PN6: 2 mg/kg) and were monitored behaviorally for 6 hours daily (n = 8 rats, KA456 group). Micropunches from the anterior SNR were obtained at PN30. RT-PCR using primers specific for the a1 - a6, b2, g1, g2L, delta GABA.sub.A receptor subunits, the rho1 GABAC receptor subunit, and beta-actin were performed. RT-PCR products were electrophoretically separated and compared by densitometry with the subunit expression of control PN30 male anterior SNR (n = 7 rats). Results were referred to beta-actin. Results: Control male PN30 anterior SNR expressed the following subunits: a1 > b2, a5, g2L > g1, a2 > a3, a4, delta, rho1. The a6 subunit was not expressed, whereas low expression of rho1 was detected in 2/7 control rats. In the PN30 KA456 male rats, the following subunits were reduced compared to controls: (1) a1 subunit (P < 0.05, unpaired t-test); (2) the a3 subunit was not present in 5/3 KA456 rats, whereas it was present in all studied controls (chi square P = 0.01); (3) the g1 subunit also tended to be reduced in KA456 males (unpaired t-test P = 0.055). No significant changes in the other subunits were seen. Conclusions: The anterior SNR of PN30 control male rats had a relative dominance of benzodiazepine-sensitive subtypes, over the extrasynaptic and steroid sensitive delta-containing subtypes. The PN30 KA456 male anterior SNR had reduced expression of the a1 and a3, and showed a trend for reduced expression of the g1 subunit. These suggest that the sensitivity to certain benzodiazepine GABA.sub.A receptor agonists - ie those selective for a1, a3, and possibly g1 - may be reduced in KA456 rat anterior SNR. Since a1-selective GABAAergic drugs can mediate anticonvulsant effects through their actions on the PN30 anterior SNR, further studies are needed to test whether the reduction of a1 subunits in KA456 rats may be a mechanism of drug refractoriness to the anticonvulsant effects of certain GABAAergic drugs, especially those selectively acting on a1-containing receptors. (Supported by NIH NINDS grants NS045243 and NS20253 and a Rett Syndrome Research Foundation grant.) Abstract: .sup.1 Maciej Gasior , .sup.1 Jessica Yankura , .sup.1 Amy French , .sup.1 Wayne Yonekawa , .sup.1 Adam Hartman , and .sup.1 Michael A. Rogawski (Epilepsy Research Section, NINDS-NIH, Bethesda, MD) Rationale: (R,S)-1,3-Butanediol (1,3-BD) is a common food additive that has a long history of safety in various species. 1,3-BD is metabolized by alcohol and aldehyde dehydrogenases in the liver to form [beta]-hydroxybutyrate, acetoacetate, and acetone, which are the same ketone bodies whose levels are elevated in the ketogenic diet. Here we sought to determine if 1,3-BD has anticonvulsant effects and could potentially be a replacement for the ketogenic diet. Methods: The anticonvulsant activity of 1,3-BD was assessed in several in vitro and in vivo models. Acute toxicity was determined in the inverted-screen test. To determine the contribution of ketosis to the anticonvulsant efficacy of 1,3-BD, animals were pretreated with the alcohol dehydrogenase inhibitor 4-methylpyrazole. In addition, we used the R- and S-enantiomers of 1,3-BD since only the R-enantiomer can be directly converted to acetone. Results: 1,3-BD attenuated epileptiform activity induced by 4-aminopyridine (55 [mu]M) in rat hippocampal slices in a concentration-dependent manner (10-100 mM). Furthermore, 1,3-BD (1-32 mmol/kg, IP) demonstrated a broad anticonvulsant profile in a variety of seizure tests in mice (iv PTZ test, electroconvulsive threshold, 6-Hz seizure model) and rats (amygdala-kindled seizures). The anticonvulsant effects of 1,3-BD occurred at doses devoid of acute toxicity. After the administration of an anticonvulsant dose of 1,3-BD, levels of the ketone body, [beta]-hydroxybutyrate, were comparable to those seen in mice exposed to the ketogenic diet for several weeks. Inhibition of 1,3-BD conversion to ketone bodies by 4-methylpyrazole (25 mg/kg) augmented the anticonvulsant effects of 1,3-BD, indicating that 1,3-BD itself has anticonvulsant activity. Like the racemic form of 1,3-BD, the R- and S-enantiomers individually demonstrated robust anticonvulsant properties, demonstrating that conversion to ketone bodies is not required for seizure protection. However, the R-enantiomer was modestly more potent suggesting that ketone bodies could contribute to the anticonvulsant activity. Conclusions: 1,3-BD and its enantiomers are protective in a broad range of seizure models. Conversion to ketone bodies is not required for efficacy, but could augment the activity of the parent. Further studies are required to assess the potential of 1,3-BD or its enantiomers in the treatment of epilepsy. (Supported by NINDS/NIH.) Abstract: .sup.1 Krista L. Gilby , .sup.2 Thomas N. Ferraro , and .sup.1 Dan C. McIntyre (Institute of Neuroscience, Carleton University, Ottawa, ON, Canada ; and Department of Psychiatry, University of Pennsylvania, Philadelphia, PA) Rationale: Epilepsy and attention deficit hyperactivity disorder (ADHD) have recently been established as true co-morbid disorders (Kanner et al., 2003). Fast (seizure-prone) and Slow (seizure-resistant) kindling inbred rat strains, derived via selective breeding, serve as an excellent model for seizure susceptibility, yet Fast rats also exhibit adult behavioral profiles indicative of relative ADHD-like behaviors. In order to begin to explore the genetic basis for relative seizure susceptibility and ADHD-like behaviors in Fast rats, we interbred the two strains and examined adult offspring (F1 hybrid) behaviors. Methods: Male Fast rats were bred with female Slow rats to create an F1 population of hybrid pups. Offspring generated by breeding Fast males with Fast females or Slow males with Slow females served as control populations. As adults, all rats were compared in several behavioral paradigms known to elicit differential responses from Fast and Slow rats including kindling, restraint and open field. Water consumption was also compared. Finally, blood was extracted from all rats and levels of non-esterfied fatty acids (NEFA), known to be deficient in Fast rats, were compared. Results: F1 hybrids showed water consumption and kindling rates that were indistinguishable from Fast rats. Remarkably, however, hybrids exhibited behaviors that were near identical to Slow rats in both the open field and restraint paradigm. Thus, hybrids appear to have maintained the relative seizure susceptibility, but not the ADHD-like behaviors, of their Fast fathers. Further, blood analysis showed that F1 hybrids exhibited the relatively higher NEFA levels characteristic of Slow rats. Conclusions: Results suggest that a primary dominant or semi-dominant genetic effect mediates the differences in kindling rate and water intake between Fast and Slow rats. Lack of a similar finding for other behavioral traits may be due to a maternal genetic influence or the influence of unknown environmental factors. Interestingly, NEFA was higher in hybrids and Slow rats compared to Fast rats, thus implying a link between ADHD-like behaviors and a relative NEFA deficiency. (Supported by CIHR and NSERC to DCM.) Abstract: .sup.1 Joseph Glykys , and .sup.1 Istvan Mody (Neuroscience IDP and Deptartment Neurology, UCLA, Los Angeles, CA) Rationale: Phasic and tonic inhibitions are mediated by synaptic and extrasynaptic GABA.sub.A receptors, respectively. Extrasynaptic receptors respond to ambient levels of GABA thought to originate in part from spillover after vesicular release. If both phasic and tonic inhibitions are activated by vesicular GABA release, in any given cell there should be a correlation between the magnitudes of the two types of inhibition. Methods: Whole-cell voltage-clamp recordings were obtained from hippocampal CA1 pyramidal cells and from molecular layer interneurons. In order to measure tonic and phasic inhibitions simultaneously over short periods of time (1 s), we developed a new method of measurement of their mean values (Imean). In addition, simulations of sIPSCs were used to model specific changes in phasic currents. Results: Simulations of sIPSC showed that this new method is highly sensitive and can detect the combined effects of small changes ([proportional to]15%) in two sIPSC properties (e.g., amplitude and decay time constant) even when the change in each individual parameter is not statistically significant. Using this method we investigated changes in phasic and tonic inhibitions during spontaneous and sucrose induced bursts of sIPSCs recorded in hippocampal interneurons and pyramidal cells. During the bursts there was a large increase ([proportional to]52 Hz) of sIPSC frequency with a concomitant change in baseline holding current. However, simulations of sIPSC bursts even at much higher frequencies than those actually recorded showed only a minimal contribution of phasic inhibition to the baseline current, thus indicating that tonic inhibition was also activated during the bursts. In addition, we found a strong cross-correlation between phasic and tonic inhibitions when vesicular GABA release was reduced by perfusing TTX + CdCl.sub.2. Conclusions: These results are consistent with a significant portion of tonic inhibition being mediated by GABA released from the very vesicles responsible for activating phasic inhibition. Therefore, phasic and tonic inhibitions can be modulated in unison resulting in a compound dampening effect on neuronal excitability. (Supported by the Gonda Fellowship to J.G. and by NS02808, NS30549 and the Coelho Endowment to I.M.) Abstract: .sup.1 Candan Gurses , .sup.1 Mehmet Kaya , .sup.1 Rivaze Kalayci , .sup.1 Nadir Arican , .sup.1 Bulent Ahishali , .sup.1 Oguzhan Ekizoglu , .sup.1 Imdat Elmas , .sup.1 Mutlu Kucuk , .sup.1 Duran Ustek , .sup.1 Bilge Bilgic , and .sup.1 Gonul Kemikler (Neurology, Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey; Physiology, Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey; Research Institute for Experimental Medicine, Istanbul University, Istanbul, Turkey; Forensic Medicine, Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey; Histology, Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey; Forensic Medicine, Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey; Pathology, Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey; and Radiotherapy, Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey) Rationale: The aim of the present study is to investigate the influence of hyperthermia on the integrity of blood-brain barrier (BBB) in rats with cortical dysplasia. Methods: In this study, 32 one month old baby Sprague-Dawley rats were used for sodium fluorescein (NaFlu) staining method and 72 more animals were used for other parameters in parallel groups. Study groups consist of a control (8), cortical dysplasia (8), hypertermia (8) and cortical dysplasia and hyperthermia (8) groups. The pregnant rats which have been given sodium pentobarbital anesthesia with the dose of 30 mg/kg intraperitonally, were administered gamma irradiation of 145 cGy from the uterus on day E 17. To induce hyperthermia (39.5 [+ or -] 0.5[degrees]C), rats were exposed to an elevated ambient temperature (55-60[degrees]C) for 30 minutes. Vital signs such as body temperature and arterial blood pressure were also recorded. Immunohistochemistry and Western blot methods were used to show any alterations in occludin protein, and electron microscopy was performed to evaluate structural properties of BBB in this experiment. On the day of the experiment there was no mortality in hyperthermic animals. Results: Blood pressure levels of rats with hyperthermia and rats with hyperthermia as well as cortical dysplasia increased significantly when compared to control values. Seizures the rats had were observed clinically; the longer the duration of hyperthermia was, the more seizures were seen. Both hyperthermia and cortical dysplasia did not change BBB permeability to NaFlu. Hyperthermia induced a generalized increase in BBB permeability to NaFlu in rats with cortical dysplasia and the highest increases (>130% over control values) were recorded in diencephalon and cerebellum regions. Western blot analysis of brain capillaries showed that the expression of the transmembrane tight junction protein occludin was not changed in response to dysplasia and hyperthermia. Conclusions: These results indicate that immature rats are resistant to hypertermic BBB disruption. However, hyperthermia may induce significant BBB permeability increase under cortical dysplastic conditions. We conclude that cortical dysplasia as an underlying pathology could increase the disruption of BBB integrity during hyperthermia. Abstract: .sup.1 Bjornar Hassel , .sup.2 Ray Dingledine , .sup.3 Leif Gjerstad , and .sup.3 Erik Tauboll (Division for Environmental Toxicology, Norwegian Defence Research Establishment, Kjeller, Norway ; Department of Pharmacology, Emory University School of Medicine, Atlanta, GA ; and Department of Neurology, Division of Clinical Neuroscience, Rikshospitalet-Radiumhospitalet, University of Oslo, Oslo, Norway) Rationale: The development of epilepsy, epileptogenesis, involves changes in gene expression. Pharmacological modulation of gene expression has been suggested as an approach to inhibit epileptogenesis. Levetiracetam (LEV) has antiepileptogenic properties, whereas phenytoin (PHT) does not. We hypothesized that regional alteration of gene expression after LEV treatment might point to brain regions that are important for epileptogenesis. Methods: Female Wistar rats were fed perorally with LEV (150 mg/kg) or PHT (75 mg/kg) twice daily for 90 days. N = 7 in each group. Affymetrix oligonucleotide-based microarrays were used to identify regional alterations in gene expression in pons/medulla oblongata, hippocampus and frontal cortex. Results: LEV treatment caused changes in the expression of 65 genes in pons/medulla oblongata, 3 in hippocampus, and 1 in frontal cortex. PHT treatment, in contrast, caused changes in the expression of 3 genes in pons/medulla oblongata, 64 genes in hippocampus, and 322 genes in frontal cortex. Very little overlap between regions or drug treatments was observed with respect to affected genes. Median serum drug concentration 4-6 h after last dose was 231 [mu]mol LEV/L and 50 [mu]mol PHT/L. Conclusions: We conclude that antiepileptic drug treatment may cause dramatic and region-specific changes in mRNA levels, possibly reflecting their different action on receptors and ion channels in the plasma membrane. The finding that gene expression was altered predominantly in pons/medulla oblongata after LEV suggests that this brain region, which i.a. harbours monoaminergic and cholinergic projection neurons, may be important for epileptogenesis. Abstract: .sup.1 Matthew H. Higgs , .sup.2 Carol A. Robbins , and 1,2,3 William J. Spain (Physiology and Biophysics, University of Washington, Seattle, WA ; Neurology, University of Washington, Seattle, WA ; and Neurology, Veterans Affairs Puget Sound Health Care System, Seattle, WA) Rationale: High-frequency spike bursts have been recorded in neocortical pyramidal neurons in vivo. Burst firing may promote epileptic discharges by increasing firing rate, synchronizing firing at high frequencies, and increasing neurotransmitter release at facilitating synapses. It is unclear whether bursts primarily reflect the pattern of synaptic input or intrinsic cellular properties. In some pyramidal neurons repetitive bursting occurs during direct current (DC) stimulation, but in most bursting is absent or appears only at the onset of current steps. In order to determine how intrinsic cellular properties shape firing patterns of layer 2/3 pyramidal neurons during input with rapid fluctuations characteristic of in vivo synaptic excitation, we applied broadband noise stimuli. Methods: Coronal slices of rat sensorimotor cortex were superfused with artificial cerebrospinal fluid (34[degrees]C) containing blockers of ionotropic glutamate and GABA receptors. Intracellular recordings were obtained from 21 regular-spiking neurons in layer 2/3 using sharp microelectrodes filled with 2.7 M KCl and 1% biocytin. The location and pyramidal morphology of most cells were confirmed by biocytin histochemistry. Neurons were stimulated with broadband current noise (1 ms exponential relaxation) in combination with a DC offset that was usually set near spike threshold. Results: All of the neurons tested fired regularly during DC stimuli, without repetitive bursting. However, 19/21 cells generated a mixture of single spikes and bursts of two or more spikes during noise stimuli. Raising the noise amplitude or the DC offset increased burst frequency, while the mean duration of intra-burst intervals remained nearly constant at approximately 10 milliseconds. The interval-dependent firing probability (hazard function) during high-noise stimulation roughly corresponded to the post-spike voltage trajectory, which is shaped by a fast afterhyperpolarization (fAHP), a fast afterdepolarization (fADP), and a medium-duration afterhyperpolarization (mAHP). Increasing the fAHP conductance by spike-triggered dynamic current clamp had little effect on regular spiking to DC stimuli but strongly reduced stochastic bursting. Conclusions: Our results suggest that intrinsic properties of layer 2/3 pyramidal neurons, including their small fAHP and prominent fADP, create an intrinsic resonance that is usually subthreshold during firing to DC stimuli but promotes bursting with an intra-burst frequency of approximately 100 Hz during noisy excitation. This frequency is similar to that of ripples observed in extracellular and EEG recordings during neocortical seizures, suggesting that high-frequency resonance in regular-spiking pyramidal neurons may contribute to epileptic discharges. (Supported by a Veterans Affairs Merit Review to W.J.S.) Abstract: .sup.1 Kristin L. Hillman , .sup.1 Van A. Doze , and .sup.1 James E. Porter (Pharmacology, Physiology and Therapeutics, University of North Dakota School of Medicine and Health Sciences, Grand Forks, ND) Rationale: Norepinephrine (NE) has been shown to have anti-epileptic effects both in vitro and in vivo. One area of the brain where this has been demonstrated is the hippocampus. We hypothesize that NE exerts its anti-epileptic effect in the hippocampus in part through activation of an [alpha].sub.1 adrenergic receptor (AR) located on a specific subset of CA1 GABAergic interneurons. Methods: Hippocampal slices prepared from young rats were used for all studies. Functional Studies: To determine the specific AR subtype expressed by CA1 interneurons, a panel of subtype specific AR agonists and antagonists was utilized. Cell-attached recordings were conducted under IR/DIC optics, with aCSF perfusion enabling drug delivery in timed increments. Action potential frequency of the interneuron was recorded, plotted as a concentration response curve, and analyzed by Schild regression. To determine if NE promotes GABA release in CA1, cell-attached and whole-cell recordings were conducted on pyramidal cells. NE challenges were made in normal aCSF and then aCSF containing GABA receptor antagonists. Molecular Studies: Cytoplasm was collected from NE-responsive CA1 interneurons and used for single-cell RT PCR. The resulting cDNA template was profiled in terms of AR subtype, neuronal cell type, and neuropeptide expression using real time PCR. Results: 6-fluoronorepinephrine (6FNE), an [alpha] AR agonist, produced a dose-dependent increase in action potential frequency in [proportional to]50% of CA1 interneurons. The response to 6FNE could be blocked by the [alpha].sub.1 AR antagonist prazosin, but not by the [alpha].sub.2 AR antagonist rauwolscine, suggesting that NE-responsive interneurons express an [alpha].sub.1 AR. Subsequent studies performed with the [alpha].sub.1 AR subtype specific antagonists 5-methylurapidil, WB 4101, L-765,314, and BMY7378 produced Schild regressions that suggested predominate functional expression of the [alpha].sub.1A AR. Molecular analysis of NE-responsive CA1 interneurons revealed transcripts for the [alpha].sub.1A and [alpha].sub.1B AR, the neuropeptide somatostatin and the GABAergic enzyme glutamate decarboxylase. Initial recordings from pyramidal cells suggest that [alpha].sub.1 AR activation decreases the activity of these cells, an effect that is attenuated by GABA blockade. Conclusions: We have identified a distinct subset of somatostatin expressing, GABAergic CA1 interneurons that respond to NE via an [alpha].sub.1A AR. Adrenergic activation of these interneurons appears to increase the amount of GABA released to pyramidal cells. These findings elucidate one mechanism by which NE may be exerting an anti-epileptic effect in the hippocampus, and furthermore identify a specific receptor subtype that may serve as a potential therapeutic target in the treatment of epilepsy. (Supported by a pre-doctoral fellowship from the American Epilepsy Society.) Abstract: .sup.1 Gregory L. Holmes , and .sup.1 Pierre-Pascal J. Lenck-Santini (Neurocience Center at Dartmouth, Dartmouth Medical School, Lebanon, NH) Rationale: Interictal spikes (IS) are morphologically defined, episodic, transient EEG discharges that occur as a result of synchronous, paroxysmal depolarization of neurons resulting in a series of action potentials. The localization of focal IS correspond closely to the site of the seizure focus and IS remain one of the most important diagnostic features on the EEG. In humans there is a clear relationship between state and IS. Inactivity, drowsiness, and sleep are potent activators of IS. The relationship between IS and state in rodents is less clear. In this experiment, we investigated the relationship between theta and IS in the rat pilocarpine model. Methods: Fifteen 60 day old rats 15 P60 rats were treated by lithium/pilocarpine and underwent status epilepticus. Among these animals, five showed spontaneous seizures and interictal spikes two months after status epilepticus induction. They were implanted with a multielectrode array driving electrodes in the CA1 region of the dorsal hippocampus so that bipolar differential recordings above and bellow the cell layer could be performed. Rats were trained to perform two types of behavioral task: 1) Pellet chasing task: rats were trained to forage for pellets randomly scattered in a circular open field arena 2) Linear track shuttle: rats were trained to shuttle between the two ends of a linear track. Hippocampal EEG theta/delta ratios, the ratio of the sum of the power-spectrum in the theta band (6-12Hz) by the sum of the power-spectrum in the delta band (1-5Hz), were performed in both tasks. Results: Strikingly, we found that more IS were observed in sessions with a low hippocampal theta/delta ratio than during periods with high ratios. Within the same session, IS appeared during low theta behaviors (either standing or walking at a speed of 10cm/sec, ratio > 4). Pepe add a figure showing this or the EEG versus spikes. Conclusions: These results show that there is a strong correlation between theta and IS. During theta activity IS are suppressed; during periods of decreased theta IS are activated. The relationship between theta and IS demonstrate that as in humans, there is a role of general arousal systems in IS modulation. IS modulation in rodents can be used to model the human condition. [figure 1] (Supported by Western Massachusetts Epilepsy Awareness Fund, Friends of Shannon McDermott, Sara Fund, and NINDS (Grants: NS27984 and NS44295).) Abstract: .sup.1 Xiaoying Huang , .sup.1 Qian Yang , and .sup.1 Jian-young Wu (Physiology and Biophysics, Georgetown University, Washington, DC) Rationale: Spiral waves are a basic feature of nature and have been observed in a variety of biological systems, but spiral dynamics have not been studied in mammalian cortex. If spirals exist in the cortex, the robust dynamics of the spiral waves may contribute to sustaining seizure activity, strategies of disrupting formation of spirals may be used to disrupt seizure activity. Methods: We used optical imaging with voltage-sensitive dyes to examine seizure-like events induced by carbachol and bicuculline in rat neocortical slices and in rat cortex in vivo. Slices from visual cortex were sectioned in a tangential plane containing cortical layers III-V. The slices were stained with an absorption dye, NK3630 before imaging, and the signals from the stained tissue were recorded by a photodiode array. In vivo imaging was performed over visual cortex in adult rats. The animals were anesthetized with urethane while a craniotomy window of 5 mm in diameter was created. The cortex was stained by a fluorescent dye RH1691, and fluorescent signals were recorded by a photodiode array. Results: The seizure-like oscillations occurred spontaneously in vitro and in vivo when the drugs were applied. The oscillations were organized spatiotemporally as propagating waves, and complex wave patterns such as spiral, plane, ring and irregular waves were observed both in slices and in vivo. These wave patterns alternated within one oscillation epoch, suggesting the patterns were dynamically organized. Spiral waves, occurred in about 50% of recording trials in slices but were rarely seen in vivo. We examined the initiation of spiral waves in slices and found that collision of two wave fronts usually occurred prior to the initiation of spiral waves. The spiral waves were very stable in slices, some sustained for up to 30 rotations, but spirals in vivo only lasted for a few cycles. In brain slices, 'phase singularity', a hallmark of spiral waves, can be clearly identified at spiral center as substantial reduction in signal amplitude and a circular phase distribution around it. Phase analysis of the spiral center and surrounding areas suggested that the amplitude reduction was caused by superimposition of multiple widely distributed phases. Conclusions: Our results suggest that spiral dynamics may exist in neocortex during seizure-like activity and may contribute to the sustaining of seizure events. (Supported by a fellowship from Epilepsy Foundation and NIH NS36447.) Abstract: .sup.1 Shin-Ichi Imamura , .sup.3 Shigeya Tanaka , .sup.2 Koichi Akaike , .sup.4 Hideshi Tojo , .sup.2 Motofumi Kasugai , .sup.2 Hideyuki Matsukubo , .sup.2 Akira Sano , and .sup.1 Kazunori Arita (Neurosurgery, Kagoshima University, Graduate School of Medical and Dental Sciences, Kagoshima, Kagoshima, Japan ; Neuropsychiatry, Kagoshima University, Graduate School of Medical and Dental Sciences, Kagoshima, Kagoshima, Japan ; Neurosurgery, Tanaka Neurosurgical Clinic, Kagoshima, Kagoshima, Japan ; and Neurosurgery, Koyokai Himawari Hospital, Kagoshima, Kagoshima, Japan) Rationale: To elusidate an antiepileptic property of gabapentine, we applied this agent to status epilepticus models induced by subcutaneous administration of kainic acid (KA). Methods: Experiments were carried out on 12 male Wistar rats undergone stereotactic implantation of electrodes in the bilateral amygdala, the bilateral dorsal hippocampus, and the bilateral sensorimotor cortex. All received subcutaneous administration of KA (10 mg/kg) to elicit status epilepticus. Thirty min after that, gabapentine (dissolved with distilled water, 30 mg/mL) was administrated intraperitoneally (2mL/kg) (gabapentine group, n = 6), while the rest received intraperitoneal administration of distilled water only (control, n = 6). Results: Ten to 20 min after KA administration, seizure discharges initially appeared in the bilateral amygdala and the bilateral dorsal hippocampus, and then propagated to the bilateral sensorimotor cortex. All rats showed status epilepticus 30 min after KA administration. In gabapentine group, seizure activity of the bilateral sensorimotor cortex decreased 15-30 min after gabapentine administration, whereas those of the bilateral amygdala and hippocampus remained unchanged. Unlike gabapentine group, seizure activity stayed unaltered in control. Conclusions: Gabapentine may show an antiepileptic effect by reducing the seizure propagation to the bilateral sensorimotor cortex. Abstract: .sup.1 Laura A. Jansen , and .sup.2 Jeffrey G. Ojemann (Division of Pediatric Neurology, Children's Hospital and Regional Medical Center and University of Washington, Seattle, WA ; and Department of Neurological Surgery, Children's Hospital and Regional Medical Center and University of Washington, Seattle, WA) Rationale: Pediatric epilepsy due to areas of focal cortical dysplasia is often intractable to anticonvulsant treatment. Many of the medications used unsuccessfully in treating this disorder, such as barbiturates, benzodiazepines, ACTH, and topiramate, are thought to exert at least a portion of their action through enhancement of inhibitory neurotransmission via GABA-A receptors. In the present study, GABA-A receptor properties in resected epileptogenic brain tissue from four children with intractable epilepsy and focal cortical dysplasia were measured to determine if this clinical resistance to pharmacologic treatment correlates with alterations in receptor responsiveness to these agents. Methods: Operative resection of the epileptogenic zones was performed, with pathologic diagnoses of focal cortical dysplasia type 2A. Portions of the epileptogenic cortex were flash frozen at the time of surgery and maintained at -80[degrees] C until use. The tissue was homogenized and subjected to a series of centrifugation steps to isolate the membrane fraction. This fraction was resuspended and injected into Xenopus oocytes, resulting in incorporation of brain membrane vesicles into the oocyte cellular membrane. Two-electrode voltage clamp analysis was performed 18-72 hours after injection, at which time GABA-A receptor currents were easily measured. Results: Functional characterization of GABA-A receptor properties was performed, including agonist potency, current-voltage relationship, current rundown, and responses to benzodiazepines, barbiturates, neurosteroids, topiramate, and zinc. Results from epileptic cortex were compared with those from non-epileptic autopsy and surgical control samples. Experimentally determined receptor properties were correlated with patient age and clinical characteristics. Conclusions: These studies of GABA-A receptor function in children with intractable epilepsy due to focal cortical dysplasia provide valuable insight into the mechanisms of anticonvulsant resistance in this disorder. (Supported by NIH NINDS K08 NS52454.) Abstract: .sup.1 Stuart Jarrett , .sup.1 Li-Ping Liang , .sup.2 Jennifer L. Hellier , .sup.2 Philip A. Williams , .sup.2 Volodymyr I. Dzhala , .sup.2 Jethro P. Jones , .sup.2 Kevin J. Staley , and .sup.1 Manisha N. Patel (Pharmaceutical Sciences, University of Colorado ; and Pediatrics and Medicine, University of Colorado, Denver, CO) Rationale: Epilepsy is a recent addition to the diverse neurological disorders in which mitochondrial oxidative stress and dysfunction has been implicated as a contributing factor. Studies in our laboratory have shown that status epilepticus (SE) produces oxidative stress that originates primarily from the mitochondria. An important consequence of mitochondrial free radical production is the oxidative damage of mitochondrial DNA (mtDNA). The goal of this study was to determine if acute or chronic epileptic seizures result in oxidative damage to mtDNA and if this activates mtDNA repair. Methods: Oxidative damage to mtDNA base 8-hydroxy-2'deoxyguanosine (8OHdG/2dG), mtDNA damage (quantitative polymerase chain reaction; QPCR), and changes in the base excision repair (BER) proteins (polymerase gamma; pol [gamma] and 8-oxoguanine glycosylase; Ogg1) were determined in mitochondria from the rat hippocampus at various times following acute kainate-induced status epilepticus (SE) or chronic kainate-induced epilepsy. To investigate acute SE, rats were injected with a single 12 mg/kg dose of kainate and sacrificed at 24h, 48h, 96h, 7d, 21d post-treatment. To investigate kainate-induced epilepsy, rats were injected with multiple low-dose injections of kainate (5 mg/kg) and sacrificed [greater than or equal to]3 months post-treatment. Rats displaying class IV-V seizures up to 6hr after the initial kainate injection (12 mg/kg; s.c.) as determined using a modified Racine's scale) were used for subsequent studies. Results: Acute kainate-induced SE produced a time-dependent increase in mitochondrial, but not nuclear 8OHdG/2dG levels and mtDNA damage in the hippocampus (not cerebellum). MtDNA lesions returned to control levels after 96 hr. MtDNA damage was accompanied by a significant increase in hippocampal pol [gamma] and Ogg1 mRNA and protein levels. In epileptic rats, mtDNA damage was significantly higher in the hippocampus compared to the cerebellum. In addition, a significant reduction in pol [gamma] protein expression was observed in the hippocampus, but not in the cerebellum. By contrast, no significant differences in Ogg1 protein levels were observed. Conclusions: These data demonstrate oxidative mtDNA damage and activation of BER following SE. In chronically epileptic rats, hippocampal mtDNA lesions occurred concurrently with a reduction of pol [gamma] repair enzyme. Together, changes in the mitochondrial BER pathway and increased mtDNA lesions provide further evidence for mitochondrial oxidative stress in the epileptic brain. Following acute SE, BER mechanisms are induced to attempt the repair mtDNA lesions. However, chronic epileptic seizures results in a failure of some BER enyzmes and consequent mtDNA lesions which could in turn contribute to the increased seizure susceptibility and development of epilepsy. (Supported by NIH RO1NS039587.) Abstract: .sup.1 Alicia L. Jensen , and .sup.1 Dominique M. Durand (Deptartment of Biomedical Engineering, Case Western Reserve University, Cleveland, OH) Rationale: The effect of deep brain stimulation (DBS), also known as high frequency stimulation (HFS), on neural elements close to the stimulating electrode remains an important question. Although computational studies have suggested that HFS has a dual effect on neural elements (inhibiting cell bodies, while exciting axons), experimental evidence of these effects is conflicting and unclear. The current study investigates the hypothesis that HFS suppresses axonal conduction. HFS application to the axon tracts themselves may provide novel, direct control for disrupting the propagation of seizure activity between brain regions. Methods: Experiments were carried out using in-vitro brain slices. Extracellular recording electrodes were positioned in the CA1 stratum pyramidale and/or alvear axon field to mointor evoked potentials. Electrical stimulation (0.5 Hz) was applied in the alveus producing compound action potentials (CAPs). Sinusoidal HFS (50Hz-10kHz) was applied to a third electrode located between the stimulating and recording electrodes. Filtering was used to remove the large stimulation artifact. Antidromic evoked field potential and compound action potential amplitude, width, and latency were analyzed prior, during, and post HFS. Extracellular potassium levels were monitored using single barreled ion-selective microelectrodes. Results: Sinusoidal stimulation suppresses activity in cell bodies (evoked potentials) and axons (compound action potentials). Suppression was dependent on HFS amplitude and frequency, but independent of stimulus duration and synaptic input. The frequency range of this effect is nearly identical to that reported for DBS, with maximal suppression between 50 and 200Hz. At 2kHz, the suppression effect disappeared. The suppression effect was independent of synaptic transmission, and could not be attributed to desynchronization or damage. Suppression of axonal conduction was associated with an increase in extracellular peak potassium concentration (r= 0.92), with complete suppression at a peak potassium threshold of 8.25mM. Bath application of elevated potassium produced significant suppression of axonal activity in 12 mM and 15mM KCl ACSF. However, 8.5 mM bath potassium levels failed to block axonal activity. Conclusions: Sinusoidal HFS suppresses axonal conduction. The frequency dependence of these suppression effects is identical to that observed for DBS in Parkinson's disease. Although the mechanism for HFS suppression is not known, damage to cells, increased inhibition, and desynchonization have been ruled out. An increase in extracellular potassium has been associated with the suppression effect; however, the potassium concentration decays significantly during stimulation while HFS suppression remains high. It is therefore likely, that, depolarization block is partially responsible for the effect. However, other mechanisms must play a role. (Supported by National Institutes of Health, the Epilepsy Foundation, and the Department of Education GAANN Fellowship program.) Abstract: .sup.1 Jenine E. John , .sup.1 Shirn L. Baptiste , .sup.1 Geza Medveczky , .sup.2 Hans von Gizycki , .sup.1 Ruben I. Kuzniecky , .sup.1 Orrin Devinsky , and .sup.1 Nandor Ludvig (Neurology, Comprehensive Epilepsy Center, NYU School of Medicine, New York, NY ; and Scientific Computing Center, SUNY Downstate Medical Center, Brooklyn, NY) Rationale: This study was part of our ongoing effort to identify an ideal drug combination for seizure-controlling subdural Hybrid Neuroprostheses. We have recently shown that in freely moving rats, epidurally delivered pentobarbital can prevent and terminate neocortical seizures induced by localized acetylcholine (Ach) application (Ludvig et al., Epilepsia, 2006; in press). The present experiments were designed to obtain information on whether epidurally applied neurotransmitters, specifically GABA and adenosine, can also exert similar effects in rats. Methods: Each rat was implanted with an epidural cup placed above the right parietal cortex. Epidural EEG electrodes were positioned adjacent to the posterior edge of the cup. After a 6-7-day recovery period, bipolar EEG recordings were conducted in a Faraday cage, using a cable with built-in operational amplifiers to eliminate movement artifacts. For focal seizure induction, 274 mM Ach was delivered into the epidural cup. After a 15-min exposure to Ach, a combination of 250 mM GABA and 5 mM adenosine was delivered into the cup and kept there for a 10-min period. Fast Fourier Transform (FFT) was used to determine the maximum and average power ([mu]V.sup.2) of the low-beta (13-20 Hz) frequency band before drug deliveries, during Ach exposure and after the delivery of the GABA/adenosine combination. Results: Ach induced EEG seizures within 390 [+ or -] 30 (mean [+ or -] S.E.M.) sec, which were accompanied by clonic convulsions. During EEG seizures, there was an increase in maximum power from 13.8 [+ or -] 10.3 to 36.2 [+ or -] 22 along with an increase in average power from 4.7 [+ or -] 2.8 to 8.9 [+ or -] 4.9. Delivery of the GABA/adenosine combination fully terminated the ongoing Ach-induced EEG seizures and convulsions within 33.5 [+ or -] 21.5 sec. This GABA/adenosine effect was reflected in the reduction of the maximum power from 36.2 [+ or -] 22 to 17.2 [+ or -] 9.2 and that of the average power from 8.9 [+ or -] 4.9 to 3.9 [+ or -] 2.3. The EEG power values after GABA/adenosine delivery were comparable to those obtained in the normal, pre-seizure state. Abnormal electrographic signs following epidural exposure to this drug combination were not observed. Conclusions: The combination of GABA and adenosine, delivered onto the surface of dura mater, can fully terminate Ach-induced neocortical EEG seizures and their behavioral manifestations. This suggests that: (1) at least a portion of the GABA and/or adenosine molecules can diffuse through the meninges into the underlying neocortical tissue, and (2) this treatment can terminate locally induced neocortical seizures. Further studies are required to thoroughly determine the efficacy of GABA and adenosine in terminating, and perhaps preventing, focal neocortical seizures upon epidural/subdural deliveries. (Supported by NYU/FACES.) Abstract: .sup.1 Motofumi Kasugai , .sup.1 Koichi Akaike , .sup.1 Shin-Ichi Imamura , .sup.1 Hideyuki Matsukubo , .sup.1 Hideshi Tojo , .sup.1 Shigeya Tanaka , and .sup.1 Akira Sano (Department of Psychiatry, Field of Social and Behavioral Medicine, Kagoshima University Graduate School of Medical and Dental, Kagishima, Kagishima, Japan; Department of Neurosurgery, Kagoshima University Graduate School of Medical and Dental, Kagishima, Kagishima, Japan; and Neurosurgery, Tanaka Neurosurgical Clinic, Kagishima, Kagishima, Japan) Rationale: Previously, we investigated seizures following microinjection of the kainic acid (KA) into the left amygdala (LA) in mice with different strains, C57BL/6 (B6) and FVB/N (FVB). We reported that mice of the FVB strain show more seizure susceptibility than those of the B6 in the KA-induced amygdalar seizures. However, the details of the genes responsible for such difference remain obscure. In the present study, we crossed male B6 with female FVB to produce an F1 generation, and made seizure models of the F1. Methods: Male B6 were crossed with female FVB to produce an F1 generation. The F1 (n = 6) underwent implantation of electrodes in the LA and the left sensorimotor cortex (LCx). Stainless cannula was also inserted into the LA for KA injection. All animals received KA injection into the LA (0.5 microg) to induce seizures, Electroencephalographic and behavioral observation was made for 48 hours, and the voltages of the spike were statically analized. Mice then were decapitated and histologic studies including analysis of the neuronal density of the CA3 in the hippocampus were carried out. Results: On electoroencephalograms (EEGs), multiple spike discharges initially appeared in the LA within 10-20 min after KA injection and mouth and facial movement were behaviorally observed. At 20-60 min after KA injection, the seizure discharge propagated to the LCx. Within 60-90 min after KA injection the amplitude of the multiple spike discharges increased. Behaviorally, the animals demonstrated seizure motor manifestation such as head nodding, forelimb clonus, rearing and falling. During ictal EEG change, all mice showed only immobilization, lasting 1-2 min and recurring every 10-20 min. At 48 hours after KA injection multiple spike discharges were disappeared. Neuronal damage in the CA3 region was moderated. Conclusions: In the KA-induced amygdalar seizures, it is likely that mice of the F1 show resistant to seizure susceptibility similar to B6. Further examination related to genetic factor will be needed. (Supported by Grant-in-Aid for Encouragement of Young Scientists from the Ministry of Education, Sciences, and Culture of Japan.) Abstract: .sup.1 Mehmet Kaya , .sup.1 Rivaze Kalayci , .sup.1 Nadir Arican , .sup.1 Candan Gurses , .sup.1 Bulent Ahishali , .sup.1 Oguzhan Ekizoglu , .sup.1 Imdat Elmas , .sup.1 Mutlu Kucuk , .sup.1 Duran Ustek , .sup.1 Bilge Bilgic , and .sup.1 Gonul Kemikler (Physiology, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey; Research Institute for Experimental Medicine, Istanbul University, Istanbul, Turkey; Forensic Medicine, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey; Neurology, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey; Histology, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey; Pathology, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey; and Radiotherapy, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey) Rationale: The cellular mechanisms underlying epileptogenesis in the setting of cortical dysplasia are still poorly understood. In this study, the changes in the blood-brain barrier (BBB) integrity after pentylenetetrazole (PTZ)-kindling were investigated in rats with cortical dysplasia. To show the changes in the expression of occludin, immunohistochemical and Western blot methods as well as electron microscopy were used. Methods: In this study, 38 young adult Sprague-Dawley rats were used for NaFlu staining method and 108 animals were used for other parameters in parallel groups. Study groups consist of a control (8), cortical dysplasia (10), kindling (10) and cortical dysplasia in addition to kindling and PTZ groups (10). The day of insemination was taken as the embryonic (E) day 0. The pregnant rats which have been given sodium pentobarbital anesthesia with the dose of 30 mg/kg, were administered gamma irradiation of 145 cGy from the uterus on day E 17. The baby rats were given 30 mg/kg PTZ three times per week intraperitoneally to induce kindling model starting from the 2.sup.nd week after birth to the 8.sup.th week. The changes in the expression of occludin, which is a tight junction protein in BBB, were shown by using immunohistochemical and Western blot methods in addition to electron microscopy. Results: While the arterial blood pressure remained at the control blood pressure values in the rats of cortical dysplasia and cortical dysplasia in addition to kindling groups, application of acute PTZ increased the blood pressure level significantly. The amount of NaFlu dye did not convey a remarkable increase in these rats. However, after seizures induced by PTZ application in the rats of cortical dysplasia and cortical dysplasia in addition to kindling groups, BBB permeability increased siginificantly. Conclusions: In the groups of rats with cortical displasia and PTZ kindled rats with cortical displasia no BBB permeability alterations were observed. However, in the dysplasia+kindling group of rats with acutely induced seizures after being given PTZ, a significant alteration of BBB permeability was seen. This shows that displasia with kindling model is more responsive to acute seizures at least in terms of BBB permeability. Abstract: .sup.1 Kevin M. Kelly , .sup.1 Hong Xu , .sup.1 Guo Yin , .sup.1 Peter Jukkola , .sup.1 Elena Kharlamov , and .sup.1 Kebin Zeng (Department of Neurology, Allegheny General Hospital, Center for Neuroscience Research, Allegheny-Singer Research Institute, Pittsburgh, PA; and Drexel University College of Medicine, Philadelphia, PA) Rationale: Biological aging is associated with an increasing incidence of functional deficits in multiple organ systems as well as a growing risk of disease. In the brain, age-related alterations in GABAergic inhibitory system may occur resulting in increased neuronal excitability and the generation of epileptiform discharges. The goal of this study was to determine whether there were significant age-related changes of GABAergic synaptic inhibition to pyramidal neurons of layer II/III of the sensorimotor cortex. Methods: Whole-cell patch clamp techniques were used to record postsynaptic currents (PSCs) and spontaneous and miniature inhibitory postsynaptic currents (sIPSCs and mIPSCs) from layer II/III pyramidal neurons in sensorimotor cortical slices of young adult (4 month old, n = 9) and aged (20 month old, n = 9) Fischer 344 rats. Analysis was performed on PSCs and sIPSCs recorded from 44 pyramidal neurons (young, n = 23; aged, n = 21) and mIPSCs recorded from 29 of the 44 pyramidal neurons (young, n = 15; aged, n = 14). Results: For aged animals, compared with young adult animals: 1) the amplitude of mIPSCs increased significantly ([proportional to]55% higher than that of young animals, 25.72 [+ or -] 2.50pA in young vs.39.87 [+ or -] 3.82pA in aged), whereas PSC and sIPSC amplitude had no significant change; 2) the 10-90% rise time of PSCs increased significantly ([proportional to]19% longer than that of young animals, 3.92 [+ or -] 0.29ms in young animals vs. 4.65 [+ or -] 0.20ms in aged animals)and sIPSCs ([proportional to]18% longer than that of young animals, 3.97 [+ or -] 0.26ms in young animals vs.4.66 [+ or -] 0.21ms in aged animals),whereas mIPSC 10-90% rise time had no significant change; 3) The PSC, sIPSC, and mIPSC decay time constant was similar; 4) The charge transfer of PSCs and sIPSCs was similar, whereas the mIPSC charge transfer demonstrated a trend toward increase ([proportional to]50% higher than that of young animals); 5) there were significant increases in the frequency of PSCs ([proportional to]130% higher than that of young animals), sIPSCs ([proportional to]127% higher than that of young animals), and mIPSCs ([proportional to]144% higher than that of young animals). Conclusions: GABA.sub.A-receptor mediated synaptic inhibition of sensorimotor cortex layer II/III pyramidal neurons is enhanced with age. This enhancement of GABAergic inhibitory synaptic transmission may lead to significant alterations in cortical signaling activity that may be associated with an increased propensity for epileptogenesis in the aging brain. (Supported by R01 NS046015 to KMK.) Abstract: .sup.1 Do Young Kim , and .sup.1 Jong M. Rho (Neurology Research, Barrow Neurological Institute, Phoenix, AZ) Rationale: It is well known that oxidative stress disrupts normal synaptic transmission and contributes to cellular injury and hyperexcitability. Seizure activity is also known to increase oxidative neuronal damage, which in turn can impair long-term potentiation (LTP), a model of learning and memory. Recently, we have shown that ketone bodies (KB) strongly suppress hydrogen peroxide ([H.sub.2]O.sub.2)-induced oxidative stress in neocortical neurons via activation of mitochondrial K.sub.ATP(mK.sub.ATP) channels (Kim et al., 2005 Soc Neurosci Abstr). In the present study, we asked whether KB and diazoxide (DZ, an activator of mK.sub.ATP channels) can protect against oxidative impairment of LTP in rat hippocampus. Methods: Acute hippocampal slices were prepared from Wistar rats (3-5 weeks), and extracellular field recordings were made in CA1 in the presence of various substrates. Schaffer collaterals were stimulated to evoke population spikes (PS) in the stratum pyramidale, and the stratum radiatum was stimulated using a theta-burst protocol (5 trains delivered at 0.2 Hz; each train consisting of five simulations, 0.1 ms, 100 Hz) to induce long-term potentiation (LTP). Results: [H.sub.2]O.sub.2 (2 mM, 15 min application alone) strongly inhibited the field population spike (PS) by over 80%. Pretreatment of KB (acetoacetate and D-[beta]-hydrobutyrate, 1 mM each) or DZ (100 [mu]M) for 10 min prior to [H.sub.2]O.sub.2 administration resulted in only a slight depression of PS amplitude (18% and 10%, respectively). The protective effects of KB and DZ were dose-dependent, and were completely reversed by 5-hydroxydecanoate (200 [mu]M; 5-HD, a mitoK.sub.ATP blocker). In additional experiments, [H.sub.2]O.sub.2 (200 [mu]M) strongly suppressed LTP induction and maintenance. Neither KB (3 mM alone or applied as a 1:1 cocktail) nor DZ exerted any significant effects on TBS-induced LTP. However, co-application of KB or DZ completely prevented [H.sub.[2]O.sub.2-]induced impairment of LTP. On the other hand, pre-treatment with 5-HD reversed the protective effects of KB and DZ. Conclusions: Our findings suggest that KB can restore oxidative stress-induced impairment of normal synaptic transmission in the CA1 region of the hippocampus. Furthermore, the underlying neuroprotective mechanisms may involve modulation of mitochondrial K.sub.ATP channels. (Supported by NIH grant NS 044846 (JMR) and the Barrow Neurological Foundation.) Abstract: .sup.1 Yeong-In Kim , .sup.2 Hong-Ki Song , .sup.1 Young-Min Sohn , .sup.3 Ji-Eun Kim , and .sup.3 Tae-Cheon Kang (Neurology, The Catholic University of Korea, Seoul, Korea ; Neurology, Hallym University, Seoul, Korea ; and Anatomy, Hallym University, Chuncheon, Gangwondo, Korea) Rationale: Riluzole may be one of candidates as supplementary medications for treatment of epilepsy. However, there is little information about the anti-epileptic effect of riluzole, since evaluation of this drug has been mostly performed in normal animals. Therefore, the comparison of effects of riluzole with valproic acid (VPA) was conducted in the pilocarpine-induced temporal lobe epilepsy model (a complex partial seizure model) and the [gamma]-hydroxybutyrate lactone (GBL)-induced absence seizure models. Methods: Following VPA or riluzole treatment, electrophysiological and immunohistochemical studies were performed in pilocarpine- or GBL-induced epilepsy models. Results: Both VPA and riluzole treatment saliently reduced vesicular glutamate transporter (VGLUT) immunoreactivities and fEPSP slope. Riluzole treatment completely inhibited pre-ictal spikes and spike-wave discharges (SWD) in pilocarpine- and GBL-induced epilepsy model, although VPA treatment partially inhibited. Both VPA and riluzole reduced paired-pulse inhibition in the pilocarpine-induced epilepsy model. In the GBL-absence seizure model, riluzole enhanced paired-pulse inhibition, but VPA reduced it. Conclusions: These results suggest that inhibition of presynaptic glutamate release by riluzole may be more effective to reduce seizure activity both in complex partial seizure and in absence seizure than VPA. Abstract: 1,2,4 Merab Kokaia , 2,4 Sara Bonde , 2,4 Katherine Jakubs , 2,4 Avtandil Nanobashvili , 2,4 Christine T. Ekdahl , 3,4 Zaal Kokaia , and 2,4 Olle Lindvall (Experimental Epilepsy Group, Section of Restorative Neurology, Wallenberg Neuroscience Center, Lund University Hospital, Lund, Sweden ; Laboratory of Neurogenesis and Cell Therapy, Section of Restorative Neurology, Wallenberg Neuroscience Center, Lund University Hospital, Lund, Sweden ; Laboratory of Neural Stem Cell Biology, Section of Restorative Neurology, Stem Cell Institute, Lund University Hospital, Lund, Sweden ; and Lund Strategic Research Center for Stem Cell Biology and Cell Therapy, Lund, Sweden) Rationale: Neurogenesis continues in the adult mammalian brain from neural stem/progenitor cells in the dentate gyrus subgranular zone (SGZ). Status epilepticus (SE) triggers increased SGZ cell proliferation. Intracellular recordings from the aberrant hilar neurons in hippocampal slices from epileptic rats indicate that they retain the intrinsic properties of granule cells. However, they exhibit abnormal pro-epileptogenic burst firing in synchrony with CA3 pyramidal cells, suggesting that the new hilar neurons contribute to the epileptic condition. The functional characteristics of the majority of the new cells generated in the epileptic brain, which are located at their normal site in the granule cell layer, have not been investigated. Methods: We have used whole-cell patch-clamp recordings to determine how a pathological environment influences the electrophysiological properties and functional integration of the new cells in the GCL. New neurons were labeled through intrahippocampal injection of RV-GFP and we compared those generated following voluntary running, i.e., a physiological stimulus, with cells formed after SE, i.e., a brain insult. Results: We found that the new cells differentiate into functionally integrated granule cells with similar intrinsic membrane properties in both running and SE environments. However, the neurons born in the epileptic brain show alterations in excitatory and inhibitory inputs, which could attenuate synaptic excitability: The excitatory synaptic drive was attenuated on the new cells born into the epileptic environment compared to those formed under physiological conditions, while the inhibitory synaptic drive on these cells was increased. Conclusions: Our study demonstrates that both a physiological stimulus and an insult to the adult brain trigger the formation of new dentate granule cells, which are functionally integrated into hippocampal neural circuitry. However, following the insult, the functional connectivity of the new neurons seems to mitigate the dysfunction in the epileptic brain. These data provide further evidence for a therapeutic potential of endogenous neurogenesis. (Supported by Swedish Research Council, and Segerfalk, Crafoord and Kock foundations.) Abstract: .sup.1 Hana Kubova (Developmental Epileptology, Institute of Physiology, Academy of Sciences, Prague, Czech Republic) Rationale: Using a model of pentylenetetrazol (PTZ)-induced seizures we examined weather withdrawal of clonazepam (CZP) and/or partial benzodiazepine (BZs) agonist Ro 19-8022 leads to changes in seizure susceptibility in immature rats. Methods: To detect development of rebound phenomena, single dose of CZP (1 mg/kg i.p.) or Ro 19-8022 (0.5 mg/kg i.p.) was injected to P12 rats. Seizures were induced by PTZ in doses of 50 (sub-threshold) or 100 (supra-threshold) mg/kg s.c. 12, 24 or 48 h later. To assess changes of seizure susceptibility after chronic administration, animals received either BZ in doses mentioned above for 5 consecutive days starting at P7. PTZ (50 mg/kg s.c.) was injected 2, 4 or 6 d after the last BZ injection. Changes in sensitivity to anticonvulsant effects of BZs after treatment discontinuation were detected in additional groups of animals 1 or 2 days after the end of chronic treatment. Animals received fully anticonvulsant dose of CZP (0.08 mg/kg i.p.) or Ro 19-2088 (0.05 mg/kg i.p.) and 10 min later PTZ in a dose of 100 mg/kg s.c. Incidence and latency of generalized tonic-clonic seizures as well as severity of seizures were registered. Results: After a single injection of tested BZs, anticonvulsant effects expressed as prolongation of GTCS latencies were still observed 12 h (Ro 19-8022) or 24 h (CZP) after injection. In intervals 24 and 48 h, respectively, latencies were however 40-60% shorter compared to untreated controls (p < 0.05). At the same time points, PTZ (50 mg/kg) induced more severe seizures (p < 0.01) with significantly higher incidence of GTCS in BZs-treated rats compared to controls (p < 0.05). Chronic administration of CZP increased both seizure severity and incidence of GTCS for 6 d after the last CZP injection compared to untreated controls (p < 0.05). In contrast, no changes in severity of epileptic phenomena were observed after chronic administration of Ro 19-8022. Neither CZP nor Ro 19-8022 administered chronically changed efficacy of BZs against seizures induced by PTZ (100 mg/kg s.c.) 1 or 2 d after the end of therapy. Conclusions: In immature rats, withdrawal of chronic CZP but not Ro 19-8022 treatment increased severity of PTZ-induced seizures. In contrast, single dose of both CZP and Ro19-8022 led to occurrence rebound phenomenon. (Supported by a research project No. LC554 of the Ministry of Education of the Czech Republic.) Abstract: .sup.1 Pawel Kudela , .sup.2 William S. Anderson , .sup.1 Piotr J. Franaszczuk , and .sup.1 Gregory K. Bergey (Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD ; and Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, MD) Rationale: Synchronized and recurrent neuronal bursting is a signature of epileptiform activity. Much work has been done on mechanisms underlying these pathological network behaviors. Several mechanisms including membrane after-hyperpolarization and depletion of synaptic transmitter have been recognized as being implicated in the mechanism of recurrent bursting. Using neuronal network models we demonstrate another plausible mechanism for recurrent bursting in disinhibited neuronal networks, caused by short-term (10-100 ms) synaptic depression. Methods: Simulated networks were composed of inhibitory (fast spiking) and excitatory (regular spiking) neurons. Excitatory neurons in response to injected external excitatory current exhibited no spike frequency adaptation. The synaptic weights in networks were dynamically modified with ongoing network activities. To simulate synaptic plasticity we used a phenomenological model in which the synaptic strength change depends on the history of pre-postsynaptic activity and is determined by the synaptic calcium residual level. The networks were activated by action potentials applied at random intervals on inputs of selected excitatory neurons (less then 25% of the total number of neurons). Neuronal bursting in these networks was induced by gradually reducing the excitatory drive to inhibitory neurons. Results: Action potentials applied at random intervals to selected neurons activated the entire network but no bursting behavior was observed. When excitatory drive to inhibitory neurons in these networks was gradually reduced, excitatory neurons exhibited steady, synchronous, and periodic neuronal bursting. Single bursts in neurons lasted for up to 60 ms and were followed by [proportional to]100 ms long periods of inactivity. This temporary cessation of single trains of APs in neurons was mediated by synaptic depression. This was confirmed by the two observations: 1) the maximum of short-term synaptic depression of presynaptic connections overlapped with burst cessation in examined excitatory neurons in these networks, and 2) when short-term depression was not simulated there were no bursting behaviors in these networks. The recovery from synaptic depression was always followed by the onset of new bursts in examined neurons, and networks returned to bursting mode. Conclusions: Our results show that short-term plasticity can be a plausible mechanism contributing to the termination of neuronal bursting. In this network model the cessation of single trains result directly from synaptic depression. We specifically utilized a calcium-regulated model of synaptic plasticity. However, other forms of synaptic plasticity matching the time scale of tens of milliseconds might be also responsible for the recurrent cessation of bursting events, including synaptic depletion. (Supported by NIH NS 51382, NS 38958.) Abstract: .sup.1 Jonathan E. Kurz , .sup.1 Ahn L. Anderson , and .sup.1 Severn B. Churn (Neurology, Virginia Commonwealth University, Richmond, VA) Rationale: Status epilepticus (SE) is a life-threatening condition charecterized by continuous seizure activity. Previous studies have documented an increase in calcineurin (CaN) activity following SE, as well as an SE-induced increase in the post-synaptic concentration of CaN (Kurz et al. J Neurochem,2001 78:304-15 and Neurobiol Dis,2003 14:483-93). The role of these changes in CaN in the pathology of SE remains unclear. This study examines a potential CaN-dependent modification of the dendritic cytoskeleton involving cofilin activation. CaN has been shown to indirectly induce dephosphorylation of cofilin, an actin depolymerization factor. The ensuing actin depolymerization can lead to a number of physiological changes that are of interest in SE. Methods: Pilocarpine injection was used to induce SE in adult male Sprague-Dawley rats. When appropriate, animals were administered 10 mg/kg cyclosporin A or 2.5 mg/kg FK506 i.v. 3h prior to SE. Animals were monitored by video-EEG. At specfic times after the initiation of seizure activity brain tissue was harvested and separated into subcellular fractions by differential centrifugation. Cofilin phosphorylation and total cofilin were assayed by western analysis. CaN activity was assayed via an RII-phosphopeptide based assay. Actin polymerization was assayed both by histochemical means and by determining the ratio of G-to F-actin after separating the two with a solubilization/centrifugation protocol. Results: The immunoreactivity of phosphocofilin decreased significantly in both hippocampal and cortical synaptosomal samples after SE. Hippocampal and cortical phosphocofilin were decreased to approximately 20% of control after 1h of SE (n = 10, p < 0.001). Total cofilin immunoreactivity remained unchanged after 1h of SE, indicating that the loss of phosphocofilin was likely due to dephosphorylation rather than overall loss of cofilin. This dephosphorylation occured as early as 10 min after SE onset, a timeframe that correlates with the timing of the increase in SPM CaN. Cofilin dephosphorylation could be partially blocked by the injection of CaN inhibitors prior to SE, indicating that this process is partially CaN-dependent. Cofilin activation could be further demonstrated by increased actin-cofilin binding on coimmunoprecipitation, which could also be blocked by CaN inhibitors. Finally, SE induced a significant CaN-dependent depolymerization of neuronal actin. Conclusions: The data demonstrate a significant, CaN-dependent activation of cofilin in SE and a subsequent CaN-mediated depolymerization of actin. This research could be applied to explain SE-induced modulation and loss of dendritic spines, as well as changes in receptor localization. Further research into this area may provide useful insights into the pathology of SE and epileptogenic mechanisms. (Supported by the Epilepsy Foundation through the American Epilepsy Society and the Lennox Trust Fund (JEK) and by R01-NS39770 (SBC).) Abstract: .sup.1 Fred A. Lado , .sup.2 Wei Sun , .sup.2 Cynthia Pan , and .sup.2 Hoby Hetherington (Neurology, Albert Einstein College of Medicine, Bronx, NY ; and Magnetic Resonance Research Center, Albert Einstein College of Medicine, Bronx, NY) Rationale: Epilepsy is a common disease disproportionately affecting children. Often, seizures develop months or years - the 'latent period'-following an initial brain injury, such as an episode of SE in early childhood. Not all individuals with comparable early life SE, however, develop epilepsy. Indeed, only a minority of children at risk subsequently develops epilepsy. To prospectively select children most likely to benefit from antiepileptogenic therapy it is necessary to identify surrogate markers. The recent advances in magnetic resonance spectroscopic imaging (MRSI) suggest that it may be possible to this tool to identify a surrogate marker of epileptogenesis. Methods: Lithium-pilocarpine status epilepticus (SE) was induced in 19-day old rat pups. Pups received intramuscular diazepam two hours after onset of SE to reduce mortality. The severity of seizures during SE were scored using the Racine scale. Pups displaying stage 5 seizures were selected for prospective repeated evaluation with MRSI 24 h, 1 wk, 2 wk, 4 wk, 6 wk after SE. MRSI data were acquired with 2ul resolution. Three voxels within each hippocampus (6 voxels per rat) were selected for analysis. Continuous video recordings are analyzed to detect spontaneous seizures in the weeks after SE. In this model, it is expected that approximately 1/3 of rats will have seizures, while the remainder do not. Results: At 24 h after SE, pups showed a 10-20% drop in N-acetyl-aspartate (NAA) to creatine (Cr) ratio in the hippocampus. Repeat scans at intermediate intervals show progressive decline in NAA/Cr in most rats undergoing SE, while control levels remain stable or rise. Preliminary evidence in some rats shows that after 6 weeks NAA/Cr may rise. Correlation of MRSI measurements with seizures detected by video recording is ongoing to determine whether the changes observed predict subsequent epileptogenesis. Conclusions: In immature rats, hippocampal NAA/Cr appears to decline for several weeks following SE. Preliminary results suggest there is partial recovery of NAA/Cr in some rats. Whether the degree of NAA/Cr decline, or possibly the extent of recovery, predict subsequent epileptogenesis will be determined as the rats are followed prospectively. (Supported by NINDS NS048149.) Abstract: .sup.1 Andre H. Lagrange , and .sup.1 Robert L. Macdonald (Department of Neurology, Vanderbilt University, Nashville, TN) Rationale: Extrasynaptic GABA.sub.A receptors (GABARs) produce small but long-lived tonic currents in response to the low levels of ambient GABA (up to [approximately equal to]1 mM) that are present outside GABAergic synapses. GABARs may also be clustered near synapses (perisynaptic), where they respond to brief surges of extrasynaptic GABA resulting from synaptic overflow. The extrasynaptic GABARs are a mixed population, although [alpha]4[beta]x[delta] and [alpha]5[beta]3[gamma]2L receptors are particularly important in the thalamus and hippocampus. To function efficiently in the extrasynaptic milieu, it has been suggested that these receptors must be non-desensitizing and sensitive to low levels of GABA. Methods: We tested this hypothesis by applying GABA with a rapid drug application system to recombinant GABARs expressed in HEK 293T cells. Results: We report that [alpha]4[beta]1[delta], [alpha]4[beta]2[delta], [alpha]4[beta]3[delta], and [alpha]5[beta]3[gamma]2L receptors responded to low levels of GABA, although the GABA EC.sub.50 varied considerably ([alpha]5[beta]3[gamma]2L > [alpha]4[beta]3[delta] > [alpha]4[beta]1[delta] > [alpha]4[beta]2[delta]). Since extrasynaptic GABA is thought to change slowly, it is not surprising that all [alpha]4[beta]x[delta] combinations activated slowly, requiring nearly 20 ms to reach peak current in response to even saturating GABA concentrations. In contrast, [alpha]5[beta]3[gamma]2L receptors activated 10 fold more rapidly. All four GABAR isoforms desensitized extensively, with residual currents that were only 30% of peak current during a 4 sec application of 1 mM GABA. Interestingly, the residual currents, generally accepted as measures of tonic inhibition, were very similar at all but the lowest GABA concentrations. However, the GABA concentration producing this maximal "tonic" current varied from 3-20 [mu]M, depending on the receptor isoform. Thus, each GABAR responded to its own unique range of GABA concentrations. Finally, [alpha]5[beta]3[gamma]2L receptors expressed a small spontaneous current (10% of the response to 1 [mu]M GABA) that was blocked by picrotoxin. This current was not recorded in cells transfected with [alpha]1[beta]3[gamma]2L, [alpha]4[beta]3[gamma]2L, [alpha]1[beta]3[delta] or [alpha]4[beta]3[delta] receptors. Conclusions: In summary, we report that the kinetic properties of individual extrasynaptic GABARs are tuned for efficient response to extrasynaptic activation and suggest that the relatively high GABA EC.sub.50 and rapid activation of [alpha]5[beta]3[gamma]2L GABARs renders them ideally suited to function as perisynaptic receptors. Furthermore, a low level of spontaneous activity could allow [alpha]5[beta]3[gamma]2L GABARs to modulate neuronal activity, even in the absence of ambient GABA. In contrast, the highly sensitive but slowly activating [alpha]4[beta]x[delta] GABARs would function well as extrasynaptic receptors. Since ambient GABA levels may be altered during development as well as by drugs and disease states, these findings raise the exciting possibility that selective modulation of extrasynaptic GABAR subtypes may modify specific components of the tonic currents to treat illnesses such as epilepsy. (Supported by K08 NS 045122 to AHL, R01 NS33300 to RLM.) Abstract: .sup.1 Ann Lam , .sup.1 Naomi Whelan , and .sup.1 Michael E. Corcoran (Anatomy and Cell Biology, University of Saskatchewan, Saskatoon, SK, Canada) Rationale: Transfer occurs when kindling of a site reduces the number of stimulations required to kindle subsequently from a secondary site. One explanation of transfer is that kindling of the first site causes other regions to become more susceptible to seizures. Typically only the rate of seizure generalization is affected in transfer, and parameters such as the afterdischarge (AD) threshold in the secondary site are unaltered. It has been reported that prekindling of some brainstem sites results in transfer to forebrain sites, associated with development of classic AD in the brainstem. In our previous studies of brainstem kindling, we observed low amplitude high frequency tightly organized AD, not classic AD. Therefore, we examined the effects of prekindling the brainstem on secondary-site kindling of forebrain sites, to determine whether transfer requires triggering of classic AD in the primary site. Methods: Male Long-Evans rats carried bipolar electrodes aimed at the nucleus reticularis pontis oralis (RPO), mesencephalic reticular formation (MRF), dorsal periaqueductal grey (dPAG), or ventrolateral periaqueductal grey (vlPAG), with a second electrode aimed at the amygdala. AD thresholds were determined at the brainstem electrode, and the rats received 1-sec trains of balanced square wave stimulation once daily for 30 days. Seven weeks after brainstem kindling the AD threshold of the amygdala was determined. Rats sham-kindled during the 30 days of brainstem stimulation served as the control group. The experimental and control groups were stimulated in the amygdala at AD threshold twice daily, until they had five Stage 5 seizures or until they received 50 stimulations. Twenty-four hr after the last kindling session the AD thresholds were again measured. Results: Prekindling of the RPO, vlPAG and MRF failed to produce transfer to the forebrain, in that kindling at the amygdala proceeded at the rate seen in naive control rats. However, prekindling of the dPAG produced positive transfer to amygdaloid kindling. Unexpectedly, all rats that were prekindled in brainstem sites failed to display a decrease in amygdaloid AD threshold when subsequently subjected to secondary-site kindling of the amygdala, in contrast to the significant decrease in amygdaloid AD threshold displayed by control rats. Conclusions: Our results suggest that prestimulation of brainstem sites results in inhibitory effects on forebrain kindling. The effects are not due to suppression of forebrain-triggered discharge and propagation to distal sites, but reflect inhibition of amygdaloid sensitization to the kindling stimulus. Because brainstem kindling involves low-voltage tightly organized AD, the triggering of classic AD in the brainstem during prekindling is not required for transfer to occur (i.e., dPAG group). Whether primary forebrain kindling affects subsequent brainstem kindling remains to be determined. (Supported by Grants from NSERC and CIHR to MEC.) Abstract: .sup.1 John G. Lamb , .sup.1 Misty D. Smith-Yockman , .sup.1 Jonathan Constance , .sup.1 Robyn Poerschke , .sup.1 Michael R. Franklin , and .sup.1 Steve H. White (Pharmacology and Toxicology, University of Utah, Salt Lake City, UT) Rationale: Polychlorinated biphenyls (PCBs) are man-made ubiquitous environmental contaminants, found as mixtures of more than 200 individual compounds. PCBs are known to cause the induction of drug metabolizing and multidrug transporter enzymes including multidrug resistant 1 (MDR1). We examined the effect of PCB exposure on the expression of MDR1a and MDR1b mRNAs in the rat liver and kidney. The effect of PCB exposure on the effectiveness of phenytoin in the maximal electroshock (MES) test was also examined. Methods: Animals were treated with either corn oil or PCBs (25 mg/kg, i.p.). After 48 hours, groups of animals were dosed with phenytoin (60 mg/kg, p.o.) and subjected to the MES test (bilateral corneal electrical stimulation 150 mA, 0.3 sec.) after two hours. Additional groups of PCB treated animals were pre-treated with clotrimazole (75 mg/kg, i.p.) 2.5 hours prior to phenytoin treatment (to inhibit cytochrome P450 activity) and subjected to the MES test, two hours after phenytoin treatment. The serum level of phenytoin was determined by HPLC analysis. Total RNA was isolated from liver and kidney tissue and the level of MDR1a, MDR1b mRNAs determined by real-time PCR analysis. Results: Significant elevations of MDR1a and MDR1b mRNAs were detected in the liver of PCB treated animals, while only MDR1a mRNA was elevated in the kidney. In the MES test, 7/8 animals were protected by phenytoin in the corn oil control group, while only 2/8 animals were protected in the PCB treated group. HPLC analysis indicated a significant decrease in the level of phenytoin in the serum of the PCB treated animals. Clotrimazole treatment did not affect the number of animals protected against MES by phenytoin (7/8 protected). In PCB treated animals clotrimazole did not fully restore the protection by phenytoin (only 5/8 protected). Conclusions: The results indicate that exposure to PCBs causes an increase in the expression of MDR1a mRNA in the liver and kidney, and MDR1b in the liver. PCB exposure also causes a significant decrease in the effectiveness of phenytoin in the MES test. Despite inhibition of cytochrome P450 activity, PCB treatment caused a decrease in the number of animals protected from MES by phenytoin, suggesting a potential role for transporters in the PCB effect. Environmental contaminants may contribute to therapy resistance in epilepsy by multiple mechanisms. (Supported by NO1-NS-4-2359.) Abstract: .sup.1 Chong L. Lee , .sup.2 James D. Frost Jr. , .sup.1 John W. Swann , and 2,3 Richard A. Hrachovy (Department of Pediatrics, Cain Foundation Labs., Baylor College of Medicine, Houston, TX ; Department of Neurology, Baylor College of Medicine, Houston, TX ; and Neurology Care Line, Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX) Rationale: Neuronal activity plays a crucial role in the proper development and maturation of the immature brain. The present project investigates the consequences of activity blockade during critical periods of brain development. Methods: Postnatal day 10 (P10) rats were implanted with a subcutaneous Alzet pump (2 or 4 weeks capacity) filled with 10 [mu]M tetrodotoxin (TTX). The infusion cannula was implanted either in the cortex or dorsal hippocampus. At the end of the infusion period (at P25 or P39) the pump was removed and animals were implanted with a series of cortical and/or hippocampal silver EEG electrodes. On the following day EEG/video recordings of 2 - 4 hours duration were initiated. Monitoring was repeated at approximately daily intervals for at least 1 week. All surgical procedures and animal care have been approved by Baylor College of Medicine. Results: To date 35 TTX treated (17 cortical -1 two weeks and 16 four weeks infusion and 18 hippocampal - 6 two weeks and 12 four weeks) and 5 vehicle infused (hippocampal - 4 weeks) animals have been implanted and monitored. All TTX treated animals exhibited high voltage multifocal interictal spiking. A subset of 4 weeks infused animals from each infusion area (3 of 16 cortical and 3 of 12 hippocampal) exhibited sporadic EEG complexes consisting of a generalized slow wave transient, a generalized voltage attenuation (2-4 seconds) or a burst of fast activity (2-4 seconds) .These complexes could occur alone or in various combinations, the most common being a high voltage slow wave complex followed by a generalized voltage attenuation with superimposed fast activity. These complexes could occur in isolation or in clusters and typically occurred on arousal from sleep. These EEG complexes were typically associated behaviorally with a brief tonic spasm, which could be preceded by a body jerk. Some of these animals also exhibited a very high voltage background EEG pattern associated with multifocal sharp wave discharges that resembled the hypsarrhythmic pattern observed in human children with infantile spasms. No seizure activity was observed in 2 week infused animals or in any of the vehicle infused animals. Conclusions: The results suggest that prolonged blockade of neuronal activity during critical periods in early life can induce seizures. Some of these animals exhibited EEG and behavioral phenotypes that are very similar to those found in human cases of infantile spasms. (Supported by NIH (NINDS) Grants: NS 18309 and NS37171.) Abstract: .sup.1 Linda H. Lee , .sup.1 Isabel Parada , and .sup.1 David A. Prince (Neurology and Neurological Sciences, Stanford University School of Medicine, Stanford, CA) Rationale: Na.sup.+-K.sup.+ ATPase (the "sodium pump) plays an important role in maintaining ionic gradients in neurons and is expressed in axons where it influences the generation of both excitatory and inhibitory synaptic events. Blockade of the pump leads to acute epileptiform activity due at least in part to decreases in GABA release. Pump activity is known to be decreased by brain trauma. We therefore assessed anatomical alterations in the pump in a rodent model of chronic posttraumatic epileptogenesis. Methods: Chronic partial isolations of sensori-motor rat cortex were made as previously described. At 7-60d after injury we quantitatively assessed the expression of the neuronal catalytic alpha 3 subunit of Na.sup.+-K.sup.+ ATPase using standard immunocytochemical techniques, fluorescent and confocal microscopy and image analysis software. Results: Alpha 3 immunoreactivity (IR) was normally prominent in the neuropil and surrounding somata of pyramidal (P) cells in all lamina. The perisomatic IR was colocalized with GAD65 IR and parvalbumin (PV) in presumed inhibitory terminals of fast-spiking (FS) interneurons. Perisomatic alpha 3 Na.sup.+-K.sup.+ ATPase IR was reduced significantly in partially isolated cortex, without obvious changes in GAD65- or PV-IR. Measurements showed that the interhemispheric differences in alpha3 subunit IR around layer V Pcell somata in partially isolated vrs contralateral cortex were10x greater than interhemispheric differences in controls (p < < 0.05). Such abnormalities were seen 7d following injury, the earliest time point examined (before onset of epileptogenesis in this model), persisted at least 2 mo., and were more marked in 7d vrs 15-30d survivors, but still significant at 15-30d. Conclusions: In the chronically injured, epileptogenic cortex, axons of FS cells, which provide a major component of somatic GABAergic postsynaptic inhibitory innervation in P cells, and the large autaptic input to FS interneurons, contain less Na.sup.+-K.sup.+ ATPase. Reductions in alpha 3 Na.sup.+-K.sup.+ ATPase might affect GABA release from inhibitory terminals through depolarization due to increases in both [Na]i and local [K+]o that would be prominent with generation of high frequency discharges such as occur in FS cells normally and during epileptiform activity. (Supported by NIH grants NS12151 and NS39579 from the NINDS.) Abstract: .sup.1 Huifang Li , and .sup.1 Robert S. Fisher (Department of Neurology, Stanford University School of Medicine, Stanford, CA) Rationale: The incidence of absence epilepsy is higher in females, however, the underlying reasons for this gender asymmetry are unknown. We reported previously that the expression of GABA.sub.A receptor subunit [gamma]2 protein levels were decreased in thalamic reticular (nRt) and ventrobasal (VB) nuclei in a group of adult rats (both male and female) after perinatal treatment with a cholesterol synthesis inhibitor (CSI) AY9944 (Li et al., 2006). We performed studies to evaluate the expression of GABA.sub.A receptor [alpha]1 and [gamma]2 subunits in thalamic nRt-VB nuclei and somatosensory cortex from male and female rats at ages p35-60 days. Methods: Experiments were carried out according to protocols approved by the Stanford Institutional Animal Care and Use Committee. Long-Evans rats were treated with AY9944 (15 mg/kg each) at p1, p5 and p9. Thalamic nRt-VB region, and somatosensory cortex were harvested at p35 and p60. Real-time PCR quantification and Western blot techniques, normalized to control group levels were performed to measure the expression of mRNA and protein levels for [alpha]1 and [gamma]2 subunits. Statistical analysis used t-tests with validation on data normalized by log transformations. Results: Real-time PCR and Western blot results showed a gender differential effect of perinatal CSI treatment on expression of GABA.sub.A receptor [alpha]1 and [gamma]2 subunit mRNA and protein levels in thalamus. At p35, [gamma]2 subunit mRNA levels were decreased in female (control, 1.00 [+ or -] 0.04 vs. CSI model, 0.77 [+ or -] 0.07, p < 0.02) but not in male thalamus. Corresponding to the decrease of mRNA expression, [gamma]2 protein levels also were decreased in female rats previously given AY9944 (control, 1.00 [+ or -] 0.07 vs. CSI model, 0.82 [+ or -] 0.04, p < 0.02). At p60, no significant difference was detected in the expression of [alpha]1 or [gamma]2 mRNAs. However, reductions of expression of [alpha]1 subunit protein levels were seen in female thalamus (control 1.00 [+ or -] 0.04 vs. CSI model 0.79 [+ or -] 0.04, p < 0.02), but not in male rats. At P60, expression of GABA.sub.A receptor [gamma]2 subunit protein also was decreased in the female, but not in the male, thalamus (control 1.00 [+ or -] 0.06 vs. CSI model 0.71 [+ or -] 0.05, p < 0.01). In contrast, the expression of [gamma]2 subunit protein in CSI model somatosensory cortex was increased in both female and male cortex at age p60, compared to age-matched control somatosensory cortex (female control 1.00 [+ or -] 0.06 vs. CSI model 1.15 [+ or -] 0.06, p < 0.03; male control 1.00 [+ or -] 0.02 vs. CSI model 1.21 [+ or -] 0.06, p < 0.02). Conclusions: Perinatal inhibition of cholesterol synthesis differentially alters GABA.sub.A receptor subunit expression in male and female rats, demonstrating a decrease of [gamma]2 subunits in female, but not male, thalamic nRt-VB nuclei. Such changes may partially explain the higher incidence of clinical absence epilepsy in female patients. (Supported by Maslah Saul MD Chair and the James and Carrie Anderson Fund for Epilepsy Research (RSF). Susan Horngren Fund and the Epilepsy Foundation (HL). Wyeth-Ayerst for supplying the AY9944 drug.) Abstract: 1,2 Jerome Niquet , 1,2 Maria-Leonor Lopez-Meraz , 1,2,3 Claude Wasterlain , and 1,2 Lucie Suchomelova (Epilepsy Research Laboratory (151), Veterans Affairs Greater Los Angeles Healthcare System, Los Angeles, CA ; Department of Neurology, David Geffen School of Medicine at UCLA, Los Angeles, CA ; and Brain Research Institute, David Geffen School of Medicine at UCLA, Los Angeles, CA) Rationale: The contribution of hyperthermia per se to status epilepticus (SE)-induced neuronal damage in the immature brain is not well understood. In adult animals, hyperthermia can aggravate seizure-induced neuronal damage (Liu et al., 1993; Lundgren et al., 1994). However, published reports did not show any additional damage caused by hyperthermic when compared to normothermic seizures in immature brain (Bender et al., 2003; Sarkisian et al., 1999; Toth et al., 1998). To study this problem, we developed a model of lithium-pilocarpine SE in rats at post-natal day 10 (P10), in which we were able to obtain seizures of very similar intensity at a body temperature of 35[degrees]C (normothermia) vs 39[degrees]C (hyperthermia). Methods: SE was induced in P10 rats with lithium (3mEq/kg, i.p.), and pilocarpine (60 mg/kg, s.c., 24 hours later). Their body temperature was maintained during 30 min of SE at 39 [+ or -] 1[degrees]C (hyperthermic animals, HT+SE) or at 35 [+ or -] 1[degrees]C (normothermic animals, NT+SE). Several measures showed seizure severity to be similar in both experimental groups. SE was terminated by diazepam (0.5 mg/kg, i.p.). Control animals received lithium and diazepam in presence of hyperthermia or normothermia. Anesthetized animals were perfused with 4% phosphate-buffered paraformaldehyde 24 hours after induction of SE. Each brain was embedded in paraffin and sectioned to 10 uM. The distribution of neuronal injury was evaluated by using Fluoro-jade B staining, whereas the expression of death markers (the active forms of initiator caspase-9 and executioner caspase-3) was assessed by fluorescence immunohistochemistry in adjacent sections. Mode of death (necrotic or apoptotic) from HT+SE and NT+SE animals was defined by electron microscopy. Results: Fluoro-jade B staining showed neuronal injury in CA1, medial amygdala, dentate gyrus inner granule cell layer and cingulate cortex from HT+SE animals. NT+SE animals displayed very mild damage in CA1 and granule cell layer, none in amygdala. No damage was detected in controls. Electron microscopy showed a mixed population of apoptotic and necrotic neurons in CA1 from HT+SE and NT+SE groups. Active caspase-3 immunoreactive neurons were seen in CA1, granule cell layer and cingulate cortex from HT+SE animals, but active caspase-9 staining was limited to granule cell layer and cingulate cortex. Caspase-3 or -9 were not activated in NT+SE animals. Conclusions: Hyperthermia markedly aggravated the severity and the extent of SE-induced neuronal injury through a caspase-3 and/or -9-dependent mechanism. (Supported by Grant NS13515 from NINDS, National Institutes of Health, by UC Mexus grant, and by the Research Service of the Veterans Health Administration.) Abstract: .sup.1 Hongtao Ma , .sup.1 Mingrui Zhao , .sup.1 Minah Suh , .sup.1 Challon Perry , and .sup.1 Theodore H. Schwartz (Neurological Surgery, Weill-Cornell Medical College, Presbyterian Hospital, New York, NY) Rationale: The hemodynamic changes associated with neuronal activity are the basis for several non-invasive imaging methods, such as fMRI, PET and SPECT. However, the precise spatiotemporal relationship between perfusion, oxygenation and the neuronal dynamic is not well understood since there is no technique for simultaneously measuring neuronal activity and perfusion in a large area of cortex with high spatial and temporal resolution. Methods: Currently, optical recording of voltage sensitive dyes (VSD) offers the best spatial ([proportional to]50 [mu]m) and temporal (1ms) resolution for recording neuronal activity from a large area of cortex (several millimeters). Intrinsic optical signal (IOS) imaging offers the best spatial ([proportional to]50 [mu]m) and temporal (tens of milliseconds) resolution for recording hemodynamic changes. We developed a technique combining VSD and IOS imaging to simultaneously measure membrane potential, perfusion and hemoglobin oxygenation in vivo. Results: Two illumination sources were employed for VSD (640 [+ or -] 10nm, for new blue dye RH-1692) and IOS (570 [+ or -] 5 or 605 [+ or -] 5) respectively. A 620 nm dichroic mirror was used to separate the VSD (>665 nm) and IOS signals (570 or 605 nm). VSD and IOS responses were simultaneously recorded with two cameras. We compared the IOS result from cortex stained with VSD and IOS result from unstained cortex and found that staining with RH-1692 has no significant effect on the hemodynamic responses. We successfully used this technique in the rat preparation to record interictal events induced with intracortical injection of bicuculline. Conclusions: Simultaneous VSD and IOS recording is a technically feasible method for studying the complex interactions between excitatory and inhibitory neuronal activity and both increases and decreases in perfusion and oxygenation. (Supported by NIH NS043799-04, NS049482-01 and the Epilepsy Foundation through the generous support of the American Epilepsy Society.) Abstract: .sup.1 Gauri H. Malthankar-Phatak , .sup.1 Daniel P. McCloskey , .sup.1 Tana M. Hintz , .sup.2 Neil J. MacLusky , and 1,3 Helen E. Scharfman (CNRRR, Helen Hayes Hospital, W. Haverstraw, NY ; Biomed. Sci., University of Guelph, Guleph, ON, Canada ; and Pharmacol. & Neurol., Columbia University, NY, NY) Rationale: It is known that testosterone influences seizures and epilepsy, but it is not clear whether there are differential effects on seizure threshold, epileptogenesis, or the chronic condition. Using the pilocarpine model of epilepsy, we tested the hypothesis that testosterone influences the initial precipitating event (status epilepticus), consequences of status (damage, sprouting), and the state of spontaneous recurrent seizures (the chronic period). Orchidectomy was used to reduce serum testosterone, and sham surgery was used as a control condition. Methods: Orchidectomized or sham adult male Sprague-Dawley rats were administered atropine methylbromide (1 mg/kg, s.c.) and pilocarpine (380 mg/kg, i.p.) 2 wks after surgery, and diazepam (5 mg/kg., i.p.) was injected 1 hr after the onset of status. Spontaneous seizures were assessed by behavioral observation. To evaluate hyperexcitability in vitro, extracellular recordings were made in hippocampal slices to detect spontaneous population bursts in area CA3, which occur in intact male rats in the chronic period. After recordings, slices were immersion-fixed, resectioned, and used for immunocytochemistry to detect neuronal loss and mossy fiber sprouting using antibodies to the neuronal marker NeuN, and neuropeptide Y, respectively. Results: There was no difference in the incidence of status in orchidectomized (59%; n = 19/32) vs. sham controls (66%, n = 10/15; p = 0.63). The incidence of CA3 bursts in orchidectomized rats was significantly lower (32%, n = 8/25, n = 4 rats), when tested 12-14 weeks after status, relative to sham and intact rats (pooled; 86%, n = 36/42, n = 7 rats; p = 0.0001). This was unlikely to have been due to differential neuronal damage or sprouting, because they appeared to be similar throughout the septotemporal axis in the two groups. If orchidectomy was conducted after the chronic period was established, it did not stop spontaneous seizures, or have any effect on the incidence of CA3 bursts in slices from these rats (orch., 92%; sham, 100%; p = 1.0). Conclusions: Orchidectomy 2 wks prior to status had no apparent effect on the incidence of status, but it did appear to reduce the incidence of CA3 bursts in slices from rats that were examined during the chronic period in the pilocarpine model. However, orchidectomy after the onset of the chronic period did not influence CA3 bursts. The results suggest that orchidectomy, and most likely testosterone, has a major effect on the development, but not the maintenance, of area CA3 network bursts in the pilocarpine model. This effect would be consistent with recent studies illustrating the influence of testosterone on synaptic remodeling in CA3, and the evidence that synaptic reorganization plays a role in epileptogenesis. (Supported by NINDS 37652, Helen Hayes Hospital Foundation, New York State Deptartment of Health.) Abstract: .sup.1 John G. Mantis , .sup.1 Nicholas C. Zimick , .sup.1 Richard McGowan , and .sup.1 Thomas N. Seyfried (Biology, Boston College, Chestnut Hill, MA) Rationale: Glucose uptake into the brain is greater during epileptic seizures than during most other brain activities suggesting a key role for glucose in the initiation and spread of seizure activity. Under conditions of fasting or caloric restriction (CR), however, brain cells can also derive energy from ketone bodies (acetoacetate and beta-hydroxybutyrate, [beta]-OHB). The high fat, low carbohydrate ketogenic diet (KD) was developed as an alternative to fasting for seizure management. We previously showed that CR and a restricted lard-based KD inhibited seizure susceptibility in epileptic EL mice. The EL mouse is a model for age-dependent multifactorial idiopathic generalized epilepsy (IGE) and expresses complex partial seizures with secondary generalization. The efficacy of the KD in managing epileptic seizures is dependent on the restriction of caloric intake that lowers glucose thereby allowing the brain to metabolize ketones for energy. KetoCal[R] (KC) is a commercially prepared soy-oil-based KD for children with a fat to protein carbohydrate ratio of 4:1 and is more palatable than lard-based or medium chain triglyceride (MCT) formulations. In this study we evaluated the antiepileptic and anticonvulsant efficacy of KC in young adult EL mice. Methods: EL mice (about 70 days of age) that experienced at least 2 complex generalized seizures were separated in three groups (n = 8 mice/group) and were fed either a standard high carbohydrate (chow) diet unrestricted (SD-UR), KC unrestricted (KC-UR), or KC restricted (KC-R) to reduce body weight by approximately 20%. Seizure susceptibility, body weights, and food intake were measured once/week over a nine-week treatment period. The handling-induced seizure test was used to evaluate seizure susceptibility in each mouse group. Plasma glucose and [beta]-OHB levels were measured once every three weeks, using enzymatic analyses. Results: Body weights, plasma glucose levels, and seizure susceptibility were significantly lower in the KC-R group over the nine-week testing period than in the SD-UR and KC-UR groups. Plasma [beta]-OHB levels were significantly higher in the KC-R group (4mM) than in the SD-UR group (0.45 mM) or the KC-UR group (1.1 mM) after week three of the study. Seizure susceptibility was lower (p < 0.05) in the KC-UR group than in the SD-UR group only at the last week of the dietary treatment despite similarities in body weight and glucose levels. Conclusions: KetoCal[R] is as effective as the lard-based KD in managing IGE in EL mice. No adverse side effects were detected in EL mice receiving KetoCal[R]. Anticonvulsant efficacy is best when KetoCal[R] is administered in restricted amounts to lower blood glucose and to elevate blood ketone bodies. We predict that KetoCal[R] will be effective in the clinic and can be used as a standard ketogenic diet therapy for both human and animal studies. (Supported by the Epilepsy Foundation, and NIH (HD39722), and the Boston College Research Expense Fund.) Abstract: .sup.1 Anatoly E. Martynyuk , .sup.2 Paul R. Carney , .sup.1 Donn M. Dennis , and .sup.3 Philip J. Laipis (Anesthesiology, University of Florida, Gainesville, FL ; Pediatrics, University of Florida, Gainesville, FL ; and Biochemistry and Molecular Biology, University of Florida, Gainesville, FL) Rationale: The mechanisms underlying epileptic activity in untreated phenylketonuria (PKU) patients are not well understood. Based on our previous findings that phenylalanine (Phe) at concentrations found in PKU mouse brain significantly and reversibly depresses glutamatergic synaptic transmission leading to a compensatory increase in the expression and density of glutamate receptors, we hypothesized that seizures in PKU mice might be observed when Phe levels are decreased and glutamatergic activity becomes facilitated. Methods: This hypothesis was tested by examining the susceptibility of PKU mice (female homozygous Pah.sup.enu2 BTBR) to audiogenic seizures (AGS) induced by sound stimulus of 118 dB (10-20 kHz). Results: Serum Phe concentrations in PKU mice were more than 10 fold higher than in heterozygous (HET) carriers or wild-type (WT) BTBR mice. Unlike HET carriers or WT mice, adult PKU mice, 8 weeks of age or older, were susceptible to AGS and exhibited dramatic responses with symptoms ranging from uncontrolled running to clonic and tonic convulsions, respiratory arrest, and death. Death could be prevented by pretreatment with oxygen-enriched air. Serum Phe levels in PKU mice could vary significantly with time of day, with an average decrease of 0.53 mM from 9 am to 5 pm. Depending on Phe levels, the same mouse could be either susceptible or resistant to AGS. In general, while adult mice with Phe levels of less than 2 mM routinely exhibited AGS, mice with Phe levels of greater than 2 mM were resistant to the stimulus. Young PKU mice consistently had significantly higher serum Phe levels and were not susceptible to AGS. Serum Phe levels in adult PKU mice could be lowered to normal, or increased to hyperphenylalaninemic concentrations within 12 hours, by switching adult PKU mice from a high to low Phe diet, or vice versa. Susceptibility to AGS disappeared after 12 hours on the low Phe diet. In contrast, more than 7 weeks on a standard diet were necessary before PKU mice, previously maintained for two weeks on a low Phe diet, regained susceptibility to AGS. Conclusions: Complex dependence of AGS on Phe levels indicates that the Phe-induced changes in glutamatergic synaptic transmission may play an important etiological role in PKU seizures. If seizure susceptibility is similarly reversible in human PKU patients, then they may face a risk of becoming susceptible to seizures after relaxing their Phe-restricted diet, perhaps especially with increasing age. Given the known mutation and the well defined biochemical background, the Pah.sup.enu2 mouse may become a new valuable model for investigating the mechanisms underlying epileptic activity in general. (Supported by NIH DK058327 (PJL), the University of Florida McKnight Brain Institute, and I. Heermann Anesthesia Foundation, Inc.) Abstract: .sup.1 Alain Matagne , .sup.1 Doru-Georg Margineanu , .sup.1 Karine Leclercq , .sup.1 Donald Dassesse , .sup.1 Nathalie Lambeng , and .sup.1 Henrik Klitgaard (CNS Research, UCB S.A., Braine-l'Alleud, Belgium) Rationale: SV2A is the most abundant isoform of SV2, a membrane glycoprotein common to all synaptic vesicles. This protein was recently reported (Lynch et al., PNAS 101:9861-66, 2004) to be the binding site of the newer generation antiepileptic drug, levetiracetam (Keppra.sup.[R]). The purpose of this study was to characterize the potential epilepsy-phenotype of mice lacking SV2A. Methods: SV2A KO mice were generated by using targeted gene disruption in S. Bajjalieh's laboratories at the University of Washington, Seattle (Crowder et al., PNAS 96:15268-73, 1999). Homozygous SV2A KO mice display severe seizures and are not viable beyond 2 weeks, thus precluding phenotyping in adults. In contrast, heterozygous SV2A KO mice (SV2A [+ or -]) are viable. In this study, we compared SV2A [+ or -] mice to wild type (WT) controls with respect to acute seizure susceptibility in response to both electrical and chemical stimulation. Evoked field potentials were recorded in the CA3, CA1 and DG areas of hippocampal slices in vitro. Different groups of animals, aged 2 months, were used for each experiment. Results: The electrical seizure threshold intensity inducing tonic hind limb extension after corneal electrode stimulation was comparable between SV2A [+ or -] and WT mice (n = 19-16/group). Similarly, there was no difference between SV2A [+ or -] and WT mice in the threshold dose inducing generalized clonic seizures after continuous intravenous infusion of pentylenetetrazole (n = 25/group). In contrast, the (stimulus intensity) - (population spike) curves of the field potentials recorded from hippocampal slices showed a higher excitability in the CA3 area in slices from SV2A [+ or -] mice compared to WT controls. Likewise, kindling acquisition induced by twice daily corneal electrode stimulations (2 mA for 3 s) was markedly accelerated in SV2A [+ or -] mice compared to WT (n = 10/group). The induction of secondary generalized motor seizures in all SV2A [+ or -] mice was reached after up to 12 repeated stimulations whereas up to 22 stimulations were necessary to reach this end point in WT mice. Conclusions: The results of this study suggest a key role of the SV2A protein in the establishment of a chronic epileptic condition, consonant with the seizure phenotype observed in SV2A homozygous mice. This supports that SV2A is the principal mechanism for the antiepileptic action of levetiracetam. (Supported by UCB S.A.) Abstract: .sup.1 Hideyuki Matsukubo , .sup.1 Koichi Akaike , .sup.2 Shin-Ichi Imamura , .sup.1 Motofumi Kasugai , .sup.1 Hideshi Tojo , .sup.3 Shigeya Tanaka , and .sup.1 Akira Sano (Department of Psychiatry, Field of Social and Behavioral Medicine, Kagoshima University Graduate School of Medical and Dental, Kagoshima, Kagoshima, Japan ; Department of Neurosurgery, Kagoshima University Graduate School of Medical and Dental, Kagoshima, Kagoshima, Japan ; and Neurosurgery, Tanaka Neurosurgical Clinic, Kagoshima, Kagoshima, Japan) Rationale: Previously, we revealed that FVB/N showed more sensitive susceptibility to seizure, by electroencephalographic and behavioral observation, compared to C57BL/6 in the kainic acid (KA)-induced amygdalar seizures. Moreover, neuronal cell death in the hippocampal CA3 region was significantly severer in FVB/N than in C57BL/6. In the present study, we investigated the serial changes of GABA and Glutamate concentration in the left hippocampus after KA microinjection into the amygdala in mice with different strains, C57BL/6 and FVB/N. Methods: Forty-eight male adult mice with two strains (n = 24, respectively) underwent a stereotactic operation using the units for stereotactic coordinates under pentobarbital anesthesia (10 mg/kg i.p.). For microinjection, a stainless steel cannula (outer diameter, 0.6 mm) with an inner needle guide (diameter, 0.3 mm) was inserted into the left amygdala (LA). On the seventh days after surgery, all animals received 0.5 microg of KA into the LA. Two hours (n = 12, respectively) or 48 hours (n = 12, respectively) after KA injection, animals were decapitated and the ipsilateral hippocampus were removed. Intrahippocampal GABA and glutamate were quantitatively measured using high performance liquid chromatography (HPLC). Data were analyzed statistically between C57BL/6 and FVB/N by using Student's t-test. In controls (n = 12, respectively), PBS was injected into the LA. They likewise were quantitatively measured using HPLC. Results: There was no significant difference concerning Glutamate and GABA concentration between two strains 2 hours after KA injection. Forty-eight hours after KA injection, however, GABA concentration of the C57BL/6 was significantly higher than that of the FVB/N, while there was no significant difference of glutamate concentration between two strains. In control, there was no significant difference between two strains concerning both GABA and Glutamate concentration 2 and 48 hours after PBS injection. Conclusions: The less seizure susceptibilities of the C57BL/6 may be associated with the intrahippocampal higher GABA concentration than that of FVB/N. This result will contribute to reveal the molecular mechanism of both strains, FVB/N and C57BL/6. (Supported by Grant-in-Aid for Encouragement of Young Scientists from the Ministry of Education, Sciences, and Culture of Japan.) Abstract: .sup.1 Andrey M. Mazarati , .sup.2 Linda Lundstrom , .sup.1 Don Shin , .sup.2 Ulo Langel , and .sup.1 Raman Sankar (Pediatrics-Neuro, David School of Medicine at UCLA, Los Angeles, CA ; and Neurochemistry, Stockholm University, Stockholm, Sweden) Rationale: Galanin is a neuropeptide with well established anticonvulsant effects in acute seizure models. No data are available onto whether galanin might inhibit epileptogenesis. In the hippocampus, galanin acts at two G protein-coupled receptor types - GalR1 and GalR2. We examined antiepileptogenic potential of selective GalR1 and GalR2 agonist to in kindling model, and examined putative downstream pathway that mediate anticonvulsant effects of galanin. Methods: Adult male Wistar rats received 24 hour long intrahippocampal infusion of a non - selective GalR1/GalR2 agonist galanin (1-29), selective GalR1 agonist synthetic peptide M617, or selective GalR2 agonist galanin (2-11). The peptides were administered alone, or combined with an inhibitor of G.sub.i/o protein pertussis toxin (PTX); blocker of G-protein coupled inwardly rectifying K.sup.+ channels (GIRK) - tertiapin Q (TPQ); G.sub.q/11 protein inhibitor [D-Arg.sup.1,D-Trp.sup.5,7,9,Leu.sup.11]-substance P (dSP); or an inhibitor of intracellular Ca.sup.2+ release dantrolene. Sixteen hours into drug delivery, the animals were subjected to rapid kindling - sixty electrical trains administered to ventral hippocampus every 5 minutes. Hippocampal afterdischarge properties before and after treatment, as well as before and after kindling, and the number of stimulations required to reach each consecutive behavioral seizure score (1 through 4) were analyzed. Results: Activation of hippocampal GalR1 delayed, but did not prevent kindling progression. Twenty four hours after kindling, M617 - treated animals exhibited the decrease of afterdischarge threshold and developed stage 4-5 seizures in response to threshold stimulation, similar to control. GalR1/GalR2 and GalR2 agonists completely prevented both the development of full motor seizures in the course of kindling, and the decrease of afterdischarge threshold 24 hours after kindling procedure. Anticonvulsant effect of GalR1 was not observed in the presence of G.sub.i/o and GIRK inhibitors. Antiepileptogenic action of GalR2 agonist was PTX- sensitive, but GIRK - independent. Furthermore, inhibition of G.sub.i/o by PTX changed antiepileptogenic effects of galanin (1-29) and of galanin (2-11) to proconvulsant. Inhibition of G.sub.q/11 and of intracellular Ca.sup.++ release eliminated proconvulsant effect of PTX+galanin (1-29) and PTX+galanin (2-11) combinations. Conclusions: We conclude that hippocampal GalR1 exert disease - modifying effect through G.sub.i-GIRK pathway. GalR2 is antiepileptogenic through G.sub.i mechanism, which does not involve GIRK. Our studies also revealed a secondary proconvulsant pathway coupled to GalR2, which involves G.sub.q/11 and intracellular Ca.sup.2+. The data are important for understanding endogenous mechanisms regulating epileptogenesis, and for the development of novel antiepileptogenic drugs. (Supported by NIH/NINDS grants NS043409 (AM) and NS046516 (RS).) Abstract: .sup.1 Guy M. McKhann II , .sup.1 Xiaoping Wu , .sup.2 Howard L. Weiner , .sup.3 Peter D. Crino , .sup.1 Robert R. Goodman , and .sup.1 Alexander A. Sosunov (Neurological Surgery, Columbia University, New York, NY ; Neurology, New York University Medical Center, New York, NY ; and Neurology, University of Pennsylvania, Philadelphia, PA) Rationale: Prominent astrogliosis is a typical feature of cortical tubers in Tuberous Sclerosis Complex (TSC). The relationship of these changes to mTOR activation, and the characteristics of tuberous astrogliosis in comparison to astrocytic changes in other human epilepsies have not been investigated. Methods: Neocortical and hippocampal specimens obtained during surgery on pharmaco-resistent MTLE (46 cases) and tuberous sclerosis complex (12 cases) were studied with immunohistochemistry. Results: Most tuber astrocytes reveal high similarity to fibrous-like astrocytes seen in areas of sclerosis in MTLE. These astrocytes differ drastically from protoplasmic astrocytes found in cortex surrounding tubers in TSC and in control human neorcortex or lateral neocortex in cases of human MTLE. Morphological and immunohistochemical analysis of sclerotic and tuber astrocytes revealed that they are both characterized by increased GFAP expression and length of cell processes, accompanied by decreased expression of glutamine synthetase and astrocyte-specific glutamate transport proteins. A subpopulation of astrocytes in tubers reveal upregulation of mTOR pathway, based on immunostaining for phospho-p70S6kinase, phospho-S6, phospho-STAT3, and phospho-4E-BP1. This mTOR activation is similar to that previously shown in giant cells (Baybis et al., 2004; Miyata et al., 2004). Acutely reactive astrocytes observed in surgically resected brain tissue from MLTE patients also revealed immunostaining for downstream components of the mTOR cascade. Conclusions: Two types of astrrocytes are found in human tubers: 1) mTOR activated cells similar to acutely reactive human astrocytes; 2) Sclerotic/gliotic astrocytes similar to human astrocytes in areas of sclerosis in MTLE. We hypothesize that activation of the mTOR pathway in tuber astrocytes due to TSC1 or TSC1 genes mutations leads to stepwise phenotypic transformation of astrocytes from reactive to sclerotic/gliotic cells. TSC may thus represent a genetic model of human astrogliosis found in other pathologies such as MTLE, stroke, and head trauma. The possible molecular mechanisms linking mTOR pathway activation, GFAP overproduction, and the appearance of fibrous-like astrocytes are being investigated. (Supported by Parents Against Childhood Epilepsy, NIH R21 NS 42334.) Abstract: .sup.1 Karen A. Menuz , .sup.1 Jessica L. O'Brien , .sup.1 Siavash Karimzadegan , .sup.1 David S. Bredt , and .sup.1 Roger A. Nicoll (Cellular and Molecular Pharmacology, University of California, San Francisco, San Francisco, CA) Rationale: Increased excitatory drive in the thalamocortical network is thought to underlie absence seizures. Mutations in channels that mediate neuronal inhibition have been found in patients with familial forms of the disorder. However, the mutation in the Stargazer mouse, an animal model for absence epilepsy, primarily affects excitatory glutamatergic transmission via AMPA receptors. To date, stargazin and its related family members, the Transmembrane AMPA receptor Regulatory Proteins (TARPs), have only been shown to regulate synaptic AMPA receptor function in excitatory neurons. However, if TARPs also regulate AMPA receptors in inhibitory GABAergic neurons, the lack of inhibition may lead to seizure activity. Therefore the aim of this study was to determine if TARPs regulate AMPA receptors in GABAergic neurons. Methods: Single and double knockout mice for TARP isoforms [gamma]2 (stargazin) and [gamma]3 were generated. Western blotting was used to analyze AMPA receptor content in different brain regions. Synaptic AMPA receptor currents were measured with standard electrophysiological techniques in acute brain slices and dissociated neurons. Results: Single [gamma]3-/- knockout mouse were indistinguishable from littermates. However, [gamma]2-/- [gamma]3-/- double knockout mice were severely impaired and generally died before the 2nd postnatal week. Western Blot analysis of [gamma]2-/- [gamma]3-/- double knockout mice found no change of AMPA receptor expression in most brain regions. Electrophysiological analysis of the GABAergic Purkinje neurons found a severe reduction of AMPA receptor mediated synaptic transmission in [gamma]2-/- mice, which was not increased in [gamma]2-/- [gamma]3-/- double knockout mice. A specific GABAergic interneuron population was found to be unaffected in single [gamma]2-/- and [gamma]3-/- knockout mice, but showed a nearly complete loss of synaptic AMPA receptors in [gamma]2-/- [gamma]3-/- double knockout mice. Conclusions: TARPs regulate AMPA receptors in GABAergic neurons, in addition to excitatory neurons. They can serve redundant roles as evidenced by the increasingly severe phenotype of double knockout mice. (Supported by Epilepsy Foundation; L'Oreal USA for Women in Science Program; NIMH.) Abstract: .sup.1 Gholam K. Motamedi , .sup.2 Michael A. Rogawski , and .sup.3 Stefano Vicini (Department of Neurology, Georgetown University, Washington, DC ; Department of Physiology & Biophysics, Georgetown University, Washington, DC ; and Epilepsy Research Section, National Institute of Neurological Disorders and Stroke, NIH, Bethesda, MD) Rationale: Cooling terminates extracellularly recorded spontaneous epileptiform discharges induced by chemoconvulsants in the hippocampal slice (G.K. Motamedi et al., Epilepsy Res., in press). Small temperature drops are effective when cooling is applied rapidly. Here we investigated the cellular mechanisms underlying the effects of rapid cooling on epileptiform activity in different regions of the hippocampus. Methods: Whole-cell patch clamp recordings were carried out from neurons in mouse hippocampal slices perfused with 50 mM 4-aminopyridine at an initial temperature of 32 [degrees]C. Rapid cooling was induced at a rate of 2-3 [degrees]C/s by switching to cold aCSF. Results: In CA1 and CA2 neurons, rapid cooling typically induced a 10-20 mV membrane depolarization associated with a large inward current, which was eliminated by 1 mM tetrodotoxin (TTX). In contrast, most CA3 neurons responded to cooling with TTX- and bicuculline (25 mM)-insensitive hyperpolarizations that were associated with a steady outward current and decreased conductance. Conclusions: Termination of epileptiform discharges in CA1 and CA2 neurons by rapid cooling may occur by synaptically-driven depolarization block. In CA3 neurons, cooling seems to inhibit a persistent depolarizing conductance, raising the possibility that closure of temperature sensitive TRP channels could mediate the suppression of discharges. (Supported by Grant # R01 NS047700 to S.V.) Abstract: .sup.1 Alberto E. Musto , and .sup.1 Nicolas G. Bazan (Neuroscience Center of Excellence, Louisiana State University-Health Sciences Center, New Orleans, LA) Rationale: Modifications of pyramidal and interneuronal network in neurological disorders are correlated by shape waves and rhythms from electroencephalogram (EEG). Unfortunately how this network participates in epileptogenesis is not yet clear. Methods: Kindling protocol is achieved by stimulating six times daily during 4 days with sub-convulsive electrical stimulations in the dorsal right hippocampus using bipolar electrode units. Spikes and behavioral responses are quantified in order to confirm the progress of kindling. The EEG obtained is recorded using Enhanced Graphics Acquisition for Analysis (RS Electronics Inc.) and offline sort spikes (Plexon Inc.), and analyzed using Neuroexplorer Software (Next Technology) (Musto and Bazan, 2006, Epilepsia, 47:267-76). Results: Spikes and semiological score increase during kindling. Different patterns are discriminated within afterdischarge. A spike-wave pattern appears immediately after the stimulation in combination with a poly-spike complex superimposed over the wave (AD1) and it is followed by another poly-spike complex (AD2). AD2 is followed by a progressive modification of synchronized spikes (AD3). The spreading depression like event (SD) is present at the end of each AD3. Theta oscillation predominates since the first two days of kindling and apparition of significative gamma oscillation is observed within AD2 and precedes the full kindled state. Conclusions: Pyramidal-interneuron network could work properly immediately after each stimulation and could be necessary, but not sufficient for epileptogenesis. The trigger of this oscillation could lead to a mechanism of desynchronisation and synchronism within neuronal network during epileptogenesis. Since a gamma rhythm precedes the generalized seizure, this could be a target for therapeutic intervention. (Supported by NIH Grant: NS 23002.) Abstract: .sup.1 Prosper N'Gouemo , .sup.2 Carl L. Faingold , and .sup.3 Martin Morad (Pediatrics, Georgetown University Medical Center, Washington, DC ; Pharmacology, Sourthern Illinois University School of Medicine, Springfield, IL ; and Pharmacology, Georgetown University Medical Center, Washington, DC) Rationale: The inferior colliculus (IC) neurons are thought to be critical in the initiation of audiogenic reflex seizures in the genetically epilepsy-prone rat (GEPR). We previously reported that high voltage-activated (HVA) Ca.sup.2+ channel current density was markedly enhanced in IC neurons of seizure-naive GEPR-3. Ca.sup.2+ currents are known to activate K.sup.+ currents that initiate the repolarization of action potential and hyperpolarize the membrane. Such Ca.sup.2+-activated K.sup.+ currents may represent intrinsic inhibitory mechanisms that would restore the resting state and maintain normal physiological excitability. Whether change in Ca.sup.2+-activated K.sup.+ current density parallel the upregulation of HVA Ca.sup.2+ channel current density reported in IC neurons of GEPR-3 is as yet unknown. Here we report on the expression of Ca.sup.2+-activated K.sup.+ currents in IC neurons and their modification in seizure-naive GEPR-3. Methods: Acutely dissociated neurons of the IC central nucleus were obtained from male seizure-naive GEPR-3s (12-15 week-old) and age-matched control Sprague-Dawley rats. The currents through K.sup.+ channels were recorded using the whole cell configuration of the patch clamp techniques. Ca.sup.2+-activated K.sup.+ currents were recorded in the presence of 2 mM extracellular Ca.sup.2+, while dialyzing the neurons with 100 nM free Ca.sup.2+ (buffered with 0.5 mM EGTA and Ca.sup.2+). Currents were normalized relative to the membrane capacitance as an estimate of current density. Results: Quantification of the current evoked at +20 mV (holding potential -50 mV) showed that the mean peak current density of total Ca.sup.2+-activated K.sup.+ current was decreased by [proportional to]40% in IC neurons of seizure-naive GEPR-3 (n = 3 cells, 3 rats) that expressed elevated HVA Ca.sup.2+ channel current density, as compared to controls (n = 8 cells, 3 rats) Quantification of the voltage dependence of total Ca.sup.2+-activated K.sup.+ current shows a downregulation of the current density at all voltages tested, on the average, by 60 [+ or -] 6% in IC neurons of the GEPR-3 (n = 3 cells, 3 rats), as compared to controls (n = 8 cells, 3 rats). We are currently examining the contribution of various fraction of the current carried by large and small conductance subtype of Ca.sup.2+-activated K.sup.+ channels. Conclusions: The data suggest that Ca.sup.2+-activated K.sup.+ current density is downregulated and parallel the enhancement of HVA Ca.sup.2+ channel current density in IC neurons of seizure-naive GEPR-3. The upregulation of HVA Ca.sup.2+ channel current density and downregulation of Ca.sup.2+-activated K.sup.+ current density may provide a cohesive mechanism for neuronal hyperexcitability that lead to seizures and mechanistic insights into therapeutic strategies for generalized epilepsy. (Supported by NIH grants NS05497 (P.N.), AA11628 (C.L.F) and HL16152 (M.M.).) Abstract: .sup.1 May Lissa Ollivier , .sup.1 Martine P. Emond , .sup.1 Pablo Lema , and .sup.1 Lionel Carmant (Research Center, CHU Sainte-Justine, Montreal, QC, Canada) Rationale: Abnormal neurogenesis secondary to status epilepticus (SE) is thought to be involved in epileptogenesis. Previous studies from our laboratory have shown that the use of the group I metabotropic glutamate receptor (mGluR) antagonist AIDA limits epileptogenesis and prevents SE-induced neurogenesis in the dentate gyrus and hilus. The goal of this study was therefore to explore the specific role of group I mGluR in the abnormal neurogenesis. In this study, group I mGluRs were activated by two agonists. The first agonist, (S)-3,5-dihydroxyphenylglycine (DHPG), activates both subtype 1 and 5 of the group I mGluRs and the second, 2-chloro-5-hydroxyphenylglycine (CHPG), selectively activates subtype 5. Methods: The agonists were given by intra-cerebro ventricular injection (ICV) in adult (P60) male Sprague-Dawley rats. The agonists were diluted in saline solution (S) and given at doses of 0.83 [mu]mol/10 [mu]l for DHPG and 3.33 [mu]mol/10 [mu]l for CHPG. Control rats received the same volume of S. To assess neurogenesis, rats of each experimental and control group were given two injections of 5-bromodeoxyuridine (BrdU, 50 mg/kg, intraperitoneal) within a two-hour interval and sacrificed 24 hrs after the second injection. This was done at three time-points after agonist injection (7, 14 and 56 days). Immunohistochemical treatment and cell counting techniques were then performed on the brain tissue. Results: No acute convulsion was observed in the different groups. Animals tolerated well their treatment. BrdU-positive cells were found in the subgranular zone (SGZ) of the dentate gyrus (DG) in all groups including controls. An increase of 64% in BrdU-positive cells was observed 7 days after CHPG injection and of 11% post DHPG compared to control rats. Fourteen days after agonist injection the levels were at 47% and 38% for CHPG and DHPG respectively. At 56 days post-injection these levels were lower than that of the controls with a decrease in BrdU-positive cells of 28% for CHPG and of 64% for DHPG compared to controls. Conclusions: These results show that the activation of group I mGluR plays a role in neurogenesis in the adult rat hippocampus. It seems that group I mGluR subtype 5 might have a more important implication than subtype 1 in the activation of neurogenesis in the adult rat hippocampus. We now need to define their contribution in SE. (Supported by Savoy Fundation.) Abstract: .sup.1 Filiz Yilmaz Onat , .sup.1 Nihan Carcak , .sup.1 Rezzan Aker , and .sup.1 Osman Ozdemir (Pharmacology and Clinical Pharmacology, Marmara University, Istanbul, Turkey; Pharmacology, Istanbul University, Istanbul, Turkey; Pharmacology and Clinical Pharmacology, Marmara University, Istanbul, Turkey; and Pharmacology, Istanbul University, Istanbul, Turkey) Rationale: Adult Genetic Absence Epilepsy Rats from Strasbourg (GAERS) show a failure to reach stage 3, 4 or 5 during the amygdala kindling process. No spike-and-wave discharges in GAERS are observed in rats younger than 30 days of age. In order to determine whether there is a relationship between the occurrence of spike-and-wave discharges and development of kindling, we examined amygdala kindling rate and spike-and-wave discharges in 20-day (PN20) and 30-day old (PN30) GAERS. Methods: Electrodes for stimulation and recording were stereotaxically implanted into basolateral amygdala and cortex of male Wistar (n = 12) and GAERS (n = 16) pups. After a recovery period of 24 h, animals were stimulated at 400 microA (80 Hz, each 1 msec in duration, for a total duration of 2 sec) every 20 min in PN20 and 90 min in PN30 groups. Animals received kindling stimulations until they achieved three stage 5 or the maximum number of stimulations (35). Seizures were scored by using Racine's 5-stage scale. EEGs from amygdala and cortex recorded continously before and after each stimulus to analyze spike-and-wave discharges and afterdischarge durations. Results: PN 20 GAERS and non-epileptic Wistar rats show no difference to achieve stage 5 during the kindling progress. Seven out of 9 GAERS animals reached stage 5 and 2 GAERS stayed at stage 2 after application of 35 stimulations in the PN30 GAERS group. The amygdala kindling rate of the 7 kindled animals in the PN30 GAERS group was significantly slower than that seen in non-epileptic animals. Afterdischarge durations of all groups were similar to each other. Spike-and-wave discharges appeared in PN30 GAERS and no spike-and-wave discharge was observed in PN20 GAERS group. The data of cumulative spike-and-wave discharge show a significant increase in post-stimulation 0-10 min period when compared to pre-stimulation period in PN30 GAERS. Kindling-induced increase in spike-and-wave discharges is specifically observed in the 7 kindled PN30 GAERS. In detail, number of spike-and-wave discharges statistically increased whereas mean duration of spike-and-wave complex did not change. Conclusions: These findings suggest that GAERS pups are more prone to develop kindled seizures than adults and resistance to kindling in GAERS is age-related. Ontogenesis of spike-and-wave discharge seems to be involved in the resistance to kindling. The increase in spike-and-wave discharges in postictal state indicates involvement of thalamus and enhanced activity of thalamic nuclei that are responsible for spike-and-wave activity. In addition, the increased spike-and-wave response to kindling stimulations might be a homeostatic mechanism to restore excitatory drive of circuitry involved in kindling progress. (Supported by Marmara University Research Committee.) Abstract: .sup.1 Gabriella Panuccio , .sup.2 Giorgio Cruccu , and 1,3 Massimo Avoli (Deptartments of Neurology and Neurosurgery, and Physiology, Montreal Neurological Institute, McGill University, Montreal, QC, Canada ; Deptartment of Neurological Sciences, La Sapienza University, Rome, Italy ; and Deptartment of Human Physiology and Pharmacology, La Sapienza University, Rome, Italy) Rationale: The cingulate cortex is a major anatomical component of the limbic system, surrounding the corpus callosum throughout its extent. Through its numerous and complex connections within and beyond the limbic system, the cingulate cortex plays different physiological roles that make of it an eclectic structure. According to Vogt et al. (Science 1979; 204:205-207; Vogt et al., Cereb Cortex, 1992; 2:435-443) two main subdivisions can be identified within the cingulate cortex: the anterior cingulate cortex (ACC) and the posterior cinguate cortex (PCC), the first being characterized by a high density of opioid receptors (Vogt et al., Exp Neurol, 1995; 135:83-92). To date, many studies have focused on the role of the ACC in pain, but little is known about its involvement in epilepsy. Here we investigate the role of mu-opiod receptors, which are capable of hyperpolarizing interneurons (Tanaka and North, J Neurosci, 1994; 14: 1106-1113), in the control of epileptiform activity in an in vitro rodent model of epilepsy. Methods: Field potential recordings were obtained from ACC deep layers in coronal slices of adult rat brains. 4-aminopyridine (4AP) 50 uM was bath-applied for [greater than or equal to] 1 hour to induce epileptiform activity. Drugs were bath-applied. Results: 4AP-induced epileptiform activity was characterized by prolonged ictal-like discharges (mean duration = 35.2 [+ or -] 4.4 s, mean interval = 310.1 [+ or -] 34.2 s; n = 9) that were preceded by a slow negative-going event. Ictal-like activity was reversibly abolished by the NMDA receptor antagonist 6-cyano-7-nitroquino-xaline-2,3-dione (CPP, 10uM) and turned into brief epileptiform events (mean duration = 2.6 [+ or -] 0.1 s, mean interval 15.2 [+ or -] 1.9 s; n = 3) when GABAergic system was blocked by picrotoxin 50uM plus CGP 55845 4uM, GABA.sub.A and GABAB receptor antagonists respectively. [D-ala2, N-Me-Phe4,Gly5-ol]-enkephalin (DAGO) 10 uM, a mu-opioid receptors agonist, decreased and eventually abolished ictal-like discharges. This effect was reversed by naloxone 10 uM, an opioid antagonist. Finally, DAGO had no apparent effect on the brief epileptiform discharges recorded after GABAergic system blockade. Conclusions: These results suggest that the ACC is capable of generating epileptiform discharges that primarily depend on NMDA receptors activation. In addition they suggest that opioid receptors may play a critical role in controlling seizures. This action may be mediated via modulation of GABAergic interneuron activity. (Supported by CIHR (Grant 8109), CURE and the Savoy Foundation.) Abstract: .sup.1 Thomas Papouin , .sup.2 Giulia Curia , .sup.2 Philippe Seguela , and 2,3 Massimo Avoli (Department of Biology, Ecole Normale Superieure, Paris, France ; Departments of Neurology and Neurosurgery, and Physiology, Montreal Neurological Institute, McGill University, Montreal, QC, Canada ; and Fisiologia e Farmacologia, Universita' Di Roma La Sapienza, Rome, Italy) Rationale: Fragile X syndrome is the most common inherited cause of mental impairment resulting from the inactivation of the fragile X mental retardation gene (fmr1) and the subsequent inability to produce the fragile X mental retardation protein (FMRP). The neurologic phenotype of fragile X patients includes epilepsy and it is well reproduced in the fmr1 knockout mice, considered a genetic model of epilepsy. The subiculum belongs to the limbic system and plays a crucial role in initiating epileptiform discharges and in gating hippocampal outputs. This gating function depends on GABA.sub.A receptor-mediated inhibition, which is important in controling hyperexcitability both at synaptic (GABA.sub.A phasic inhibition) and extrasynaptic (GABA.sub.A tonic inhibition) levels. In preliminary electrophysiological studies we have found a loss of GABAergic tonic inhibition in the subiculum of seizure prone fragile X knockout mice. In addition, altered expression of [alpha]5 and [delta] GABA.sub.A receptor subunits, which compose receptors mediating tonic inhibition, have been reported in other models of epilepsy. Thereby, in the present study we investigated at the molecular level whether a possible loss of GABAergic tonic inhibition occurs in the subiculum of fragile X versus wild type (wt) mice. Methods: We performed real time RT-PCR experiments with RNAs extracted from subiculum of age-matched fragile X and wt mice. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and TATA box binding protein (TBP) housekeeping genes - reported as suitable for RNA quantification (Radonic et al. 2003) - were used as reference genes. Using the Pfaffl method (Pfaffl, 2001), we calculated R values (level of expression relative to control) for GABA.sub.A[alpha]5 and [delta] subunits genes. Results: GABA.sub.A[alpha]5 and [delta] subunits genes were consistantly underexpressed in fragile X mice: for GABA.sub.A[alpha]5 R = 0.779 [+ or -] 0.022 with GAPDH (n = 8) and 0.729 [+ or -] 0.017 with TBP (n = 3); for GABA.sub.A[delta] R = 0.829 [+ or -] 0.021 with GAPDH (n = 8) and 0.808 [+ or -] 0.033 with TBP (n = 3). So the [alpha]5 subunit gene expression was decreased by 22% (GAPDH) to 27% (TBP), and the [delta] subunit gene expression was decreased by 17% (GAPDH) to 20% (TBP) in fragile X versus wt mice. This mild but reproductible decrease was significant: mean R values were different from 1 with a T-test (p values < 0.05). Conclusions: Our data demonstrate that a down-regulation of subunits involved in GABAergic tonic inhibition occurs at the level of RNA in the subiculum of fragile X knockout mice versus control. (Supported by CIHR, FXRFC, CURE and Savoy Foundation.) Abstract: .sup.1 Manisha N. Patel , and .sup.1 Li-Ping Liang (Pharmaceutical Sciences, University of Colorado Health Sciences Center, Denver, CO) Rationale: Free iron is an important catalyst for the initiation and propagation of free radical reactions and has been implicated in the pathogenesis of a variety of neuronal disorders. Studies in our laboratory have shown that status epilepticus (SE) results in increased mitochondrial superoxide production. The toxicity of superoxide radical is mediated in part via the release of redox-active iron from cellular iron stores including labile iron-sulfur centers. The role of intracellular iron in mediating seizure-induced brain damage is unknown.The goal of this study was to determine whether kainate-induced SE produces changes in mitochondrial iron levels, whether this correlates with mitochondrial and cellular oxidative stress and if SE-induced oxidative stress can be ameliorated with a cell-permeable iron chelator. Methods: Rats were injected with with 12 mg/kg kainic acid and hippocampal mitochondria were analyzed for chelatable (free) iron measured by the bleomycin method 24, 48 and 96 hr post-KA. N,N'-bis (2-hydroxybenzyl) ethylenediamine-N,N'-diacetic acid (HBED), a synthetic lipophilic cell-permeable chelator was used as an iron chelator. Brain HBED concentrations, reduced and oxidized glutathione levels and the oxidative DNA lesion 8-hydroxy-2'-deoxyguanosine (8OHdG/2DG) were measured by an HPLC method coupled with electrochemical detection. Results: Kainate-induced SE in rats resulted in a time-dependent increase in chelatable iron in mitochondrial fractions of the hippocampus. The time-course of the iron changes paralleled the inactivation of mitochondrial aconitase, DNA oxidation (8OHDdG/2DG) and glutathione depletion. Systemically administered HBED, a synthetic iron chelator ameliorated SE-induced changes in chelatable iron and oxidative stress. Furthermore, measurement of brain HBED levels confirmed its penetration in hippocampal tissue. Conclusions: These results suggest a role for mitochondrial iron in the pathogenesis of SE-induced brain damage and iron chelation as a novel therapeutic approach for the treatment of epileptic brain damage. (Supported by NIH R01NS039587.) Abstract: .sup.1 Peter R. Patrylo , .sup.1 Isha Tyagi , .sup.1 Amy Willingham , .sup.2 Sung Lee , and .sup.2 Anne Williamson (Physiology, Southern Illinois University School of Medicine, Carbondale, IL ; and Neurosurgery, Yale University School of Medicine, New Haven, CT) Rationale: The elderly have the highest incidence and prevalence for seizure disorders with up to 50% of cases having no identifiable antecedent. Rodents also exhibit an aging-related increase in seizure susceptibility with aged rats having a higher frequency of wet-dog shakes and a shorter latency to onset for hippocampal-associated seizures and status epilepticus during kainate-treatment. Based on kainate's mode of action, we hypothesize that these differences reflect pro-convulsive changes in limbic circuitry. To test this hypothesis, electrophysiological recordings were performed in brain slices. The dentate gyrus was our model region since it is involved in generating wet-dog shakes and limbic seizures, and it is affected preferentially with age. Methods: Field potential and whole-cell recordings were used to assess dentate circuitry and function in brain slices from adult (3-8 mo.) and aged (22-32 mo.) Fischer 344 rats. Results: No aging-related differences were noted in the maximal amplitude of orthodromic evoked field potential responses, or paired-pulse inhibition and facilitation (10 and 50 ms interpulse intervals). However, 5 Hz orthodromic stimulation (10 s) evoked multiple population spikes in 40% of aged slices (n = 10) vs. 0% of adult slices (n = 9; p = 0.033; chi-square). Further, when the dentate's capacity to filter 5 Hz afferent input was assessed by recording in CA3, approximately 30% of aged slices (n = 21) exhibited multiple spikes during the train which often progressed into spontaneous interictal-like events. Similar activity was not seen in adult slices (n = 16; p = 0.036; chi-square). Whole cell recordings revealed no aging-related differences in the membrane properties of granule cells within the linear portion of the IV curve (adult n = 22; aged n = 26), or the conductances and reversal potentials of the fast and slow IPSPs evoked with molecular layer stimulation (adult n = 17; aged n = 21). In contrast, aged granule cells (n = 21) did exhibit a 50% reduction in the frequency of spontaneous IPSPs, compared to adults (n = 20; p 10 mV amplitude) compared to adult granule cells (n = 18; p < 0.05; ANOVA). Conclusions: These data reveal that the capacity of the dentate gyrus to respond to and filter afferent input is altered in a pro-convulsive manner in aged rats. Further, these aging-related changes are associated with a compromise in inhibition and an increase in local excitatory activity. These changes could contribute to the enhanced susceptibility to wet-dog shakes and limbic seizures in aged rats during systemic kainate treatment and could contribute to the compromise in spatial learning and memory seen with age. (Supported by NIA and EFA.) Abstract: .sup.1 Bi-wen Peng , .sup.1 William A. Swan , and .sup.1 Russell M. Sanchez (Department of Pharmacology, University of Texas Health Science Center, San Antonio, TX) Rationale: A-type K.sup.+ channels can critically determine neuronal excitability and shape spiking patterns. Pharmacological blockade of A-type currents (I.sub.A) induces limbic seizures (Juhng et al., 1999 Epilepsy Res 23:675), and decreased or absent I.sub.A in principal hippocampal neurons may promote seizures in multiple epilepsy models (Bernard, et al., 2004 Science 305:532; Castro, et al., 2001 J Neurosci 21: 6626). Hypoxia is a major cause of neonatal encephalopathy and seizures, and can lead to epilepsy, but the consequences of an acute episode of seizure-inducing hypoxia are not fully understood. In the current study, we asked whether I.sub.A may be altered in hippocampal neurons following neonatal seizure-inducing hypoxia. We examined this in dentate granule cells, as the dentate gyrus is widely believed to gate the propagation of limbic seizures. Methods: Spontaneous seizures were induced by global hypoxia in P10 Long-Evans rat pups (5-7%O.sub.2 for 15 min; rectal temp. maintained at 33-34 deg. C.). Whole-cell voltage- and current-clamp recordings were obtained from visually identified dentate gyrus granule cells in hippocampal slices excised 30 min to 7 days post-hypoxia. The magnitude and voltage-dependence of I.sub.A, and spiking behavior was compared between control and hypoxia-treated groups. Results: I.sub.A was observed in all recorded dentate granule cells (n = 57 control; n = 79 hypoxia). In the hypoxia-treated group, I.sub.A amplitude was decreased to 61% of control at maximal activation with no change in the voltage-dependence of activation or steady-state inactivation. Under current-clamp, depolarizing current injection from near rest (-60 to -65 mV) elicited similar spike patterns between the two groups. However, depolarizing steps to the same absolute current from a negative conditioning potential (-80 mV) to partially remove steady-state I.sub.A inactivation revealed a longer delay to initial spike onset in the control (51 [+ or -] 3 ms) compared to the hypoxia-treated (15 [+ or -] 1 ms) group. This decreased spike latency also was associated with an increased spike frequency in the hypoxia-treated group under this experimental condition. No clear time-dependent differences were observed across days post-hypoxia. Conclusions: Our data indicated that I.sub.A was decreased in dentate granule cells acutely and for at least one week following perinatal seizure-inducing hypoxia. Additionally, this decrease in I.sub.A was associated with an increased propensity for more rapid action potential firing in response to depolarization from hyperpolarization within a physiological range. Given that the dentate gyrus is widely believed to gate the propagation of limbic epileptiform activity, the observed decrease in I.sub.A may decrease this gating function by increasing the intrinsic excitability of dentate granule cells. (Supported by an EFA Postdoctoral Research Training Fellowship (BP) and PHS Grant NS 047385 (RMS).) Abstract: .sup.1 Zechun Peng , .sup.1 Christine S. Huang , and 1,2 Carolyn R. Houser (Neurobiology, David Geffen School of Medicine at UCLA, Los Angeles, CA ; and Research Service, VA Greater Los Angeles Healthcare System, Los Angeles, CA) Rationale: Extracellular signal-regulated kinases (ERKs) appear to be particularly important in activity-dependent forms of plasticity. While previous studies have reported a significant increase in phosphorylated ERK (pERK) levels in the hippocampus following the induction of severe seizures, it is unknown whether ERK is activated during spontaneous seizures. The goals of this study were to determine if ERK phosphorylation increases after spontaneous seizures and describe the time course and distribution of pERK labeling. Methods: Adult C57BL/6 mice were injected with pilocarpine (320 mg/kg) to induce status epilepticus that was then stopped by injection of diazepam (5 mg/kg) 2 h after seizure onset. Beginning two weeks after status epilepticus, animals were observed for spontaneous behavioral seizures and perfused transcardially with fixative at intervals between 2 min and 24 h after the onset of spontaneous seizures. Tissue containing the hippocampal formation was processed for immunohistochemical localization of pERK. Results: In control mice, moderate to strong pERK immunoreactivity was detected in cell bodies and dendrites of neurons that were scattered throughout the hippocampal formation. Labeled neurons were most prominent in the CA1 pyramidal cell layer, subiculum and entorhinal cortex. In mice sacrificed 24 h and longer after spontaneous seizures (n = 22), pERK labeling was substantially decreased throughout the hippocampal formation. However, as early as 2 min after the onset of a spontaneous seizure (n = 5), the cell bodies of many dentate granule cells were moderately to strongly labeled. Increased pERK immunoreactivity was also evident in the entorhinal cortex, subiculum and pyramidal cell layer of the hippocampus. At this time point, pERK labeling in the neocortex was also increased but to a lesser extent than in the hippocampal formation. By 5 min after the onset of spontaneous seizure (n = 3), moderate to strong pERK labeling was evident in virtually all dentate granule cells, the majority of CA1 pyramidal cells and numerous neurons in the entorhinal cortex. At 15 min after spontaneous seizures (n = 6), pERK labeling in these regions was lower than at earlier time points. By 30 min and longer after spontaneous seizures, pERK-immunoreactivity had returned to very low levels, similar to that in animals at 24 h or more after spontaneous seizures. Conclusions: ERK was activated in neurons throughout the hippocampal formation near the time of spontaneous seizures. This seizure-related phosphorylation of ERK could be detected early but for only a short period. The functional significance of such ERK activation is unknown but could be part of the seizure initiation process and potentially contribute to long-term plastic changes in the epileptic animals. (Supported by VA Medical Research Funds and NIH Grant NS046524 to CRH.) Abstract: .sup.1 Heidrun Potschka , .sup.1 Judith Winter , .sup.2 Bjoern Bauer , .sup.2 Anika M.S. Hartz , and .sup.2 David S. Miller (Deptartment of Pharmacology, Toxicology, and Pharmacy, University of Veterinary Medicine, Hannover, Germany ; and Laboratory of Pharmacology and Chemistry, National Institute of Environmental Health Sciences/National Institutes of Health, Research Triangle Park, NC) Rationale: Accumulating data indicate that seizure-induced upregulation of the multidrug transporter, P-glycoprotein (Pgp), can contribute to pharmacoresistance. Enhanced Pgp expression in brain capillary endothelial cells seems to limit brain penetration of antiepileptic drugs. Thus, interaction with factors mediating seizure-induced upregulation of Pgp may be one strategy to limit the development of pharmacoresistance. There is evidence that cyclooxygenase (COX) may be involved in seizure-induced upregulation of Pgp. On one hand COX is induced by seizure activity and on the other hand COX can induce Pgp expression. In the present study, we tested whether COX inhibition can prevent induction of Pgp by seizure activity. Methods: Status epilepticus (SE) was induced in rats by fractionated pilocarpine administration (10 mg/kg i.p. every 30 min). Eighteen animals received daily i.p. injections of the COX inhibitor indomethacin (2.5 mg/kg i.p. 2 times daily) starting at the day before status epilepticus induction. Twelve rats received corresponding vehicle injections. SE duration was limited to 90 min using diazepam. Rats were perfused two days following SE. Pgp expression was analyzed in the hippocampus and adjacent cortex by immunhistochemistry. Expression of both groups with SE was compared with that of an additional control group (n = 8) which received vehicle injections, but in which no SE was induced. Results: Severity of SE was not influenced by indomethacin treatment. However, exitus during SE and at the day following SE was more frequent in the group of indomethacin treated rats (44%) as compared to vehicle treated rats (8%). Pgp expression was significantly upregulated in the hippocampus and adjacent cortex following SE. Indomethacin treatment prevented this upregulation both in the hippocampus and the cortex. A pronounced indomethacin-mediated neuroprotection was obvious in the piriform cortex by visual inspection of the brain sections. Currently, we are studying the degree of neurodegeneration in the hippocampus by neuronal cell counts to evaluate whether neuroprotection was also achieved in this brain region. Conclusions: The data indicate that cyclooxygenase plays a critical role in seizure-induced upregulation of Pgp. Inhibition of inflammatory cascades may be a valid strategy to prevent Pgp-mediated pharmacoresistance. Further validation of this strategy is necessary with a focus on safety. (Supported by National Institute of Environmental Health Sciences, Division of Intramural Research.) Abstract: .sup.1 Kim L. Powell , .sup.1 Caroline Ng , .sup.2 Sheng-Hong Xu , .sup.1 Terence J. O'Brien , .sup.2 David A. Williams , and .sup.3 Chris A. Reid (Departments of Medicine and Surgery (RMH/WH), University of Melbourne, Melbourne, Victoria, Australia ; Department of Physiology, University of Melbourne, Melbourne, Victoria, Australia ; and Howard Florey Institute, Melbourne, Victoria, Australia) Rationale: Inducible changes in ion channel expression influencing neuronal excitability are emerging as potential mechanisms in human epilepsies. A relative new comer to the list of ion channels involved in epilepsy is the hyperpolarization-activated cyclic nucleotide-gated cation channels (HCN). Several studies in chronic human epilepsy and various animal models of temporal lobe epilepsy (TLE) have reported alterations in the HCN expression and function in the hippocampus. However, these changes have only being reported in immature animals or models with significant neuronal loss or in end stage epileptic patients. Therefore, whether these reported changes in HCN expression are a critical component of epileptogenesis in uncertain. Methods: In this study we have measured HCN1 and HCN1 mRNA expression in two different animal models of TLE; the amygdala kindling model (associated with minimal cell loss) and the kainic acid (KA) post-status epilepticus model (associated with significant cell death) during early epileptogenesis. 7-8 week old female rats were studied. The kindled group were electrically stimulated in the left amygdala until 'fully kindled' (5 class V seizures according to the racine classification). The KA group received i.p. KA injections (2.5-5 mg/kg) hourly until continuous seizures were observed on EEG. At designated time points, brains were extracted and hippocampal subregions (CA1 & CA3) were sectioned. Real time PCR was performed using gene expression assays (Applied Biosystems) and analysis was performed using the [DELTA]C.sub.T quantification method. Results: In the CA1 region of the hippocampus, HCN1 mRNA expression was significantly reduced in post-KA animals compared to controls. In contrast, HCN1 mRNA expression did not differ between control and kindled animals. There was no difference in HCN2 mRNA expression in either epileptogenic model. In the CA3 region of the hippocampus, HCN1 mRNA expression was significantly reduced in post-KA animals compared to controls. In contrast, HCN1 mRNA expression did not differ between control and kindled animals. There was no difference in HCN2 mRNA expression in either epileptogenic model. Conclusions: A reduction in HCN1 mRNA expression in the CA1 or CA3 region of the hippocampus would alter the balance of HCN1:HCN2 ratio in these neurons and thus the firing properties of these neurons would change. However, since no changes in HCN mRNA expression was observed in the kindled group we question whether the changes we have observed in the post-KA animals is a true epileptogenic response or a non-specific response to cell death post status epilepticus. (Supported by NH&MRC project grant 406640.) Abstract: .sup.1 Robrecht Raedt , .sup.1 Annelies Van Dycke , .sup.2 Tom Boterberg , .sup.3 Ann-Marie Alborn , .sup.1 Tim De Smedt , .sup.1 Tine Wyckhuys , .sup.1 Veerle De Herdt , .sup.1 Liesbeth Waterschoot , .sup.1 Kristl Vonck , .sup.3 Eriksson Peter , .sup.3 Elinor Ben-Menachem , and .sup.1 Paul Boon (Department of Neurology, Ghent University Hospital, Ghent, Belgium ; Department of Radiotherapy, Ghent University Hospital, Ghent, Belgium ; and Institute of Clinical Neuroscience, Sahlgrenska University Hospital, Goteborg, Sweden) Rationale: A common feature in kindling and status epilepticus models for Temporal Lobe Epilepsy (TLE) is enhanced neurogenesis in the subgranular layer of the hippocampus. Although there is evidence that some of these newborn neurons migrate towards ectopic locations and have abnormal firing behaviour, no clear relationship between enhanced neurogenesis and epileptogenesis has been demonstrated. In the present study we investigated the effect of blocking neurogenesis by low dose whole brain [gamma]-irradiation on hippocampal kindling epileptogenesis. Methods: Female rats (n = 16) were implanted with a bipolar recording and stimulation electrode in the right dorsal hippocampus. One week after implantation half of the rats were subjected to low-dose (8 Gy) whole brain irradiation. One or two days after irradiation, rats were subjected to a rapid alternate day kindling acquisition protocol and received 48 stimulations spread over 4 alternating days. One and two weeks after termination of kindling acquisition, rats were retested by giving ten additional stimulations per test. The afterdischarge threshold, afterdischarge duration and the clinical seizure severity (Racine stages 1-5) were compared between irradiated and non-irradiated rats. Results: All rats displayed afterdischarges and clinical seizures with an increasing length and severity, until they displayed five or more consecutive stage 5 seizures and were labelled fully kindled. At kindling day 3 and 4 mean seizure severity in irradiated rats was significantly worse compared to non-irradiated rats. This difference was evident because seizures in irradiated rats developed more quickly to the most severe seizure state compared to seizures in non-irradiated rats. In line with these differences, afterdischarge threshold at kindling day 4 was significantly lower in irradiated compared to non-irradiated rats. All these differences were temporary and could no longer be demonstrated during retest stimulations. Conclusions: Low-dose whole brain irradiation, administered in order to block neurogenesis, leads to a lowered threshold to electrically evoke an epileptic event one week after irradiation. At this time point, irradiated rats also displayed more rapidly more severe seizures compared to non-irradiated rats. These effects were only temporary and no longer evident two and three weeks after irradiation. We can conclude that blocking neurogenesis by low-dose whole brain irradiation has mild accelerating but temporary effects on epileptogenesis. (Supported by the Institute for the Promotion of Innovation through Science and Technology (IWT) in Flanders and UCB Pharma.) Abstract: .sup.1 YogendraSinh H. Raol , .sup.2 Madeline E. Rhodes , .sup.2 Cheryl A. Frye , and .sup.1 Amy R. Brooks-Kayal (Division of Neurology, Children's Hospital of Philadelphia, Philadelphia, PA ; and Department of Psychology, University at Albany, Albany, NY) Rationale: A number of studies suggest that progesterone and its metabolite 5[alpha]-pregnane-3[alpha]-ol-20-one (3[alpha],5[alpha]-THP) act as endogenous anticonvulsants. 3[alpha],5[alpha]-THP is a positive modulator of GABA(A) receptors and reduced endogenous levels of 3[alpha],5[alpha]-THP increase seizure susceptibility. Studies from various laboratories suggest that early-life seizures increase susceptibility to development of seizures in later life. In the current study we induced prolonged status-epilepticus (SE) at postnatal day 10 (P10) and measured levels of progesterone and 3[alpha],5[alpha]-THP in plasma and hippocampus immediately after SE and in adulthood. Methods: Sprague-Dawley rat pups at P9 were injected intraperitoneally with 3mEq/kg lithium followed 14-18 hrs later by 60 mg/kg pilocarpine. Control rats were treated identically to that of seizure rats but were injected with 0.9% saline on P10. Trunk blood and brains were collected 6 hours after the onset of SE or in adulthood for measurement of progesterone and 3[alpha],5[alpha]-THP in plasma and hippocampus by radioimmunoassay. Results: 3[alpha],5[alpha]-THP levels in hippocampus increased significantly (p < 0.0002) in rats that were sacrificed 6 hours after SE induction. Plasma and hippocampal levels of progesterone and plasma 3[alpha],5[alpha]-THP were not different in SE as compared to control rats. Rats that had experienced SE at P10 and were sacrificed in adulthood had lower levels of 3[alpha],5[alpha]-THP in plasma as compared to control rats (p < 0.013). Levels of 3[alpha],5[alpha]-THP in hippocampus and progesterone levels in plasma and brain were not significantly different between seizure and control rats. Conclusions: These results suggest that 3[alpha],5[alpha]-THP levels increase locally in association with P10 SE, possibly as an adaptive mechanism to inhibit seizures. Long-term decreases in circulating levels of 3[alpha],5[alpha]-THP in adult rats that had experienced SE in early-life may contribute to increased seizure susceptibility in these rats. (Supported by NIH NS38595 to ABK and MH-067698 to CAF.) Abstract: .sup.1 Adi Cymerblit , and 1,2 Yitzhak Schiller (Physiology, Technion Medical School, Haifa, Israel ; and Neurology, Rambam Medical Center, Haifa, Israel) Rationale: Seizures are associated with hyper-synchronous electrical activity in the EEG recordings. However, the network dynamics at the single neuron level during the development and maintenance of seizures remains largely unknown. In this study we studied the firing behavior of single neurons, and monitored synchronization between the different recorded neurons during developing and ongoing seizures evoked by pilocarpine and picrotoxin. Methods: Multi-electrode extra-cellular recordings of single units and local field potentials were performed from the hippocampus of anesthsized (urethane) and awake rats under control conditions and after administration of pilocarpine and picrotoxin. The two convulsants were administered either by systemic (I.P.) or local (intra-hippocampal) injection. In addition simultaneous intracellular whole-cell recordings were performed. Results: Immediately after administration of pilocarpine and picrotoxin, prior to initiation of clinical or electrographic seizures, the frequency of spikes decreased, and synchronization between neurons, as measured by the cross-correlogram, markedly diminished or disappeared all together. In contrast after clinical and electrographic seizures initiated the frequency of spikes gradually increased, neurons tended to fire in bursts and the inter-neuron synchronization markedly increased. The cross-correlograms during the clinical seizure revealed higher values and wider time-windows as compared to control conditions. Similar results were obtained in anesthetized and awake rats; in systemic (I.P.) and local intra-hippocampal administration of the convulsants, and with pilocarpine and picrotoxin. Simultaneous intracellular whole-cell recording from anesthetized rats demonstrated the intracellular correlate of developing seizures. Conclusions: In this study we showed that during the development of pharmacologically induced seizures synchronization of firing between different neurons in the hippocampus decreased dramatically, while only later during clinical seizures inter-neuronal synchronization gradually increased. These findings may promote our understanding of the network dynamics responsible for seizure initiation and maintenance. Moreover, in future it may serve as the basis for detection of impending seizures in human patients. (Supported by ISF foundation; Rapapport family Foundation.) Abstract: .sup.1 Ulrich Schridde , .sup.2 Manjula Khubchandani , .sup.2 Basavaraju G. Sanganahalli , .sup.1 Dario J. Englot , .sup.2 Fahmeed Hyder , and 1,3 Hal Blumenfeld (Neurol., Yale University Sch. Med., New Haven, CT ; Diagnos. Radiol., Yale University, New Haven, CT ; and Neurobiol., Yale University, New Haven, CT) Rationale: Recently, fMRI became an important tool to study brain activity in epilepsy by investigating increases/decreases in the blood oxygenation level dependent (BOLD) response, known to be associated with changes in cerebral blood flow (CBF) and -volume (CBV). Although increases in the BOLD response have been linked to increases in neuronal activity, decreases of the signal, although often present, are generally neglected and still poorly understood. A detailed understanding of the relationship and causes of the different hemodynamic and neuronal signals is however essential for interpreting neuroimaging data in epilepsy and also has important implications for its treatment. Methods: 26 Wistar rats were injected with bicuculline (0.5 mg/kg, iv) to induce generalized tonic-clonic seizures. Increased/decreased brain activity during seizures were investigated by combining fMRI and CBV imaging with neurophysiological recordings (field and unit activity), and local CBF measurements using laser Doppler flowmetry. Results: The fMRI showed widespread BOLD and CBV increases throughout the cortex, basal ganglia and thalamus during seizures. Neurophysiological recordings, likewise, showed increased CBF, local field potential power and neuronal firing rate in cortical areas. Surprisingly, however, most of the rats had a decreased BOLD response in the hippocampus, which was accompanied by increased CBF and neuronal activity. Conclusions: Our results show that during generalized seizures the brain shows rather heterogeneous BOLD activity with increases/decreases in distinct networks. While our data confirm that increases in the BOLD signal during seizures correlate with increases in CBF, CBV and neuronal activity, we found that in the hippocampus increases in the latter three variables are associated with a decreased BOLD response. This observation argues against the hypothesis of a vascular steal phenomenon as cause for a decreased BOLD signal. Although the cause for the local dissociation of the signals is not known, one possibility is that the neuronal oxygen consumption in the hippocampus during seizures exceeds its supply; which is currently under investigation in our laboratory. In conclusion, our data show that fMRI can reveal important information about the pathophysiology of epilepsy and its treatment, but that the interpretation of the neuroimaging signals during seizures should proceed with caution, and might be different depending on brain region. (Supported by NIH R01 NS049307 and the Blattmachr Fund (HB); NIH R01 MH67528 and DC03710 (FH).) Abstract: .sup.1 Arjune Sen , .sup.2 Lillian Martinian , .sup.1 Kate Chandler , .sup.3 Tom Jacobs , .sup.1 Matthew C. Walker , .sup.3 Margareta Nikolic , 1,2 Maria Thom , and .sup.1 Sanjay M. Sisodiya (Department of Clinical and Experimental Epilepsy, The Institute of Neurology, London, United Kingdom ; Division of Neuropathology, The Institute of Neurology, London, United Kingdom ; and Department of Cellular and Molecular Neuroscience, Imperial College, London, United Kingdom) Rationale: Hippocampal sclerosis (HS) is amongst the most common causes of medically refractory epilepsy in adults. Histologically HS is characterised by segmental neuronal loss, but the molecular mechanisms underlying this cell loss remain uncertain. Deregulation of the serine-threonine kinase cyclin dependent kinase 5 (cdk5) by p25, the calpain-mediated cleavage product of the physiological cdk5 activator p35, occurs in several cell death paradigms. Recently we demonstrated subfield-specific deregulation of cdk5 in human HS while animal models have shown that abnormal phosphorylation of the NMDA receptor might contribute to epileptogenesis. Intriguingly, when cdk5 is deregulated it phosphorylates NMDA receptors resulting in excitotoxicity. We therefore hypothesised that there might be abnormal cdk5-mediated phosphorylation of the NMDA receptor in HS. Methods: Surgically-resected cases of HS with adjacent histologically-normal temporal lobe and hippocampi from rats sacrificed at intervals following pilocarpine-induced status epilepticus were examined for cdk5, p25/p35, NeuN, NMDA and cdk5 phosphorylated NMDA receptor using immunohistochemistry, confocal microscopy, western blots and kinase assays. Results: Animal models demonstrated a time-dependent increase in the ratio of p25:p35 and in p25-cdk5 activity following administration of pilocarpine. p25-cdk5 phosphorylated NMDA was not detected in control animals, but was strongly detected seven days after pilocarpine-induced status epilepticus and persisted until at least five months after pilocarpine treatment. Furthermore, despite neuronal loss, hippocampi from patients with HS had increased levels of p25-cdk5 phosphorylated NMDA compared to adjacent histologically-normal temporal lobe. Conclusions: Our results suggest that cdk5 might contribute to the neuronal loss, and potentially the epileptogenicity, of HS through direct phosphorylation of the NMDA receptor. (Supported by Medical Research Council, UK, Guarantors of Brain and The Wellcome Trust.) Abstract: .sup.1 Sevgi Turker Gorgulu , .sup.1 Nihan Kara , .sup.1 Zeynep Gunes , .sup.1 Nurbay Ates , .sup.1 Gul Ilbay , and .sup.1 Feride Severcan (Biology, Middle East Technical University, Ankara, Turkey; Biology, Middle East Technical University, Ankara, Turkey; Biology, Middle East Technical University, Ankara, Turkey; Physiology, Kocaeli University, Kocaeli, Turkey; Physiology, Kocaeli University, Kocaeli, Turkey; and Biology, Middle East Technical University, Ankara, Turkey) Rationale: The epilepsies are a heterogeneous group of symptom complexes, whose common features is the recurrence of seizures (Brown et al., 1993). It is known that epileptic activity induce many molecular and structural changes in the brain (Eid et al., 2004). In order to evaluate these changes we aimed to investigate the consequences of a convulsant agent called pentylenetetrazol (PTZ). PTZ is a chemical frequently used in the study of epileptic seizures since it produces a reliable discriminative stimulus (De Boer et al. 1982; Bradford 1989; De Deyn and Macdonald 1989). The present study was focused to analyze the changes resulted from sub-consulvant and consulvant dose of PTZ on rat brain homogenate at molecular level by using Fourier transform infrared (FTIR) spectroscopy. Methods: Experimental animals were divided into three groups as control (n = 6), sub-convulsant (n = 6) and convulsant (n = 6). The treated groups were made epileptic by a single administration of sub-convulsant (25 mg/kg) and convulsant (60 mg/kg) dose of PTZ. Following PTZ administration, behavioral observations were carried out to determine epileptic activity for 30 min. Subsequently, in the FTIR spectra, the shift in peak positions, the change in bandwidths and the intensity/area values of the bands were determined and compared in between control and treated groups. Results: The results revealed that the concentration of unsaturated lipids decreased for both treated samples. The lipid/protein ratio slightly increased for sub-convulsant group whereas this ratio decreased for convulsant group. An increase in the lipid order and a decrease in the fluidity of membrane were observed for low dose of PTZ's group. However, for high dose of PTZ's opposite effects on the order and fluidity were observed. In addition, the changes in hydrogen bonding capacity of head groups of membrane lipids as well as a decrease in the protein concentration caused by the administration of PTZ were obtained. Conclusions: Consequently, these results show that the early alteration at molecular level resulted from chemically induced epileptic activity can be detected by FTIR. Moreover, the molecular effects of unconvulsive and convulsive epileptic seizures can be compared by this method. Brown S. et al., Seizure (1993) 2: 91-103. Bradford H. F. Epilepsia (1989) 30: 17-25. De Boer T., et al., Brain Res. (1982) 253: 153-160. De Deyn P. P. and Macdonald R. L. Epilepsia (1989) 30: 17-25. Eid et al., (2004) 3, 77 Lancet Neurology 28-37. Abstract: .sup.1 Maria Sitges , and .sup.1 Vladimir Nekrassov (Department of Cellular Biology and Physiology, Biomedical Research Institute, UNAM, Mexico; and Basic and Applied Research Division, National Rehabilitation Institute, SSA, Mexico) Rationale: Several of the most effective antiepileptic drugs (AEDs) are believed to stop the paroxysmal neuronal activity acting as Na.sup.+ channel blockers. However, no single study comparing in parallel their effects on brain Na.sup.+ channel-mediated responses is available.In this work the effects of increasing concentrations of CBZ, PHT, LTG, OXC and Top on the release of the excitatory amino acid neurotransmitter, glutamate (Glu), induced by the selective activation of Na.sup.+ channels with the Na.sup.+ channel opener toxin, veratridine, were investigated; and compared with the effect of the memory enhancer with antiepileptic potential, VPC (ethyl apovincamine-22-oate). Previously we have shown that VPC is a potent inhibitor of the epileptic cortical activity induced by the convulsing agents, pentylenetetrazole and 4-aminopyridine in the guinea pig in vivo (Epilepsy Res. 2004; 60: 63-71; Clin. Neurophysiol. 2004; 115: 2711-17), that VPC inhibits the tetrodotoxin sensitive fraction of the rise in internal Na.sup.+ induced by 4-amonipyridine (Neurochem. Int. 2005; 46: 533-540)) in vitro, and that VPC and CBZ presynaptic effects in vitro are additive (Neurochem. Int. 2006; 49: 55-61). In this investigation the efficacy and potency of several of the most widely used AEDs and VPC on Glu release to veratridine was compared. Methods: These experiments were done following methods that we have described previously (J. Neurosci. Res.1995; 40: 613-621; Neurochem. Res. 1996; 21: 889-895). Results: Under equivalent experimental conditions, CBZ, PHT, LTG and OXC are required in the range from 150 to 1500 uM in order to progressively inhibit the release of [3.sup.H]Glu induced by veratridine,VPC is required in a much lower range of concentrations (from 1.5 to 15 uM), and Top was able to inhibit the veratridine-induced response in about 20% only at the highest dose tested (1500 uM). Conclusions: It is concluded that, with the exception of Top, the mechanism of action of all the AEDs tested involves a reduction of the carrier-mediated fraction of Glu release through a decrease in cerebral presynaptic Na.sup.+ channels permeability. Considering that the high doses of AEDs required to control seizures are frequently accompanied by adverse secondary effects, the much higher potency of VPC for blocking presynaptic Na.sup.+ channels could be advantageous in the treatment of epilepsy. [figure 1] (Supported by: This work was financially supported by Psicofarma S.A. de C.V. and by project P-48695 from SEP-CONACYT.) Abstract: .sup.1 Alexander A. Sosunov , .sup.1 Xiaoping Wu , .sup.2 Casper Caspersen , .sup.1 Robert R. Goodman , and .sup.1 Guy M. McKhann II (Neurological Surgery, Columbia University, New York, NY ; and Pediatrics, Columbia University, New York, NY) Rationale: Uptake and metabolism of synaptically released glutamate by astrocytes is critical for brain homeostasis and protection of neurons against overexcitation. To test the hypothesis that glutamate uptake and metabolism are deteriorated in gliotic protoplasmic astrocytes, we carried out immunohistochemical and electrophysiological analysis of surgically resected brain tissue in patients with pharmaco-resistant temporal lobe epilepsy and Tuberous Sclerosis Complex (TSC). Methods: Specimens of neocortex and hippocampi from pharmaco-resistent MTLE (n = 44) and cortical tubers resected from epileptic TSC patients (n = 12) were studied with immunohistochemical and patch-clamp electrophysiological techniques. Results: Areas of severe gliosis/sclerosis typical of cortical tubers in TSC and hippocampal sclerosis in MTLE were composed of "scar" astrocytes characterized by high upregulation of GFAP and S100 beta and less upregulation of CD44 and alphaB-crystallin. These cells revealed dramatic reduction in the expression of glutamine synthetase (GS) and astrocytic glutamate transporters proteins (GT: EAAT1 and EAAT2). In addition, scar astrocytes underwent conspicuous alterations in their "gross" morphology consisting of transformation into fibrous-like astrocytes accompanied by elongation of cell processes but loss of fine lamellipodial structures. Whole cell patch-clamp recordings of scar astrocytes showed dramatic decrease in inward glutamate transport currents, in comparison to control protoplasmic astrocytes. Conclusions: Downregulation of GS and GT expression proceeds in parallel with progression of gliotic changes in protoplasmic astrocytes. We hypothesize that these glutamate metabolic alterations may not be pathology specific, but rather may constitute a common effect of profound reconstruction of the cellular cytoskeleton during the development of gliosis. Astrocytic lamellipodia may be the main subcellular compartment involved in glutamate uptake by protoplasmic astrocytes. These structures are significantly lost in association with GFAP upregulation in gliosis. (Supported by Parents Against Childhood Epilepsy; R21 NS 42334.) Abstract: .sup.1 Matt Stead , .sup.1 Fredric B. Meyer , .sup.1 Cindy L. Nelson , .sup.1 Stephen J. Goerss , and .sup.1 Gregory Worrell (Neurology, Mayo Clinic, Rochester, MN) Rationale: High frequency oscillations (>= 40 Hz) in epileptogenic brain are not yet understood, but are a recognized phenomenon in a large proportion of patients with epilepsy. DC potential shifts have been noted in animal models of seizure but are not routinely recorded in humans with epilepsy. The physiologic effect of extracellularly mediated DC potentials on transmembrane voltage-gated ion channels suggests itself as a potential contributor to ictogenesis, and thus worthy of investigation. Methods: To investigate the potential relationship between high and low frequency oscillations in epilepsy we recorded continuous unfiltered EEG from from 15 patients implanted with intracranial electrodes as part of their evaluation for epilepsy surgery. The acquistion system used had a dynamic range of +132 to -132 mV, no hardware filtering, and a per channel digitization rate of 32556 Hz. The data were recorded from Pt/Ir clinical intracranial electrodes in both depth and grid/strip configurations. Nine of the patients had mesial temporal onset, the remaining 6 had neocortical onset seizures. Results: High frequency oscillations, defined here as > = 40Hz, were observed in the clinically determined seizure onset leads in all 15 patients. The seizure onset frequency ranged from 45 to 143 Hz. In 14 of the 15 patients the high frequency discharge was cotemporaneous with a large scale DC shift ranging from 0.6 to 1.7 mV. The shift occasionally preceded the high frequency discharge by many seconds as shown in the accompanying figure, however in some cases the high frequency discharge appeared to precede the DC shift. Conclusions: There is high degree of association between the DC shifts and high frequency oscillations during seizure. Further investigations into the cause of this association may lead to a better understanding of ictogenesis and open novel avenues for treatment. [figure 1] (Supported by: Salary support for Dr. Stead was provided by the Epilepsy Foundation of America.) Abstract: .sup.1 Waldemar B. Swiercz , .sup.1 Phill A. Williams , .sup.2 Edward F. Dudek , and .sup.1 Kevin J. Staley (Pediatrics Neurology, University of Colorado Health Sciences Center, Denver, CO ; and Physiology, University of Utah School of Medicine, Salt Lake City, UT) Rationale: We used a computational model of simultaneous discharges of a population of pyramidal cells in the hippocampal area CA3 to estimate the relationship between the network connectivity and the probability of spontaneous bursts. Prior investigations (Traub and Miles, Neuronal Networks of the Hippocampus, Cambridge University Press, 1991; Lytton et al. Computer models of hippocampal circuit changes of the kindling model of epilepsy Artif Intell Med. 13:81-97 1998) have supported a nonlinear relationship between connectivity and the probability of synchronous population discharges. In our experiments we wished to simulate the growth of recurrent synaptic connections after an epileptogenic insult (sprouting) in order to model the effects on sprouting on network activity. Methods: Our artificial network model employed an array of 10000 pyramidal cells and 225 interneurons. Interactions between the neurons take place via a randomly generated mesh of synapses that includes activity-dependent depression. The strength of glutamatergic synaptic connections is modifiable according to rules governing long-term synaptic plasticity. The network was initially generated with sparsely connected neurons, simulating the network prior to sprouting. Then after each experiment we increased network connectivity to simulate sprouting of synaptic connections. This was accomplished by increasing the maximum feasible radius of synaptic connections and randomly generating additional glutamatergic synapses onto principal cells and interneurons. Results: We performed experiments for networks with various connectivities, focusing on spontaneous bursting behavior. An average number of incoming connections per neuron in separate experiments ranged between 35 and 208. The relationship between the connectivity and the average incoming connections count was close to linear. Interestingly the relationship between network connectivity and spontaneous bursting behavior was very nonlinear. This agrees with our findings regarding seizure frequency vs. time following an epileptogenic injury, (see abstract by Williams PA et al.), and with prior modeling studies employing more detailed neuronal models (Traub and Miles 1991). Conclusions: The rapid increase in seizure probability that eventually follows an epileptogenic insult may be the result of a more linear increase in local connectivity arising from axonal sprouting. The modeling data suggest that a similar and cotemporaneous change in the pattern of interictal spikes should also be evident. Further, the evolution of interictal spikes over time may reflect changes in the underlying epileptogenic network that may help in clinical decision making regarding long-term seizure probability. These hypotheses are being tested using continuous radiotelemetric EEG monitoring. (Supported by NIH (NINDS).) Abstract: .sup.1 Feng Ru Tang , .sup.1 Dong Liang Ma , and .sup.1 Shwn Chin Chia (Epilepsy Research Lab, National Neuroscience Institute, Singapore, Singapore) Rationale: It has been shown that different isoforms of protein kinase C (PKC) are involved in the signal transduction of the 3 groups of metabotropic glutamate receptors (mGluRs) In animal model of seizure, different subtypes of PKC isoforms have been reported to play different roles in epileptogenesis. However, no similar study has been done in the hippocampus from patients with temporal lobe epilepsy. Methods: Immunocytochemical study at both light and electron microscopic level, double immunostaining, and RT-PCR were used. Results: In the hippocampus from patients with mesial temporal lobe epilepsy, PKC[alpha], PKC[beta], PKC[gamma], PKC[zeta], PKC[eta] and PKCϵ immunopositive products decreased in the stratum pyramidale of CA1 and CA3 areas. However, no obvious change was found in CA2 area. PKC[gamma] and PKCϵ immunopositive products increased in the granule cells of the dentate gyrus, however, PKC[zeta], and PKC[eta] immunopositive products decrease in these cells. Double labeling immunocytochemistry showed the co-localization of PKC[alpha], PKCg, PKC[beta]2 with metabotropic glutamate receptor 5 (mGluR5), G-protein subtype G.sub.q/11[alpha] and phospholipids C beta1 subtype (PLC[beta]1) in pyramidal neurons in CA1-3 areas. Under electron microscopic level, PKC[delta], PKC[eta], PKC[zeta], or PKCϵ was located in the post-synaptic elements. However, PKC[gamma] was found in both pre- and post-synaptic elements. Gene expression of different isoforms of PKC such as PKC[alpha], PKC[beta], PKC[gamma], PKC[zeta], PKC[eta] and PKCϵ has been studied by RT-PCR in the hippocampus of the "normal" human samples. We showed relative lower expression of PKC[beta] when compared to other PKC isoforms. Conclusions: PKC[gamma] and PKCϵ may be related to the hyperactivity of granular cells and involved in epileptogenesis. PKC[alpha], PKCg, PKC2 may be involved in mGluR5, G.sub.q/11[alpha] and PLC[beta]1 related signal transduction in pyramidal neurons. (Supported by a research grant from National Medical Research Council (No: NMRC/0731/2003).) Abstract: .sup.1 Sevgi Turker Gorgulu , .sup.1 Feride Severcan , .sup.1 Stephen Wassall , and .sup.1 William Stillwell (Biology, Middle East Technical University, Ankara, Turkey; Biology, Middle East Technical University, Ankara, Turkey; Physics, Indiana University Purdue University, Indianapolis, IN; and Biology, Indiana University Purdue University, Indianapolis, IN) Rationale: Although there have been progresses to understand the mechanisms underlying epilepsy, the cellular and molecular basis of epilepsy still is not known yet very well (Patel, 2004; McCorry et al., 2004; Wolfgang and Dieter, 2002). Therefore, epileptic seizures remain as uncontrolled in at least 30% of all the cases. In order to promote new advances in the development of antiepileptic drugs a better understanding of cellular and molecular mechanisms of epilepsy should be the ultimate goal (Avanzini and Franceschetti, 2003). This can be achieved by beginning to evaluate epilepsy-induced changes in the brain cells. However, beside the consequences of epileptic activity, any possible effect of experimental convulsants on the cells should be under consideration. In order to evaluate the effects of convulsants we initially aimed to investigate the interaction of model membranes with a convulsant agent called picrotoxin (PTX). PTX is a noncompetitive antagonist at GABA-A receptors and it blocks the GABA-activated chloride ionophore (Mazarati et al., 2004). Methods: This study was performed to characterize the effects of PTX on some parameters of dipalmitoylphosphotidylcholine (DPPC) model membrane system. We used differential scanning calorimetry (DSC), electrospin resonance (ESR) spectroscopy and steady-state fluorescence spectroscopy. All studies were carried out with three different concentrations of PTX (1 mol%, 12 mol% and 24 mol%). Results: The DSC results revealed that PTX has little effect on the melting behavior of DPPC. Indeed, ESR and steady-state fluorescence data indicated that PTX does not significantly affect the order and fluidity of DPPC model membranes. Conclusions: All these results suggest that PTX does not directly interact with membrane lipids, which means that it only acts by binding the membrane protein. Therefore, it can be concluded that the molecular and cellular alterations of chemically induced seizures are only resulted from the consequences of epileptic activity not from convulsant itself. (Supported by Avanzini G., and Franceschetti S., Lancet Neurology 2003; 2: 33-42. Manisha Patel M., Free Radical Biology & Medicine (2004) 10: 1951-1962. Mazarati, A.M., Halaszi, E., Telegdy, G., Brain Res. (1992) 589: 164- 166. McCorry D., Chadwick D., and Marson A., Lancet Neurol (2004) 3: 729-35 Wolfgang L., and Dieter S., Epilepsy Research (2002) 50: 3-16.) Abstract: .sup.1 Yuto Ueda , .sup.1 Taku Doi , .sup.1 Keiko Nagatomo , .sup.2 Mayuko Takaki , .sup.4 Akira Nakajima , and .sup.3 James Willmore (Psychiatry, Miyazaki Medical College, University of Miyazaki, Miyazaki, Japan ; 1st Physiology, Miyazaki Medical College, University of Miyazaki, Miyazaki, Japan ; Pharmacology and Physiology, Saint Louis University School of Medicine, St. Louis, MO ; and Section of Chemistry, Department of Medical Science, Faculty of Medicine, University of Miyazaki, Miyazaki, Japan) Rationale: EAAC-1 functions in the re-uptake of extracellular glutamate that then leads to synthesis of GABA. Knockdown of EAAC-1 by anti-sense induces convulsive seizures in rats. GTRAP3-18 suppresses the binding affinity of glutamate with EAAC1 on the GABAergic neuron. The aim in this study was to measure the changes in GTRAP-18 expression induced by PTZ-kindling, and evaluate PTZ-kindling development in rats with knockdown of GTRAP3-18. Methods: Exp.1; Male Sprague-Dawley rats were kindled by repetitive i.p. injection of PTZ (16 mg/ml in saline, 40 mg/kg) at the rate of three times a week. Control animals were injected at the same rate with an equal volume of saline. Behavior was observed and scored following each injection. Animals were considered fully kindled after having two consecutive stage 5 seizures. Only fully kindled animals were used for the experiment. Six animals in each group were anesthetized with sodium pentobarbital and killed by decapitation on day 14 of the experiment. PTZ animals were sacrificed following a last stage 5 seizure. A crude membrane fraction extracted from both hippocampi was used for western blot followed by hybridized with GTRAP3-18 polyclonal antibody. Exp. 2; Rats underwent i.c.v. injection of oligonuleotide of anti-sense primer (n = 12) and sense primer (n = 12) for GTRAP3-18. Following these injections half of the rats in each group were used to measure the knockdown ratio of GTRAP3-18, while the remaining rats were used for PTZ-kindling. Results: Exp.1; In PTZ-kindled rats, GTRAP3-18 was reduced to 89% of control. Exp.2; GTRAP3-18 was decreased by about 40% of control in rats receiving the anti-sense oligonuleotide. Further, in these animals, PTZ seizure duration was significantly prolonged, latency more brief and kindling onset earlier when compared to control animals. Conclusions: Prior work with microdialysis from GTRAP3-18 antisense treated rats has shown that extracellular GABA concentration is markedly elevated. In this experiment, PTZ kindling over 14 days caused decrease in GTRAP3-18 protein. This down regulation of GTRAP3-18 may be a fundamental response to kindling to enhance GABA synthesis in compensation for the enduring process of epileptogenesis. However, GTRAP3-18 knockdown was associated with worse responses to PTZ. Our observation suggests that there is a pathological reversal of GABA action between inhibition and excitation that is induced by changes in Cl- gradient with enhanced GABA release. (Supported by Grant-in-Aid for Scientific Research (C) (2) (18591297) from the Ministry of Education, Science, Sport and Culture, Japan (to Y.U.).) Abstract: 1,2 Jana Veliskova , .sup.1 James G. Heida , and 1,2,3 Solomon L. Moshe (Departments of Neurology ; Neuroscience ; and Pediatrics, Albert Einstein College of Medicine, Bronx, NY) Rationale: Seizures affect males more often than females. The substantia nigra pars reticulata (SNR) belongs to an endogenous seizure-controlling network. Previous studies suggest that the SNR has age- and sex-specific features in terms of seizure control via activation of the GABA.sub.A receptor system. The sexual differentiation of the SNR is apparent during early development. At postnatal day 15 (P15), bilateral SNR infusions of the GABA.sub.A receptor agonist muscimol have a proconvulsant effect in males and no effect on flurothyl clonic seizure threshold in females. Exposure to testosterone during early postnatal development determines the male SNR response to muscimol. Here we investigated whether the testosterone effects on development of the male SNR response to muscimol are mediated via activation of androgen receptors during early postnatal development. Methods: In order to block the androgen receptors, male rat pups were injected daily with the androgen receptor antagonist flutamide (10 mg/kg i.p.) at P0-P2. Previously we have found that testosterone exposure during this time period leads to the male, proconvulsant effects of SNR muscimol infusions. At P13, we implanted cannulae in the SNR bilaterally using stereotaxic surgery. At P15, muscimol (100 ng/0.25 [mu]l per side) or saline (0.25[mu]l per side) was infused bilaterally 30 minutes prior to flurothyl seizure testing. The clonic seizure threshold was determined as the amount of flurothyl necessary to induce the seizure. Subsequently cannulae placements were confirmed histologically. Only rats with bilateral SNR canulae placements were used for statistical evaluation. Results: Muscimol infusions in the SNR of P15 male rats treated with flutamide at P0-P2 had no effect on flurothyl seizure threshold compared to saline-infused controls. This is in contrast to previously published data in naive male rats, in which SNR muscimol infusions have a proconvulsant effect on flurothyl clonic seizures. Conclusions: The results of the present study suggest that activation of androgen receptors during the early postnatal period participates in the development of proconvulsant SNR responses to muscimol. Understanding the mechanisms involved in sex-dependent sensitivity to seizures may lead to development of sex-specific treatment strategies. (Supported by NIH Grant NS 20253 and the Heffer Family Medical Foundation.) Abstract: .sup.1 Teresa Ravizza , .sup.1 Barbara Gagliardi , .sup.1 Francesco Noe' , .sup.2 Karin Boer , .sup.1 Nicola Marchi , .sup.2 Eleonora Aronica , and .sup.1 Annamaria Vezzani (Neuroscience, Mario Negri Institute for Pharmacological Research, Milano, Italy ; and Neuropathology, Academic Medical Center, University of Amsterdam, Amsterdam, Netherlands) Rationale: Inflammatory mediators are rapidly synthesized by glia and neurons in the rodent brain during seizures. The proinflammatory cytokine interleukin(IL)1beta increases epileptic activity and contributes to neurodegeneration. The aim of this study was to explore the relative contribution of innate and adaptive immune mechanisms in perpetuating brain inflammation during epileptogenesis and in the chronic phase of spontaneous seizures (SS). Methods: The immunohistochemical pattern of microglia (CD-11b/c, HLA-DR), monocytes/macrophages (ED-1, CD68), granulocytes (HIS48), T-(CD3, CD4, CD8) and B-lymphocytes (CD45RA), NK (NKR-P1A) was assessed in rats, 4-36 h (acute phase) and 3-7 d (epileptogenesis) after the onset of pilocarpine-induced status epilepticus (SE) and during SS, and in hippocampal specimens from medically intractable TLE with or without hippocampal sclerosis (HS). Results: Activated IL-1beta-expressing microglia and astrocytes were observed in rat forebrain within 4h after SE while abundant parenchymal macrophages were found after 18h. Both cell populations persisted until SS. Granulocytes appeared between 18h-3d only. Scarse B and T-lymphocytes and NK cells were found associated with microvessels, rarely in brain parenchyma. No changes occurred in pilocarpine+phenobarbital treated rats. In human TLE with HS, IL-1beta-expressing microglia and astrocytes, and clusters of perivascular monocytes/macrophages occurred in the hippocampus; T-lymphocytes were scarse and B cells were lacking. Conclusions: Activated glial cells expressing IL-1beta sustain the early phases of inflammation after SE and, toghether with macrophages, contribute to its persistence both in rat and human epileptic tissue. Granulocytes are transiently recruited, while B-, T-, NK-cells are scarse or absent. These findings, together with functional studies in experimental models, indicate that innate, but not adaptive, immunity significantly contributes to epileptogenesis in TLE. (Supported by Fondazione Mariani Onlus and Negri-Weizmann Programme (A.V.) and National Epilepsy Fund (E.A., K.B.)) Abstract: 1,2 Roi Ann Wallis , and 1,2 Kimberly L. Panizzon (Neurology, VA Greater Los Angeles Healthcare System, North Hills, CA ; and Neurology, David Geffin UCLA School of Medicine, Los Angeles, CA) Rationale: Mitochondria play a central role in energy production and processes of cell death. These organelles also regulate Ca.sup.2+ homeostasis, and thereby modulate neuronal excitability and synaptic transmission. More recently, mitochondrial dysfunction has been shown to occur in seizure foci from humans and experimental models of epilepsy. In addition, a number mitochondrial DNA mutations have been identified which lead to the inhibition of the mitochondrial respiratory chain. Because neuronal injury during prolonged seizure activity may occur via modulation of mitochondrial channels and mitochondrial membrane depolarization, we hypothesized that the opening of mitochondrial ATP-sensitive potassium (mitoK.sub.ATP) channels would provide protection against CA1 depolarization-induced injury. Methods: Using paired rat hippocampal slices, we monitored the CA1 orthodromic and antidromic population spike (PS) amplitude during depolarization injury with and without chromakalim treatment. To induce depolarization injury, slices were exposed to 25 mM KCl for 8 min. Treatment with chromakalim was begun 30 minutes prior to KCl exposure and continued for the first 15 min. of recovery. Results: Chromakalim, an opener of the mitoK.sub.ATP channel, provided robust neuroprotection of CA1 PS amplitude in hippocampal slices subjected to depolarization-induced injury. Slices exposed to 25 mM KCl demonstrated rapid loss of evoked response with a mean CA1 orthodromic and antidromic recovery of only 11% 3 and 13% 2, respectively. Treatment with chromakalim (300 uM) provided significant protection against this depolarization injury with CA1 orthodromic and antidromic PS amplitude recovering to 94% 3 and 95% 3. Treatment with 300 uM chromakalim during depolarization injury also produced significant recovery of mean excitatory post-synaptic potential slope (94%[+ or -] 6) after depolarization when compared to paired, unmedicated slices which did not recover (0%[+ or -] 0). Mean fiber volley responses were slightly resistant to depolarization injury with a mean recovery of 32%[+ or -] 2 in paired, unmedicated slices. In contrast, treatment with chromakalim showed full recovery with a mean of 100%[+ or -] 0. Conclusions: These studies demonstrate that chromakalim, a mitoK.sub.ATP channel opener, provides neuroprotection against CA1 depolarization injury. In addition, these data suggest that the use of agents that modulate the mitoK.sub.ATP channel may provide a useful strategy in the prevention of brain injury during status epilepticus. (Supported by VA Research Service.) Abstract: .sup.1 H. Jurgen Wenzel , .sup.1 Kimberley D. Katleba , .sup.1 Mareike E. Wenzel , .sup.2 James S. Trimmer , and .sup.1 Philip A. Schwartzkroin (Neurological Surgery, University of California, Davis, Davis, CA ; and Pharmacology, University of California, Davis, Davis, CA) Rationale: Gene therapy offers an alternative strategy for treating some CNS disorders, but has been relatively unexplored as a therapy for epilepsy. As background for a gene therapy approach, it is important to gain some understanding about the details of expression for those genes exogenously introduced into the brain. Toward this end, we have begun to characterize the efficiency, cell type specificity, and subcellular localization of Kv potassium channel subunits introduced into neurons in vitro. These studies will provide essential background for testing gene therapy approaches in intact preparations. Methods: Organotypic hippocampal slice cultures (OHSC) and dissociated primary neuronal cell cultures were prepared from wildtype mice and Kv1.1 knock-out mice (lacking the Kv1.1[alpha] subunit). Cultures were maintained for seven days in vitro, and then infected via an adenoviral vector or an HSV1 vector containing voltage-gated potassium (Kv) channel subunit genes (e.g., Kv1.1, Kv[beta]2, Kv1.4, Kv2.1). Cultures were fixed at 1-3 weeks post-infection, and subsequently processed for immunofluorescence and electron microscopic observation. Immuno-reacted cultures were analyzed for cell-type specific expression, subcellular distribution, and co-localization of Kv channel subunit expression. Results: In both primary dissociated cultures and OHSCs, adenoviral vectors successfully infected a variety of neuron types with Kv channel subunits. Exogenously-expressed subunits were observed in vitro at several days post-injection. Adenoviral gene transfer appeared to target larger populations of neurons than the HSV1 vector. Subcellular localization of exogenous subunits did not always reproduce the patterns seen for endogenous subunits. For example, endogenous Kv1.1 and Kv1.4 subunits are expressed only in the axons and terminals of hippocampal pyramidal and granule cells; however, these subunits appeared primarily in cell somata when expressed via viral infection. In the relatively complex and well-organized organotypic cultures, injections of adenoviral vectors resulted in infection of neurons primarily around the site of the injection. In these OHSCs, pyramidal cells, granule cells, mossy cells, and interneurons expressed the exogenous gene product. Conclusions: Viral vector-mediated infection with genes for Kv channel subunits resulted in successful in vitro infection of limited neuronal populations. Expression patterns of exogenous Kv subunits did not always mimick the subcellular localization patterns seen in vivo. Further studies are needed to optimize infection efficiency and to reproduce appropriate localization of expression for specific channel subunits. (Supported by American Epilepsy Society Research Initiative Award.) Abstract: .sup.1 Philip Williams , .sup.1 Andrew White , .sup.1 Kevin Staley , and .sup.2 Edward Dudek (Neurology and Pediatrics, University of Colorado, Denver, Health Sciences Center, Denver, CO ; and Physiology, School of Medicine, University of Utah, Salt Lake City, UT) Rationale: The latent period is usually considered to be a seizure-fee interval between an epileptogenic injury and the development of chronic seizures. If the chronic seizures occur at a stable frequency, then post-injury seizure probability can be considered a step function. We hypothesize 1) post-injury seizure probability is a continuous function and that at some point the seizure frequency goes through an exponential growth phase, and 2) that seizures tend to occur in clusters throughout the period of continuously increasing seizure frequency. Methods: We obtained nearly continuous electroencephalographic recordings using radiotelemetry in adult rats for 100 days after kainate-induced status epilepticus (n = 9). Results: In all 9 kainate-treated rats the seizures continuously increased with time after a mean latent period of 8 days, and all rats experienced an exponential increase in the seizure frequency at variable times after the latent period. A frequency histogram of the inter-seizure intervals throughout the period of seizure progression shows a distinct peak of intervals occurring at very short times. An analysis of the inter-seizure interval distribution compared to both a Poisson distribution and matched randomly generated intervals shows that the inter-seizure interval distribution is significantly different from a Poisson and random distribution. Conclusions: Seizure frequency continuously increases as a function of time after injury, suggesting that progressive molecular and cellular changes underlying seizure generation (e.g., axon sprouting and increased recurrent excitation) are steadily increasing with time after the injury. The exponential increase in seizure probability may reflect nonlinearities in the network response to steady increases in sprouting (see abstract by Swiercz et al.). These data also indicate that spontaneous seizures tend to occur in clusters throughout the period of increasing seizure frequency. (Supported by NIH.) Abstract: .sup.1 Nese Dericioglu , and .sup.2 Anne Williamson (Institute of Neurological Sciences, Hacettepe University, Ankara, Turkey ; and Neurosurgery, Yale University School of Med., New Haven, CT) Rationale: Epileptic tissue is characterized by a number of metabolic alterations. We have been investigating the effects of altered metabolism on GABAergic neurotransmission; here we have examined the physiological effects of blocking GABA synthesis using slices from both control and epileptic rats. We have used 3-mercaptoproprionic acid (3-MPA) a selective glutamic acid decarboxylase (GAD) inhibitor for these studies. Our primary hypothesis is that alterations in excitability will only be seen under activated conditions as GABA stores will need to be depleted. Methods: Adult male rats were used for these studies. We used the systemic kainic acid (KA) model of temporal lobe epilepsy. Hippocampal slices were prepared and maintained using standard methods. Field potential recordings were obtained from the CA1 region. Stimuli were delivered to stratum radiatum at an intensity twice that needed to evoke a 1 mV population spike. 3-MPA was washed on and recordings were made in different slices. The effect of repetitive stimulation was determined by giving 10 Hz, 10 sec trains and monitoring both evoked and spontaneous activity. Results: In control animals, 3-MPA was not associated with significant hyperexcitability in the absence of afferent input even after several hours exposure. However, following a series of stimulus trains there was a significant enhancement of the duration of the evoked response and the development of short bursts of population spikes. We assayed the role of neuronal GABA transport by blocking GABA uptake with NO-711. There was a further prolongation of the evoked response in NO-711 and 3- MPA demonstrating that neuronal transport can contribute to synaptic inhibition. However, we were unable to elicit spontaneous activity under either recording condition. We found that raising excitability slightly by bath applying 5 mM K+ ACSF, was associated with the development of interical-like activity in the presence of 3-MPA in all slices studied, however. These events became prolonged seizure-like events following repetitive stimulation. In KA-treated rats there was a significantly greater effect of 3-MPA on the duration of the evoked responses compared to controls. In addition, intracellular recordings demonstrated the development of spontaneous activity, unlike control animals. Conclusions: These data demonstrate that blocking GABA synthesis does not significantly alter neuronal circuit excitability in the absence of robust afferent input and/or a global elevation in tissue excitability. Thus, neurons appear to have plentiful reserves of GABA under baseline conditions in vitro. In epileptic animals, however, there appears to be a reduced inhibitory reserve under conditions when GABA synthesis is reduced. However, this may reflect the higher baseline levels of activity. (Supported by NIH NS045792 to AW and a Scientific and Technical Research Council of Turkey Fellowship to ND.) Abstract: .sup.1 Amy Willingham , .sup.1 Ashish Tokhi , .sup.1 Ryan Jones , and .sup.1 Peter R. Patrylo (Physiology, Southern Illinois University School of Medicine, Carbondale, IL) Rationale: The elderly have an enhanced incidence and prevalence of seizure disorders with up to 50% of cases having no identifiable antecedent. This has lead to the hypothesis that aging of the central nervous system (CNS) itself may be epileptogenic. To begin to assess this hypothesis, our laboratory compared the susceptibility of adult and aged rats to kainate-induced seizures and status epilepticus. This study revealed that aged rats have a greater frequency of wet-dog shakes and shorter latencies to onset for all classes of seizures and status epilepticus during systemic kainate-treatment. This experiment did not, however, distinguish whether this aging-related change is due to altered pharmacokinetics, blood brain barrier, and/or pharmacodynamics. Consequently, in this study the threshold for kainate induced seizures was examined in vitro. Methods: Hippocampal slices from adult (4-8 mo.) and aged (22-26 mo.) Fischer 344 rats were examined using field potential recordings from CA3 to assess activity evoked with single or paired orthodromic stimuli. The threshold for kainate-induced epileptiform activity was examined by increasing the kainate concentration (12.5, 25, 50, and 100 nM) while monitoring evoked and spontaneous activity. Results: In aCSF, no difference in the maximal amplitude of the evoked response was noted in aged (n = 24; 6.6 [+ or -] 0.6 mV) vs. adult slices (n = 25; 6 [+ or -] 0.5 mV; p > 0.4; ANOVA). Evoked activity usually entailed a population spike followed by a positive field potential shift that often had a secondary small amplitude spike. In two of the aged slices, but none of those from adults, stimulating at a supra maximal intensity elicited an epileptiform burst. When paired stimuli were delivered using a 50 ms interpulse interval, 92% of aged slices (n = 13) and 31% of adult slices (n = 16) showed facilitation of the second spike (p < 0.001; chi-square). To assess the threshold for kainate-induced epileptiform activity, 10 slices from adult and 10 slices from aged rats were examined. In total, 80% of adult and 90% of aged slices showed epileptiform activity by the time the kainate concentration reached 100 nM. The kainate concentrations required for evoked (9.7 [+ or -] 2.8 nM vs. 21.9 [+ or -] 4.6 nM; p < 0.05; ANOVA) and spontaneous (20.3 [+ or -] 4.7 nM vs. 34.4 [+ or -] 4.6 nM; p < 0.05; ANOVA) epileptiform activity were significantly lower in slices from aged rodents, compared to adults. Conclusions: These data indicate that the aged CNS itself has an enhanced susceptibility to kainate-induced epileptiform activity that could contribute to the increase in wet-dog shakes and reduced latency to onset for class I/II seizures and status epilepticus in aged rodents during systemic kainate treatment. Further, if these pharmacodynamic changes also occur in the human CNS during aging, they could contribute to the increased incidence and prevalence of idiopathic seizure disorders. (Supported by the Epilepsy Foundation of America and NIA.) Abstract: .sup.6 L. James Willmore , .sup.1 Yuto Ueda , 2,3 Hidekatsu Yokoyama , .sup.4 Akira Nakajima , .sup.5 Maiko Takaki , .sup.1 Keiko Nagatomo , and .sup.1 Taku Doi (Department of Psychiatry, University of Miyazaki, Miyazaki, Japan ; National Institute of Advanced Industrial Science and Technology (AIST), Nagoya, Jordan ; Institute for Life Support Technology, Yamagata, Japan ; Section of Chemistry, Department of Medical Science, University of Miyazaki, Miyazaki, Japan ; Department of 1st Physiology, University of Miyazaki, Miyazaki, Japan ; and Departments of Neurology and Pharmacology and Physiology, Saint Louis University School of Medicine, Saint Louis, MO) Rationale: Epileptogenesis is associated with collapse of glutamate regulation, alteration in redox state and generation of free radicals. However, measurement of reductants in brain tissue extracts is problematic because of intrinsic free radical scavenging. Using in vivo methods, we measured hippocampal antioxidant ability from the decay rate of i.p injection of nitroxide radical with 700 MHz Electron Paramagnetic Resonance (EPR) spectroscopy during amygdalar ferric chloride-induced chronic recurrent seizures in rats. Methods: We used Region-Selected Intensity Determination (RSID) methods for estimation of nitroxide decay ratio. Hippocampal antioxidant function was estimated based on the half-life of the EPR signal of the blood-brain barrier-permeable nitroxide radical. Rats underwent unilateral right amygdalar injection with aqueous ferric chloride. Recording from chronically implanted depth electrodes showed development of epileptiform discharges, followed by recurrent seizures arising from amygdalar regions with propagation into both hippocampi. EPR spectroscopy with RSID estimation was performed at 5 days, 15 days and 30 days after FeCl.sub.3 injection. Results: In bilateral dorsal hippocampal regions, in vivo antioxidant ability was significantly decreased at 5 days and 30 days in animals with chronic, recurrent seizures when compared to controls. Conclusions: We observed impairment of antioxidant function in the hippocampi of rats with epileptogenesis induced by amygdalar ferric chloride injection. Of interest, unilateral induction of a chronic seizure focus was associated with bilateral alterations both early after injection and late. We wonder if hippocampal collapse of glutamate regulation during epileptogenesis might be followed by consumption of endogenous antioxidants. (Supported by: Grant-in-Aid for Scientific Research (C)(2)(16591146) from the Ministry of Education, Science, Sport and Culture, Japan (to Y.U.).) Abstract: .sup.1 Yannan Ouyang , .sup.1 Xiao-Feng Yang , .sup.1 Xiao Yan Hu , .sup.1 Ebru Erbayat-Altay , .sup.1 Linghui Zeng , .sup.1 Jin-Moo Lee , and .sup.1 Michael Wong (Neurology, Washington University, Saint Louis, MO) Rationale: Seizures may injure the brain and contribute to neurological and cognitive deficits of epilepsy patients. Although it is well-established that seizures may cause neuronal death under some conditions, seizures may also induce more subtle "non-lethal" pathophysiological changes in neuronal structure and function, including abnormalities in synaptic transmission. In particular, seizures have been shown to cause morphological changes in dendrites and dendritic spines, but the molecular mechanisms mediating these structural changes are poorly understood. Actin represents a major structural protein of dendrites and actin filaments (F-actin) can be depolymerized by the regulatory molecule, cofilin, leading to structural changes in dendrites. Thus, we examined changes in F-actin and cofilin in hippocampus due to 4-aminopyridine (4-AP)-induced seizure activity in vivo and in vitro. Methods: Seizures were induced with intrahippocampal injections of 4-AP in mice in vivo. Epileptiform activity was elicited by superfusion of 4-AP over hippocampal slices in vitro. The gross morphology of dendrites and the number of dendritic spines following one hour of seizure activity was assessed by confocal imaging of fixed hippocampal sections from mice with in vivo seizures or by time-lapse multiphoton imaging of living hippocampal slices during in vitro epileptiform activity. Changes in F-actin levels in hippocampus following seizures was assessed by the phalloidin-rhodamine immunolabeling technique. Total actin, total cofilin, and phosphorylated (inactive) cofilin (p-cofilin) were measured following seizures by Western blotting. Results: Following one hour of in vivo or in vitro seizure activity, there was no significant difference in the gross morphological properties of dendrites and number of dendritic spines in hippocampus. However, hippocampal seizures led to a significant decrease (30%) in F-actin levels, indicating a depolymerization of actin filaments. Correspondingly, seizure activity caused a significant decrease (64%) in p-cofilin, but no change in total cofilin levels, implying an increase in unphosporylated (active) cofilin, which should depolymerize F-actin. By comparison, elevated extracellular potassium (50 mM KCl) induced a rapid swelling and loss of spines in hippocampal slices and a strong decrease (93%) in p-cofilin. Conclusions: Seizures may induce activation of cofilin and depolymerization of F-actin. Although no gross changes in dendritic structure or spine number were observed with seizures over a short time period, these changes in actin dynamics with seizures could be related to dendritic spine loss over a longer time period, more subtle changes in spine morphology, or non-structural roles of actin in synaptic transmission. Additional studies are required to define the functional and clinical relevance of changes in actin polymerization in seizure-induced brain injury. (Supported by NIH K02 NS045583 (MW).) Abstract: .sup.1 Hong Xu , .sup.1 Guo Yin , .sup.1 Peter Jukkola , .sup.1 Elena Kharlamov , .sup.1 Kebin Zeng , and .sup.1 Kevin M. Kelly (Department of Neurology, Allegheny General Hospital, Center for Neuroscience Research, Allegheny-Singer Research Institute, Pittsburgh, PA; and Drexel University College of Medicine Philadelphia, PA, Philadelphia, PA) Rationale: Acute ischemic infarction of the neocortex is typically followed by functional reorganization of cortical circuitry. In the periinfarction area, GABAergic-mediated inhibitory neurotransmission is impaired and neuronal activity is enhanced, changes that may predispose to epileptogenesis and poststroke epilepsy. We used the model of cortical photothrombosis and brain infarction to examine the changes of GABA.sub.A receptor-mediated inhibitory synaptic input to pyramidal neurons of layer II-III of ipsilateral sensorimotor cortex 1-2 mm lateral to the infarct 7-10 days after photothrombosis of 4-month-old Fischer 344 rats. Methods: Whole cell patch-clamp techniques were used to record postsynaptic currents (PSCs), and spontaneous and miniature inhibitory postsynaptic currents (sIPSCs and mIPSCs) of pyramidal neurons of layer II-III of ipsilateral sensorimotor cortex 1-2 mm lateral to the infarct. Analysis was performed on PSCs and sIPSCs recorded from 46 pyramidal neurons (lesioned, n = 23; control, n = 23) and mIPSCs recorded from 33 of the 46 pyramidal neurons (lesioned, n = 18; control, n = 15). Results: In lesioned animals, PSC amplitude was unchanged compared with controls, whereas decreases were observed for sIPSC ([proportional to]23% lower than that of control, 24.93 [+ or -] 2.23 pA in lesioned vs.32.54 [+ or -] 2.98 pA in controls) and mIPSC ([proportional to]22% lower than that of control, 20.12 [+ or -] 1.27 pA in lesioned vs. 25.72 [+ or -] 2.50 pA in controls) amplitude. There was an increase in the 10-90% rise time of PSCs ([proportional to]31% longer than that of control), sIPSCs ([proportional to]36% longer than that of control) and mIPSCs ([proportional to]29% longer than that of control) in lesioned animals, whereas decay time constant and charge transfer were not significantly different compared with controls. There was an increase in PSC frequency ([proportional to]41% higher than that of control, 3.10 [+ or -] 0.28 Hz in lesioned vs. 2.17 [+ or -] 0.23 Hz in controls) in lesioned animals, whereas sIPSC ([proportional to]21% higher) and mIPSC ([proportional to]18% higher) frequency demonstrated only trends toward increases. Additionally, sIPSC and mIPSC conductance was decreased (23% and 21.7%, respectively). Conclusions: These findings indicate that cortical photothrombosis results in alterations of GABAergic inhibition of pyramidal neurons in close proximity to the infarct, which likely contribute to the increased excitatory tone of the area following ischemic infarction and may predispose it to the mechanisms of epileptogenesis. (Supported by R01 NS046015 to KMK.) Abstract: .sup.1 Xiao-Feng Yang , .sup.1 Liu Lin Thio , .sup.1 Steven M. Rothman , .sup.1 Aryan E. Weisenfeld , and .sup.1 Kelvin A. Yamada (Neurology, Washington University, Saint Louis, MO) Rationale: Previous studies have demonstrated that leptin inhibits epileptiform-like activity in rat hippocampal neurons. Therefore, we examined whether leptin acts as an anticonvulsant in an in vivo rat model of focal neocortical seizures. Methods: Focal neocortical seizures were induced by the microinjection of 1 [mu]L artificial cerebrospinal fluid containing 12.5 mM 4-aminopyridine (4-AP, 12.5 nanomoles) into the motor cortex of halothane-anesthetized, 4-6 week old male Sprague-Dawley rats. Leptin-treated rats were co-injected with 39 [mu]M leptin (39 picomoles). A screw electrode was placed symmetrically over each hemisphere and used to record the electroencephalogram (EEG) in a referential montage. We recorded EEG prior to the injection of 4-AP and continued recording for one-hour after the injection. The number of seizures, the average seizure duration, and the cumulative seizure duration during the one hour recording period after 4-AP injection were determined by analyzing the EEG using commercial software (pClamp 9, Axon Instruments). Results: Leptin reduced the number of seizures from 33.8 [+ or -] 15.8 seizures/hour to 10.2 [+ or -] 3.3 seizures/hour (n = 5 rats/each group, p = 0.012). Leptin decreased the average seizure duration from 34.3 [+ or -] 4.9 seconds to 14.5 [+ or -] 6.0 seconds (p < 0.001). It also decreased the cumulative seizure duration from 1141 [+ or -] 558 seconds to 156 [+ or -] 104 seconds (p < 0.01). Conclusions: These results show that leptin has anticonvulsant effects in the in vivo 4-AP model of severe focal neocortical seizures. Thus, leptin and other leptin receptor agonists may be clinically useful anticonvulsants. (Supported by JDRF 1-2004-594 and NIH NS 32636 (KAY).) Abstract: .sup.1 Audrey S. Yee , .sup.2 Lihong Diao , and 1,2 Kevin J. Staley (Pediatrics, University of Colorado at Denver and Health Sciences Center, Denver, CO ; and Neurology, University of Colorado at Denver and Health Sciences Center, Denver, CO) Rationale: The spontaneously active in-vitro CA3 slice preparation is an important model of interictal spikes. The temporal pattern of interictal spikes reflects seizure probability in experimental epilepsy. We investigated determinants of CA3 burst timing to understand mechanisms of interictal timing. In numerous paradigms used to initiate bursts, we noted a significant increase in the interburst interval in the 1.sup.st hour following burst onset. The mechanisms for this finding are unclear and we investigate them here. Methods: We used rat hippocampal slices (male Sprague-Dawley 4-8 weeks, 400 uM sections prepared with a tissue chopper, slices cut and stored in ACSF) and extracellular field recordings to record bursting from the CA3 pyramidal cell layer. This bursting was induced in modified ACSF (3.3 K.sup.+, 0.9 Mg.sup.2+, 1.5 Ca.sup.2+) by addition of: (1) Pictrotoxin (PTX) 100 uM + CGP 1 uM (2) Bicuculline Methiodine (BMI) 50 uM + CGP or (3) Gabazine (GBZ) 10 uM + CGP to block GABA.sub.A and GABA.sub.B receptors. In other experiments, slices were co-treated with a glutamate precursor, glutamine (10 uM) or pretreated with the NMDA receptor (NMDAR) antagonist, d-APV 100 uM, prior to GABA.sub.A and GABA.sub.B blockade. Changes in burst frequency (measured as interburst interval) and burst length were examined. Results: 1. Synchronized network activity can occur in the presence of postsynaptic inhibition. 2. Burst frequency increased during the first hour following burst initiation. 3. The initial burst frequency increase was present regardless of the type of GABA.sub.A receptor blockade (BMI, PTX, and GBZ). 4. This initial burst frequency increase occurred in the presence of glutamine replenishment. 5. This initial burst frequency increase occurs in the presence of NMDAR blockade, suggesting that burst-related synaptic plasticity does not explain the initial increase in burst frequency. 6. When the degree of post-synaptic inhibition was increased (see concentration- response curve), initial burst frequency increase was diminished. Conclusions: These results suggest that inhomogeneous GABA.sub.A receptor function is reflected in altered burst frequency in the 1.sup.st hour after burst induction. Similar inhomogeneities may be present in-vivo early in epileptogenesis. (Supported by NINDS.) Abstract: .sup.1 Nicole A. Young , .sup.1 Corey Flynn , .sup.1 Lana J. Ozen , and .sup.1 G. Campbell Teskey (Psychology, University of Calgary, Calgary, AB, Canada) Rationale: Functional alterations in neural networks are thought to underlie both susceptibility to seizures and interictal behaviours. Our laboratory has previously shown that electrical kindling of the rat neocortex, amygdala or hippocampus results in a dramatic expansion of motor maps. The aim of the current study was to determine if those observations extend to another model of epilepsy. In this study we employed the pilocarpine-induced status epilepticus model that results in extended seizure activity and, after a maturation interval, can ultimately lead to spontaneous seizures. Methods: Young adult male rats were injected with atropine methyl-nitrate (10.0 mg/kg) 30 minutes prior to convulsant treatment in order to block peripheral affects. Pilocarpine (320.0 mg/kg) was administered intraperitoneally at a dose that elicited seizures in all rats with approximately 50% expressing status epilepticus. Control rats received physiological saline. All rats were videotaped for 2 hours following convulsant administration or control. Seizure behaviours were quantified by duration and type. Diazepam (10.0 mg/kg) was administered to all rats 1.5 hours following convulsant or control injections to terminate seizure activity and promote survival. Rats had high-resolution intracortical microstimulation movement representations derived using standard methodologies at either 48 hours, 1 week, 2 weeks or 3 weeks following treatment. Results: Our preliminary results show that rats have motor map expansion following pilocarpine treatment, regardless of whether status epilepticus occurred or not. Conclusions: These results demonstrate that a single episode of seizure activity induced by a convulsant agent can alter the functional expression of motor maps. (Supported by Natural Sciences and Engineering Research Council of Canada.) Abstract: .sup.1 Hai Xia Zhang , and 1,2 Liu Lin Thio (Neurology, Washington University, St. Louis, MO ; and Pediatric Neurology, St. Louis Children's Hospital, St. Louis, MO) Rationale: Although Zn.sup.2+ is a potent and efficacious biphasic modulator of [alpha].sub.2-containing strychnine sensitive glycine receptors (GlyRs), the physiological role of this modulation is unknown. Therefore, we examined whether potentiating Zn.sup.2+ concentrations enhance GlyR mediated inhibition of neuronal bursting induced by a Mg.sup.2+ free extracellular solution. Methods: Current clamp recordings from cultured embryonic mouse hippocampal neurons were obtained using the whole-cell patch clamp technique. Bursts consisting of 2-10 second depolarizing envelopes with superimposed action potentials were induced by omitting extracellular Mg.sup.2+ ("0" mM Mg.sup.2+). The basal Zn.sup.2+ concentration in our extracellular solution was 0.8 [mu]M. A nominally Zn.sup.2+ free extracellular solution ("0"[mu]M Zn.sup.2+) was obtained by adding 10 mM tricine, a Zn.sup.2+ chelator. Ten [mu]M D-serine was added to the extracellular solution to saturate the NMDA receptor glycine binding site. Statistical comparisons were made by ANOVA. Results: In an extracellular solution containing 1 mM Mg.sup.2++"0"[mu]M or 0.8 [mu]M Zn.sup.2+, no bursts occurred and action potentials occurred at 0.2 [+ or -] 0.1 Hz (S.E). In "0" mM Mg.sup.2++ 0.8 [mu]M Zn.sup.2+, 11 [+ or -] 0.8 bursts occurred per minute, and the action potential frequency was 3.5 [+ or -] 0.4 Hz (Figure). Omitting Mg.sup.2+ and Zn.sup.2+ from the extracellular solution did not change burst or action potential frequency, which were 12 [+ or -] 0.8 per minute and 2.7 [+ or -] 0.4 Hz, respectively. In "0"[mu]M Mg.sup.2++ 0.8 [mu]M Zn.sup.2+, 20 [mu]M glycine decreased burst frequency from 11 [+ or -] 1.3 per minute to 1.4 [+ or -] 0.6 per minute (p < 0.05) and decreased action potential frequency from 2.9 [+ or -] 0.8 Hz to 0.3 [+ or -] 0.1 Hz (p < 0.05) (Figure, top panel). In "0" Mg.sup.2++"0"[mu]M Zn.sup.2+, 20 [mu]M glycine decreased burst frequency from 13 [+ or -] 0.8 per minute to 7.8 [+ or -] 0.8 per minute (p < 0.05) and decreased action potential frequency from 2.6 [+ or -] 0.5 Hz to 1.9 [+ or -] 0.5 Hz (p < 0.05) (Figure, bottom panel). Although baseline burst and action potential frequency in 0.8 [mu]M and "0"[mu]M Zn.sup.2+ did not differ, 20 [mu]M glycine decreased burst and action potential frequency in 0.8 [mu]M Zn.sup.2+ more than in "0"[mu]M Zn.sup.2+ (p < 0.05). Similar results were obtained with 200 [mu]M taurine. These inhibitory effects of glycine and taurine were blocked by 1 [mu]M strychnine. Conclusions: These results indicate that potentiating Zn.sup.2+ concentrations enhance GlyR mediated depression of neuronal hyperexcitability in pathological conditions. Thus, designing drugs that modulate GlyRs directly or via Zn.sup.2+ may lead to novel anticonvulsants. [figure 1] (Supported by NIH and Washington University Department of Neurology.) Abstract: .sup.1 Mingrui Zhao , .sup.1 Hongtao Ma , .sup.1 Challon Perry , .sup.1 Minah Suh , and .sup.1 Theodore H. Schwartz (Neurological Surgery, Weill Cornell Medical College, New York Presbyterian Hospital, New York, NY) Rationale: Epileptic events elicit an increase in neuronal metabolism and an increase in cerebral blood flow (CBF), bringing oxygenated hemoglobin to the activated neurons. Whether CBF and oxygenation are adequate to meet metabolic demand in the focus and surround is unknown. Methods: We induced acute focal seizures in the rat neocortex by injection of 4-aminopyridine (4-AP, 15 mM, 0.5 [mu]l). The local field potential was recorded to identify the ictal discharge. The partial pressure of tissue oxygen (pO2) and CBF were measured with two Clark-style polarographic oxygen microelectrodes and two probes of laser Doppler flowmetry located in the focus and surround. Optical spectroscopic imaging was used to measure deoxyhemoglobin (Hbr), oxyhemoglobin (HbO2) and total hemoglobin (Hbt). Results: The duration of ictal discharge ranged from 50 to 200 sec (n = 15 rats). CBF in the focus increased 150.0 [+ or -] 13.7% (p < 0.05, n = 4 rats) but decreased in the surround by 90.9 [+ or -] 2.5% (p < 0.03, n = 4 rats). Tissue pO2 decreased by 64.3 [+ or -] 7.8% (p < 0.05, n = 5 rats) in the focus during the ictal discharge and then increased at the termination of ictal event. In the surround tissue pO2 increased by 120.5 [+ or -] 5.1% (p < 0.05, n = 5 rats). The optical spectroscopic analysis showed the spatial extent of the perfusion and oxymetric responses. Conclusions: These results demonstrated a prolonged epileptic dip in both tissue and hemoglobin oxygenation during ictal events in spite of a dramatic increase in CBF in the region of focus. In the surround, pO2 increased in spite of a decrease in CBF, consistent with a decrease in neuronal activity and metabolism. (Supported by NIH NS043799-04 and NS049482-01].) Abstract: .sup.1 Shuangping Zhao , .sup.1 Steve S. Shinmei , .sup.1 Ernesto R. Aviles Jr. , and 1,2,3 Denson G. Fujikawa (Neurology, VA Greater Los Angeles Healthcare System, North Hills, CA ; Neurology, UCLA Geffen School of Medicine, Los Angeles, CA ; and Brain Research Institute, UCLA Geffen School of Medicine, Los Angeles, CA) Rationale: Seizure-induced neuronal death is morphologically necrotic and caspase-independent, despite internucleosomal DNA cleavage (DNA laddering), a programmed process. We studied whether seizure-induced neuronal necrosis involves translocation of mitochondrial apoptosis-inducing factor (AIF) and endonuclease G (endoG) to the nucleus and cytochrome c (cyt c) to the cytoplasm, where AIF and endoG could be involved in large-scale (50 kb) and internucleosomal DNA fragmentation respectively. In caspase-independent, excitotoxic neuronal death, cyt c translocation from mitochondria to cytoplasm may occur because of non-specific outer membrane rupture from mitochondrial swelling. Methods: Adult Wistar rats were subjected to 3-h lithium-pilocarpine-induced status epilepticus (LPCSE), after which seizures were stopped with diazepam and phenobarbital. Six and 24 h later rats underwent in situ transcardiac perfusion-fixation of their brains, with subsequent removal and processing for H & E staining, TUNEL and AIF, endoG and cyt c immunoreactivity (n = 3 in each group). Twenty-six brain regions were assessed for the degree of neuronal death, using H & E and TUNEL, and for translocation of AIF, endoG and cyt c, by brightfield immunohistochemical and immunofluorescent light-microscopic examination of the piriform cortex, which is one of the most damaged regions in the brain. Results: The extent of neuronal necrosis and TUNEL positivity was similar to what we have described previously. Twenty-four h after SE, 19 of 26 brain regions had significant numbers of acidophilic neurons by H & E staining, and 10 of 26 regions had TUNEL-positive necrotic neurons. In the piriform cortical neurons of control rats AIF, endoG and cyt c appeared in a punctate distribution in the cytoplasm. Unexpectedly, 6 and 24 h after SE, cyt c, as well as AIF and endoG, had translocated to the pyknotic nuclei of necrotic neurons, and there was diffuse staining in the cytoplasm. Thus, in addition to the nuclear translocation of lysosomal cathepsin B and DNase II, which we reported last year, nuclear translocation of mitochondrial AIF, endoG and cyt c also occurred. Conclusions: Nuclear AIF and endoG could contribute to DNA fragmentation. Nuclear translocation of cyt c was unexpected, and has only been reported previously in HeLa cells in culture exposed to apoptotic stimuli, where it was associated with release of acetylated histone H2A from nucleus to cytoplasm and chromatin condensation. We are currently investigating earlier time points to determine when nuclear translocation first appears. Our results suggest that caspase-independent programmed mechanisms contribute to seizure-induced neuronal necrosis. (Supported by Department of Veterans Affairs.) Abstract: .sup.1 Jokubas Ziburkus , .sup.1 John R. Cressman , .sup.1 Rozangela Lazaneo , and 1,2 Schiff J. Schiff (Krasnow Institute, George Mason University, Fairfax, VA ; and Departments of Neurosurgery, Engineering Science and Mechanics, The Pennsylvania State University, University Park, PA) Rationale: We recently described a novel pattern of interleaving excitatory-inhibitory (EI) activity (Ziburkus et al., J Neurophys., 2006) and calculated dynamic EI conductances during spontaneous in vitro seizures. The goal of the present study is to describe pharmacological effects, including gap junction blockade on inhibitory and excitatory conductances and EI network synchrony. Methods: Simultaneous single and dual voltage clamp whole-cell and extracellular recordings were performed in the CA1 region of rat transverse hippocampal slice preparations (n = P16-P30). To induce seizures, slices were bathed in 100-200 [mu]M 4-Aminopyridine in decreased magnesium (0.6 mM). Intracellular solution contained blockers for voltage-gated conductances. Excitatory conductances were recorded at -80 to -50 mV and inhibitory at 0 to +40 mV. Dynamic reversal potentials for interictal bursts and seizures were calculated using varying holding potential (-80 mV to +40 mV). Post-hoc, the cells were stained and identified using fluorescence conjugated antibody cocktails (neurobiotin, somatostatin, parvalbumin, and mGLUR1). Following immunocytochemical testing, the sections were processed for peroxidase-based neurobiotin visualization used for digital morphological reconstructions of the recorded cells. Results: We performed measurements of dynamic synaptic inhibitory (Gi) and excitatory (Ge) conductances in E (n = 12) and I (n = 12) cells during seizures. During the initiation of the ictal events, inhibitory conductances were even or greater to the excitatory conductances. Inhibitory conductances peaked 1-2 seconds before the ictal discharge of pyramidal cells. During the persistent stage of the ictal event, Gi transiently diminished, and then increased with Ge during the post-ictal bursting. Bursts and ictal events were observed in the presence of any one of NMDA (APV), AMPA (CNQX), or GABAa (picrotoxin or bicuculline) receptor blockers. A cocktail of excitatory blockers was necessary to completely abolish bursting and ictal events. Gap junction blockers (mefloquine, n = 14 and carbenoxolone, n = 12) had profound effects on E and I network synchronization, inhibitory currents, and seizure blockade. We measured effects of the individual blockers on burst dynamics and synchronization. Mefloquine preferentially affected inhibitory network synchronization (n = 6), and both carbenoxolone and mefloquine had as strong of an effect on seizures as a mixture of excitatory conductance blockers. Conclusions: We report mechanisms of the first comparative interneuron-principal cell conductance study in a spontaneous in vitro seizure model. Gap junction blockers exhibit a potent effect on neuronal network synchronization and block 4-AP seizures. (Supported by Epilepsy Foundation Fellowship (JZ) and NIH grants: K02MH01493, and RO1MH50006.) Abstract: .sup.1 Tiina Alapirtti , .sup.1 Janne Hulkkonen , .sup.1 Riikka Makinen , and .sup.1 Jukka Peltola (Department of Neurology and Rehabilitation, Tampere Univercity Hospital, Tampere, Finland; Department of Clinical Physiology, Tampere Univercity Hospital, Tampere, Finland; Department of Clinical Neurophysiology, Tampere Univercity Hospital, Tampere, Finland; and Department of Neurology and Rehabilitation, Tampere Univercity Hospital, Tampere, Finland) Rationale: Experimental and clinical studies have shown that prolonged seizures result in increased cytokine production in the central nervous system. However, the effect of seizure type on cytokine production is not well established. The purpose of this study is to examine the concentrations of two key cytokines in patients with temporal and extratemporal epilepsy undergoing a video-EEG study. Methods: Here we determined the levels of interleukin-6 (IL-6) and interleukin-1 receptor antagonist (IL-1RA) in serum from 20 patients before and after seizure during continuous video-EEG monitoring in Tampere University Hospital. In addition to baseline the blood samples were taken 3, 6, 12 and 24 hours after the seizure. The definition of the seizure focus and classification of epilepsy was based on concordant findings in this investigation, and on MRI examination. Patients were divided into two groups: temporal lobe epilepsy (n = 11), and extratemporal epilepsy (n = 9). Results: In the patients with seizures originating from temporal (p < 0.01) and extratemporal lobes (p < 0.05) there was a statistically significant increase in the levels of IL-6 production after the seizure. Moreover, the concentrations of IL-6 were significantly higher in temporal lobe group compared with extratemporal group (p = 0,016). In the patients with temporal seizures there was a significant decrease in the levels of IL-1RA production (p < 0.05) but not in the patients with extratemporal seizures. Conclusions: The present study demonstrates that single seizures in chronic epilepsy are followed by activation of cytokine network. The production of IL-1RA and IL-6 are influenced by the origin of the seizures from either temporal or extratemporal lobes. Experimental studies have demonstrated that IL-1RA may act as an anticonvulsant and neuroprotectant, whereas IL-6 seems to be proconvulsant and neurotoxic. Differential activation of these key cytokines may modify central nervous system reactions to seizures. Abstract: .sup.1 Mohammed Ali Alhakami , .sup.1 Neelan Pillay , and .sup.1 Sam Wiebe (Clinical Neurosciences, University of Calgary, Calgary, AB, Canada) Rationale: The coexistence of localization related and generalized epilepsy is rare ( Methods: Eighteen patients fulfilled our criteria, which required the coexistence of interictal focal and truly generalized EDs in the same or in different EEGs. Generalized EDs with shifting focal fragments, and focal discharges with secondary bilateral synchrony were excluded. We reviewed seizure semiology, history of febrile seizures, family history of epilepsy, AEDs used, type of investigations, whether epilepsy surgery was done and seizures outcomes. Results: Eleven patients (61%) were female. Eight were < 10 years old. Generalized EDs consisted of [greater than or equal to]3Hz spike-wave and polyspike-wave. Focal spikes had a variety of distributions. All but two patients had video EEG monitoring. Five patients (27.7%) had a family history of epilepsy, and the neurological examination was abnormal in two patients (11%). Ten patients (55.5%) had normal MRI. Five patients had epilepsy surgery, two of whom had mesial temporal sclerosis(MTS). One patient with left temporal EDs and MTS, underwent a conventional temporal lobe resection. Postoperatively, the patient developed seizures of generalized semiology, confirmed with video EEG. Seizure control was poor in the majority of our patients. Conclusions: The coexistence of focal and strictly defined generalized EDs may be a marker of poor seizure control. In patients with localization related epilepsy, this EEG pattern may indicate poor postoperative seizure outcomes. Rarely, failed epilepsy surgery may indicate the emergence or presence of generalized epilepsy. Abstract: .sup.1 Gail D. Anderson , .sup.2 Nancy R. Temkin , .sup.3 Ramon Diaz-Arrostia , .sup.2 Carol Farhrenbruch , .sup.4 Michael Edenfield , .sup.5 John W. Miller , and .sup.4 S.M. Hossein Sadrzadeh (Pharmacy, University of Washington, Seattle, WA ; Neurological Surgery, University of Washington, Seattle, WA ; Neurology, University of Texas SW Med. School, Dallas, TX ; and Laboratory Medicine, University of Washington, Seattle, WA) Rationale: Post-traumatic seizures (PTS) are a complication of head trauma. Hemorrhagic injury to the cortex has an increased risk of PTS, possibly due to reaction of free hemoglobin (Hb) with lipid rich environment. Hb and iron can enhance oxidative stress. The brain is susceptible to oxidative damage due to is high oxygen consumption, high lipid content and low antioxidant activity. Haptoglobin (Hp) is a plasma glycoprotein that removes free Hb and prevents Hb-induced damage. Hp has three different phenotypes, HP-1-, Hp 2-1 and Hp 2-2. Hp 1-1 is biologically the most and Hp 2-2 is the least active in Hb binding, antioxidant activity. Polymorphisms of apoliprotein E (APOE) plays an important role in neuropathological finding in patients with head injuries. Of the three common alleles, e2, e3 and e4, the presence of e4 has been shown to have a less favorable outcome. We hypothesized that patients who develop PTS have a 1)lower concentration of haptoglobin and/or a higher frequency of low capacity haptoglobin phenotype; 2)higher oxidant stress as reflected in increased 8-isoprostane (8-isoP), a lower total antioxidant capacity (TAOC); 3)higher frequency of APOE e4 allele. Methods: Patients were recruited from our valproate prophylaxis study (Temkin et al. J Neurosurg 1999;91:593-600). Of the 51 who had developed PTS, 25 cases and 26 controls, matched for age, sex, PTS risk factor and treatment group. Baseline information on seizure history, current alcohol use and medications, and 50 ml of blood were obtained. Plasma/serum were used for the biochemical analysis and blood cells for genetic analysis. Statistical analysis was performed using unpaired t-test for the biochemical markers and Chi-square analysis for the genotyping. Results: PTS Cases vs.Controls 8-IsoP: 246 [+ or -] 170 pg/ml vs 172 [+ or -] 152 pg/ml (p = 0.1) TAOC: 1.2 [+ or -] 0.1 mM vs 1.1 [+ or -] 0.1 mM (p = 0.38) Hp: 122 [+ or -] 39 mg/dl vs 142 [+ or -] 69 mg/dl (p = 0.22) Ferritin 96 [+ or -] 87 [mu]g/ml vs 154 [+ or -] 131 [mu]g/ml (p = 0.069). Conclusions: We found no difference in Hp concentration or phenotype, oxidative activity or capacity, or APOE genotype in the head trauma patients who developed PTS. There was a trend towards an increased oxidative stress and decreased ferritin concentrations. In contrast to microhemorrhagic gradual brain injury that may be mediated by free Hb/iron, direct physical damage in this case may be the most contributing factor in developing PTS. (Supported by Research Initiative Award from the American Epilepsy Society.) Abstract: .sup.1 David J. Anschel , .sup.1 Brian Pike , .sup.2 Sylvia Dolce , and .sup.3 Armin Scwartzman (Neurology, State University of New York at Stony Brook, Stony Brook, NY ; Art for Health Program, Stanford University Medical Center, Stanford, CA ; and Statistics, Stanford University, Stanford, CA) Rationale: Language shapes all aspects of human interaction. The assessment of written language may provide a unique insight into the functioning of the human mind and brain. Methods: 60 subjects were recruited from a Comprehensive Epilepsy Center. The group consisted of 39 females and 21 males. Subjects were placed into the following four categories: non-epileptic events (NEE); seizures (Sz); partial seizures (PartSz); complex partial seizures with temporal focus (CPS-TF). Each subject was given a simple writing task to complete. The essays were scored using four unique scales: Subject - whether the writing concentrated upon physical or emotional aspects of the seizures. Focus - how much of the essay was on task. Ratio - a measure of detail. Word - total word count. Results: Patients with nonepileptic events (NEE) tended to write essays containing both physical and emotional components (p = 0.058), while those with partial seizures (PartSz) did not (p = 0.006). Essays concentrating upon emotional aspects, increased the likelihood of a partial seizure diagnosis being made (p = 0.059). The variance of Ratio was higher in the non-NEE group than in the NEE group (p = 0.0003), while for the PartSz group, the variance of Ratio was higher than in the non-PartSz group (p = 0.000008). Conclusions: The results of the present study have revealed a relationship between the differences of writing content in those with NEE and those with seizures, especially partial onset seizures. The findings with regard to the variable "Subject" not only give insight into the mind of these patients, but provide the preliminary basis for the development of a diagnostic tool utilizing a simple writing task. With further refinement and validation such a task may one day become an aid to replacing costly video-EEG studies in select patients. The findings with respect to "Ratio" may also eventually play a role in this later purpose, but yield a different sort of information. The significant difference in variance between groups may be due to the greater degree of structural brain damage likely to be found in patients with partial seizures compared to those with primarily generalized epilepsy or NEE. (Supported by Office of Community and Patient Relations and the Department of Neurology at Stanford University Medical Center and through the generous and kind assistance of Ms. Jeanne Kennedy. Armin Schwartzman was supported by a William R. and Sara Hart Kimball Stanford Graduate Fellowship.) Abstract: .sup.1 Nabil J. Azar , .sup.1 Tania F. Tayah , and .sup.1 Bassel W. Abou-Khalil (Neurology, Vanderbilt University Medical Center, Nashville, TN) Rationale: The distinction between epileptic and non-epileptic seizures can be very challenging in the absence of EEG-video monitoring with recorded seizures. Discriminating ictal features described in the literature often cannot be used in the clinic while obtaining the history. Overwhelmed observers often miss ictal details. Postictal signs could prove easier to identify. We observed that the postictal breathing after non-epileptic seizures with generalized motor activity is considerably different from the postictal breathing pattern of generalized tonic-clonic seizures. We evaluated this feature in consecutive patients admitted for EEG-video monitoring. Methods: We reviewed seizures recorded with video-EEG in consecutive adult patients with epilepsy and patients with non-epileptic psychogenic seizures. We reviewed 22 generalized tonic-clonic seizures in 15 patients with epilepsy and 22 convulsive non-epileptic seizures in 15 patients with pure psychogenic seizures. We analyzed the altered postictal breathing pattern for rate, depth, regularity, loudness, and duration. We used the Fisher exact test to compare groups for categorical variables and the student t-test to compare them for continuous variables. Results: The breathing after generalized tonic-clonic seizures was deep with prolonged inspiratory and expiratory phases, regular, and usually loud (except for two short seizures). The breathing after convulsive non-epileptic seizures was characterized by increased respiratory rate or hyperpnea with short inspiratory and expiratory phases, as can be noted after exercise. The breathing was often irregular, with brief pauses. The altered breathing lasted longer after epileptic tonic-clonic seizures. The two groups differed significantly in loudness of postictal respiration, postictal snoring (only with epileptic seizures), respiratory rate (mean of 30/min for the epileptic group and 44/min for the psychogenic group) and duration of altered breathing (mean of 6 min for the epileptic group and 1.6 min for the psychogenic group) (p < 0.00001 for all features). Conclusions: The postictal breathing pattern can help differentiate generalized tonic-clonic seizures from non-epileptic psychogenic seizures and may be helpful to the practitioner obtaining a seizure history in the clinic setting, or witnessing a seizure. Abstract: .sup.1 Marino M. Bianchin , .sup.1 Tonicarlo R. Velasco , .sup.1 David Araujo Jr. , .sup.1 Veriano Alexandre Jr. , .sup.1 Lauro Wichert-Ana , .sup.1 Vera C. Terra-Bustamante , .sup.1 Ana Paula P. Martins , .sup.1 Antonio C. dos Santos , .sup.1 Roger Walz , .sup.1 Joao A. Assirati Jr. , .sup.1 Carlos G. Carlotti Jr. , .sup.1 Osvaldo M. Takayanagui , and .sup.1 Americo C. Sakamoto (CIREP-Department of Neurology, Psychiatry, and Medical Psychology, Ribeirao Preto School of Medicine, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil) Rationale: Observations in our population indicate that in endemic areas to neurocysticercosis (NCC), NCC is more common in patients with mesial temporal lobe epilepsy associated with hippocampal sclerosis (MTLE-HS) than in other forms of focal medically intractable epilepsy, suggesting a possible association between these two diseases. We designed this study in order to further explore this association. Methods: Retrospective cohort of 191 patients with MTLE-HS from Brazil. We evaluated demographic, clinical, and neuroimaging findings of patients with MTLE-HS plus NCC and compared them with those of patients with MTLE-HS without NCC. The magnitude of differences was measured by the O.R. (95% C.I.). Adjusted O.R. was estimated by logistic regression. Logistic regression was also used to predicting NCC in MTLE-HS patients. Results: We observed chronic NCC radiological findings in 68 (35.6%) MTLE-HS patients. In these patients, NCC was significantly more common in women (45 patients 45.9%) than in men (23 patients, 24.7%), (adjusted O.R. = 3.33; 95% C.I. = 1.67-6.62 p < 0.001), in patients with no history of an initial precipitating injury (IPI) before 5 years-old, (adjusted O.R. = 3.04; 1.38-6.72; p < 0.01), and in patients presenting bilateral temporal interictal spikes during pre-surgical video-EEG, (adjusted O.R. = 2.36; 95% C.I. = 1.19-4.66; p < 0.02). In endemic areas similar to ours, the possibility of finding NCC in refractory MTLE-HS (Prob(NCC) can be calculated by the following equation: Prob(NCC) = 1/(1+e.sup.-v), where V = (-0.038 - 1.2sex + 1.304IPI + 0.95iEEG). This equation was able to correctly predict the presence of neuroimaging findings of NCC in 71% of our surgical MTLE-HS patients. Conclusions: Taken together, our results suggest that NCC might be associated with epileptogenesis in mesial temporal lobe epilepsy in susceptible patients, thus being a possible risk factor for development of MTLE-HS. Indeed, if such association were only coincidental, occurring because of the relative high prevalence of both pathologies in the same region, differences between these two groups of patients were expected to be null. However, some important differences were observed and some associations were possible. Differences were observed for women, for those patients without an IPI history before 5 years-old, and for those patients with bi-temporal interictal EEG spikes. These results might help us to better understand the interactions between NCC and MTLE-HS, improving our insights on the possible factors that might explain the association of NCC and refractory MTLE-HS in some world areas. (Supported by FAPESP.) Abstract: .sup.1 Frank D. Boesebeck , .sup.1 Stefan Freermann , .sup.1 Christoph Kellinghaus , .sup.1 Moddel Gabriel , and .sup.1 Evers Stefan (Neurological Department, University Hospital of Muenster, Muenster, Germany) Rationale: Paroxysmal neurological deficits following epileptic seizures or other medical conditions frequent lead to admission on a Neuro-ICU. The etiological misinterpretation of such events may cause a delayed begin of treatment or an inappropriate utilization of diagnostic- and ICU-capacities. Methods: The data of 208 patients being admitted to a Neuro-ICU because of episodic neurological deficits, loss of consciousness and unclear motor phenomena have retrospectively been analyzed. The initially suspected diagnosis in the neurological emergency room was compared to the final diagnosis and the proportion of misdiagnosis was related to the patients' history and diagnostic data respectively. Results: 13.9% of the episodes initially suspected as epileptic seizures proved to be non-epileptic events (NEE). 76% of those patients with "misdiagnosed" NEE had a prior history of epilepsy. The rate of misdiagnosed NEE was higher in patients with regular EEG-findings (16.7% vs. 8.7%) and regular CT findings (21.2% vs. 11.4%). 15.6% of the epileptic seizures were falsely interpreted to be non-epileptic (6.7% with status epilepticus). The rate of misdiagnosed epileptic Seizures (MES) was higher in patients with newly diagnosed epilepsy if compared to known epilepsy (11.2% vs. 1.4%), whereas the rate of misdiagnosed status epilepticus was independent from a prior history of epilepsy (6% vs. 5.4%). The rate of MES was significantly higher in patients with pathologic CT findings (18.9% vs. 1.8%) and pathologic MRI findings (16.3% vs. 3.3%) respectively. MES were most frequent in patients with hydrocephalus, microangiopathy in CT/MRI and prior neurovascular insults. Conclusions: Our data reveille a known epilepsy and the lack of pathological findings on EEG, CT to be risk factors for the misdiagnoses of non-epileptic paroxysmal events, whereas no prior history of epilepsy and pathologic findings on CT and MRI more often caused a misinterpretation of epileptic seizures to be other neurological disorders. Abstract: .sup.1 Maya R. Carter , .sup.1 Allan Krumholz , .sup.1 Tricia Ting , and .sup.1 Jennifer Hopp (Neurology, University of Maryland, Baltimore, MD) Rationale: Nocturnal Frontal Lobe Epilepsy (NFLE) is established as a distinct clinical epileptic syndrome with a spectrum of clinical presentations within the heterogeneous group of paroxysmal sleep-related disturbances. Pregnancy is well known to exacerbate seizures in many patients with epilepsy, but there are no reports of the effects of pregnancy on seizures in patients with NLFE. Methods: We report two women with NFLE documented by clinical history and video-EEG monitoring with severe seizure exacerbations during pregnancy. Both patients had long histories of mainly nocturnal episodes of motor jerking or spasm involving mainly the hand and arm on one side and nocturnal generalized tonic-clonic seizures. Both patients were eventually successfully treated with carbamazepine with nearly complete resolution of symptoms. However, during their first pregnancies both patients experienced severe exacerbations of their nocturnal seizures requiring hospitalizations. Additionally, both patients were on high doses of folic acid. Results: An exacerbation occurred in the 3.sup.rd trimester of pregnancy in one patient taking stable doses of Carbatrol, with therapeutic carbamazepine blood levels, despite increasing the dosage of Carbatrol. During the exacerbation she experienced several distressing nocturnal seizures. Upon admission to the hospital her seizures were eventually controlled upon substitution with a short acting carbamazepine preparation, given more frequently than her Carbatrol, and the addition of phenytoin. Events subsequently improved and she was tapered from phenytoin and maintained on the short acting carbamazepine with therapeutic blood levels. She delivered a healthy child without complication, after which she resumed taking Carbatrol without problems. The second patient took carbamazepine only at bedtime with therapeutic blood levels, when her seizures suddenly severely worsened in the first trimester of her pregnancy, despite modest increases in the dosages. She was admitted to the hospital and the dosage was increased and spread out and her folic acid dosage was reduced. Her episodes ceased and she eventually was able to return to the full dosage of folic acid. She and her baby are doing well in the 3rd trimester of pregnancy. Of note is that in the past she had exacerbations of her seizures with the use of birth control pills. Conclusions: Nocturnal Frontal Lobe Epilepsy as demonstrated by these two patients is a distinct epilepsy syndrome which may be exacerbated during pregnancy and exhibits characteristic clinical features. In particular, these patients may develop severe exacerbations of motor symptoms and seizures during pregnancy. Potential mechanisms include alterations in antiepileptic drug metabolism, hormonal factors that lower seizure threshold, and the effects of folic acid. The potential for pregnancy to exacerbate NFLE should be recognized and warrants further study and investigation. Abstract: .sup.1 Hyunmi Choi , .sup.1 Dionysios Pandis , .sup.1 Gary Heiman , and .sup.1 W. Allen Hauser (Neurology, Columbia University, New York, NY; Neurology, Columbia University, New York, NY; Epidemiology, Mailman School of Public Health at Columbia University, New York, NY; and Neurology/Epidemiology, Columbia Universiy, New York, NY) Rationale: Although much is known about the rate of seizure remission in adults that have undergone epilepsy surgery or children with newly diagnosed epilepsy, prognosis for adults with intractable epilepsy is not clearly defined. Population-based or community-based study would require large sample size to detect those with intractable epilepsy that subsequently experience remission. Therefore, examining patients followed at a tertiary clinic setting offers a unique opportunity to study the prognosis in intractable epilepsy patients. Our objectives were to determine the frequency of seizure remission and to evaluate the influence of clinical factors on a 6- month seizure remission among a cohort of adults with intractable epilepsy followed at a tertiary clinic setting. Methods: This was a retrospective cohort study. We reviewed all adult patients seen in 2001 at the Columbia University Epilepsy Center. Subjects met our criteria if 1) they had failed [greater than or equal to]2 AEDs at the index date in 2001, 2) if they had on average [greater than or equal to]1 seizure per month for 3 consecutive months prior to the index date, and 3) enough follow-up time to experience study endpoint. Those who met our criteria of intractability were followed from their index visit in 2001 until their most recent visit to determine if they experienced a 6-month seizure free period. Crude remission rates were determined, overall and for subgroups, and logistic regression analysis was used to establish the independent significance of the clinical variables, which included history of prior epilepsy surgery, status epilepticus, age of onset, epilepsy classification, developmental delay, and etiology. Analysis was conducted stratifying the subjects by their status of epilepsy surgery following entry into the study. Results: Out of 1389 patients seen at the center in 2001, 191 subjects met our criteria of intractable epilepsy. Mean follow-up of these subjects was 3.5 years. The subjects had the following epilepsy types: Localization-related epilepsy 158 (83%) Primary generalized epilepsy 10 (5%) Symptomatic generalized epilepsy 20 (10%) Both LRE and generalized epilepsy 2 (1%) Unclassifiable 1 (1%) Thirty four subjects had epilepsy surgery after entering the study of whom 21(61.7%) experienced a 6-month remission. Of 157 subjects that did not have epilepsy surgery, 32(20.3%) experienced a 6-month remission during the follow-up. None of the clinical variables were significant predictors of a 6-month remission in either group. Conclusions: Subjects not eligible or refusing epilepsy surgery achieved a 6-month remission at a rate of 6% per year. We found no significant predictors of 6-month remission, likely related to power. We suggest a multicenter study with systematic follow-up of intractable cases to address this issue. (Supported by K12 RR017648.) Abstract: .sup.1 Ji Soo Choi , .sup.1 Gregory L. Krauss , and .sup.1 Ronald P. Lesser (Neurology, Johns Hopkins University, Baltimore, MD) Rationale: Seizure response dogs have recently been used to assist patients during seizures and to provide companionship for patients' coping with a chronic condition. We found that 4 of 6 patients with seizure response dogs at our center had, on further investigation, psychogenic nonepileptic seizures (PNES), and no evidence for epilepsy. These cases illustrate the importance of establishing a firm diagnosis of epilepsy for patients obtaining seizure response dogs. Methods: We evaluated seizure histories, medical evaluations and psychiatric diagnosis of patients with seizure response dogs and PNES. We determined the response of dogs to PNES and their support behaviors. Results: Three patients had specially trained seizure response dogs; a fourth patient's family pet learned to bark during seizures. The four patients reported their dogs reliably anticipated their seizures: in two cases licking them prior to the episodes; two patient's dogs barked prior to their episodes. Three patients had videoEEG which documented PNES; a fourth patient had multiple normal EEG and somatization syndrome. Patients' antiepilepsy drugs were discontinued; their PNES stopped after diagnosis and counseling. Conclusions: Seizure response dogs help vulnerable patients during seizures and help them cope with their chronic condition. Two of our patients with epilepsy are supported by service dogs that stay with them during their unconscious episodes. We found, however, that four other patients with seizure response dogs had PNES, and no evidence for epilepsy. These cases show that patients with abnormal illness behaviors may seek service animals for support and demonstrate the importance of establishing an accurate diagnosis of epilepsy before patients obtain epileptic seizure response dogs. While patients may benefit from emotional support provided by pets, therapy with service animals for primarily psychiatric conditions should be based on individual assessments and formulations regarding these conditions. Abstract: 1,2,3 Damian E. Consalvo , .sup.1 Marcelo Kauffman , .sup.1 Pablo A. Salgado , and 1,3 Kochen S. Silvia (Epilepsy Center, Ramos Mejia Hospital, Buenos Aires, Argentina ; FEMIEN Foundation, Argentina ; and University of Buenos Aires, Argentina) Rationale: In patients with the syndrome of mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS), extrahippocampal abnormalities (EHA) are frequently described on MRI studies. The aim of the study was to evaluate the frequency of EHA in patients with MTLE-HS, and to correlate these findings with some of its clinical characteristics. Methods: MTLE-HS patients, diagnosed by electro-clinical and high resolution MRI findings were examined. The presence of temporal lobe atrophy (TLA), parahippocampal gyrus atrophy (PHA), fornix atrophy (FA) and mammillary bodies atrophy (MBA) on MRI studies were considered EHA by visual analysis. The variables studied were age, age at the beginning of the epilepsy, time of evolution, history of febrile convulsions (FC), familial epilepsy, status epilepticus, secondary generalization of the seizures and response to the pharmacological treatment. Results: Fifty four patients were included. Fifty of them (92.5%) showed EHA on MRI studies. They were divided in 4 groups. Group 1: TLA patients, 18 cases (33.3%); Group 2: PHA patients, 44 cases (81.4%); Group 3: FA patients, 29 cases (53.7%); and Group 4: MBA patients, 23 cases (42.5%). EHA were found in 14 of 17 (82.3%) normal controls. Anyone showed TLA or PHA. Three controls showed FA (p < 0.02) and 2 MBA (p < 0.04). A strong association between PHA and the history of FC was observed (p = 0,002). There were no differences in the rest of the variables analyzed. Conclusions: The presence of EHA suggests that the causal noxa of the syndrome of MTLE-HS also affects other areas widely connected with the hippocampus. The association between PHA and FC introduce a hypothesis of a physiopathological mechanism of common affectation. Abstract: .sup.1 Charles L. Dalmagro , .sup.1 Ricardo Guarnieri , .sup.1 Maria Priscila Cescato , .sup.1 Tonicarlo R. Velasco , .sup.1 Veriano Alexandre Jr. , .sup.1 Vera C. Terra-Bustamante , .sup.1 Lauro Wichert-Ana , .sup.1 Joao A. Assirati Jr. , .sup.1 Carlos G. Carlotti Jr. , .sup.1 Marino M. Bianchin , .sup.1 Jayme E. Hallak , and .sup.1 Americo C. Sakamoto (CIREP-Department of Neurology, Psychiatry, and Medical Psychology, Ribeirao Preto School of Medicine, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil) Rationale: Some reports point to the inverse correlation between depression and the presence of generalized seizures in focal symptomatic epilepsy. However, this issue is still controversial. In order to evaluate whether MTLE-HS patients with few tonic-clonic episodes have differences in the frequency of depressive or other psychiatric symptoms, we conducted the present study. Methods: In a retrospective case-control study we analyzed presurgical data of a group of 81 homogeneous MTLE-HS. Patients were divided in two sub-groups: a) group with rare generalizations (n = 40): no episodes to less than one episode of secondarily generalized tonic-clonic seizure per month; b) group with frequent generalization (n = 41): at least one or more generalized seizures per month. Patients were matched other varibles Psychiatric diagnosis was established by psychiatrists with experience in psychiatric disorders associated with epilepsy, and according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition. Patients were further classified for positive or negative psychiatric diagnosis and further sub-divided according to symptoms of depression or other psychiatric diagnostics. The magnitude of associations was measured by the O.R. (95% CI). Results: Psychiatric diagnosis was observed in 45 of the 81 patients (55%). In the group with rare generalization we observed 18 patients with psychiatric symptoms and 22 patients without psychiatric co-morbidities. In the group of frequent generalizations we observed 18 patients with psychiatric symptoms and 23 patients without psychiatric co-morbidities (O.R. = 0.96; 95% C.I = 0.40 2.30; p = 1.00). Regarding only depressive symptoms, in the group with rare generalizations, we observed 9 patients with depressive co-morbidities, and 31 patients without any history of depressive co-morbidities. In the group of frequent generalizations we observed 8 patients with depressive co-morbidities and 33 patients without any psychiatric depressive co-morbidities (O.R. = 0.83; C.I = 0.29-2.44; p = 0.79). Conclusions: Our results do not support the hypothesis that secondary generalized tonic-clonic seizures affect the frequency of depressive symptoms, depression, or other psychiatric co-morbidities in MTLE-HS patients. However, further studies with larger samples are necessary to confirm our results and to assess the impact of surgery in seizure control and its influences in psychiatric co-morbidities. (Supported by FAPESP) Abstract: .sup.1 Sandeep Dutta , .sup.1 Balakrishna S. Hosmane , and .sup.1 Ronald C. Reed (Clinical Pharmacokinetics, Abbott Laboratories, Abbott Park, IL; Division of Statistics, Northern Illinois University, DeKalb, IL; and Neuroscience, Abbott Laboratories, Abbott Park, IL) Rationale: The development of specific epilepsy subtypes may be gender specific (Epilepsia 2005; 46: 956-60), and it has been speculated that response to antiepileptic drug (AED) treatment is potentially influenced by gender. Gender influences the pharmacokinetic disposition of drugs in general (Clin Pharmacokinet 2003; 42: 107-21); this concern extends to AEDs (Epilepsia 2005; 46-Suppl 6: 46). While gender is associated with alterations in rat valproic acid (VPA) glucuronidation (Drug Metab Dispos 1996; 24: 367-70), limited information exists concerning the impact of gender on VPA disposition in humans. The objective of this analysis was to examine the effect of gender on VPA pharmacokinetics in adults. Methods: Plasma VPA pharmacokinetic parameters from 5 published multiple dose studies (Clin Drug Investigation 2004, 24 (9): 495-508) in male (n = 106) and female (n = 64) healthy subjects and epilepsy patients on drug-metabolizing hepatic enzyme-inducing AEDs receiving 875-5000 mg/day doses of conventional delayed-release enteric-coated divalproex sodium (divalproex) and extended-release divalproex sodium (divalproex-ER) tablets were analyzed. Statistical analysis was performed in SAS at a significance level of 0.05. A linear mixed model with heterogeneous compound symmetry structure across studies to account for correlation among multiple observations within a subject was used for assessing the gender effect on the response. The model included effects for study, formulation, gender, dose level, an interaction between gender and formulation; body weight (ranging from 45 to 139 kg) was a covariate. Data for divalproex and divalproex-ER were analyzed separately since interaction between gender and formulation was significant (p < 0.0027). Results: Gender did not have a significant influence on VPA pharmacokinetics following divalproex or divalproex-ER administration. VPA least squares mean values for males vs. females, respectively, were: (a) for exposure (AUC24 [mg.h/L]) 1800 vs. 1762 (p = 0.5148) for divalproex and 1593 vs. 1636 (p = 0.5284) for divalproex-ER; (b) for maximum concentration (Cmax [ mg/L]) 101.8 vs. 102.4 (p = 0.8262) for divalproex and 81.0 vs. 83.9 (p = 0.3230) for divalproex-ER; (c) for minimum concentration (Cmin [ mg/L]) 52.6 vs. 51.9 (p = 0.7894) for divalproex and 47.2 vs. 49.1 (p = 0.5470) for divalproex-ER; and (d) for degree of peak-trough fluctuation (DFL) 0.635 vs. 0.660 (p = 0.3687) for divalproex and 0.439 vs. 0.476 (p = 0.2241) for divalproex-ER. Conclusions: No significant differences in VPA pharmacokinetic parameters were observed between male and female participants receiving divalproex or divalproex-ER after accounting for dose, differences in body weight, study effect and multiple observations within a subject/patient. (Supported by Abbott Laboratories.) Abstract: .sup.1 John O. Elliott , .sup.2 Mercedes P. Jacobson , and .sup.3 Brenda Seals (Public Health, Ohio State University, Columbus, OH ; Neurology, Temple University, Philadelphia, PA ; and Public Health, Temple University, Philadelphia, PA) Rationale: It is well known in the neurology literature that anti-epileptic drugs (AEDs) lead to bone loss. Several large epidemiologic studies have found twice the fracture rate in persons with epilepsy compared to non-epilepsy populations. To date no one has attempted to assess knowledge of osteoprotective behaviors in persons with epilepsy. Methods: The Osteoporosis Knowledge Test (OKT), a validated, 24-item test, was administered to 94 epilepsy patients. Kim, et al. developed the OKT to measure knowledge of risk factors for osteoporosis and strategies for prevention related to calcium and exercise. Results: Mean age of participants was 45 years with 20 years of AED exposure. There were 66 females and 28 males of which 50 were Caucasian and 44 were non-Caucasian. No significant differences related to age or gender for the OKT were found. One-way ANOVA of ethnicity found non-Caucasians had much lower knowledge for calcium (F = 8.15, p = .005) and exercise (F = 7.71, p = .007). Those reporting > 12 years of education had higher knowledge scores for calcium (F = 39.25, p = .000) and exercise (F = 17.09, p = .000). Subjects reporting an income above $30,001 had higher levels of knowledge for calcium (F = 3.93, p = .024) and exercise (F = 3.46, p = .037). Subjects who were seizure free > 1 year also had higher knowledge related to calcium (F = 6.76, p = .002) and exercise (F = 3.59, p = .032) compared to those reporting seizures within the last year. Significant differences in calcium knowledge were noted for individuals previously diagnosed with bone loss (F = 4.98, p = .028). In addition, for those reporting a family history of osteoporosis, knowledge was higher for calcium (F = 5.58, p = .005) and exercise (F = 6.23, p = .003). Subjects taking calcium supplements had higher knowledge for calcium (F = 5.35, p = .023) but not for exercise. No significant differences were found based on fracture history. Conclusions: Overall, this study adds to the understanding of patient knowledge related to bone loss in epilepsy. Age and gender appear to have little impact on knowledge related to osteoporosis, compared with socioeconomic factors such as years of education, yearly income and ethnicity. This mirrors what has been found in the medical literature for non-epilepsy populations. One area that may help improve knowledge is the use of culturally relevant education materials. Patient education materials that focus on facts and use medical terms foreign to the general public (disease history, etiology, symptoms, treatment options and side effects) are less effective and tend to be written at a level that exceeds the readability of most patients. Clinically, patients may need to achieve longer term seizure freedom before osteoprotective information can be recommended. Low socioeconomic status may place people at a disadvantage for prevention recommendations. Abstract: .sup.1 Siegward M. Elsas , .sup.1 Garrett White , .sup.1 Carole-Anne Randall-Stitt , and .sup.1 Sonia Munoz (Department of Neurology, Oregon Health and Science University, Portland, OR) Rationale: This is a pilot study to evaluate physiologic and potential therapeutic effects of a meditative relaxation technique in patients with complex partial epilepsy. Methods: Two patients with a 30 vs. 45 year history of complex partial and generalized seizures practiced meditative relaxation for 30 min per day over 12 months in the context of a behavioral program for seizure control. Three serial EEGs were obtained during simulated stress and meditative relaxation. In addition to epileptic spikes, changes in breathing and heartrate and tonic EMG were obtained as measures of body tension. Results: At baseline, the patients had 4.2 vs. 2.7 GTC and 6.9 vs. 1.9 CPS seizures on average per month despite optimal medical and surgical epilepsy care. Twelve months after initiation of the behavioral program, patient 1 was seizure free for 8 months and patient 2 had 0.15 GTC and 0 CPS on average per month. In both patients, body tension measures in the first serial EEG showed poor compliance with the meditative relaxation. In both patients, the second and third serial EEGs showed increased epileptic spikes and decreased body tension during meditative relaxation compared to the simulated stress periods. Conclusions: Meditative relaxation appears to increase epileptic spikes in the EEG but may reduce seizures for patients with complex partial epilepsy in the context of a behavioral program. It is known that sufficient sleep is essential for maintaining good seizure control, even though sleep temporarily increases epileptic spikes. In similarity to sleep, it is possible that the temporary increase in spikes during meditative relaxation practice may be beneficial to epilepsy patients. (Supported by Medical Research Foundation of Oregon; National Center for Complementary and Alternative Medicine at NIH) Abstract: .sup.1 Bernt A. Engelsen , .sup.1 Charalampos Tzoulis , .sup.2 Massimo Zeviani , .sup.1 Bjorn Karlsen , .sup.1 Atle Lillebo , .sup.3 Liv Laegreid , .sup.4 Aasly Jan , and .sup.1 Laurence Bindoff (Institute of Clinical Medicine and Molecular Medicine, University of Bergen and University Hospital of Haukeland, Bergen, Norway ; Unit of Molecular Neurogenetics Pierfranco and Luisa Mariana Center for the Study of Children's Mitochondrial Disorders, National Neurological Institute "C.Besta", Milano, Italy ; Department of Pediatrics, University Hospital of Bergen, Bergen, Norway ; and Department of Neurology, St. Olavs University Hospital, Trondheim, Norway) Rationale: The epileptic semiology of 19 norwegian patients (13 families) with mutations in the polymerase-[gamma] (POLG) gene is presented for the first time with EEG findings. A broader clinical symptomatology has been presented previously, but not a detailed description of the epilepsy. The patients were either homozygous for the 1399 G > A or 2243 G > C mutation, or compound heterozygotes for either of the two mutations. Methods: Data from patients with therapy refractory status epilepticus were collected over several years, and tested for various mitochondrial mutations and delesions. After examination of several candidate areas mutations in POLG were found and subsequent candidate patients tested. Results: Irrespective of the genotype the epilepsies developed a similar semiology with initial features of occipital lobe epilepsy followed by more secondary generalized features including focal motor phenomena and GTC seizures. Occipital seizure phenomena included, flickering coloured lights, visual loss, nystagmus (horizontal or vertical), dysmorphopsia, micro-macropsia, palinopsia and most patients had simple motor phenomena suggesting frontal lobe seizure origin or spread. SPS, CPS, myoclonic seizures and frequent convulsive status epilepticus were observed in the epileptic syndrome. All but one patient developed status epilepticus. Ten of 19 patients are dead, all related to prolonged status epilepticus, icluding 2 with liver failure. Epilepsy was the most common presentation of the full clinical syndrome, often with headache. The mean age at epilepsy debut was 18,4 years (6-58). The broader clinical semiology of most of these patients have been presented (Wintherthun et al. Neurology 2005;64:1204-8 and Tzoulis et al. Brain 2006, in press). The epilepsy and consequences of valproate treatment has proven the most important factor for fatal outcome. Conclusions: Occipital lobe epilepsy is a hallmark of POLG mutations in the present patient population, and epilepsy an important factor in fatal outcome. (Supported by Helse-Vest, Bergen. Norway.) Abstract: .sup.1 Berend Feddersen , .sup.1 Margret Kilian , .sup.1 Cordula Mauerer , .sup.1 Christian Vollmer , .sup.1 Jan Ricken , and .sup.1 Soheyl Noachtar (Epilepsy Center Department of Neurology, University of Munich, Munich, Germany) Rationale: Most seizures of patients with temporal lobe epilepsy (TLE) are characterized by manual and oral automatisms preceded by abdominal aura. Head version, hand dystonia and secondary generalization reflect spread of epileptic seizure activity. The aim of the present study was to evaluate whether different spread pattern in temporal lobe epilepsy are more likely to generalize secondarily than others. Methods: We searched the data base of the University of Munich Epilepsy Monitoring Unit for the terms dystonia, version and temporal lobe epilepsy (TLE). The first consecutive 62 TLE patients with seizure evolutions characterized by either dystonia (n = 24) or head version (n = 38) were included in the study. MRI and high quality EEG-videos recordings were available in all patients. 82 seizure evolutions with dystonia were intraindividually compared with 68 seizures of these patients without dystonia. In addition, 65 seizures with version were compared with 163 seizures of the same patients without version. All seizure evolutions were classified prospectively independent of the other clinical results based on a semiological seizure classification system. Head version was classfied independent of secondary generalization of a given seizure. Results: The rate of secondary generalization was significantly higher in seizures with version (86%, 56 of 65) than seizures with dystonia (17%, 14 of 82) (p < 0.001). 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 higher in seizures with dystonia (17% vs. 3%). However, the difference in rate of generalization was much higher in the seizures with version: 12% of the seizures generalized secondarily if no version was present as opposed to 86% in seizures with version. 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 a generator of dystonia. These findings could support an inhibiting role of the basal ganglia for spread of epileptic activity. Abstract: .sup.1 Felicia Ferguson , .sup.1 Martin C. Salinsky , and .sup.1 David Spencer (Neurology, Oregon Health & Science University, Portland, OR) Rationale: Psychogenic non-epileptic seizures (PS) are a common, often disabling condition, frequently confused with epilepsy, and usually treated with antiepileptic drugs (AEDs) prior to diagnosis. The gold standard for diagnosing PS is inpatient video-EEG monitoring, where PS represent 20-50% of discharge diagnoses. There can be no effective intervention for PS without first establishing the diagnosis. The past decade has seen extensive efforts to educate medical students, primary care givers and neurologists about PS, emphasizing early referral in order to prevent the social and medical morbidity of this potentially treatable condition. To assess the effects of this effort we reviewed records from two cohorts of proven PS patients, one from 1989-1993 and the other from 2003-2005. Methods: Since 1989, our inpatient video-EEG monitoring unit has maintained a database of all admissions and discharge diagnosis. A series of 46 consecutive patients with PS was selected, beginning with January 1989. The diagnosis required at least two typical spells with either alteration of consciousness and/or bilateral motor activity and no EEG changes, absence of epileptic seizures, and team consensus. A second cohort of 46 consecutive PS patients was selected beginning with admissions from January 2003, using identical criteria. All records were abstracted for age, sex, number of years between onset of the currently active PS and admission for video-EEG, number of AEDs at the time of admission, and other features. Results: The two cohorts did not significantly differ with regard to age, or proportion of women. The median delay between onset of PS and diagnostic admission was 5.0 years for the 1989 cohort and 2.3 years for the 2003 cohort (p < 0.01; Wilcoxon test). For the 1989 cohort, 80% of patients were on AEDs at the time of admission and 24% were on 2 or more AEDs. Results for the 2003 cohort were nearly identical (mean of 1.2 AEDs at admission for each group; NS). Conclusions: Our results show that patients with PS are now being referred for diagnostic evaluation much earlier than was the case 12 years earlier. During this time period there were no changes in the availability of inpatient monitoring in Oregon. Therefore, a likely explanation is the impact of efforts, local and national (AES), to educate students and physicians about PS. Despite progress, two years remains a considerable lag between onset of PS and diagnostic referral, and there has been no reduction in the use of AEDs to treat PS prior to diagnosis. Abstract: .sup.1 David M. Ficker , .sup.1 Cynthia Hughes , .sup.1 Michael D. Privitera , and .sup.2 Rakesh Shukla (Department of Neurology, University of Cincinnati Academic Health Center, Cincinnati, OH ; and Department of Environmental Health, University of Cincinnati Academic Health Center, Cincinnati, OH) Rationale: To determine which components of quality of life (HRQOL) are different between subjects who were seizure free and not seizure free at last follow up in a cohort of newly diagnosed epilepsy. Methods: We prospectively measure the QOLIE-89, Profile of Mood States (POMS) and Adverse Events Profile in patients within three months of a new diagnosis of epilepsy. Serial measures are performed every four months. Inclusion criteria are: age between 18-64, able to read and speak English and a minimum tenth grade education. Exclusion criteria are: inability to complete forms, presence of other neurologic or medical disorders that may affect HRQOL, prior craniotomy and use of medications that can affect the central nervous system (other than AEDs). A t-test was performed comparing QOL measures at last follow up between subjects who were seizure free and those not seizure free. Results: We enrolled 53 subjects with newly diagnosed epilepsy (22 males, 32 females). Follow-up data was available on 40 patients (mean duration of follow-up 1.3 years). At last follow-up, 65% were seizure free and 35% were not seizure free. Table 1 summarizes differences on QOL measures between these groups. Conclusions: These results suggest that a seizure free state is associated with a higher QOL and improved anxiety symptoms. We found non-significant trends toward superior scores on other POMS subscales in the seizure free group. TABLE OMITTED Abstract: .sup.1 Maria B. Viaggio , .sup.1 Maria E. Fontela , and .sup.1 Alfredo Thomson (Epilepsy Section, Department of Neurology, Hospital Frances, Buenos Aires, Argentina) Rationale: Non Convulsive Status Epilepticus (NCSE) assembles different electrographic and clinical syndromes. Although consciousness impairment is its outstanding feature, several other manifestations such as behavioral and cognitive changes, turns it difficult to diagnose. Due to this pleomorfism, great knowledge of its clinical presentation and high suspicious index is mandatory for prompt diagnosis. Methods: Records of patients with diagnosis of Status Epilepticus admitted into our center between 1996 and 2005 were reviewed. According to clinical and electrographic criteria, cases of NCSE were selected. Semiology, demographic data, aetiology, referral latencies and delays in diagnosis, treatment given and mortality were considered and submitted to statistical analysis. Results: 63 patients were included, 3 of them presented more than 1 NCSE, totaling 68 NCSE. The study group mean age was 57 ([+ or -]21), in it 35 patients were female. Main presenting features were: consciousness impairment, confusion, behavioral changes and prolonged postictal state. NCSE were classified as acute symptomatic: 26 cases, of remote aetiology: 30 cases and idiopathic or cryptogenic: 12 cases. Mean referral latency was 65 hours (few minutes to 1 month). On admission into our center, suspicious of NCSE arouse: immediately in 54 cases (79,41%), in 9 between 24 and 72 hours and in 5 cases around 96 hours (due to time required to rule out other aetiologies related to history of dementia, stroke, aphasia and psychiatric problems of these patients). Time to control status after treatment started, was 24 hours. The larger delay in referral the longer the time to control status (p = 0.0005). 11 cases were classified as refractory. Five patients died (7,3% of the study group). 4 deaths were related to the underlying disease. While new onset NCSE were 38 cases (with a mean age of 61.2 [+ or -] 21.3), NCSE among epileptic patients resulted in 25 cases (mean age: 50.8 [+ or -] 18.4); (difference in ages between these sub-groups determined: p = 0.03). Non epileptic patients revealed longer delays when looking for medical intervention; and among them NCSE of acute symptomatic aetiology was more frequent (p = 0.003). Conclusions: Our study population represents one of the largest groups when considering prior reports in the literature. Two main pieces of conclusions could be obtained from its analysis: In first place: Considerable delays in referral were observed probably due to confusing heterogeneous clinical presentation; this determined longer times to control status. Low mortality rate could be related to early diagnosis and few refractory cases. Secondly: in our study population NCSE appeared "de novo" mainly among older patients and specially related to an underlying acute process. Absence of history of epilepsy resulted in longer delays in looking for medical help. We emphasize that: due to its potential control with prompt treatment and because of its heterogeneous presentation, high suspicious index could improve outcome of NCSE. Abstract: .sup.1 Nathan B. Fountain , and .sup.1 Miao Liu (Comprehensive Epilepsy Program, University of Virginia, Charlottesville, VA) Rationale: The rate of becoming seizure free (SF) in established epilepsy is unknown. It is often assumed that SF is unlikely to be achieved once epilepsy is established. An important principle of treatment is that seizures respond soon after an intervention or not at all and that there is a very low background rate of becoming SF that continues even if the intervention is ineffective. We hypothesized (incorrectly) that most patients would become SF at the time of an intervention and then would become SF at a constant very low rate afterward. Methods: Standardized data were prospectively collected by epileptologists from patients in the UVA Epilepsy Clinic from April 1998 to November 2004. Data were acquired directly from patients and medical records, and refined, cleaned, and updated at each clinic visit. Analysis was limited to subjects with definite epilepsy > 5 years old with 3 or more SZ frequency entries to construct SZ frequency curves. Frequency of each SZ type was collected. For each subject, monthly SZ frequency was plotted over time throughout the study period. SF was defined as a total frequency of 0 for at least 6 mo. Duration of SF was defined by the interval between an intervention (surgery, adding a drug, increasing a drug dose, or other) and the last seizure. Results: From 1638 subjects, 988 had 3 or more SZ frequency entries. 231 were excluded because SF was present from the first entry. Therefore, 757 had active seizures at entry to the study and represented the population that could become SF. Of these, 32% (243) became SF and 68% (514) continued to have seizures. Prior to becoming SF, 34% (84) had a history of SF so that a 6 month interval of SF may not be different from their baseline. However, 50% (121) had a high baseline SZ frequency of > 1/mo while only 16% (38) had SZ frequency of < 1/mo. suggesting that SF was not due to low baseline SZ frequency. SF developed at the onset of an intervention in 54%, as expected. a cumulative plot of SF over time revealed a significant minority (25%) became SF over the next 24 mo., a rate of 13%/year. The rate of SF subsequently flattened after 24 mo. to a new baseline rate of only about 1-2% per year. Conclusions: A surprisingly significant minority of patients will become SF in the 2 years after the last intervention, suggesting it may take months for therapy to become effective in some situations. This suggests there are brain changes that occur over long durations, possibly in response to therapy. The low rate of SF subsequent to 2 years after an intervention is likely the baseline rate of SF in epilepsy since it occurred without any recent intervention. [figure 1] (Supported by University of Virginia.) Abstract: .sup.1 Marlise Frey , .sup.1 Andres M. Kanner , and .sup.1 Glenn T. Stebbins (Rush Epilepsy Center, Rush University Med Ctr, Chicago, IL) Rationale: Several retrospective studies have reported significant increases in apparent clearance (Cl) of lamotrigine (LTG) during pregnancy. Such an increase has been shown to be progressive until the 32nd week of gestation, followed by a rapid return postpartum to preconception levels. Failure to adjust LTG doses to the increase in Cl yields lower serum concentrations and may result in seizures. The aim of our study was to evaluate the LTG dose adjustments required across trimesters (TMs) to maintain preconception levels in a cohort remaining seizure free during pregnancy. Methods: A retrospective chart review study resulted in four patients meeting the following criteria: stable dose of LTG and seizure free state for at least one month prior to conception, LTG monotherapy or one adjunct AED with no known pharmacokinetic interactions. Adjunct AED dose was not adjusted during pregnancy. No concomitant drug therapy interfering with metabolic pathway of LTG. At least one documented LTG serum level at steady state (Css) and wt. per TM. No known seizures during pregnancy. LTG dose adjustments were only made in an attempt to maintain preconception levels. For each TM the following mean values were recorded for each pt.: wt., dose (in mg/kg), and LTG Css. LTG Cl was calculated with the equation: CL = dose (mg/kg/day)/LTG Css. Data were analyzed using a Friedman's ANOVA with post-hoc comparisons using a Wilcoxon Signed Ranks Test. LTG values were normalized for weight (LTG/wt). This value was our primary outcome variable. We also compared serum LTG levels across TMs to assess if preconception values were maintained. Results: Serum LTG levels (median per TM: 1st = 5.6, range 5.8; 2nd = 7.6, range 5.8; 3rd = 7.6, range 12.6) did not change significantly across TMs (X.sup.2= 1.50, p = .47). LTG dose/wt (median per TM: 1st = 8.6 mg/kg, range 5.9 mg/kg; 2nd = 12.5 mg/kg, range 14.3 mg/kg; 3rd = 14.3 mg/kg, range 12.9 mg/kg) was significantly different across TMs (X.sup.2= 8.0, p = .02). Post-hoc comparisons revealed significant increase in LTG dose/wt from 1st to 2nd TM (Z =-1.83, p = .03 one-tailed) and between 2nd and 3rd TM (Z =-1.83, p = .03 one-tailed). Median increase in dose/wt from 1st to 2nd TM was 3.91 mg/kg/day (range = 8.4), and median increase from 2nd to 3rd TM was 1.96 mg/kg/day (range = 1.61). Conclusions: Increments of LTG dose across TMs can maintain preconception LTG levels throughout pregnancy. The required increase appeared greater between the 1st and 2nd TM than between the 2nd and 3rd TM, although our limited sample size precludes formal statistical analysis of these differences. This may appear counterintuitive as increased blood volume in the 3rd TM would seem to necessitate a larger increase in LTG dose. However, we controlled for this by normalizing LTG dose for wt. (LTG/wt). These results suggest that significant increases in LTG dose/wt are required to maintain preconception LTG serum levels across TM. Presently prospective data is being collected to expand on these findings. Abstract: .sup.1 Sandra S. Funayama , .sup.1 Edna A. Monteiro , .sup.1 Vera C. Terra-Bustamante , .sup.1 Ricardo Guarnieri , .sup.1 Otavio I. de Faria , .sup.1 Tonicarlo R. Velasco , .sup.1 Veriano Alexandre Jr. , .sup.1 Lauro Wichert-Ana , .sup.1 Joao A. Assirati Jr. , .sup.1 Carlos G. Carlotti Jr. , .sup.2 Li Li Min , .sup.1 Marino M. Bianchin , .sup.1 Jaime E. Hallak , and .sup.1 Americo C. Sakamoto (Department of CIREP-Neurology, Psychiatry, and Medical Psychology, Ribeirao Preto School of Medicine, University of Sao Paulo at Ribeirao, Sao Paulo, Brazil ; and Department of Neurology, UNICAMP-State University of Campinas, Campinas, Sao Paulo, Brazil) Rationale: A large proportion of patients with mesial temporal lobe epilepsy associated with hippocampal sclerosis (MTLE-HS) can be rendered seizure-free after surgery. In these patients, the benefit of surgical treatment seems to extend beyond seizure control, affecting many other aspects of their health and quality of life. Some of these benefits might be quite subjective and because of that might remain underevaluated. Recently, we have validated the Subjective Handicap of Epilepsy scale (SHE) for Brazilian patients. Here, we use SHE scale as a tool to further evaluate the benefits of surgical seizure control in refractory MTLE-HS. Methods: A validated version for Brazilian patients of the SHE scale was administered to 74 patients with refractory MTLE-HS during the presurgical evaluation, and to another 89 surgically treated MTLE-HS patients with good seizure control, in a case-control study. To assess seizure outcome we applied Engel's scale and we definied good seizure control as patients with Engel classes 1. The patients were matched by pre-surgical age, neurophysiological data, and neuroimaging variables. They were also matched by socio-economic level. For numeric variables, we used Mann-Whitney test. For categorical variables we applied Fisher's exact test. A Multivariate Analysis of Variance (MANOVA) with Wilks' Lambda multivariate test was used to assess mean differences of the six sub-scales scores of SHE scale. For those, results are expressed in mean (95% C.I.), and were significant if p < 0.05. Results: The multivariate test showed significant differences between the two groups of patients (Walk's Lambda = 0,045; F(6,156) = 554.17; p < 0.001; observed power = 1.00). The mean results [95 C.I.] of the six sub-scales scores of SHE scale is presented in Table. Conclusions: Post-surgical good seizure control improves all the six sub-scales scores of SHE in refractory MTLE-HS. Our results strongly support that good post-surgical seizure control in medically refractory MTLE-HS improves health quality of life. (Supported by FAPESP.) Abstract: .sup.1 Helena Gauffin , .sup.1 Lena Raty , and .sup.1 Birgitta Axelsson-Soderfeldt (Department of Neurology and Locomotion, Division of Neurology, Faculty of Health Sciences,Linkoping University, Linkoping, Sweden) Rationale: The aim of this study was to compare seizure frequency and antiepileptic drug (AED) treatment with the situation five years earlier in a group of young people with uncomplicated epilepsy. Methods: We followed 99 young people with uncomplicated epilepsy from 1999 to 2004. Data was collected from medical records from swedish hospitals and from self-reported questionnaires.We collected data about medication and seizure frequency. Results: One third of the patients were still using old anti-epileptic drugs (Phenytoin, Karbamazepin or Valproate) compared to 67% five years earlier. The number of patients using new antiepileptic drugs or a combination of drugs had increased from 23 to 41%. Despite of these changes of medication the seizure frequency had not improven significantly. However, the number of patients not using antiepileptic drugs had increased from 9 to 19% during these five years.These patients were also seizure free. Conclusions: There have been considerable changes in the use of anti-epileptic drugs during the last few years in this group and that reflects a shift in AED use in Sweden during this period. In this study we could not demonstrate that these changes of medication have improved the seizure frequency for young people with uncomplicated epilepsy. Abstract: .sup.1 Ian L. Goldsmith , and .sup.1 Amit Verma (Department of Neurology, Peter Kelleway Section of Neurophysiology, Baylor College of Medicine, Houston, TX) Rationale: The intracarotid amobarbital procedure (IAP) remains an important part of presurgical evaluation. The initial doses of amobarbital required to induce an adequate clinical response varies in adults. Variation in dosing in 2 patients with diffuse unilateral hemispheric abnormalities, who underwent IAP as part of a presurgical evaluation, is reported. Methods: 5 intracarotid amobarbital injections were performed in 2 adult presurgical patients with diffuse unilateral hemispheric abnormalities. The patients had previously undergone prolonged video EEG monitoring, MRI head imaging and neuropsychological testing. Angiography was performed first to determine the degree of crossfilling. EEG monitoring was performed throughout the procedure. Injected doses of amobarbital ranged from 62.5 to 150 mg. Effectiveness of dosing was determined by the presence of a contralateral hemiplegia, ipsilateral slowing on EEG and the absence of obtundation that precluded adequate clinical testing. Testing was performed to determine lateralization of language and memory function. Results: The 2 patients studied had previously performed MRI imaging demonstrating diffuse left hemispheric abnormalites. In the first a multilobar neuronal migration abnormality was seen and in the second multilobar encephalomalacia was present. Neuropsychological testing in both patients was nonlateralizing. In both patients the IAP was performed with injection of the affected hemisphere first. An initial injection of 100 mg was increased to 125 mg in the first patient and to 150 mg in the second in order to achieve an effective dose. In injecting the contralateral, unaffected hemisphere, the first patient had an inital dose of 100 mg given that resulted in obtundation that precluded any clinical testing. This injection was later repeated at an initial dose of 50 mg that was then increased to 62.5 mg to achieve the desired effect with only brief obtundation. The second patient was given an initial dose of 75 mg that was effective with minimal obtundation. Conclusions: In the 2 patients with diffuse unilateral hemispheric abnormalities who underwent a bilateral IAP a significant difference in the required amobarbital dosing was observed. The effective dose of amobarbital was significantly less for the uninvolved hemisphere and was significantly higher for the affected hemisphere. Abstract: .sup.1 Zakaria Hakma , .sup.1 Mercedes P. Jacobson , and .sup.1 Ausim Azizi (Neurology, Temple University Hospital, Philadelphia, PA) Rationale: Nontraumatic intracerebral hemorrhage is bleeding into the parenchyma of the brain that may extend into the ventricles and, in rare cases, the subarachnoid space. Each year, approximately 37,000 to 52,400 people in the United States have an intracerebral hemorrhage. Seizures associated with nontraumatic ICH have not been fully characterized. There is little information relating seizures to clinical presentation, course, or outcome in ICH and the existing data do not define the role of anticonvulsants during the acute hospitalization for ICH. This study addresses these issues through a retrospective evaluation of patients with hypertensive ICH. Methods: We analyzed 123 consecutive admissions to our NICU between 12/2005 and 5/2006. Subjects with head trauma, brain neoplasm, aneurysms or any etiology to explain the ICH besides hypertension (HTN) were excluded. 78 ICH met the criteria. Clinical features, timing of seizures relative to hemorrhage onset, factors predisposing to ICH, including HTN, bleeding diathesis (thrombocytopenia or anticoagulation therapy), worsening of clinical and radiographic findings after the seizure were assessed. Results: Demographics: Median age 62.8, range 35 to 90 years. Ethnicity: African American 59%, Caucasian 25%, Hispanic 11.5% and Asian 4% . 85.9% had Chronic HTN and 33% had prior CVA; 4 subjects had history of seizures. Average BP on admission was 170/95. Fourteen patients were coagulopathic. Clinical presenting features were: change in mental status 35%, Hemiparesis 25.6%, Headache18%, Aphasia 5%, and Seizure 2.5%. ICH location: Basal ganglia/thalamus 51.2%, Lobar 41%, Cerebellar 2.5%, and Pons 3.8%. Intraventricular extension occurred in 38.4%. Of 78 subjects, 2 had seizures, both upon presentation, both with prior seizures. 75 subjects were evaluated for AED prophylaxis (AEDP). 23 (30%) received AEDP. Lobar hemorrhages comprised 46% of AEDP but only 32% of the no AEDP group. Average ICU stay for all subjects was 7.8 days compared to subjects who received AED 10.6 days and subjects who did not receive AED 6.2 days. Adverse effects attributed to AEDP included fever, abnormal liver functions and hypotension. Based on one-way ANOVA there was a significant difference on length of stay between subjects who did not receive AED prophylactic and those who did (F = 5.52, p = 0.022). Conclusions: In this series of HTN ICH, the incidence of acute seizures was 2.6% and those subjects with seizures had antecedent epilepsy. This contrasts to previous reports. Features of our population would suggest a greater tendency to seizure. Subjects with seizures had their ictal event at or shortly after, hemorrhage onset, a finding previously reported. In our series 23 subjects received prophylactic AED therapy; however none of the remaining 55 subjects had clinical seizures during their hospital stay. Additionally, those who received AEDs had longer ICU stays and increased non-neurologic complications. Abstract: .sup.1 Nathan C. Hantke , .sup.1 Michael J. Doherty , and .sup.1 Alan M. Haltiner (Neuroscience Institute/Epilepsy Center, Swedish Medical Center, Seattle, WA) Rationale: The incidence of psychogenic nonepileptic seizures (PNES) varies with demographic, psychological and clinical history variables (i.e. gender, MMPI profiles, age of seizure onset, etc). Medication use may also differ; PNES patients generally take fewer AEDS than patients with epileptic seizures (ES). Few studies have explored the utilization of other medications within these two populations. This study evaluated the types of medications used by patients with PNES and ES. Methods: Hospital records of a consecutive series of adult patients who completed video-EEG monitoring and were diagnosed with ES (n = 178) or PNES (n = 170) were prospectively reviewed. Patients with both pseudoseizure and an epilepsy diagnosis were excluded. Medication variables examined include: the type and number of different prescribed medications, including current AEDs, previously tried AEDs, medications for psychiatric symptoms (typical and atypical SSRI, tricyclics, anti-psychotics, anxiolytics), pain medications (both narcotic and non-narcotic), and others. Results: PNES patients were on nearly twice as many total medications at the time of admission (M = 6.0) than ES patients (M = 3.2, p < .01). PNES patients were more likely to be taking psychiatric medications, anti-hypertensives, inhalers, gastric-reflux medications, and analgesics(all p < 0.01). Nearly half of the PNES group took analgesics (46%) compared to only 11% of patients with ES (p < 0.01). Narcotic pain medication use was common in the PNES (32%) population but rare in the ES group (3.4%, p < 0.01). Patients with epilepsy were currently taking more AEDS and were tried on more in the past than PNES patients (p < 0.01). However, the duration of illness was longer in the ES group (p < 0.01). When the total number of AEDS tried was divided by the duration of the disorder, PNES patients tried more AEDS per unit of time than ES patients (p < 0.05). 28% of all PNES patients took benzodiazepines compared to only 7% of ES patients (p < .01). Benzodiazapine monotherapy for "seizures" was reported exclusively in the PNES group (p < 0.01). Conclusions: Our findings demonstrate increased utilization of a variety of different medications by patients with PNES as compared to ES. The increased medication usage might reflect increased physician utilization, drug intolerance by psychogenic patients, and other markers of somatization. PNES patients were not only taking more AEDS per duration of their "epilepsy", but also more commonly use benzodiazepines as monotherapy. The increased benzodiazepine usage among PNES patients could reflect physicians' attempts to treat both psychiatric symptoms such as anxiety and presumed epilepsy. In combination with other clinic data, the medication profiles may provide clues about the likelihood of diagnosing PNES or ES. Abstract: .sup.1 Nicole C. Hank , and .sup.1 Steve S. Chung (Barrow Neurological Institute, Neurology, Phoenix, AZ) Rationale: Patients who volunteer for phase II and phase III clinical trials can gain access to investigated drugs long before these compounds are approved by the FDA. Patients participating in double blind, placebo-controlled trials of new medications are invited to partake in an open label extension study. Patients are openly given the active medication at this stage, regardless of their assignment in the initial trial, as long as the study permits. In order to determine if the open label extension phase has an impact on patient enrollment, we surveyed study participants who are currently participating in open label extension trials at our epilepsy center. Methods: All epilepsy patients who are currently participating in one out of 6 phase II or III trials at the Barrow Neurological Institute were included in the study. Questionnaires were administered to patients during their clinic visits or via email. Participating patients were asked questions regarding how open-label extension trials impacted their decision on participation and continuation in a study. In addition, patients were asked to rate their reasons on a 1-5 scale (5 being the most influential) why they would remain in the study in case of potential adverse events or no benefit in the initial placebo-controlled phase. Results: All 41 patients replied to questionnaires. The median age of patients was 32, ranging from 19 to 64 years old. Almost two-third of them were female (68% compared to 32% of males). Thirty-seven of the 41 study participants (90%) answered that they decided to participate in a trial because of the offered open label extension phase. Ninety-Five percent (35 out of 37) admitted that they would have not participated in a research trial if an open label extension had not been offered. Despite 56% (23 out of 41) of patients experienced medication related adverse events (AEs), 87% of them participated in the open label extension phase. All the patients who experienced little or no medication related AEs also proceeded to an open extension. Even though 15 out of 41 patients experienced no significant seizure improvement during the initial placebo controlled phase, all of them participated in the open extension phase. The majority of the patients (83%) identified that the available open label extension phase was one of most influential reasons for remaining in clinical trials (X = 4.2). Conclusions: Open label extension phase is an important and motivating factor why patients would initially participate in clinical trials. Regardless of success of the initial placebo controlled phase, the majority of patients would participate in an open label extension phases. Thus, offering open label phase could significantly increase the patient recruitment in the early phase of clinical trials. Abstract: .sup.1 Cynthia Harden , .sup.1 Blagovest Nikolov , .sup.1 Douglas Labar , .sup.2 Kristen Fowler , .sup.3 Michael Sperling , .sup.3 Joyce Liporace , .sup.4 Page Pennell , .sup.2 Donald Schomer , .sup.4 Melanee Newman , .sup.3 Gwen Taylor , and .sup.2 Andrew Herzog (Neurology, Weill Medical Collage of Cornell University, New York, NY ; Harvard Neuroendocrine Unit, Beth Israel Deaconess Medical Center, Boston, MA ; Neurology, Thomas Jefferson Hospital, Philadelphia, PA ; and Neurology, Emory University School of Medicine, Atlanta, GA) Rationale: Seizure severity is an important aspect of epilepsy, however the relationship between quality of life in epilepsy and seizure severity has been incompletely explored. Further, since various seizure severity scales have been used in clinical research, this association is limited by the scope of the seizure severity scale. Using a dataset from the baseline phase of clinical treatment trial, the relationship between seizure severity and aspects of quality of life were evaluated. Methods: Data obtained at the initial visit of the baseline phase of subjects enrolled in the double-blind, placebo-controlled, randomized trial entitled "Progesterone Therapy for Women with Epilepsy" were used. Seizure severity was assessed using the National Hospital Seizure Severity Scale (NHS3) which consists of 7 questions with 2-4 degrees of responses and scores that range from 1-27 (higher is more severe). This scale focuses on events that occur during the seizure and injuries, rather than postictal recovery. Quality of life was assessed using the QOLIE-31. Statistical methods were Spearman's bivariate correlations and linear regression. Results: One hundred eighteen subjects were evaluated; all were women between the ages 18-45 years. Domains of the QOLIE-31 that significantly correlated with seizure severity were seizure worry (r = 0.265, p = 0.004) and social functioning (r = 0.280, p = 0.002). The overall QOL and cognitive domain each slightly correlated with seizure severity (r = 0.212, p = 0.021; r = 0.205, p = 0.026, respectively). When controlling for the effect of a potential confounding variable, the Beck Depression Inventory, the regression of seizure severity with the QOLIE seizure worry and social functioning remained significant (p = 0.004; p = 0.001, respectively). Conclusions: Seizure severity using the NHS3 correlated with the seizure worry and social functioning domains of the QOLIE-31. These findings indicate that severe and potentially injurious seizure behaviors affect specific aspects of quality of life in epilepsy. Furthermore, these effects of seizure severity on quality of life persist even after accounting for the influence of depression and likely contribute to anxiety and socially avoidant behavior for persons with epilepsy. (Supported by NINDS: NS39466.) Abstract: .sup.2 Andrew G. Herzog , .sup.1 Mark Quigg , .sup.2 Kristen M. Fowler , and Progesterone Trial Study Group (Neurology, University of Virginia, Charlottesville, VA ; and Neurology, Beth Israel Deaconess Hospital, Harvard University, Boston, MA) Rationale: Estimates of the prevalence of catamenial epilepsy differ in definitions used to designate catamenial epilepsy. Past investigations have focused on the relationship between seizure occurrence and either menstrual phase or reproductive hormone levels. In contrast, few studies have evaluated the periodicity of seizures in women to test the hypothesis that they have circalunar or ultralunar patterns of occurrence independent of a priori hormonal considerations. In the present study we determined seizure periodicity according to a "menstrual clock" provided by a common phase marker of the onset of menstrual bleeding. Methods: Baseline, pretreatment data from seizure and menstrual diaries of [proportional to]3 months duration were obtained from women enrolled in a trial of hormonal therapy for localization-related epilepsy. Midluteal serum progesterone levels classified each cycle as ovulatory (>5ng/ml) (OC) or anovulatory (AC). To group seizure data, durations of individual cycles were normalized to a common menstrual phase and period. The start for each cycle was identified as the onset on menstrual bleeding (day 1). Intervening days were then converted to 28 normalized cycle days. Normalized time series were then combined to create mean seizures per menstrual day divided between OC and AC. Rhythms of seizure occurrence were then characterized with cosinor-nonlinear least squares analysis that iteratively tested all potential rhythms for best-fit described by period, amplitude, phase, and mean within designated confidence limits. Rhythms were designated significant when the lower confidence limit of the amplitude > 0 by single-tailed p-value < 0.05. Results: 100 patients provided 3344 seizures within 293 cycles. The duration of cycles before normalization ranged from 20-37 days (mean [+ or -] standard deviation, 27.92 [+ or -] 2.89). 77% cycles were OC (2587/3344); 20% were AC (673/3344). 3% of seizures occurred during cycles with indeterminate status and were discarded. Seizures during OC displayed a significant circalunar rhythm of occurrence of 31.7 days (26.9-36.5, 90% confidence interval) with peak phase of occurrence at onset of menses. Seizures during AC not only featured a circalunar rhythm of 28.6 (26.4-30.7) days that peaked at menses onset but were also characterized with ultralunar rhythms of 14.0 (13.6-14.3) days and 9.0 (8.9-9.1) days. Conclusions: This study determined that seizures in women with localization-related epilepsy occur in a circalunar rhythm during both ovulatory and anovulatory menstrual cycles. Ovulatory and anovulatory cycles differ in seizure timing with the seizures in the latter occurring in ultralunar rhythms in addition to the predominant circalunar rhythm. This finding supports the existence of catamenial epilepsy and differences in patterns of seizure occurrence between ovulatory and anovulatory cycles. Abstract: .sup.1 Yue-Loong Hsin , .sup.1 Min-Fei Chuang , and .sup.2 Tomor Harnod (Department of Neurology, Buddhist Tzu Chi General Hospital, Hualien, Taiwan ; and Department of Neurosurgery, Buddhist Tzu Chi General Hospital, Hualien, Taiwan) Rationale: To study the clinical grounds, neuroimaing and EEG activity from a three-generation of family with autosomal dominance of adult-onset myoclonic epilepsy. Methods: A 3-generation family of 10 affected members in east-Taiwan was studied with video-EEG and brain MRI. Results: Autosomal dominant hereditary was showed. The myoclonic seizure semiology is characterized by bilaterally symmetric jerks of the shoulders, arms, or legs. Ictally, the consciousness is not affected. The ictal EEG demonstrated polyspike-wave. Rare generalized tonic-clonic seizures occur in 2 patients. No absence is reported. The myoclonic seizures do not related to sleepiness. And no involuntary movement, such as chreoasthetosis or astasia was accompanied. No structural disorder was demonstrated. A lower dose of valproic acid (200 to 500 mg/D) is required to stop the myoclonic seizure. Conclusions: Mutations of the GABRA1 gene, CACNB4 gene and CLCN2 gene have been described in the autosomal dominance inheritance patterns of juvenile myoclonic epilepsy (range, 10 to 20 years). Late-adult onset of myoclonic epilepsy is not a recognized epileptic syndrome. This is a distinct familial epileptic syndrome different from recognized myoclonic epilepsies in infancy and early childhood, or myoclonias and absences in childhood and adolescence. Abstract: .sup.1 Kerry L. Hulsing , .sup.1 Linda Selwa , and .sup.1 Karuna Mandava (Department of Neurology, University of Michigan, Ann Arbor, MI) Rationale: Employment status is a significant issue for people with epilepsy. Employment has been described as an important factor in their quality of life, yet high unemployment rates have been previously reported. The purpose of this study was to investigate employment status and factors associated with employment in patients with refractory epilepsy. Methods: Data were collected from 44 adult epilepsy patients undergoing evaluation for epilepsy surgery at the University of Michigan. Data was obtained via self-completed questionaire and from the patient's medical record. Students were excluded. Results: The unemployment rate in this population was 72%. Overall quality of life was rated higher in the employed group, but did not reach statistical significance. There was not a significant difference in age of onset of seizures or seizure frequency between the employed and unemployed. Those that were employed had attained a higher level of education on average, with 40% having a college degree, compared to 11% in the unemployed population. The importance of employment to quality of life was rated highly, with 75% of the population reporting it as very important. 17% of the population reported that they continued to drive. Lack of seizure control was ranked as the primary variable hindering employment by 96% of the unemployed. Driving status alone had the highest odds of predicting employment status (p = 0.001). Conclusions: Employment status is an important quality of life issue for people with epilepsy, yet the unemployment rate in those with medically refractory epilepsy is much higher than that of the general population. Awareness of this issue may be helpful to epileptologists in caring for this patient population. Abstract: .sup.1 Jaromir Janousek , .sup.1 Arkady Barber , and .sup.1 Pavel Klein (Epilepsy, Mid-Atlantic Epilepsy and Sleep Center, Bethesda, MD) Rationale: Estrogens are generally thought to have a proconvulsant effect.In animals, estrogens activate spike discharges and lower seizure threshold. In humans, catamenial seizure exacerbation at the time of ovulation has been suggested to be due to a peri-ovulatory surge of serum estradiol level (Herzog AG et al., Epilepsia 1997). In the only such study in humans, intravenous administration of conjugated estrogens (premarin) was reported to activate epileptiform discharges and seizures in some women with epilepsy (Logothesis J, et al., Neurol 1959). More recently, oral administration of premarin in perimenopausal women with epilepsy indicated dose-related seizure exacerbation with premarin (Harden C, et al., Epilepsia 2004). The purpose of this study was to determine whether the reported proconvulsant effect of estrogens could be used to activate seizures in women with epilepsy undergoing EEG monitoring for pre-surgical work up. Methods: Women of reproductive age with medically refractory localization related epilepsy (LRE) were evaluated. Subjects undergoing in-patient long term video-scalp EEG (LTVEEG) monitoring were injected with 10 mg premarin i.v. at 8 am during the first or second day of monitoring, before AED discontinuation. No changes in AEDs were made for 24 hours after premarin injection. In two subjects, a second injection of 50 mg was given 24 hours after the 10 mg injection. All subjects had been evaluated for presence of catamenial seizure exacerbation using both subjects' subjective impression and previously described criteria (Herzog AG et al., Ann Neurol 2004). Serum levels of estradiol, estrone and progesterone were measured immediately before and 1 and 2 hours after premarin injection. EEG was recorded continuously and stored for manual and automated spike and seizure analysis. EEG was analysed for spike and seizure frequency using 30 minute bins during 12 hours before and 24 hours after the injection. Results: 6 women (age range 23-45) were evaluated. 4 women had subjective impression of catamenial seizure exacerbation, (3 Type I, perimenstrual exacerbation, 1 Type 1 and 2, perimenstrual and periovulatory exacerbation), but this could be confirmed with prospective seizure diary documentation in only 1. 4 subjects had TLE (left, n = 2; right, n = 1; bitemporal, n = 1), 2 had extratemporal LRE. 0/6 women had a seizure during 24 hours after premarin injection with either 10 mg (n = 6) or 50 mg (n = 2 of the initial 6). There was no change in interictal spike frequency following premarin injection, including absence of spikes for 24 hours after premarin injection in 2 women. Conclusions: Intravenous injection of conjugated estrogens did not activate ictal or interictal epileptiform activity in this small study of women with medically refractory epilepsy. Estrogen injection may not be a useful activating procedure in these patients. A larger study is needed to confirm these preliminary findings. Abstract: .sup.1 Sigmund Jenssen , .sup.1 Eve Khlyavich , and .sup.1 Jyoti Pillai (Neurology, Drexel Medical College, Philadelphia, PA) Rationale: Some patients whose seizures routinely is preceded by an aura claim they are able at times to delay or abort the seizure. It is unclear if this subjective experience is real or if the experience somehow is caused by the seizure itself. We wanted to explore if there is any relationship between this claim and seizure type, the presumed seizure focus and the aura. Methods: After approval by IRB was obtained patients answered a questionnaire in the presence of the doctor and their charts were reviewed. Only patients with verified epilepsy were included and all had to have a reliable aura. We correlated the claim of influence over seizures with the epileptic focus according to EEG and/or brain MRI, duration and quality of aura and presence of generalized tonic-clonic seizures using Chi Square. Results: 65 patients were included. Mean age was 45 years (range 20 to 70) of which 20 were males. 28 patients (43%) reported that they some times could delay or abort seizures voluntarily. The methods included directing the thought content (20 patients), moving the body (16 patients) or stimulating the body (14 patients). More patients who reported some influence had extratemporal focus (p < 0.001). No correlation was found with type and duration of aura or presence of generalized tonic clonic seizures. Conclusions: A substantial portion of patients who have seizures with aura claim they are able to influence the progression of their seizures. This claim was more common among patients with extratemporal focus. A possible explanation could be that afferent inputs into the cerebral cortex exert an inhibitory effect thereby delaying the propagation of an exitatory epileptic focus. Further research may provide insight into the mechanisms that may underlie this phenomenon. Abstract: .sup.1 Barbara C. Jobst , .sup.1 Karen L. Gilbert , .sup.1 Vijay M. Thadani , and .sup.1 Gregory L. Holmes (Section of Neurology, Dartmouth-Hitchcock Medical Center, Lebanon, NH) Rationale: Pregnancy registries are increasingly established to correlate malformation rate with use of AEDs in women with epilepsy. Other seizure and AED related problems remain unanswered in the treatment of epileptic pregnant women. Seizures during pregnancy are assumed to be detrimental, but significantly increased adverse outcomes have not been systematically studied. Methods: Retrospective review of 35 pregnancies that were followed in a highly specialized pregnancy and seizure clinic in an academic institution. Seizure occurrence, complication rate, abortion rate, malformation rate, preconceptive counseling and AED adjustments and levels were recorded during pregnancy. Results: Of 35 pregnancies in 34 women, there were 22 deliveries, one first-trimester abortion and one 20-week abortion due to severe neural tube defect in a pregnancy on lamotrigine monotherapy (LTG). 11 women have not delivered yet. There were 2/35 (5.7%) true malformations (neural tube defect, craniosynostosis) and 11/33 viable pregnancies were complicated by other than seizures (oliohydramion/low birth weight in three, rash due to AEDs in three, HELLP syndrome, bleeding, severe migraines, prematurity). 23 of 35 (65%) of all and 14 of 22 (63%) of completed pregnancies were complicated by seizures. This reflects the specialized referral pattern. Seizure occurred equally in the first, second or third trimester (p > 0.05). Of the six pregnancies with malformations, prematurity or low birth weight, four had seizures during pregnancy, two had generalized tonic-clonic convulsions (GTC). Overall 7/35 (20%) pregnancies were complicated by GTCs with two off-spring related complications (low-birth weight and neural tube defect). Only one patient had seizures during delivery (4.5%). Three previously well controlled patients had reoccurrence of myoclonic seizures when switched to lamotrigine monotherapy prepregnancy. In 17/35 (48.5%) of pregnancies, AEDs needed to be changed either due to seizure exacerbation, side effects or decreasing levels. LTG (n = 10) required regular adjustment of dosing due to decreasing levels, carbamazepine, valproic acid, phenytoin, levetiracetam and topiramate levels were more stable. Data on oxcarbazepine was insufficient. Three patients had their first seizure during pregnancy and all of those patients were not treated with AEDs. 12 (34.2%) of patients underwent prepregnancy counseling. Conclusions: Management of seizures during pregnancy remains challenging. The impact of seizures in pregnancies with or without AEDs needs to be studied in larger studies. Systematic investigation of AED adjustments during pregnancy is required to assess utility. Myoclonic epilepsies treated with LTG alone can be problematic in pregnancy. Specialized clinics can increase preconceptive intervention and improve AED management. Abstract: .sup.1 Luydmila Jovine , .sup.1 Blagovest Nikolov , .sup.1 Douglas Labar , .sup.2 Stephen Ferrando , .sup.1 Padmaja Kandula , .sup.1 Sona Narula , and .sup.1 Cynthia Harden (Neurology, Weill Medical College of Cornell University, New York, NY ; and Psychiatry, Weill Medical College of Cornell University, New York, NY) Rationale: Patients with non-epileptic psychogenic seizures (NES) frequently have affective disorders, anxiety disorders and personality disorders. Further, the presence of personality disorders in NES patients is associated with intractability. However, the personality disorders of NES patients that impair coping skills and contribute to the occurrence of NES are not fully understood. We sought to determine the type of personality disorders associated with NES patients compared to epilepsy patients by prospectively evaluating adult patients admitted for video-EEG. Methods: Consecutive adult patients admitted to the hospital for video-EEG monitoring found to have NES were studied; the epilepsy comparator group was derived using a simultaneously admitted patient with findings on video-EEG confirming the diagnosis of epilepsy. No patients who had findings of both disorders on video-EEG were studied. Patients who could communicate in English and without significant cognitive impairment were included. Personality was assessed using the Structured Clinical Interview for DSM-IV Axis II Personality Disorders. Personality disorders were then divided into personality clusters described in the DSM-IV, which are based on functional status: A (paranoid, schizotypal, schizoid), B (borderline, histrionic, antisocial, narcissistic) or C (avoidant, dependent, obsessive-compulsive). Results: Sixteen NES subjects were studied, 15 were women. Sixteen epilepsy subjects were studied, 10 were women. Three subjects in the NES group had no personality disorder, compared to four in the epilepsy group. Twelve subjects in the NES group met criteria for either Cluster A or B type personality disorders, or both, compared to six in the epilepsy group. However, only one NES patient met criteria for Cluster C personality disorder, compared to six in the epilepsy group. The distribution of having a Cluster A and/or B personality disorder compared to Cluster C or no personality disorder was significantly different between NES and epilepsy groups (chi-square p = 0.03). Conclusions: NES patients are more likely meet criteria for a personality disorder that is more dysfunctional, in the Cluster A or B group, compared to epilepsy patients, who are more likely to have Cluster C personality disorders, predominantly obsessive-compulsive and avoidant. We suggest that the Cluster A or B personality features contribute to the occurrence of NES, while Cluster C behaviors in epilepsy patients may be in part a reaction to the experience of having chronic recurrent seizures. Abstract: .sup.1 Eleanor B. Justen , .sup.1 Glenn Catalano , and .sup.1 William O. Tatum IV (Department of Psychiatry, UMDNJ-Robert Wood Johnson Medical School, Piscataway, NJ; Department of Psychiatry, University of South Florida, Tampa, FL; and Department of Neurology, University of South Florida, Tampa, FL) Rationale: Psychiatric profiles of patients with PNES have been described. Limited comparison exists between historical self-reported psychiatric profiles with formal in-patient assessment. We compared patient self-reported psychiatric diagnosis with formal in-patient psychiatric evaluations preformed at the time of PNES diagnosis. Methods: Twenty six patients with confirmed PNES by inpatient VEEG done at a tertiary care epilepsy center were reviewed. The patients demographic information, history of headaches, suicide attempts, history of abuse, insomnia, chronic pain, self reported psychiatric problems, psychiatric medications, Anti-epileptic drugs, and psychiatric evaluation were collected and compared. Axis I diagnosis was made according to DSM-IV TR criteria. Statistical comparison was done looking at relative frequency of symptoms. The number of patients that self reported depression was compared to the number of patients to receive an Axis I diagnosis using the Mann-Whitney test. Data is reported as median (mean [+ or -] SD). Results: The patients were 42.3% males and had a median age of 43.5 years (40.4 [+ or -] 15.5). Only 19 of 26 patient received formal psychiatric evaluations while in hospital. Of the 19 patients, 89.5% had self reported mood disorder compared to 84.2% that received a formal inpatient Axis I diagnosis showing no statistical difference (p = 0.783). 76.9% of all patients self reported a history of psychiatric symptoms. Vegetative symptoms such as headaches and insomnia were reported in 73% and 58% respectively. Seven patients (27%) reported suicide attempts in their lifetime. 18 of the 26 (69%) were already taking SSRIs at the time of evaluation, with 53.8% having been exposed to multiple psychiatric medications. Only 4% required altered medications during their hospitalization. Conclusions: We conclude that self-reported psychiatric diagnoses are concordant with psychiatric evaluation in patient PNES diagnoses. The majority of PNES patients were currently taking psychotropic medication at the time of in-patient diagnosis, and required no modification following formal psychiatric evaluation. The primary focus for ongoing psychiatric treatment in PNES patient should be provided on an outpatient basis. Abstract: .sup.3 Buchwaldt Fredrik , .sup.2 Jan-Anders Ahnlide , .sup.2 Ingmar Rosen , and .sup.1 Kristina Kallen (Neurology, University Hospital, Lund, Sweden ; Clinical Neurophysiology, University Hospital, Lund, Sweden ; and Neurology, University Hospital, Malmo, Sweden) Rationale: To assess the additional value of subtraction ictal SPECT co-registrated to MRI (SISCOM) for localisation of the epileptogenic zone in epilepsy surgery. Methods: We retrospectively analyzed 49 consecutive epilepsy patients who had undergone invasive monitoring between January 2000 and Mars 2006 and had presurgical examination with high resolution 1.5T MRI or higher field strength. Attempts for ictal SPECT as part of the pre-surgical work-up was made for 29 patients, two patients had no seizures during ictal SPECT preparedness, SISCOM analysis showed areas with significant hyperperfusions in 26 patients. By SISCOM analysis interictal and ictal images were normalized, subtracted and coregistered with MRI. The standard deviation of the ictal hyperperfusion was measured in each hemisphere and the hemisphere with the least standard deviation was used as reference. Hyperperfusions exceeding 3.5 SD were considered significant. Two independent epileptologists evaluated if SISCOM results (A) altered the hypothesis and extended the strategy for electrode placement at invasive recording, or (B) did not alter the strategy. We defined that SISCOM had an impact on seizure outcome if the seizure onset zone was seen in electrodes overlying a brain region with a significant hyperperfusion. Engel class I-II was considered favorable outcome at one-year postoperative follow-up. SISCOM was considered a prerequisite for favorable outcome when concordant with invasive ictal onset in extended electrodes. Results: Twenty-six patients were offered surgery; 20 had localizing data from SISCOM and > 1 y follow-up whereof 40% had favorable outcome. A. SISCOM findings were complementary to other localizing non-invasive findings and altered and extended the strategy for electrode placement at invasive recording in 15 patients. SISCOM was a prerequisite for seizure outcome in all 6 patients with favorable outcome. Nine patients had poor results from surgery; SISCOM was concordant with invasive EEG in 6 patients, and discordant with invasive EEG with no impact on outcome in 3 patients. B. SISCOM findings did not change the hypothesis for electrode placement at invasive recording in 5 patients, but SISCOM was confirmatory with other localizing data in the 2 patients with favorable outcome. Three patients had poor results from surgery; SISCOM was concordant with invasive EEG in 1 patient, and discordant with invasive EEG and with no impact on results in 2 patients. When SISCOM was concordant with invasive EEG the predictive value for favorable outcome was 53%. Conclusions: SISCOM is valuable for the identification of the epileptogenic zone in patients with drug resistant epilepsy scheduled for invasive video-EEG. In all patients with favorable seizure outcome at one year follow-up SISCOM analysis was either a prerequisite for the favorable result or concordant with other localizing data. Abstract: 1,2 Maria Kaartinen , .sup.3 Anna-Maija Haapala , .sup.5 Markku Maki , .sup.4 Jani Raitanen , .sup.1 Tapani Keranen , .sup.6 Katri Kaukinen , and .sup.1 Jukka Peltola (Department of Neurology and Rehabilitation, Tampere University Hospital, Tampere, Finland ; Medical School, University of Tampere, Tampere, Finland ; Department of Clinical Microbiology, Center for Laboratory Medicine, Tampere University Hospital, Tampere, Finland ; School of Public Health, University of Tampere, Tampere, Finland ; Department of Pediatrics, Tampere University Hospital, Tampere, Finland ; and Department of Gastroenterology and Alimentary Tract Surgery, Tampere University Hospital, Tampere, Finland) Rationale: Previously, several studies have shown an increased prevalence of antiphospholipid (aPL) and anti-glutamic acid decarboxylase (GAD) antibodies in patients with refractory localization-related epilepsy (RLRE). Gliadin antibodies (AGA) have been associated with brain pathology, most commonly in patients with cerebellar degeneration. We have recently described an association between AGA antibodies, gluten sensitivity and temporal lobe epilepsy with hippocampal sclerosis. The first objective of the present study is to compare the frequencies of aPL or GAD antibodies in epilepsy patients with and without AGA. The secondary objective is to compare the frequencies of celiac disease (CD) related antibodies between RLRE and generalized epilepsy (GE). Methods: We measured CD related antibodies [AGA, anti-tissue transglutaminase and anti-endomysium antibodies] in 50 consecutive patients with GE syndromes and 48 consecutive patients with RLRE. The frequencies of aPL and GAD antibodies have been reported previously. Results: Patients with AGA-positivity had a greater prevalence of aPL antibodies and GAD-antibodies. IgG-class AGA were significantly associated with GAD-antibodies (p = 0,042) and IgG-class aPL (p = 0,018). IgA-class AGA were associated with IgG-class aPL (p = 0,029). 50% of the patients with IgG-class AGA harbour also IgG-class aPL, which is five times more than in AGA-negative patients. In patients with IgG-class AGA-positivity 33% (nine times more common than in AGA-negative patients) had also antibodies against GAD. In multiple regression analysis, age, sex, and duration of epilepsy were not associated with the prevalence of antibodies in patients with AGA-positivity. The frequencies of all measured antibodies were higher in RLRE compared with GE except with AGA-antibodies, which were more common in GE group. Conclusions: In the present study we demonstrated that both aPL antibodies and GAD antibodies were strongly associated with AGA. The co-expression of CD related autoantibodies and anti-GAD autoantibodies is of special interest, because increased expression of various autoantibodies including anti-GAD autoantibodies has been reported in patients with CD and moreover, in patients with cerebellar degeneration and AGA. This study provides additional evidence of immunological activation in epilepsy: sensitivity to dietary gluten may explain partly the presence of antineuronal antibodies such as aPL and GAD. Abstract: .sup.1 Joong Koo Kang , .sup.1 Eun Mi Lee , .sup.2 Jung Kyo Lee , .sup.2 Seok Ho Hong , and .sup.1 Sang Ahm Lee (Department of Neurology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Korea ; and Department of Neurosurgery, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Korea) Rationale: Although all patterns and frequencies of seizure onset were recorded in both temporal and extratemporal locations, some specific ictal patterns during intracranial recordings may help to predict surgical outcome. The purpose of this study is to compare the intracranial ictal EEG (ictal EEG) patterns in patients with temporal lobe epilepsy (TLE) to find the ictal patterns suggesting good surgical outcome. Methods: We analyzed total 128 seizures in 33 TLE patient underwent invasive intracranial study using bilateral temporal depth and subdural electrodes. Of the 33 patents, 26 were underwent temporal lobectmy with amygdallohippocampectomy, in whom 21 were 'seizure-free' and 5 were 'not seizure-free' with minimum 1 year follow-up. Seven were not operated. We divided the patients into 3 groups as 'seizure-free', 'not seizure-free' and 'non-operation'. We analyzed each ictal EEG charicteristics based on presence or absence of peri-ictal discharges, ictal distribution, involved number of electrodes, waveform pattern, onset frequency and interhemispheric propagation time (IHPT). Statistical analysis used chai-squre, Fisher's exact test and ANOVA test using SPSS 12.0 version. Results: There are no differences in dermographic data including sex, onset age, duration of epilepsy, risk factor and hippocampal atrophy on MRI among 3 groups. But all analyzed ictal EEG charicteristics have statistically significant difference among 3 groups. The presence of peri-ictal discharges was exclusively observed in 'seizure-free'. The lesser number of involved elcetrodes with ictal distribution of 'depth only' or 'depth and medial strip electrode' were associated with 'seizure-free' but 'non-operation' had spatially wider distribution of ictal onset involving 'diffuse strip and depth' or 'strip only'. Higher onset frequency above beta range more correlated to 'seizure-free' while slow onset frequency lesser than alpha in others. Ictal onset patterns with fast spike trains were exclusively observed in 'seizure-free', whereas pattern with rhythmic activity was common in 'not-seizure-free' and 'non-operation'. IHPT was 22. 46 [+ or -] 19.44 seconds(s) in 'seizure-free', 18.56 [+ or -] 16.74 s in 'not seizure-free', but 5.96 [+ or -] 8.04 s in 'non-operation. IHPT was signifiantly shorter in 'non-operation'. Conclusions: Intracranial ictal EEG patterns are significantly different among 'seizure-free',''not seizure-free' and 'non-operation'. These findings suggest that analysis of ictal EEG patterns help to predict surgical outcome. Abstract: .sup.1 Makoto Kawai , .sup.2 Daniel Yoshor , .sup.1 Ian L. Goldsmith , and .sup.1 Amit Verma (Peter Kellaway Section of Neurophysiology, Department of Neurology, Baylor College of Medicine, Houston, TX ; and Department of Neurosurgery, Baylor College of Medicine, Houston, TX) Rationale: Vagus nerve stimulation (VNS) has been reported to improve daytime sleepiness in patients with epilepsy when measured at 3 months post-treatment, however the long term effects of VNS on excessive sleepiness are unknown. To assess whether VNS results in sustained and clinically relevant improvements in excessive daytime sleepiness (EDS), we measured patients' daytime sleepiness levels before and after the use of VNS. Methods: Patients seen by a single epileptologist (AV) in the Baylor Comprehensive Epilepsy Center completed questionnaires for the Epworth Sleepiness Scale (ESS) prior to beginning VNS and at periodic intervals thereafter. The ESS is a standardized and validated scale used to assess sleepiness. A score of greater than 10 is indicative of excessive daytime sleepiness (EDS). To determine if VNS results in a clinically significant improvement in daytime functioning, we examined the effect of VNS on sleepiness in patients who suffered from EDS prior to VNS. Only improvements in the ESS to levels less than 10 were considered clinically meaningful. Medication dosage and seizure frequency were evaluated to determine if they influenced ESS scores. Results: Eighteen consecutive patients with refractory epilepsy who underwent implantation of a vagus nerve stimulator were asked to participate in this study and 14 complied. The mean age of patients who participated in the study was 32 years (range 17-63; Male:5, Female:9). Mean follow-up period was 13 months (range 3-32 months). The mean pre-implantation ESS score for all enrolled patients was 11.8. The mean post-implantation scores at approximately 3 months, 6 months and 12 months were 9.8, 9.6 and 10.1 respectively. Eight of the 14 patients had pre-implantation EDS (ESS score greater than 10). Of these eight EDS patients, four never exhibited meaningful improvement in their ESS at any time-point after VNS was initiated, while two had transient improvements and only two had sustained improvement at last follow-up. This could not be explained by medication changes or changes in seizure frequency. Conclusions: VNS may reduce daytime sleepiness in patients with epilepsy. The effects in our patients were not as robust as have been previously reported and the effects may also not be sustained. (Supported by Peter Kellaway, PhD Foundation for Research. We would like to thank Elizabeth Miae Kim for her help with data collection.) Abstract: .sup.1 Sung Eun Kim , .sup.1 Min Ah Kim , and .sup.1 Kang Min Park (Neurology, Busan Paik Hospital, Busan, Korea) Rationale: Although seizures are important neurological manifestations or complications of CNS infection, there have been only a few studies. The purpose of this study was to determine the incidence, etiology, and clinical factors to predict acute symptomatic seizures with CNS infections. Methods: We retrospectively reviewed the medical records of patients with CNS infections from 2000 to 2005. We excluded patients with brain abscess, neurocysticercosis, and undetermined CNS infections, and 147 patients were finally included for this study. The clinical variables were analyzed between with and without acute symptomatic seizures. Results: Of 147 patients, 23.2% (34/147) had acute symptomatic seizures. Univariate analysis revealed a significant relation between clinical variables and occurrence of acute symptomatic seizure; encephalitis as etiology of CNS infection [73.5%(25/34) vs 16.8%(19/113), p < 0.0001 by X.sup.2 test], Glasgow coma scale (GCS) [less than or equal to] 12 at admission [73.5% (25/34) vs 28.3% (32/113), p 42 years {OR 4.4, [95% CI (1.26-15.66)], p = 0.02}, encephalitis {OR 13.9, [95% CI (4.60-42.25)], p < 0.0001}, and GCS [less than or equal to]12 at admission {OR 10.3, [95% CI (3.12-34.28)], p = 0.0001} were still significant clinical variables to predict the occurrence of acute symptomatic seizures. Conclusions: Encephalitis and GCS [less than or equal to]12 at admission are the significant clinical variables to predict the occurrence of acute symptomatic seizures in CNS infection, suggesting that the more parenchymal damaged, the more vulnerable to acute symptomatic seizures occur. Abstract: .sup.1 Beth Korby , .sup.1 Gerald L. Moriarty , .sup.1 Deanna L. Dickens , .sup.1 El-Hadi Mouderres , and .sup.1 Patricia E. Penovich (Minnesota Epilepsy Group, PA, Minnesota Epilepsy Group, PA, St. Paul, MN) Rationale: Young women with epilepsy are eager to enjoy full productive lives including bearing children. There has long been a fear regarding the effect of the pregnancy on seizures and the effect of epilepsy and its treatment on the fetus/infant. Over the last five years numerous registries have been developed within the USA and globally to evaluate the effects of AEDs on pregnancy outcomes (North American Registry, EURAP, Australian and UK Registries). We evaluated our pregnancy outcomes before and after the approval of the new AEDs. Methods: Questionnaires were sent to women in our practice who were pregnant between 1995 and the present. Questions and information were given in written format and by telephone communication. Women who wished to participate signed consents and returned the information in pre-stamped return packets. Also included was a self-addressed, prepaid envelope to receive a complete report from the authors when the project is finished. Questions regarding all pregnancies were asked including age, folate use, AED use, seizures during pregnancy, delivery type, any birth/delivery complications, fetal loss, ultrasound results, and developmental status of the child with each pregnancy. For this study, "old AEDs" were defined as those approved prior to 1990 and "new AEDs" were those approved after 1990. Results: 98 questionnaires were sent out 38% have been returned thus far. 24 patients have reported information on 32 pregnancies, one with twins. 16 babies (48%) were exposed to old AEDs, 13 (39%) to new AEDs; 4 to polytherapy with old AEDs (12%). AEDs were CBZ 30%, PB 3%, PHT 3%, VPA 3%, GBP 3%, LTG 15%, LEV 12%, OXC 3%, TPM 3%. Polytherapy was PB + VPA in 2, CBZ + LTG in 1, CBZ + PB in 1. There were 29 (88%) live births. Chromosomal abnormalities were present in 2 of 4 unsuccessful pregnancies: 1 tubal (CBZ + PB), 1 stillbirth (CBZ), and 2 miscarriages (CBZ, LEV)). 13 women had seizures at some time during pregnancy: 3 had a single GTC, 8 had CPS [+ or -] SPS, 2 had only SPS. There were no cases of status. There were no major congenital abnormalities. Folate was supplemented in 62% of live births at doses between 1-4 mg/day. There were 8 (28%) abnormal outcomes in the live birth group defined as delayed development for speech or motor milestones, requiring therapy and special programming. All abnormal outcomes were seen in mothers on monotherapy and occurred with folate supplementation: 5 (62%) were with old AEDs and 3 (37%) were with new AEDs: 3 with CBZ, 3 with LTG, 1 with PB, 1 with VPA. Conclusions: Women with epilepsy can have successful pregnancies without fear of major malformations. More subtle developmental delays and need for early special services and therapy for the child may be more prevalent than previously recognized with both old and new AEDs. Awareness of this potential concern may direct a choice in AED(s), as well as allow the parents to advocate for the child earlier. Abstract: .sup.1 Vitaliy Koss , .sup.1 Peyman Andalib , .sup.1 Sigmund Jenssen , and .sup.1 Jyoti Pillai (Neurology, Drexel Medical College Hahnemann University Hospital, Philadelphia, PA) Rationale: The elderly in United States are representing a greater percentage of the general population than before. This age group has increased risk of developing epilepsy as compared to those who are younger. Previous reports indicate that in this age group the clinical presentation of new onset epilepsy is frequently varied and obvious epileptic seizures are sometimes not reported. Epileptiform activity in the EEG (EA) is highly specific for epilepsy when acute symptomatic conditions and PLEDS are excluded. Methods: We reviewed chief complaint (CC) and EEG findings in patients minimum 65 years old with EA referred to the neurophysiology laboratory at Hahnemann University Hospital between October 2004 and April 2006 for possible new onset seizures. We compared them to control group (patients younger then 65 years). Patients with known additional acute CNS event, PLEDS, recent medication change, severe metabolic derangement or prior history of epilepsy were excluded. Results: See Table for results. Fall, electrographic diffuse slowing, organic brain syndrome and medical comorbidities was more frequent (p < 0.05) in the elderly group, while there was a trend (p < 0.10) of confusion and temporal lobe spikes to be more frequent in this group as compared to controls. Conclusions: Fall, confusion, diffuse slowing in the EEG and temporal lobe spikes are more common with onset of epilepsy in the elderly compared to younger adults. Organic brain syndrome and medical comorbidities are also more common. Abstract: .sup.1 Katia Lin , .sup.2 Henrique Carrete Jr. , .sup.1 Jaime Lin , .sup.1 Luis Otavio S.F. Caboclo , .sup.1 Pedro Alessandro L. Oliveira , .sup.1 Americo C. Sakamoto , and .sup.1 Elza Marcia T. Yacubian (Department of Neurology and Neurosurgery, Uuniversidade Federal de Sao Paulo (UNIFESP/EPM), Sao Paulo, SP, Brazil ; and Departamento de Diagnostico Por Imagem, Universidade Federal de Sao Paulo (UNIFESP/EPM), Sao Paulo, SP, Brazil) Rationale: Unilateral facial paresis has been observed in patients with temporal lobe epilepsy. We aimed to assess the frequency and significance of this clinical sign in a well-defined cohort of mesial temporal lobe epilepsy (MTLE) patients. Methods: One hundred consecutive patients with MRI findings consistent with mesial temporal sclerosis (MTS) and concordant electroclinical data underwent facial motor examination at rest, with voluntary expression and spontaneous smiling. Hippocampal, amygdaloid and temporal pole volumetric measures were acquired in these exams. Thirty healthy subjects, matched through age and sex, were taken as controls. Results: Facial paresis was found in 46 patients. In 41 (89.13%) of them it was visualized at rest, with voluntary and emotional expression, characterizing true facial motor paresis (FP). In 33/46 (71.74%) patients, FP was contralateral to the side of MTS. There was no correlation between presence and lateralization of FP and neither hippocampal, amygdaloid nor temporal pole volumes. Simple febrile seizures as initial precipitating injury (IPI) were observed in 11 (68.75%) of patients without FP while complex febrile seizures occurred in 12 (70.59%) of patients with FP (p = 0.03). Also, presence of FP was positively associated with age of onset of secondary generalized tonic-clonic seizures (SGTCS) and their duration. Conclusions: FP is useful as a clinical sign with lateralizing value in MTLE and was significantly associated with history of complex febrile seizures as IPI and a longer duration of GTCS, indicating probably more widespread disease. (Supported by FAPESP, CAPES/CNPq.) Abstract: .sup.1 Lucretia Long , and .sup.1 Holly Helfant (Neurology, The Ohio State University, Columbus, OH; and Neurology/Neurosurgery, Mount Carmel, Columbus, OH) Rationale: Clinical and developmental outcomes in the offspring of pregnant women with epilepsy taking novel anticonvulsants are unknown. Although researchers continue to accrue data through pregnancy registries, practitioner's caring for women of childbearing age may benefit from preliminary data. This case series summarizes the clinical and developmental outcomes in five offspring of women with epilepsy, all of whom took levetiracetam monotherapy during pregnancy. Offspring ages range from 4 months to 4 years. Methods: A retrospective chart review from maternal and pediatric medical records regarding clinical outcomes in the offspring of women taking levetiracetam monotherapy was conducted. Developmental, cognitive and physical milestone data were collected. Maternal information included gestational seizure frequency, serum concentration levels and other relevant data. Results: Developmental milestones in five offspring of pregnant women with epilepsy taking levetiracetam monotherapy were reported. The women with epilepsy were seizure free during pregnancy. Maternal serum concentrations were therapeutic. All women took folic acid preconception and throughout their pregnancy. Data analysis revealed no cognitive or developmental abnormalities. Two siblings were diagnosed with cryptorchidism. No malignancies or additional physical or cognitive abnormalities were noted. No major or minor birth defects were recorded. Conclusions: This case series provides preliminary data. Due to the retrospective nature and small sample size, caution should be taken when interpreting the above outcomes. While future prospective studies with an adequate sample are mandatory, this case series may be encouraging for those seeking preliminary data on developmental outcomes in the offspring of pregnant women with epilepsy taking levetiracetam. Abstract: 1,2,3 Morten I. Lossius , .sup.4 Jan E. Erikssen , .sup.5 Petter Mowinckel , and .sup.6 Leif Gjerstad (Research and Education, National Centre for Epilepsy Rikshospitalet, Sandvika, Norway ; Helse Ost Health Services Centre, Akershus University Hospital, Lorenskog, Norway ; Department of Neurology, Akershus University Hospital, Lorenskog, Norway ; Department of Preventive Medicine, Akershus University Hospital, Lorenskog, Norway ; Department of Paediatrics, Division Woman and Child, Ulleval University Hospital, Oslo, Norway ; and Department of Neurology, Rikshospitalet, Oslo, Norway) Rationale: Sudden unexpected death in epilepsy (SUDEP) may be related to impaired autonomic cardiac control. The cause of such impairment is not known: is it the specific type of epilepsy, ictal or interictal epileptiform discharges or the treatment with antiepileptic drugs (AEDs)? The aim of our study was to assess heart rate variability (HRV) by spectral analysis in patients with epilepsy before and after withdrawal of AEDs. Methods: The study was a prospective randomised and double blinded withdrawal study of seizure free patients on AED monotherapy. 24 h ambulatory ECG was performed before AED withdrawal and 4 months after withdrawal. Mean of the standard deviation of all filtered RR intervals for all 5 minute segments of the analyses (SDNN-i) and different power analysis were studied. Results: HRV at the baseline did not differ significantly between the withdrawal group (n = 57) and the non-withdrawal (n = 69). The SDNN-i in the withdrawal group improved significantly in patients < 45 years (6.2ms, p = 0.05). Apart from high frequency power the other power variables also increased after withdrawal, but not to a statistically significant extent. The heart rate variability decreased with age in both groups, but less in the withdrawal group. Interictal epileptiform activity did not significantly influence the heart rate variability. Conclusions: We found a significant increase in SDNN-i in the patients below 45 years after AED withdrawal. This indicates a potential positive effect of AED-withdrawal in this age group. This finding may be important in understanding the pathophysiology of SUDEP which occurs most frequently in the age group 20- 45 years. (Supported by Norwegian Foundation for Health and Rehabilitation (EXTRA FUND), The Norwegian Epilepsy-association and The Norwegian Chapter of the International League Against Epilepsy.) Abstract: .sup.1 Edward H. Maa , and .sup.1 Mark C. Spitz (Neurology, University of Colorado Health Sciences Center, Denver, CO) Rationale: Twenty million people vacation in the American Rockies and Sierras annually. The physiologic consequences of transient environmental hypoxia have been responsible for everything from poor sleep to pulmonary edema. And on occasion, a first-time seizure or worsening of well controlled epilepsy erupts in this setting. Epidemiologic data for high altitude seizures at the moderate altitudes (>8000 ft) of our American resort towns does not exist. Methods: A retrospective chart review examining the risk factors in subjects with new or worsening seizures upon arrival to the moderate altitudes of Colorado resort towns was performed. Electronic records from Summit County Emergency Department (a cachement area including Breckenridge, Arapahoe Basin, and Keystone ski resorts) were reviewed from January 2001 to October 2005 for the ICD9 diagnosis codes for seizures and epilepsy. Results: A total of 64 individual subjects suffered one or more seizures, 28 (43%) of which occurred in subjects visiting from elevations less than 8000 ft. Sixteen of the 28 (57%) subjects from lower elevations had an average POx reading of 82.6%. Twelve of the 36 (33%) subjects from Summit County averaged POx of 87.3%. Alcohol abuse was noted in 3/28 low elevation subjects and 6/36 Summit County subjects. Headaches, sleep disturbances, and tachycardia at presentation were more frequent in low elevation subjects. CT abnormalities were seen more frequently in Summit County subjects and more commonly associated with EtOH abuse. Newly found CT lesions in low elevation subjects without a history of prior seizure was not seen. CXR abnormalities were seen in 2/28 low elevation subjects and 1/36 Summit County subjects. Electrolyte disturbances, infections, and pharmaco-nonadherence did not appear to show any differences between the two groups. Conclusions: Subjects who experienced new or worsening seizures, from elevations less than 8000 ft, were more likely to show signs of hypoxia-related physiologic changes. Compared with subjects from elevations of 8000 ft or greater, these subjects demonstrated lower average oxygen saturations, increased tachycardia, and more sleep disturbances and headaches. Transient exposure to hypobaric hypoxia is suspected of altering cerebral physiology in such a way as to increase the risk of developing a seizure. Mechanisms are yet unknown, and further research, including a prospective collection of epidemiologic data is required. Abstract: .sup.1 Debra MacGarvie , .sup.1 Nina Politzer , .sup.1 Mary Pat MacAndrews , and .sup.1 Martin del Campo (Epilepsy Clinic, University Health Network, Toronto Western Hospital, Toronto, ON, Canada) Rationale: The purpose of this study is to determine if dietary intervention can be an effective and sustainable method of seizure reduction when used in addition to standard drug therapy. The ketogenic diet (KD)has been used in children with varying degrees of success; however, the diet has been very difficult for adults to follow due its unpalatable nature. The Atkins Diet (AD) which like the (KD), achieves a state of ketosis may be an alternative. Methods: Patients with pharmacoresistant epilepsy were randomized into one of two treatment groups- either basic or intensive dietary counselling and initiated with a 20 gram carbohydrate diet. Menu plans and AD guidelines were provided. A control group was aged-matched to groups 1 and 2 and consisted of patients who were unable to follow the AD. Patients were assessed by dietitian, physician, psychologist and nurse. Cholesterol, triglicerides and fasting blood sugar were measured at baseline and at 6 months. Food intake records, ketosis, weight and Body Mass Index (BMI) were evaluated. Various aspects of seizure quality were collected, including seizure frequency, severity and recovery time. In addition, we measured cognitive status and health-related quality oif life (QOLIE-31). Results: Seven patients in the treatment groups have completed 6 months on the AD. All patients completed food records that showed 90% compliance with diet and acceptable caloric intake. All patients who had BMI over 30 achieved a healthier BMI of under 30. Cholesterol and trigliceride levels remained in acceptable ranges. All patients achieved ketosis. More than half reported a reduction in seizure intensity as well as a shorter recovery time after a seizure. There was a moderate improvement in attention and learning efficiency, both below average at baseline and at the six month follow-up, but no change in quality of life ratings. Conclusions: The AD can achieve sustainable ketosis over time. The great majority of patients showed acceptable compliance. Fifty percent of the cohort studied showed worthwhile improvement in their seizure disorder. Although results are preliminary given the small sample completing the protocol at this point, they suggest the AD may be a useful adjunct in the treatment of refractory epilepsy. Abstract: .sup.1 Fumisuke Matsuo (Neurology, University of Utah School of Medicine, Salt Lake City, UT) Rationale: Epilepsy is defined by unprovoked recurrences of epileptic seizures, but its natural history varies widely. Even within the same epilepsy syndrome, some patients will remit and can discontinue AED (antiepileptic drug) treatment, while others will prove refractory. Prompt control of seizure recurrences has been recognized as the most significant predictor of remission, but some can enjoy clinical seizure control, only when AED prophylaxis is continued. The latter group of patients is not well characterized, even though they benefit most from the epileptologist's expertise. Methods: The author's epileptology practice data were reviewed for patients with the following characteristics: 1) follow-up record spanning a period longer than 10 years, 2) the diagnosis of epilepsy has been based not only on clinical semiology, often confirmed by EEG findings, but also on sporadic clinical recurrences, when AED treatment is interrupted, and 3) interative AED trials led to a terminal prophylactic single AED regimen. Results: A total of 6530 adult patients had been screened prior to 1996 by the author, based at a University Hospital. Also generated were a total of 43101 AED treatment history entries for 916 patients, in 377of whom computer medical record was accessible for the most recent visit. The 10-year time span criterion narrowed the list down to 87. After patients with multiple developmental handicaps or unconfirmed epilepsy diagnosis, and those having undergone resection brain surgery were excluded, there remained 31 patients (31/377 = 8.2%). Their terminal AED regimens consisted of phenytoin (15 patients), carbamazepine (12), phenobarbital (3) or valproic acid (2). Conclusions: As a consultative service, the clinic has not consistently extended clinical follow-up, once a degree of seizure control was attained. The result (8.2%) is therefore assumed to underestimate the prevalence of this significant group of patients. The time span criterion, not the lack of efficacy, excluded monotherapy with a second generation AED. Also excluded was a group of patients on 2 AED with an essentially similar degree of excellent clinical control. This may explain a smaller group size of valproic acid monotherapy. This group of patients does not seem to require aggressive AED dosing nor report AED-related adverse symptoms, but represents the failure of efforts to wean off AED. They come to recognize, and fear, inevitable clinical recurrences off AED. They are typically protective of their AED regimen, and may well opt not to continue to work with the epileptologist. Sporadic seizure recurrences seem to occur often during attempts to replace AED with a second generation agent. Abstract: .sup.1 Paul H. McCabe , .sup.1 Daniel Chehebar , and .sup.1 Matthew Eccher (Neurology, Penn State, Milton S. Hershey Medical Center, Hershey, PA) Rationale: With the exception of pregabalin, all of the second generation antiepilepsy drugs (AEDs) have been available for over 6 years. Many reports on retention and seizure response are done on an individual basis for the different AEDs from different institutions/authors. This makes comparison of the different AEDs difficult, as variation in reporting styles, differences in use of the drugs, and differences in methods can vary significantly from one center to another. It is also unlikely that a large study doing a direct side-to-side comparison of all the newer AEDs will be performed. We present the overall results of use of the newer AEDs from the same institution by the same authors. Methods: All patients treated at Penn State, Milton S. Hershey Medical Center were included. Charts of patients treated from January 1st, 1993 - June 30th, 2005 were reviewed from the time of the first exposure to each drug until the final visit for each patient. Patients treated less than 6 months for any of the drugs (aside from discontinuation) were excluded. The following drugs were investigated: gabapentin (GPN), lamotrigine (LTG), topiramate (TPM), oxcarbazepine (OXC), levetiracetam (LEV), and zonisamide (ZNS). Many patients were exposed to several of the newer AEDs and therefore data from each drug is expressed as patient exposures, rather than patients. Results: During the study, there were a total of 1262 patient exposures to the second generation AEDs. Breakdown is as follows: GPN-96 patients, LTG-494 patients, TPM-232 patients, OXC-146 patients, LEV-294 patients, and ZNS-133 patients. Seizure-freedom was most likely to occur with LTG, TPM, or LEV, and least likely with GPN; ranging from 5% to 32%. Response varied in part depending on seizure type and how early in the treatment regimen patients were exposed to each drug. Despite many patients not becoming seizure-free, the majority chose to remain on the newer agents. Retention rates by the end of the study varied from 27% to 84% with 27% of patients remaining on GPB, 84.4% remaining on LTG, 84.1% remaining on TPM, 61.6% remaining on OXC, 66.3% on LEV, and 84.2% on ZNS. This suggests better tolerance with the newer AEDs, although this could have been better supported had Quality of Life studies been included in our study. Conclusions: Despite studies suggesting that the chance of becoming seizure-free after failing 2 AEDs is very low (roughly 3%), our study does demonstate that several of the newer AEDs carry a much higher likelihood of patients becoming seizure-free. Despite many patients not becoming seizure-free, retention rates at our center were also higher than other reports, suggesting better tolerance to the newer AEDs. Since all patients were under the care of the same authors at the same institution, there is less variability than comparing trials from different patients groups/centers. Despite this, the authors acknowledge that bias can still exist. Abstract: .sup.1 Romila Mushtaq (Department of Neurology, Medical College of Wisconsin, Milwaukee, WI) Rationale: Sexual dysfunction in women can be categorized into four major categories: sexual desire disorders, sexual arousal disorders, orgasmic disorders (either primary or secondary), and sexual pain disorders. Current evidence supports the idea that women with epilepsy have sexual dysfunction. We propose to assess the prevalence of sexual dysfunction in women who have epilepsy and further analyze types of sexual disorders and relation to types of epilepsy, seizure frequency, and use of antiepileptic medications. Methods: A prospective study was initiated in the outpatient epilepsy clinic of the author. All female patients were instructed to complete a questionnaire as a part of a comprehensive evaluation of review of systems pertaining to women with epilepsy. The patients were specifically questioned regarding: 1) age of onset of epilepsy, 2) type of epilepsy 3) antiepileptic medications currently being prescribed 4) frequency of seizures, and 5) whether the patient was sexually active. All patients were also questioned regarding symptoms of: 1) decreased libido, 2) pain during intercourse, 3) difficulty becoming aroused, and 4) difficulty or inability to reach orgasm. Results: A total of 109/160 patients completed the questions pertaining to sexuality, and 62 (57%) reported being sexually active at the time of the office visit. Of those women who were sexually active, 46(74%) denied symptoms of sexual dysfunction. The other 26% of women reported difficulty or inability to reach orgasm and decreased libido as the most frequent complaints of sexual dysfunction. A total of 47(43%) women were not sexually active, and 34 (72%) of these patients denied symptoms of sexual dysfunction. Of the 28% of women with sexual dysfunction in this group, decreased libido was the most common symptom. Poorly controlled seizures was the only statistically significant factor seen in patients who were not sexually active compared to the women who were sexually active. Note: data collection is ongoing at the time of the submission of this abstract. Conclusions: Women with epilepsy may have sexual dysfunction, the etiology of which has been proposed to be due to complex factors such as disturbance of the hypothalamic-pituitary axis, effects of anti-epileptic medications, effects of seizures, psychological issues, and social influences. Our study shows that whether a woman with epilepsy is sexually active or not, approximately 26-28% report symptoms of sexual dysfunction. Women with poorly controlled seizures were less likely to be sexually active. Symptoms of sexual arousal and orgasmic disorders were the most commonly reported symptoms amongst both groups of women. No statistical significance was present to correlate factors of type of epilepsy or antiepileptic medications utilized with women who reported sexual dysfunction. Abstract: .sup.1 Ruth E. Nemire , .sup.2 Marinelli Vega , .sup.2 Dalia Lorenzo , and .sup.2 R. Eugene Ramsay (Pharmacy Practice, Nova Southeastern University, Ft. Lauderdale, FL ; and Neurology, University of Miami, Miami, FL) Rationale: Pregabalin (PGB) is approved for add on therapy for the treatment of partial onset epilepsy. Three double blind placebo controlled trials were completed in over 1052 patients. This report from one epilepsy center provides long term outcome information on patients with refractory seizures treated with PGB. Methods: Patients were entered into an open label treatment protocol to obtain information on long term safety and efficacy of PGB. Patients entered the study either after completing a double blind controlled efficacy trial or as de novo patients. During the open label trial patients were requalified to continue if they had a greater than 50% reduction in seizures. The dose of pregabalin could be titrated to a total maintenance dose of 600mg per day. Patients were allowed to continue treatment with 1-3 other medications in addition to the pregabalin. Doses of other antiepileptic medications could be adjusted during treatment. Results: Thirty six total patients were entered into an open label trial. Twenty three (64%) patients dropped for reasons of lack of efficacy, withdrew consent, or adverse events. Thirteen patients (36%) continued in the open label trial. During the trial 6 (16%) patients did not meet requalification criteria and 1 patient dropped out. Six patients were followed until the close of the study in 2005. Three males and three females completed the open label trial. Ages ranged from 28 y.o. to 62 y.o.; two patients were of Hispanic descent, three Caucasian and 1 African American. Seizures began in all patients between age 1 and age 18. Patients had been treated for epilepsy between 4 and 48 years with the majority over 40 years of treatment. Four patients were on 2 antiepileptic drugs when starting the trial 2 patients were on lamotrigine only at enrollment. Patients completing the trial were followed from four to six years total treatment. Three patients were seizure free and the others a marked reduction of seizures ranging from fifty to ninety percent per month. Two patients had an increase in seizure free months. Sixty seven percent of the patients completing the trial continued to have a further reduction in seizures from 48% to 100%. Adverse events reported were few and coincide with data reported from double blind placebo controlled trials. Two patients reported no adverse events, one reported weight gain and double vision was reported by 3 patients at some point during the course of the trial but they continued on drug. Conclusions: Patients maintained on pregablin for greater than 4 years continued to have a greater than 50% reduction in seizure frequency, elimination of other antiepileptic drug doses were achieved during this time period, and patients had few or no adverse effects. Abstract: .sup.1 Mordekhay Medvedovsky , 1,2 Svetlana Kipervasser , .sup.3 Batya B. Davidovici , .sup.4 Tatiana Vander , and 1,2 Miri Y. Neufeld (EEG and Epilepsy Unit, Department of Neurology, Tel-Aviv Sourasky Medical Center, Tel-Aviv, Israel ; Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel ; Department of Dermatology, Kaplan Medical Center, Rechovot, Israel ; and Rehabilitation Center, Beit Loevenstein, Raanana, Israel) Rationale: Video-EEG monitoring (VEM) is a vital diagnostic tool in the pre-operative work-up of patients suffering from intractable focal epilepsy. During VEM, anti-epileptic drug (AED) dosages are often reduced in order to induce focal seizures in patients; an unavoidable side effect can sometimes be generalized tonic-clonic seizures (GTCS). The goal of this study was to answer two questions: 1) What are the risk factors for GTCS in patients with focal epilepsy during VEM? 2) At what stage in the reduction of AEDs will GTCS most likely occur? Methods: We retrospectively reviewed the records of all adult patients (>18y) hospitalized in the VEM Unit in the Tel-Aviv Sourasky Medical Center from 2002-2005, with focal epilepsy who also had GTCS. A comparative group was created by reviewing the records of the first patient of each month who had focal epilepsy but who did not have a GTCS during VEM hospitalization. The drug reduction in both groups was done using identical protocol. Results: Fifty-one patients with GTCS were identified and compared with 48 patients in the control group. Men were more prevalent in the group with GTCS - 34 (61%) - than in the control group - 22 (39%) (p = 0.037). More patients with GTCS during VEM also had a history of GTCS - 45 (58%)/33 (42%) (p = 0.018); a history of simple partial seizures (SPS) was less prevalent in the group with GTCS - 33 (45%)/40 (54.8%) p = 0.035). There was a clustering of GTCS on the seventh day from the start of AED reduction (8 patients). Women had GTCS earlier (5.75 days, SD 3.53) than men (8.15 days, SD 6.083) (p = 0.033). Patients with a history of febrile convulsions had GTCS later (9.35 days, SD 7.71) compared with patients with no history of febrile convulsions (6.79 days, SD 4.59) (p = 0.027). Conclusions: The risk factors for GTCS in patients with focal epilepsy during VEM were shown to be: male gender, history of GTCS and absence of history of SPS; women and patients with no history of febrile convulsions have GTCS earlier, and the risk of GTCS is highest on the seventh day from the onset of AED reduction. These results provide a helpful tool in predicting risk factors for GTCS during drug dosage reduction in VEM. Abstract: .sup.1 Young Joo No , .sup.1 Sang-Ahm Lee , and .sup.1 Joong Koo Kang (Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea) Rationale: Several studies have revealed the increased incidence of atypical language dominance in patients with left hemisphere epileptic foci. We retrospectively investigated the incidence and related factors for language dominance shift, determined by intracarotid amobarbital procedure (IAP), in patients with hemispheric focal epilepsies. Methods: We included 222 left epileptic patients whose language dominance was determined by IAP at Asan Medical Center from 1994 to 2004. The items of language test in IAP included spontaneous speech (6 items), understanding (2 items), and repetition (2 items). Language lateralization index (LI) was computed according to the formula L = (Score IAP right -Score IAP left)/(Score IAP right + Score IAP left). Clinical information was obtained from medical records including age, gender, age at onset of epilepsy, duration of epilepsy, frequency of seizures, risk factors, onset age of risk factors, and lateralization of MRI or EEG. Results: Of the 222 patients (male 110 patients, 49.5%), complete left language dominance was 142 patients (64.0%), and complete right hemispheric language dominance was 29 patients (13.1%). Seizure onset age, onset age of risk factors, handedness and MRI lesions (hippocampal atrophy or left extensive lesion) had statistically significant association with atypical language dominance. On a linear regression analysis, the significant predictors of the atypical language dominancy were handedness and left extensive lesion (R.sup.2= .64). Conclusions: Atypical language dominancy in patients with left epileptic foci was highly correlated with non-right handedness and extensive lesion on the left hemisphere. Abstract: .sup.1 Ketil B. Olsen , .sup.1 Erik Tauboll , and .sup.1 Leif Gjerstad (Department of Neurology, Division of Clinical Neuroscience, Rikshospitalet-Radiumhospitalet, University of Oslo, Oslo, Norway) Rationale: Intravenous VPA has been approved for treatment of benzodiazepine-resistant status epilepticus (SE) in Norway. We are now performing a prospective registration of the effect of i.v. VPA in SE and serial attacks (SA) from 6 Norwegian hospitals. Methods: Data have been obtained from the first 41 adult patients, 23 women and 18 men, treated with i.v. VPA after having been unsuccessfully treated with benzodiazepines for SE/SA. The protocol suggested a VPA loading dose of 25 mg/kg over 30 min, followed by continuous infusion of 100 mg/hour for 24 hours. Results: 29 patients were diagnosed with SE, 12 with SA. 19 had generalised tonic-clonic while 16 had complex-partial seizures as their main SE/SA seizure type. 6 had mixed or other type of seizures. Median VPA bolus dose was 1800 mg (range 700-2500). Median time to treatment of SE/SA for all patients was 3 h (range 0.3-72 h). For patients with complex partial SE/SA, median time to start of VPA was 4 h (range 0-72 h). Effect, defined as the lack of need of anaesthesia (barbiturates or propofol), showed that VPA was effective in 31 of 41 cases (76%). Of the 10 cases requiring anaesthesia, 5 had a significant delay before starting treatment (20 to 72 hours). Moderate hypotension during bolus dose infusion was seen in one patient, but corrected easily with i.v. saline without complications. Otherwise, no side-effects were reported. Conclusions: VPA is safe and effective in the treatment of benzodiazepine-resistant SE/SA. SE/SA was stopped in 31 of 41 cases (76%). Lack of effect may be related to delayed treatment and too low loading dose. Abstract: .sup.1 Alison M. Pack , .sup.2 Anne R. Davis , .sup.2 Jordana Kritzer , .sup.1 Ava Yoon , .sup.2 Adele Camus , and .sup.1 Silvia Done (Neurology, Columbia University, New York, NY ; and Obstetrics and Gynecology, Columbia University, New York, NY) Rationale: Pre-pregnancy planning is important for women with epilepsy to optimize fetal outcome. In addition, women with epilepsy experience higher pregnancy-related risks than their healthy peers. Published guidelines recommend highly effective contraception; however, no published data describes contraceptive practices in this population. Methods: We conducted a cross-sectional questionnaire study among a convenience sample of reproductive-age women (18 to 44) with epilepsy attending outpatient clinics at an urban, academic medical center. Questionnaires were administered at 3 clinics. Two included patients in a private office with predominantly commercial insurance and 1 was a multi-provider clinic of mostly Medicaid insurance. Our questionnaire assessed demographics, reproductive history, antiepileptic drug use, sexual behavior, and contraceptive use. The questionnaire was offered in both English and Spanish and conducted in a private area. Results: We approached 425 women to obtain 148 (35%) completed questionnaires; many were excluded due to age (n = 198). Participants had a mean age of 32 years (SD [+ or -] 8), 32% spoke Spanish and 32% described themselves as Hispanic. Insurance included commercial (55%), Medicaid/Medicare (41%), and self-pay (6%). Participants reported a wide range of educational attainment and income. Fifty-five percent (n = 78) reported having vaginal intercourse in the last 30 days. Of these sexually active women, 59 (76%) used contraception, 11 (14%) were pregnant, and 2 (3%) sought pregnancy. Contraceptive methods included male condoms (n = 21, 36%), withdrawal (n = 18, 31%), oral contraception (n = 16, 27%), male or female sterilization (n = 10, 17%), patch (n = 3, 5%), rhythm (n = 3, 5%), intrauterine device (n = 1, 2%), depo-provera (n = 1, 2%), and other (n = 11, 19%). Some women used multiple methods. Of those using hormonal methods (oral contraception, patch, depo-provera) (n = 21), 6 (29%) concomitantly took an enzyme inducing antiepileptic drug with known increasing susceptibility to pregnancy. Women reported 181 pregnancies and 91 (50%) were unplanned. Conclusions: In this sample, many women with epilepsy at risk for getting pregnant rely on low-efficacy contraceptive methods. Similar to the general population, 50% of their pregnancies were unplanned. These findings illustrate the need for continued education of women with epilepsy regarding effective contraception. Abstract: .sup.1 Erin Phillip , .sup.1 Shahram Shahrokshi , and .sup.1 Neelan Pillay (Clinical Neurosciences, University of Calgary, Calgary, AB, Canada) Rationale: Psychosis in epilepsy is more commonly postictal and less often ictal or related to treatment. However de novo psychosis upon administration of antiepileptic medications is not uncommon. De novo psychosis on rapid withdrawal of antiepileptic medications in patients with epilepsy in the seizure monitoring unit (SMU) has not been reported. The purpose of this article is to investigate a possible link between withdrawal of carbamazepine and the onset of psychotic symptoms. Methods: We describe the clinical course of two patients in whom rapid or abrupt withdrawal of carbamazepine was followed by new onset of psychotic symptoms. We analysed patient specific factors such as history, age, gender, date/time since last seizure and length of use of carbamazepine as well as the rate of withdrawal and related these to the onset of symptoms. The symptoms displayed included aggression, disorientation, delusions/hallucinations and violent behaviour. Neither patient had a previous psychiatric history. One patient has frontal lobe epilepsy with nocturnal complex partial seizures. This patient was admitted to the SMU unit for further localization of seizure foci using invasive subdural and depth electrode monitoring. During the admission carbamazepine was withdrawn rapidly over 4 days from 1200 mg/day to zero. The second patient has temporal lobe epilepsy with complex partial seizures and has had previous left temporal lobectomy. This patient abruptly stopped taking carbamazepine on her own and was admitted to the SMU immediately following onset of psychosis. There were also 2 previous admissions for psychosis on abrupt withdrawal of carbamazepine. Results: Carbamazepine was withdrawn rapidly in 1 patient and abruptly in the 2.sup.nd patient. Both patients became aggressive, disoriented and delusional and did not return to baseline behaviour for several days. At onset of psychosis neither of the patients was on any other anti epileptic medications. No clinical or subclinical seizures were recorded before, during or after the psychosis on video EEG telemetry. MRI scans were normal. Conclusions: We believe that these cases indicate a possible link between withdrawal of carbamazepine and new onset of psychotic symptoms. Carbamazepine withdrawal followed by de novo psychosis in epilepsy patients has not been reported. These cases indicate a need for extra caution when withdrawing carbamazepine rapidly in the seizure monitoring unit or for other reasons. Abstract: .sup.1 Neelan Pillay , .sup.1 Jose Zenteno-Tellez , .sup.1 Sam Wiebe , and .sup.1 Paolo Federico (Clinical Neurosciences, University of Calgary, Calgary, AB, Canada) Rationale: General neurologists, and epileptologists, need to interpret and apply driving guidelines for patients with seizures and epilepsy. Physicians are expected to know and follow provincial regulations, which have variable reporting requirements. Our purpose was to determine clinicians' familiarity with driver guideline restrictions, and the differences in application of restrictions between epileptologists and general neurologists. Methods: A modified motor vehicle driving questionnaire based on the Canadian Medical Association (CMA) Guidelines was mailed out to 33 neurologists in Calgary, Canada. The conditions listed were syncope with no obvious neurological or cardiac cause; aura or simple partial seizure unchanged for more than one year; unprovoked first seizure; epilepsy on treatment and seizure free; seizure during sleep or awakening with normal EEG; simple partial seizures (aura); seizure recurrence on medication withdrawal or change under medical direction; alcohol withdrawal seizures. The questionnaire asked to recommend duration of driving restrictions for various license classes (class 5 = private car or small truck and 1-4 = professional). Auras and simple partial seizures were categorized separately but defined as non disabling sensory or motor seizures. Results are expressed as median and range. Groups are compared with the rank-sum test. Results: 76% (25/33) of the questionnaires were returned. The responses were analyzed for the entire group and for epileptologist separately. For syncope, the entire group's responses were concordant with the CMA guidelines, but epileptologists recommended driving earlier (within 1 month) if the examination and investigations were normal. Neurologists and epileptologists agreed with CMA guidelines in patients with a solitary seizure and in those with only auras. However, both groups would restrict driving for 12 months, as opposed to the recommended 6 month. There is consensus for AED withdrawal seizures, alcohol related seizures. Interestingly, both groups would not restrict driving for an aura but would restrict driving for 12 months for "simple partial seizures". More disagreement between neurologists and epileptologists occurred in restriction for commercial licenses, but these were not statistically significant. Conclusions: In an unprovoked first seizure general neurologists and epileptologists agree with guidelines' restrictions, but in cases of syncope, epileptologists are less restrictive. Neurologists and epileptologists interpreted an aura to be different from a simple partial seizure. In general, neurologists and epileptologists are more restrictive for epilepsy than the recommended CMA guidelines. Differences between neurologists and epileptologists were not significant. Abstract: .sup.1 Annkathrin Poepel , .sup.1 Christian Hoppe , .sup.1 Deirdre Cooper-Mahkorn , and .sup.1 Christian E. Elger (Epileptology, University of Bonn, Bonn, Germany; Epileptology, University of Bonn, Bonn, Germany; Epileptology, University of Bonn, Bonn, Germany; and Epileptology, University of Bonn, Bonn, Germany) Rationale: To compare patients' seizure documentation with seizure diaries and seizure count documented by in-patient video-EEG-monitoring. In addition, the benefit of daily reminding patients to carefully document all seizures was assessed in a randomized controlled trial (RCT). Methods: Prospective investigation of all in-patients with a diagnosis of epilepsy who received video-EEG-monitoring for at least 24 hours (Oct 2004 - Jul 2005). In a RCT patients received a conventional seizure diary without (group A) or including daily personal reminding (B). Based on video documented semiology and EEG data all monitored seizures were classified as simple-partial seizures (SPS, including auras), complex-partial seizures (CPS) and generalized tonic clonic seizures (GTCS). Pre-ictal EEG-periods were classified regarding the patient's state of vigilance. Results: We recorded a total of 613 classifiable seizures (190 SPS, 371 CPS, 52 GTCS) in 95 patients (mean age 39.3 years, 43 female). Seizure documentation quality of patients was different depending on seizure type and pre-ictal state of vigilance. Particularly, CPS (72%) and seizures beginning in sleep (86%) were at risk not to be documented by the patient. 59 patients (62%) had an incomplete seizure documentation including 41 patients (43%) who documented no seizure. Incompletely documenting patients had significantly less SPS but significantly more CPS during video-EEG-monitoring as compared to completely documenting patients (Mann-Whitney, p < .01). Furthermore they had more seizures beginning in sleep, higher risk for ammonshorn sclerosis, lower educational level and lower verbal learning performance. Regarding RCT group B (with reminding) yielded a higher rate of documented seizures as compared to group A but this difference was not significant. Conclusions: In this sample of in-patients with chronic epilepsies, a high rate of patients with incomplete documentation (62%) as well as a high rate of undocumented seizures (55%) could be revealed during video-EEG-monitoring. Regarding seizure types, CPS were the most common subtype in our study (61%). These, however, were documented most unreliably (72% missing in patients' documentation). Seizures occurring from sleep were less reliably documented than seizures occurring from a waking state (86% versus 31% missing in patients documentation). Seizure frequency, particularly frequency of CPS and frequency of seizures in sleep, ammonshorn sclerosis, low educational level and low verbal learning performance were negative prognostic factors for correct seizure documentation. Although conditions in this investigation are not entirely comparable with out-patient condition at home, evaluations of the status and course of epilepsy in individual patients and clinical studies should consider that seizure diaries provide highly subjective data. Abstract: .sup.1 Nina Politzer , and .sup.1 Peter L. Carlen (Epilepsy Program, University Health Network,Toronto Western Hospital, Toronto, ON, Canada) Rationale: In our tertiary care epilepsy clinic we have many pharmacologically intractable patients that are not surgical candidates, who have failed most antiepilepsy drugs. Magnesium is known to block seizures in eclampsia. It is anticonvulsant in several in vitro models, and its mechanisms of action include decreasing neuronal excitability by decreasing membrane surface charge and blocking NMDA receptors. Therefore we decided to administer magnesium supplements to these intractable patients. Methods: This observational study was designed as a preliminary unblinded test of the hypothesis that magnesium supplementation would decrease the frequency of intractable seizures. Patients were selected on the basis of having pharmacologically intractable seizures. Patients were followed as per their routinely booked clinic visits. They were asked to purchase over the counter magnesium supplements, with preference given to magnesium oxide, 420 mg. Results: To date we have followed 20 patients taking magnesium supplements ranging from 250 mg od to 420 mg qid per day. No side effects have been noted to date from the magnesium administration. Five of the 20 patients reported greater than 50% reduction in seizure frequency, and 2 reported virtually no seizures without stressors or forgetting their medications. No patients reported worsening of seizures and some reported a calming effect, but no obvious seizure reduction. Conclusions: This safe and cheap additional pharmacotherapy deserves a serious double blinded trial in intractable epilepsy. There are many questions to be considered including the mechanism(s) of action, alterations in serum (and brain) magnesium, the best formulation (as a salt, as a chelate), the proper dosing, possible side effects, and the patient characteristics of responders and nonresponders. Abstract: .sup.1 Gulseren Aktas , .sup.2 Kader Karli Oguz , .sup.3 Nese Dericioglu , and .sup.1 Serap Saygi (Neurology, Hacettepe University, Ankara, Ankara, Turkey ; Radiology, Hacettepe University, Ankara, Ankara, Turkey ; and Institute of Neurological Sciences and Psychiatry, Hacettepe University, Ankara, Ankara, Turkey) Rationale: Recently, abnormalities of the shape and positioning of the hippocampal formation have been recognized increasingly on MRI. The significance of this finding is not well known since it can be detected in patients investigated for various reasons. Few data exists in the literature regarding the presence of this abnormality in patients with seizures. We aimed to investigate the clinical and EEG findings in patients with seizures whose cranial MRI revealed the presence of hippocampal shape abnormalities. Methods: We retrospectively analyzed the clinical and laboratory findings of patients who were investigated at our epilepsy clinic for seizures between 1995-2006 and had cranial MRI that clearly demonstrated abnormalities of the shape of the hippocampus. Patients with any accompanying lesions (malformation of cortical development, tumor, encephalomalacia, etc) were excluded. Results: Twenty one patients [11M, 10F; age: 17-69 years (mean 33)] were included in the study. Seizure onset varied between 1-58 years (mean 21.5) and occurred in adulthood in 15 (71%) patients. Seizure characteristics were as follows: simple partial (n = 9), complex partial (n = 17) and secondary generalized tonic-clonic (n = 10). Twelve (57.1%) patients were seizure free on monotherapy with anticonvulsants. Eight (38.1%) patients reported risk factors for seizures (febrile convulsion, head trauma, perinatal injury, family history of epilepsy). Cranial MRI was initially reported as normal in 11 (52.4%) patients. Abnormal shape of the hippocampus was evident on the left side in 16 (76.1%), right side in 3 (14.3%) and bilaterally in 2 patients (9.5%). Eight (38.1%) patients had normal or non-specific EEG recordings. Pathological EEG findings were concordant with MRI results in 9 (42.8%) patients. Conclusions: Although uncommon when compared to hippocampal sclerosis, hippocampal shape abnormalities may be associated with epilepsy. Seizure onset is usually late in this group and more than half of the patients are seizure free on single anticonvulsant. EEG may fail to demonstrate pathological findings in a considerable number of patients. The occurrence of this abnormality mostly on the left side may somehow be related to development of the dominant hemisphere and deserves further attention. Radiologists should be aware of hippocampal shape abnormalities since half of the cases may be overlooked on MRI. Abstract: .sup.1 Johnny Salameh , and .sup.1 Sanjay P. Singh (The Nebraska Epilepsy Center, Deptartment of Neurological Sciences, University of Nebraska Medical Center, Omaha, NE) Rationale: Mesial Temporal Sclerosis (MTS) is the most common cause of refractory temporal lobe epilepsy. It is also the one most often treated surgically, because medical treatment fails in upto75% of cases, and surgical treatment is successful in about 58%.sup.1. This study was undertaken to determine the efficacy of Levetiracetam in the treatment of Temporal Lobe Epilepsy in patients with Mesial Temporal Sclerosis. Methods: This is a retrospective study looking at our database of over 1000 epilepsy patients. The patients were classified as having Temporal Lobe Epilepsy and Mesial Temporal Sclerosis based on the following criteria: A. Video-EEG confirmation of Temporal Lobe Epilepsy by ictal and interictal recordings and no other contradictory data and MRI showing mesial temporal sclerosis. Or B. Electroencephalogram showing epileptiform discharges consistent with Temporal Lobe Epilepsy, clinical features consistent with Temporal Lobe Epilepsy and MRI showing mesial temporal sclerosis. No contradictory data. Thirty patients fulfilled the above criteria and were started on Levetiracetam. The patients were followed for an average of 20 months from the time of starting Levetiracetam. No other antiepileptic drug changes were made during this period in these patients. One patient could not afford the medication after being started on Levetiracetam and three patients had no follow up after Levetiracetam was started. Therefore, the total number of patients were 26. The outcome was classified as: Class 1: Seizure free Class 2: > 50% reduction in seizure frequency Class 3: < 50% reduction in seizure frequency Patients in Class 1 and 2 were designated as responders. Results: Twenty-one of the 26 patients (80.7%) with Temporal Lobe Epilepsy and MTS had a greater than 50% reduction in seizures. Ten of the 26 (38.4%) patients became seizure free. The average number of antiepileptic drugs these patients had failed is 3.13 before starting Levetiracetam. Outcome: Class 1: 10 patients (38.5%) Class 2: 11 patients (42.3%) Class 3: 5 patients (19.2%) 38.5% of patients with Temporal Lobe Epilepsy and MTS, who had failed an average of more than 3 antiepileptic medications, became seizure free on treatment with Levetiracetam. The responder rate was 80.7%. The responder rate for Levetiracetam in all partial seizure is 39.8% to 44%.sup.2. Conclusions: This study clearly demonstrates a unique efficacy profile of Levetiracetam in Temporal Lobe Epilepsy with MTS. Large randomized controlled studies are needed to study this important efficacy profile. References Spencer, SS When should temporal-lobe epilepsy be treated surgically? Lancet Neurol 2002 Volume: 1 Issue: 6 375-382 Cereghino, JJ Biton, V Abou-Khalil, B Dreifuss, F Gauer, LJ and Leppik, I Levetiracetam for partial seizures: results of a double-blind randomized clinical trial. Neurology 2000 Volume: 55 Issue: 2 236-242 Abstract: .sup.1 Hermann Stephan , .sup.2 Andreas Schreiner , .sup.3 Anja Mueller , and .sup.3 Barbara Schauble (Neurology Clinic, Epilepsy Center Erlangen, Erlangen, Germany ; Medical and Scientific Affairs, Janssen-Cilag EMEA, Neuss, Germany ; and Medical and Scientific Affairs, Janssen-Cilag, Neuss, Germany) Rationale: To assess seizure reduction and adverse events in different age groups of elderly patients ([greater than or equal to] 60) with epilepsy treated with topiramate (TOPAMAX, TPM). Methods: In this prospective, open label, multicenter phase IV flexible dose study, elderly patients [ < 65, 65 - 74 and [greater than or equal to]75 years (yrs)] with epilepsy irrespective of seizure type were included and followed for 12 months. Seizure frequency and adverse events were documented at each visit. Results: 107 patients (53% male, mean age 69 [+ or -] 7 years) were enrolled, 32 patients 74 yrs. 102 patients had at least one seizure during the retrospective baseline encompassing a 3 months period. Mean duration of epilepsy was 9.5 yrs. Most frequent seizure types at baseline were generalized tonic-clonic (58%) and complex partial (25%). Mean seizure frequency was 3.5 [+ or -] 14.6 and decreased to 1.6 [+ or -] 7.7 (p < 0.0001) versus baseline. At endpoint, the mean TPM dose was 98 mg/day in monotherapy and 153 mg/day in add-on. Mean dose was lowest (86 mg/day) and the percentage of monotherapy was highest in the oldest group. Overall, 78% of patients were responders (seizure reduction [greater than or equal to]50%) and 44% remained seizure-free throughout the study. The highest percentage of responders was in the youngest group (96%). 46 patients (43%) had at least one treatment-emergent adverse event (TEAE). The number of TEAEs was highest amongst the oldest group. TEAEs [greater than or equal to] 5% overall were somnolence (9.4%), dizziness (7.5%), paraesthesia (5.6%), and cognitive complaints (10.3%). Main reasons for study discontinuation (40% overall) were a TEAE (15.9%) or loss to follow-up (12.2%). Conclusions: In elderly patients with epilepsy, TPM was well tolerated among all age groups andwas associated with a significant decrease in seizure frequency. Doses used were slightly lower than the recommended target doses for adults. Abstract: 1,2 Alex M. Taylor , .sup.1 Deborah A. Cahn-Weiner , .sup.1 David Filippi , and .sup.1 Paul A. Garcia (Neurology, UC San Francisco, San Francisco, CA ; and CSPP, San Francisco, CA) Rationale: Measures of personality functioning are frequently included in the evaluation of patients suspected of having nonepileptic seizure (NES) disorder. Previous research has indicated that the Personality Assessment Inventory (PAI) provides moderate discrimination between patients with NES and those with epileptic seizures (ES).sup.1. In addition to personality assessment, evaluation of mood functioning using the Profile of Mood States - Depression scale (POMS-D) may serve to classify patients with seizure disorders who endorse depressive symptoms.sup.2. The purpose of this study was to compare the discriminative properties of these measures in classifying patients with NES and ES. Methods: Patients were recruited upon admission to the video-EEG (v-EEG) monitoring unit, and completed the POMS and PAI prior to receiving their diagnosis. Twenty-eight patients (20 female, 8 male) with definitive diagnoses were included in the analysis. NES and ES patients were matched on age and education. v-EEG monitoring resulted in 20 diagnoses of ES and 8 diagnoses of NES. Results: There was a borderline significant difference between groups on the POMS-D (p = .08), with the NES group endorsing more depression symptoms. On the PAI, the NES and ES groups differed significantly on the Somatic Complaints-Conversion (SOM-C) subscale, Antisocial Features-Antisocial Behaviors subscale, and Aggression-Verbal Aggression subscale (all p's < .05). The only clinically elevated scale was in the NES group on SOM-C (Mean T-score 72.7). In a discriminant function analysis (DFA) correcting for group size probabilities with v-EEG confirmed diagnosis as the dependent variable, the three PAI subscales provided an overall classification accuracy of 71.4%. In a separate DFA, the POMS-D provided an overall classification accuracy of 75%. When entered jointly, the PAI subscales and POMS provided an overall classification accuracy of 82.1% Conclusions: Theories addressing the underlying etiology of NES have included both personality and mood factors. Psychological measures such as the PAI and POMS are routinely used in the work-up of both NES and ES patients when psychiatric features are suspected. These results suggest that used jointly, these measures can provide fairly accurate discrimination between these patient groups and may serve as useful screening instruments. Wagner et al. Use of the Personality Assessment Inventory as an efficacious and cost-effective diagnostic tool for nonepileptic seizures. Epilepsy Behav 2005;7:301-304. Griffith et al. Measuring depressive symptoms among treatment-resistant seizure disorder patients: POMS Depression scale as an alternative to the BDI-II. Epilepsy Behav 2005;7:266-272. Abstract: .sup.1 Jose F. Tellez-Zenteno , and .sup.1 Samuel Wiebe (Department of Clinical Neurosciences, University of Calgary, Calgary, AB, Canada) Rationale: Quality of life is lower in people with epilepsy than in the general population. Its association with seizures, adverse drug effects and mood disorders has been documented. Despite a high level of somatic comorbidity in epilepsy, its contribution to Health Related Quality of Life (HRQOL) has not been studied. The objective of this study is to asses the impact of comorbid chronic conditions (eg., heart diseases, diabetes, arthritis and migraine) on HRQOL in people with epilepsy in the general population. Methods: Data were collected as part of the 19996-1997 Canadian Community Health Survey. HRQOL was assessed using the Health Utilities Index Mark 3 (HUI-3). The HUI3 maps any one of the vectors of eight health attribute levels into a summary health value ranging between -0.36 and 1.0. The most preferred health level (perfect health) is rated as 1.0 and death is rated 0.0, while negative scores reflect health states considered worse than death. Respondents (n = 129921) were classified based on the presence or absence of epilepsy, diabetes, migraine, arthritis, and possible combinations. Results: The overall HUI-3 score for the general population (n = 129921) was 0.86, and it was significantly higher than that of the studied chronic conditions (p < 0.05). The mean HUI-3 score was 0.70 (CI.sub.95 0.67-0.73) in persons with epilepsy (n = 803), 0.74 (CI.sub.95 0.73-0.75) in diabetes (n = 6180), 0.79 (CI.sub.95 0.78-0.80) in migraine (n = 11915), 0.72 (CI.sub.95 0.71-0.73) in arthritis (n = 23908) and 0.71 (CI.sub.95 0.70-0.72) in heart disease (n = 7725). The HUI-3 score was 0.45 (CI.sub.95 0.31-0.59) in persons with epilepsy and diabetes (n = 46), 0.59 (CI.sub.95 0.49-0.69) in epilepsy and heart disease (n = 87), 0.60 (CI.sub.95 0.52-0.68) in epilepsy and migraine (n = 146), and 0.56 (CI.sub.95 0.49-0.63) in epilepsy and arthritis (n = 199). All were significantly lower than in the overall epilepsy population. Conclusions: In addition to the effect of epilepsy on quality of life, the presence of common, chronic somatic comorbid conditions has a major impact on the quality of life of persons with epilepsy. This is important when reporting HRQOL of patients with epilepsy, and when assessing the impact of epilepsy therapies on HRQOL. Abstract: .sup.1 Annamaria M. Terracciano , .sup.2 Gennaro Barbato , .sup.2 Pasquale Alfieri , .sup.2 Rossana Arlomede , .sup.1 Loredana Nargi , and .sup.2 Giacomo Visco (Neurologic Rehabilitation, Clinica "S.Maria del Pozzo", Somma Vesuviana, Napoli, Italy ; and Neurology, ASL NA 1, Naples, Naples, Italy) Rationale: To evaluate efficacy and tolerability of LEV in elderly patients with newly diagnosed seizures and neuroradiologic findings of Chronic Cerebrovascular Disease (previously asymptomatic). Methods: elderly patients with onset Epilepsy at onset (Partial seizures). LEV as Monotherapy. Lev commenced at 250 mg/die,titrated weekly by 250 mg to the final dose. 12 months follow-up: visit every two montlhs, seizure diary check,EEG, complete blood sampling,Mini Mental State Examination(MMSE), Short Form Health Survey (SF12)and caregiver Neuropsychiatric Inventory (NPI) questionnaires. Outcome assessments. Change in seizure frequency. Quality of life (by MMSE,SF12 e NPI). Tolerability by evaluating LEV-related AEs. Results: 33 patients were included and all of them completed the 12 months study and achieved seizure freedom. 27(81.8%) with LEV dose 1000-1500mg daily. 6 (18.1%) needed add-on therapy with other AED: Topiramate, Lamotrigine, Oxcarbazepine. LEV showed no interactions with any concomitant drug used for other patologies(Diabetes Mellitus, Hypertension,Mood Depression,etc) MMSE, SF12 and NPI score did not worsen. 31 patients showed AEs during the titration phase: somnolence(n = 5), headache(n = 23), dizziness(n = 2),mental confusion(n = 1). Conclusions: LEV showed god efficacy and tolerability in treatment of Epilepsy in elderly patients with Chronic Cerebrovascular Disease. Abstract: .sup.1 Farrah M. Thomas , .sup.2 Noah Webster , .sup.1 Peggy Crawford , and .sup.1 Nancy Foldvary (Neurology, Cleveland Clinic, Cleveland, OH ; and Sociology, Case Western Reserve University, Cleveland, OH) Rationale: Patients with epilepsy experience varying types of sleep difficulties including increased latency to sleep onset, increased number and duration of awakenings, decreased sleep efficiency, and sleep apnea that contribute to excessive daytime sleepiness (EDS), known to negatively affect quality of life. EDS is commonly attributed to the effects of seizures and/or anti-epileptic drugs (AEDs). It is possible that individuals attempt to compensate for EDS by using caffeine, a mild stimulant and the most widely used psychoactive drug in the world. The purpose of this study is to investigate factors associated with caffeine consumption in individuals with epilepsy. Methods: Participants are two hundred and twenty-five adults (44% male, 88% Caucasian, 44% married) aged 18-75 with medically intractable epilepsy referred for surgery evaluation at the Cleveland Clinic. They participated in a health psychology evaluation that included chart review for demographic, disease, and health-related information; clinical interview concerning current and past psychological status; and self-report measures of mood, fatigue, and daytime sleepiness. Results: Daily caffeine consumption ranges from 0 to 320 oz, with 20% consuming no caffeine, 38% 24 oz or more, and 3% 100 oz or more. Fifteen percent of participants are on one AED, 64% two AEDs, 18% three, and 2% four or more. Sleep ranges from 2 to 11hrs, with an average of 7hrs/night. On a subset of 49 patients, mean scores on the Fatigue Severity Scale (FSS) and Epworth Sleepiness Scale (ESS) are 4 and 8, respectively. Quantity of caffeine consumption is not associated with gender, the number of AEDs, hours of sleep/night, fatigue level, or daytime sleepiness. Greater caffeine consumption is associated with regular use of nicotine and alcohol but not with illicit drug use (current or past) or exercise. Lastly, individuals who are currently employed full/part-time (42%) are more likely to consume larger quantities of caffeine. Conclusions: These results suggest that excessive consumption of caffeine is common among individuals with medically intractable epilepsy, particularly those who use other substances or are employed. Surprisingly, caffeine use is not associated with the number of AEDs, fatigue, daytime sleepiness, or the number of hours of sleep/night suggesting large quantities of caffeine are consumed for reasons other than fatigue or daytime sleepiness. For example, there is a strong behavioral association among caffeine, tobacco, and alcohol use. Individuals may also use large quantities of caffeine in an attempt to counteract perceived mental fatigue. In future studies, it may be helpful to ask patients why they consume caffeine. Subsequent analyses will examine the association between specific AEDs and caffeine use. We will continue to examine the association between fatigue, daytime sleepiness, sleep characteristics (i.e., sleep onset and disruptions), and caffeine as additional data becomes available. Abstract: .sup.1 Paolo Tinuper , .sup.1 Francesca Bisulli , .sup.1 Luca Vignatelli , .sup.1 Francesca Pittau , .sup.1 Sara Musho-Ilbeh , .sup.1 Federica Provini , .sup.1 Ilaria Naldi , and .sup.1 Pasquale Montagna (Department of Neurological Sciences, University of Bologna, Bologna, Italy) Rationale: Video EEG-polysomnography is the reference standard in the differential diagnosis between nocturnal frontal lobe seizures (NFLS) and non-epileptic paroxysmal sleep disorders. Nevertheless it is expensive and useful only in case of frequent attacks. On the other hand, the diagnostic accuracy of clinical history, that could be useful in patients with rare events, is unknown. The aim of this study is to measure the diagnostic accuracy of clinical history in distinguishing NFLS from non-epileptic paroxysmal sleep disorders. Methods: Three categories of subjects were eligible for this study: 1) patients with NFLS; 2) patients with parasomnias; 3) patients with other sleep disorders confounding for NFLS. Clinical, neuroradiological and interictal and ictal video-polysomnographic recordings from the database of the Sleep Center of the Department of Neurological Sciences, University of Bologna, were retrospectively reviewed. Eligible subjects were classified as having "NFLS" or "NOT-NFLS" by sleep medicine and epilepsy experts. Moreover, 5 major and 13 minor anamnestic criteria were selected to be tested in the diagnosis of NFLS. On the basis of these criteria a questionnaire was devised and administered by telephone to 103 patients by a young newly qualified doctor blinded to the aim of the study and the final diagnosis of patients. The diagnostic parameters (i.e. sensitivity, specificity, likelihood ratios, diagnostic odds ratio - DOR) were calculated for each criterion. Results: Forty-four subjects had documented NFLS; 59 subjects were classified as "NOT-NFLS". Generally, sensitivity appeared to be moderate (range 0.2-0.6) and specificity was high (0.9-1) for most of the possible criteria associations. The higher diagnostic accuracy, measured by the DOR, was obtained combining the first criterion (defining the semeiology of the epileptic event) with one of the following features (in decreasing order): 1. presence of vocalization (DOR 76.5), 2. less than 2 minutes' duration (42.8), 3. aura preceding motor behaviour (38.9), 4. positive personal history of tonic-clonic seizures during sleep (25.5), 5. stereotypy of motor phenomena (20.9); 6. recurrence of seizures every night for at least one month (20.4). Conclusions: From this preliminary study we selected one major and six minor clinical criteria for the anamnestic diagnosis of NFLS. The diagnostic accuracy of these criteria appeared to have moderate sensitivity and high specificity. A further prospective validation is needed. Abstract: .sup.1 Michelle Tisch , .sup.1 Linda Allen , and .sup.1 Romila Mushtaq (Department of Neurology, Medical College of Wisconsin, Milwaukee, WI) Rationale: Menarche on average occurs 2.5 years after the onset of puberty in healthy female adolescents. The normal age range for menarche is reported from 11 to 16.5 years with an average of 13.0 to 13.5 years at onset. Catamenial epilepsy has been described as increase in seizure frequency before or during the first few days of the menstruation and/or seizures at the time of ovulation. We proposed to analyze the age of menarche in female patients diagnosed with epilepsy to assess if epilepsy may influence the age of menarche. In addition we further assessed age of menarche in those patients with catamenial epilepsy and compared the age of menarche to those patients who did not have catamenial epilepsy. Methods: A prospective study was initiated in the outpatient epilepsy clinic of senior author. All female patients were instructed to complete a questionnaire as a part of a comprehensive evaluation of review of systems for issues pertaining to women with epilepsy. The patients were specifically questioned regarding: 1)age of onset of epilepsy, 2) age of menarche, 3) if seizures occurred in the 7 days prior to the menstruation, 4)if seizures occurred during the first 3-4 days of menstruation, and 5) if seizures occurred mid-cycle. The findings were confirmed by the physician during the interview. Results: A total of 83 patients were identified with an average age of menarche of 12.68 years. Of the 83 patients, 29 (35%) had catamenial epilepsy with average age of menarche of 12.53 years. 54 patients (65%) did not have catamenial epilepsy with average of menarche of 12.78 years. There was no statistical significance in the difference of age of menarche in the two groups. Age of onset of epilepsy in patients with catamenial epilepsy was 17.37 years compared to 27.8 years in patients without catamenial epilepsy. Of the 29 patients with catamenial epilepsy, only 9 (31%) experienced onset of epilepsy prior to or during the year of menarche. Note: Data collection is ongoing at the time of the submission of this abstract and the total number of patients included in the data analysis is expected to be approximately 150. Conclusions: In our study, females with epilepsy experience an earlier age of menarche compared to reported average age of menarche in healthy female adolescents. This may be due to multiple factors influencing the neuroendocrine axis such as seizures and anti-epileptic medications. There is, however, no difference in age of menarche in women with catamenial epilepsy compared to those women who do not have catamenial epilepsy. Patients with catamenial epilepsy had younger age of onset of epilepsy as compared to those women who did not have catamenial epilepsy. There was no relationship of epilepsy onset prior to and during the year of menarche as a factor associated with the development of catamential epilepsy. Abstract: .sup.1 Heber L. Varela , .sup.1 Denise S. Taylor , and .sup.1 Selim R. Benbadis (Neurology & Neurosurgery, University of South Florida & James A Haley Veterans Hospital, Tampa, FL) Rationale: The gold standard for diagnosis of psychogenic non-epileptic seizures (PNES) is EEG-video monitoring. EEG-video monitoring is usually prolonged and inpatient, but the availability of this procedure for veterans is limited. This study thought to evaluate the yield of short-term outpatient EEG-video monitoring for the diagnosis of PNES in a V.A. population. Methods: We reviewed the data on all short-term outpatient EEG-video monitoring performed at our V.A. hospital over a 2-year period. Short-term EEG-video monitoring was performed with induction according to a published protocol [Benbadis et al, 2000]. Briefly, induction is performed without a placebo, using hyperventilation, photic stimulation, and verbal suggestion. This was performed on patients in whom there was a clinical suspicion of PNES on clinical grounds. Results: A total of 52 short-term EEG-video monitoring sessions were performed. Of those, 40 patients (77%) were men. In 35 patients (67%) the procedure recorded the habitual episode and resulted in a clear diagnosis of PNES. The procedure was inconclusive in 17 patients (33%), either because a non-habitual event was induced (7 patients, 14%), or no event was recorded (10 patients, 19%). Conclusions: The yield of EEG-video monitoring with induction in a (predominantly male) V.A. population is high, and comparable to a non-V.A. population [Benbadis et al, 2000]. Abstract: .sup.1 Nandakumar B. Vittal , .sup.2 Eric R. Siegel , and .sup.1 Naim I. Haddad (Neurology, University of Arkansas for Medical Sciences, Little Rock, AR ; and Biostatistics, University of Arkansas for Medical Sciencs, Little Rock, AR) Rationale: Sensory simple partial seizures (SSPS) were extensively studied as auras preceding complex partial (CPS) and secondarily generalized seizures (GS). However, they have not been well characterized in epilepsy patients presenting with these seizures as the predominant or only type. The objective of this study is to describe our experience with this particular group of patients. Methods: The adult epilepsy database was searched for patients diagnosed with SSPS, excluding the ones with current CPS or GS. The epileptic nature of the spells had to be supported by EEG and/or the presence of a compatible epileptogenic lesion on brain MRI. We examined whether these patients had other seizure types (CPS or GS) in the past and determined the distribution of epileptic foci between the two cerebral hemispheres. In those patients who had never experienced CPS or GS, we determined the interval between onset of symptoms and their correct characterization as epileptic. Results: Twenty one patients, nine men and 12 women, were selected. Their age ranged from 25 to 79 years. Three (14%) were included just based on the presence of clear ictal and/or interictal epileptiform discharges on EEG, nine (43%) had appropriate epileptogenic lesions on brain MRI, and the remaining nine (43%) had combined EEG and MRI abnormalities. Twelve subjects (57%) had never experienced other seizure types, while nine (43%) reported rare GS/CPS in the past. The epileptogenic focus lateralized to the right cerebral hemisphere in 15 patients (71%) and to the left in six (29%), p = 0.08. In the subjects with no history of GS or CPS, the delay to the correct diagnosis ranged from three weeks to five years (mean = 16 months), with seven of the 12 patients waiting for more than one year to be adequately diagnosed and treated. Conclusions: Practitioners should be aware of the potential for SSPS to occur as the exclusive or main seizure type in some epilepsy patients. The appropriate work up, diagnosis and treatment are often delayed in such circumstances. In addition, we noted a trend suggesting that right hemispheric foci are overrepresented in this particular population. Abstract: .sup.1 Noah Webster , .sup.2 Peggy Crawford , and .sup.2 Farrah M. Thomas (Sociology, Case Western Reserve University, Cleveland, OH ; and Neurology, Cleveland Clinic, Cleveland, OH) Rationale: Patients with epilepsy generally are discouraged from driving due to the risk of experiencing a seizure, losing control, and causing harm to themselves and/or others. Driving regulations for epilepsy patients vary across the U.S.; some states requiring physicians to report patients to motor vehicle bureaus and/or requiring patients to be seizure-free for varying periods of time, while others have no regulations at all. Few studies examine why patients with epilepsy disregard medical advice and continue to drive. The purpose of this study is to examine demographic, disease, and health-related correlates of driving status among medically intractable epilepsy patients. Methods: Participants are two hundred and twenty-five adults (44% male, 88% Caucasian, 44% married, 42% employed full or part-time) aged 18-75 with medically intractable epilepsy referred for surgery evaluation at the Cleveland Clinic. They participated in a health psychology evaluation that included a chart review for demographic, disease, and health-related information. Results: Nineteen percent of the patients continue to drive against medical advice. Current driving status is not associated with age, gender, marital or employment status, duration of epilepsy, seizure frequency, or the use of tobacco, alcohol, or illicit drugs. However, patients on disability are less likely to be currently driving. Descriptive data on a subset of 49 patients indicate 11 of the 12 patients currently driving have never been in an epilepsy-related motor vehicle accident (MVA), while 13 of the 37 not currently driving have been in at least 1. Only 2 of the 12 (17%) currently driving report limited social support, and 11 of the 37 (30%) not driving report limited support. Conclusions: A fair number of patients with medically intractable epilepsy continue to drive against medical advice, but those in an epilepsy-related MVA are less likely to be currently driving. Do these patients drive out of necessity (i.e. limited social support or employment) or because they are generally nonadherent? Results suggest the answer is not that straightforward. While employment status is not associated, patients on disability are less likely to be driving, possibly due to the severity of seizures. What would increase epilepsy patients' adherence to the recommendation against driving? Anecdotally, our patients report community neurologists are less restrictive about driving than neurologists affiliated with epilepsy centers. This suggests more consistent guidelines and follow-up are needed. Ideally an epilepsy-related MVA should not be needed for adherence. Subsequent analyses will examine how personality factors (i.e. anti-social or risk taking characteristics) are related to driving status, and when data are available we will re-examine how driving status relates to epilepsy-related MVAs and social support. Abstract: .sup.1 David Wheeler , .sup.1 John O. Elliott , .sup.2 Mercedes P. Jacobson , and .sup.2 Jose Cabassa (Public Health, Ohio State University, Columbus, OH ; and Neurology, Temple University, Philadelphia, PA) Rationale: In 1997 the Centers for Disease Control defined epilepsy as a public health issue and specific goals related to access, quality of care, surveillance and public education were developed. These continue to be issues where progress has been slow. A 2005 health disparities report from the Agency for Healthcare Research and Quality revealed that significant gaps still exist in non-Caucasians for many chronic health conditions. The US government report Healthy People 2010 recommended that 90% of all major National, State and local health systems promote nationwide use of Geographic Information Systems (GIS) to map health related data. The process of geocoding (assigning map coordinates) allows researchers to examine where people reside in relation to distribution of health, socioeconomic conditions and proximity to environmental hazards. This is the first use of GIS technology to map epilepsy related health data. Methods: Three years (2002-2004) of inpatient/outpatient data were obtained from the 5 hospitals within the Temple University Health System (TUHS). The catchment is primarily North and Northeast Philadelphia - an area of significant ethnic diversity and varying socio-economic status. The dataset included 24,406 contacts with the TUHS. Data included demographics, address, up to 5 ICD-9 codes and insurance type. Results: Fifty-four percent of patient contacts were males (mean age = 47.4) and 46% were female (mean age = 48.9). Thirty-two percent were Caucasian, 47% African American, 20% Unknown and 33% Latino residents. There were 665 patient contacts within the ICD-9 codes 345 to 345.81 and 817 with a diagnosis of 345.9. There 22,207 patient contacts (91%) coded for seizure NOS code (780.39). The most frequent co-morbidities were hypertension (n = 2322), diabetes (n = 1158), heart failure (n = 794), other urinary problems (n = 774), non-compliance with medical treatment (n = 707), COPD (n = 545), volume depletion (n = 522), asthma (n = 494) and depression NOS (n = 459). Conclusions: This study adds to the understanding of co-morbidities in persons with seizures/epilepsy. Statistical modeling will be used to explain similarities and differences. Culturally relevant outreach to these populations is critical and GIS technology is useful in demonstrating where educational programs for patients and health professionals should be focused. Materials based on medical terms foreign to the lay public are less effective and tend to be written at a level that exceeds the readability of most patients. Since patients of low socioeconomic status are less likely to access internet resources, other community-based outreach needs to be investigated. Abstract: .sup.1 Tarakad Ramachandran , .sup.1 Robert Beach , and .sup.1 Alireza Yarahmadi (Neurology, SUNY Upstate Medical University, Syracuse, NY) Rationale: Seizure is one of the most common serious neurological conditions encountered in the emergency departments. There is limited information about current use of emergent Brain CT in the evaluation of breakthrough seizure and its yield. Methods: Retrospective review, over a period of two years, of medical records of 250 emergency room admissions of adult patients (ages 18-59) with known history of epilepsy and breakthrough seizure(s) while on antiepileptic medications.The patients with diagnosis of status epilepticus have been excluded. Results: Of the 250 patients with epilepsy, emergent brain CT was performed in 130(52%) cases. The indication for emergent brain CT in addition to epilepsy were fall in 56 (43.1%), altered mental status in 24 (18.5%), lethargy in 19(14.6%), prolonged seizure in 18 (13.8%), recurrent seizures in 11 (8.5%), and headaches in 2 patients (1.5%). Type of seizures included 87 (66.9%) patients with generalized tonic clonic, 33 (25.4%) complex partial with secondary generalization, 2 (1.5%) myoclonic, and 8 (6.1%) had different types of seizures. 98 (75.4%) patients had previous emergent brain CT(s) in our institution. Of the 130 patients scanned 86 (66%) had a normal CT brain, evidence of an old stroke was evident in 8 (6%), small vessel disease in 9 (6.9%), post operative changes in 11 (8.4%), remote trauma in 5 (3.8%), previously diagnosed primary brain tumor in 4 (3%), and nonspecific findings in 7 (5.3%) patients. Emergent brain CT had minor or no impact on the therapy decision-making in any of the 130 patients. Conclusions: The cost of multiple brain CTs in patients with established diagnosis of epilepsy is not trivial, and the benefits have not been determined. The potential cost savings of a more rational approach, multiplied by the large number of patients present to ED with breakthrough seizure, could be huge. Development of criteria or evidence-based guidelines is needed to justify necessity and value of emergent brain CT in this patient group. Abstract: .sup.1 Amir M. Arain , .sup.1 Aley M. Hamadani , and .sup.1 Bassel W. Abou-Khalil (Neurology, Vanderbilt University Medical Center, Nashville, TN) Rationale: To evaluate the predictors of seizure remission after the diagnosis of psychogenic nonepileptic seizures (PNES) is made with Video EEG monitoring. Methods: We retrospectively analyzed the records (2000-2005) of 39 patients with psychogenic seizures who returned for follow up after diagnosis on video EEG monitoring. They represented 24% of 165 who received this diagnosis on video EEG monitoring, even though all patients were asked to return for follow-up. Neuropsychological testing was ordered in all patients, but only 9 had this testing. The status of seizure control was determined at follow up (1 week-4years, mean 7months). We compared the group with residual seizures at follow up and the seizure-free group for the following variables: demographic variables, age at spell onset, level of education, marital status, history of abuse, type and number of recorded events, history of drug abuse, and evidence of social support as suggested by an accompanying family member. We used Fisher's exact test for categorical variables and the t-test for continuous variables. Results: There were 17 males and 22 females with an age range from 20-82 years (mean 39.5 years). Ten patients (26%) had achieved seizure remission, while 74% continued to have psychogenic seizures. Patients who were seizure-free at follow up were more likely to be accompanied by a family member to their first clinic visit. No other variables distinguished the two groups. Conclusions: Patients with psychogenic seizures have poor compliance with follow up after their diagnosis. The majority continue to have spells at follow up. Patients with a better outcome had evidence of greater psychosocial support. Abstract: .sup.1 Jorge J. Asconape , and .sup.1 Jan C. McGee (Neurology, Loyola University Chicago, Maywood, IL) Rationale: Humans have individual differences in the timing of their daily activities falling between two extremes: morning types (larks) and evening types (owls). The normal work schedules, determined by light/dark cycles, tend to favor morning types and could lead to chronic sleep deprivation in evening types. As a group, patients with idiopathic generalized epilepsy (IGE) show a tendency for being "owls". This particular circadian sleep pattern is often not recognized by treating physicians but can have important implications in the diagnosis and managment of these patients. Methods: We adminstered the English version of the Horne-Ostberg self-assesment questionnaire.sup.1 to 12 patients with idiopathic generalized epilepsy to determine their morningness and eveningness chronotypes. All patients included had both clinical and EEG findings consistent with IGE. Results: Of the 12 patients, 7 had juvenile myoclonic epilepsy, 3 had IGE with generalized tonic-clonic seizures only, and 2 had juvenile absence epilepsy. Mean age at the time of testing was 19.6 years (r: 15-25 years). The mean test score was 39.1 (r: 25-64) which corresponds to a moderately evening type. Two patients (16.7%) were scored as definitely evening type, 5 (41.7%) as moderately evening type, 4 (33.3%) as neither and 1 (8.3%) as a moderately morning type. When patients were asked to classify themselves as either a "morning" versus "evening" type, 7 (58.3%) indicated a definitely evening type, 4 (33.3%) a moderately evening type and 1 (8.3%) a moderately morning type. Conclusions: Data from this preliminary series suggests that the evening chronotype predominates in patients with IGE. Inquiring about chronotypes may be helpful in the initial diagnosis of IGE. More important, eveningness has been associated with chronic sleep deprivation, a well-known seizure precipitating factor in IGE, so its recognition may be important for clinical management. Reference Horne, JA Ostberg, O. A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. Int J Chronobiol 1976 Volume: 4 97-110 Abstract: .sup.1 Olga Bogdanova , .sup.2 Henry Buchtel , and .sup.1 Daniela Minecan (Neurology, University of Michigan, Ann Arbor, MI ; and Psychiatry, University of Michigan, Ann Arbor, MI) Rationale: The Intracarotid Amytal testing (WADA test) has been used as part of the presurgical evaluation of patients with temporal lobe epilepsy(TLE). The purpose is to confirm language lateralization and identify patients at risk for memory impairment after the resective surgery. The ideal surgical candidate has better memory scores in the hemisphere contralateral to the expected surgery side. The concern is in cases with reversed WADA memory asymmetry scores. The memory outcome has been previously analyzed for patients with left anterior temporal lobectomies(ATL) and reversed scores. The significance of symmetrical WADA memory scores(irrespective of the side of surgery) and reversed WADA memory asymmetry scores for patients who undergo right ATL has not been previously evaluated. Methods: Patients were selected from those with temporal lobe epilepsy and who underwent complete presurgical evaluation at the University of Michigan Medical Center between 2000 and 2005. We included patients who had symmetrical and reversed WADA memory asymmetry scores and who were seizure free at least at one year follow up after the resective surgery. A total of 14 patients were identified. Age range was between 9 and 56 years at the time of the WADA test. Six patients were males. All patients underwent neuropsychometric evaluation as part of the presurgical evaluation and at three months after the resective surgery. Seven patients also had another set of neuropsychometric evaluation performed 12 to 13 months after the surgery date. Nine patients had symmetrical WADA memory scores and two of these patients had right temporal lobectomies. The other five patients had reversed WADA memory asymmetry scores. In this group, four patients underwent left temporal lobectomies. We compared 3 month post-operative verbal and pictorial memory change scores. Results: None of the patients with symmetrical WADA memory scores(n = 9) have shown any significant change in the memory outcome at the post-surgical neuropsychometric evaluation, irrespective of their site of resection. One patient with left temporal lobectomy showed slightly improved pictorial memory. All patients with reversed WADA memory asymmetry scores (n = 5) had a poorer verbal and/or pictorial memory outcome, despite the fact that they were also seizure free. The decline was considered statistically significant in all of them, again irrespective of the site of surgery. Conclusions: Dominant or non-dominant anterior temporal lobectomy (ATL) patients with reversed WADA memory asymmetry scores are at significant risk for post-operative memory deficits, despite complete seizure control. However, symmetrical WADA memory scores do not pose a threat for either the verbal or the pictorial memory scores of the ATL patients. A larger cohort of patients and longer post-operative follow up may be useful to confirm these results. Abstract: .sup.1 Christine Dong , .sup.1 Eric Pina-Garza , and .sup.1 Bassel Abou-Khalil (Neurology, Vanderbilt Medical Center, Nashville, TN; Neurology, Vanderbilt Medical Center, Nashville, TN; and Neurology, Vanderbilt Medical Center, Nashville, TN) Rationale: Pentobarbital is a widely used for the treatment of refractory status epilepticus. The EEG is usually continuously monitored during pentobarbital administration and during the withdrawal period and is the basis for the decision to resume pentobarbital or continue its withdrawal. We observed a pattern of periodic sharp EEG discharges related to pentobarbital withdrawal. This pattern is transient with continued withdrawal. It is not associated with clinical seizure activity and is not predictive of recurrence of status epilepticus. This pattern is important to describe and recognize because it may result in unnecessary resumption of pentobarbital. Methods: We recognized a periodic pattern during recovery from pentobarbital coma in 5 patients. EEG was recorded intermittently in all, for 20-30 minutes every 8 hours in the first patient, using paper EEG, and for 5 minutes every 30 minutes for the remaining patients (using digital EEG recording). We reviewed the EEG pattern during pentobarbital withdrawal. We also reviewed the clinical history, including outcome in all patients. Results: The first patient received pentobarbital for increased intracranial pressure after traumatic brain injury and the other four received pentobarbital for refractory status epilepticus. During pentobarbital withdrawal all developed a pattern of sharply contoured periodic complexes, typically with a recurrence rate of one every 1-2 seconds. In the first patient, this was interpreted as electrographic status epilepticus. However, after several repeated trials of pentobarbital, it was realized that this may not be a seizure pattern and pentobarbital was fully withdrawn with no clinical seizure activity. With the remaining patients, the pattern was recognized earlier, thus avoiding prolonged pentobarbital treatment. In all patients, the periodic pattern evolved into more continuous EEG activity that was not ictal. This was associated with clinical recovery, though not necessarily to pre-status baseline. Conclusions: A periodic EEG pattern may occur in association with pentobarbital withdrawal. When such a pattern is not associated with clinical seizure activity, it should not be a reason to resume pentobarbital since it is usually not an ictal pattern. [figure 1] Abstract: .sup.1 Fariba Farhidvash , .sup.1 Amir M. Arain , .sup.2 Pradumna P. Singh , and .sup.1 Bassel W. Abou-Khalil (Neurology, Vanderbilt University Medical Center, Nashville, TN ; and Neurology, Meharry Medical College, Nashville, TN) Rationale: Access to healthcare is unsatisfactory in indigent patients with epilepsy. Multiple factors hinder in the access to healthcare including absence of insurance. Many epilepsy patients use the hospital emergency room as a primary or secondary source of care for their seizure disorder. Many indigent patients do not see an epilepsy specialist for long term care. We wished to find the factors that affect the access to healthcare in these indigent epilepsy patients. Methods: We obtained data of patients seen in the ER for seizures in 2002 and 2003, in two institutions, Vanderbilt University Hospital (VUH) and Metro Nashville General Hospital (MNGH), seeking to identify patients who had visited the emergency room more than once. Vanderbilt is a tertiary care institution while MNGH serves the more underserved population. We also collected insurance information of these patients, and compared them to the outpatient VUH epilepsy clinic data. We did a chart review to identify these patients who finally followed up in the epilepsy clinic. Results: A total of 702 patients visited the VUH ER and 446 patients visited the MNGH ER for seizures. Of the VUH patients, 125 patients visited the ER more than once. Of these patients who visited ER multiple times, 69 (55.2%) finally followed up in the epilepsy clinic and of those, 24 patients (34.8%) were already followed by an epileptologist. Patients who visited the ER mulitple times, 31 (25%) carried private or commercial insurance, 91 (73%) carried Medicare or Medicaid, and 3 had no insurance. The 69 patients who followed up in the clinic, 19 (61.3%) carried private insurance, 50 (55%) carried Medicare/Medicaid. The average duration between the first ER visit to the first clinic visit was 196 days with those who carried private insurance averaged 91 days and those with Medicare/Medicaid averaged 236 days. Of the 446 MGNH patients, 157 (35%) visited the ER more than once. Of these 157 patients, 36 (23%) finally followed up in the outpatient Neurology clinic. Of these patients who made multiple ER visits, 0.9% patients had commercial insurance, 38% carried Medicaid, 12% Medicare and 36% were self pay. None of the commercial insurance carriers followed up in clinic while 31 (86%) of the Medicaid/Medicare carriers and 3 (8%) of the uninsured patients had. The average duration of time between the first ER visit and the first clinic visit was 73 days, 71 days for the Medcare/Medicaid carriers and 84 days for the uninsured Conclusions: Among indegent patients using ER for treatment of epilepsy is not adequate and there is still considerable lack of long term follow up of seizures in outpatient setting. Overall this results in disparity of care in indigent patients and adds tremendous cost to society. This study emphasizes the importance of better epilepsy management and educational and social programs for those with epilepsy. Abstract: .sup.1 Jacqueline A. French , .sup.2 Nancy R. Temkin , .sup.3 Jeffrey S. Nye , and .sup.3 Filip De Ridder (Department of Neurology, Hospital of University of Pennsylvania, Philadelphia, PA ; University of Washington, Seattle, WA ; and Johnson & Johnson Pharmaceutical Research & Development, L.L.C., Raritan, NJ) Rationale: Over the past two decades, the efficacy endpoint in clinical trials of new AEDs as adjunctive therapy has been reduction in baseline seizure frequency, requiring [greater than or equal to]4-wk baselines and up to 12-wk treatment periods. Time to n-th seizure (T.sub.n) has been proposed as an alternative endpoint that could reduce study duration, eliminate the baseline period, and broaden inclusion criteria, thereby speeding enrollment. Using data from TPM clinical trials, we compared T.sub.n and median% reduction in baseline seizure frequency for their abilities to detect treatment effects. Methods: Two methods were used to compare endpoints: 1) reanalysis of TPM clinical trial data using T.sub.n as efficacy endpoint; 2) a simulation study with a model based on TPM data. Data were from six randomized, double-blind, placebo-controlled trials with topiramate as adjunctive therapy (200-1000 mg/day) in refractory partial-onset seizures. T.sub.n for 3, 6, 9, and 12 seizures (T.sub.3, T.sub.6, T.sub.9, T.sub.12) were estimated using Kaplan-Meier analyses. Cox proportional hazard regression, stratified by baseline 28-day seizure frequency (22), was used to estimate the hazard ratio for each study arm (dose) vs placebo. Simulation study was performed as 12-wk dose-ranging trial with 4-wk baseline assessing 6 doses to yield treatment effects (difference from placebo in median% seizure reduction) of 0-50%. Results: T.sub.n analysis showed superiority of TPM over placebo, consistent with original analyses of median% seizure reduction. Simulation study results showed a linear relationship between T.sub.n hazard ratio and median% seizure reduction. T.sub.n analysis was generally less powerful than seizure rate analysis, although power increased with number of seizures (n[greater than or equal to]6) and with stratification according to baseline seizure frequency. Adequate power of the stratified T.sub.6 analysis for detecting clinically meaningful treatment effects (30% difference from placebo) could be attained with a reasonable sample size (80 patients/arm). A study with a prospective baseline and 12-wk follow-up would require 36 subjects to achieve the same power. In TPM clinical trials, 80% of patients experienced their 6.sup.th seizure within 6 wks of study initiation. Conclusions: Stratified T.sub.n analysis, with n = 6, can provide adequate power to detect clinically relevant effect sizes of an AED as adjunctive therapy in a sample of 80 patients with refractory partial-onset seizures. The potential advantages of reducing clinical trial duration from 16 wks to 6 wks and potentially eliminating the prospective baseline period must be weighed against potential disadvantages such as need for larger sample sizes. (Supported by Johnson & Johnson Pharmaceutical Research & Development, L.L.C., Raritan, NJ.) Abstract: .sup.1 Daniel Friedman , .sup.2 Evren Burakgazi , and .sup.1 Susan T. Herman (Department of Neurology, Hospital of the University of Pennsylvania, Philadelphia, PA ; and Department of Neurology, Virginia Commonwealth Unversity, Richmond, VA) Rationale: Subacute sclerosing panencephalitis (SSPE) is a rare but devastating late neurological complication of measles infection characterized by a progressive encephalopathy, myoclonus, and focal neurologic deficits occuring 7-10 years after initial infection. Survival from onset is typically 1-3 years. There is no proven effective therapy. We report a 15-year old girl with SSPE who had marked improvement in atonic seizures and neurocognitive status with zonisamide and interferon-beta 1a. Methods: Case report. Results: A 15-year-old girl presented to our epilepsy center with seizures and cognitive decline. At 19 months, a febrile exanthem illness diagnosed as Kawasaki disease was treated with intravenous immunoglobulins. At age 13 she developed nocturnal frontal lobe complex partial seizures, successfully controlled with lamotrigine. Eighteen months later, her family noticed personality changes and inattentiveness. Cognitive decline continued over the next 8 months; full scale IQ decreased from 101 to 77. At age 15, she began to have atonic seizures and falls. Video EEG monitoring showed frequent atonic episodes, brief behavioral arrests with slow eye blinks, and myoclonic jerks, concurrent with periodic generalized high-voltage slow wave complexes occurring once every 5 - 6 seconds. Laboratory data included markedly elevated IgG, IgG index, and IgG synthesis rate with positive CSF oligoclonal bands. Serum and CSF measles IgG were significantly elevated. MRI of the brain revealed subtle bilateral occipital white matter T2 hyperintensities. Functional MRI synchronized to the abnormal blinks showed increased BOLD signal and perfusion in bilateral putamina and cingulate cortex. Zonisamide 250 mg daily was added to her regimen. Atonic episodes remitted completely, and EEG 1 month later showed no periodic slow wave complexes. On isoprinosine 3000 mg/day, neurologic status remained stable for 6 months. Cognitive status then declined again (FSIQ 40, modified mini-mental status exam (MMSE) 20/57) and new MRI T2 hyperintense lesions appeared in bilateral basal ganglia. She was started on interferon-beta 1a 44mcg subcutaneously three times weekly. Her Neurologic Disability Index declined from 24% to 6%, and MMSE improved to 36/57. She continues to improve 1 year after subcutaneous interferon was initiated. Conclusions: Myoclonic and atonic seizures, hallmarks of SSPE, are usually refractory to standard antiepileptic drugs. Our case demonstrates a rapid and complete response to zonisamide which was sustained over nearly two years, suggesting a new avenue for therapy of these refractory seizures. In addition, our patient showed sustained improvement in neurocognitive function with subcutaneous interferon beta 1a. This therapy should be considered in SSPE patients who have the failed the current standards of care. Abstract: .sup.1 Christian P. Hansen , .sup.1 Rene Mathiasen , .sup.1 Jorgen Alving , .sup.1 Birgitte H. Jensen , .sup.1 Ellen H. Jensen , and .sup.1 Signe Madsen (Management, Danish Epilepsy Centre, Dianalund, Dianalund, Denmark) Rationale: Patient safety is of increasing concern in epilepsy centres. Various devices for detetction of epileptic seizures are available on the market. We decided to test the clinical value of bed alarms and pulse oxymeters in the Danish Epilepsy Centre, Dianalund. Methods: In-patients at increased risk of generalized convulsive seizures were supervised by video-cameras during the time spent in bed. The signals from the video-cameras were transmitted to a central supervision unit where a nurse was watching the screens with patients under supervision. In case of seizure, the supervising nurse notified the ward with the patient. The patient under supervision in bed carried a pulse oxymeter, and the bed was equipped with a seizure alarm. The staff in the ward registered alarms and compared with the messages from the nurse watching the patient on the screen. The sensitivity of the alarm (probability of alarming in case of seizure) and the occurrence of false alarms were registered. Patients were included from April 2004 to September 2005. In a second inclusion period from October 2005 to April 2006 we tried to improve the use of bed alarms by adjustment of alarm parameters. Results: In the first period 83 patients, aged 1-63 years, had 1332 epileptic seizures. The sensitivity of bed alarms was 18.7% in case of tonic-clonic seizures, 1.2% with tonic seizures, and 3.6% with focal motor seizures. The sensitivity of pulse oxymeters was 22.5% in tonic-clonic seizures, and 9.2% in tonic seizures. The sensitivity of both alarms were lower for other seizure types. The combined sensitivity of pulse oxymeters and bed alarms was 26% for tonic-clonic seizures and smaller for other seizure types. In average, there were 0.75 false alarms from the bed alarm per 24 hours, and 2.19 false alarms from the pulse oxymeters per 24 hours. In the second inclusion period it was not possible to improve the sensitivity or specificity of either alarm type. Conclusions: In our hands the combined use of bed alarms and pulse oxymeters detects a quarter of tonic-clonic seizures in epilepsy patients at increased seizure risk and a smaller fraction of other seizure types. There is a need for development of better technology for the provision of safety for epilepsy patients. Abstract: .sup.1 Lara E. Jeha , .sup.1 Imad Najm , .sup.2 Bill Bingaman , .sup.1 Peter Widdess-Walsh , and .sup.1 Hans Luders (Neurology, Section of Epilepsy, Cleveland Clinic Foundation, Cleveland, OH ; and Neurosurgery, Cleveland Clinic Foundation, Cleveland, OH) Rationale: Frontal lobe epilepsy (FLE) is the second most common cause of intractable focal epilepsy. Yet, little is known about the predictors of surgical success with frontal lobectomy. Methods: We reviewed patients who underwent a frontal lobectomy at our institution between 1995 and 2003 for the treatment of intractable FLE. Endpoints included Engel score at last follow-up, and time to any seizure recurrence. Overall seizure freedom was assessed using a Kaplan-Meyer survival analysis, and multivariable logistic regression. Results: Seventy patients were identified. Thirty (43%) were female. Age at surgery ranged from one to 57 yrs(median 18.5 yrs), with a median of age at seizure onset of 6 yrs. Duration of follow-up ranged from one to 15 yrs (median 3.1 yrs). MRI showed frontal lobe lesions in 58%, and was normal in 26%. In 16% of the cases, MRI lesions extended beyond the frontal lobe, showing mesial temporal sclerosis in 8(12%). The most common etiology was a malformation of cortical development seen in 60% (with normal imaging in 29% of those), followed by tumor (19%), encephalomalacea from stroke or trauma (10%), cryptogenic with normal MRI and pathology (10%), and vascular malformation (1%). Thirty nine patients had at least one seizure recurrence during the whole duration of follow-up. Kaplan Meyer survival analysis estimated a 56% probability of seizure-freedom at 1 year, 45% at 3 years, and 30% at 5 years. [figure 1]Engel scores at the last follow-up were Engel 1 in 56%, Engel 2 in 7%, Engel 3 in 16%, and Engel 4 in 21%. Univariate screening identified mesial temporal sclerosis on preoperative MRI, extrafrontal MRI abnormalities, etiology, use of subdural electrodes, and immediate postoperative seizures as predictors of postoperative seizure recurrence. Only extrafrontal MRI abnormalities and immediate postoperative seizures remained significant upon using multiple logistic regression. Conclusions: Frontal lobectomy is an effective treatment in intractable FLE. Extrafrontal MRI abnormalities and immediate postoperative seizures identify the patient population with the worst outcome. Abstract: .sup.1 Sam A. Kabbani (Neurology, East Tennessee Neurology Clinic, P.C., Knoxville, TN) Rationale: Use of a single broad-spectrum antiepileptic drug (AED) that is both efficacious and safe is optimal in epileptic patients. An AED displaying these properties and given as a first-line agent could eliminate trial and error of multiple therapies which often must be tried before the most effective therapy is determined. Levetiracetam (LEV) has previously shown efficacy and safety when used as monotherapy. Methods: Single-center, retrospective review of 98 epileptic patients (69 females, 29 males; mean age 39.5 years, rage 15-75 years) who were treated with LEV monotherapy over a one-year period. Medical records were reviewed for seizure type and etiology, history of prior AED use, imaging results, and LEV dosage. Efficacy and tolerability were assessed by reduction in seizure frequency and by reported adverse effects. Results: The most common seizure types were generalized (n = 49) and complex partial (with or without secondary generalization, n = 45). Fifty-one patients (52%) were given LEV as the first-line AED; 34 wre taking 1 AED and 13 were taking 2 AEDs at study initiation. The most frequently used AEDs included divalproex sodium (11%) and phenytoin (9%). LEV dosage was 3000 mg/day in 46 patients (47%) and 1500 mg/day in 36 (37%); titration to this dosage occurred over 1 week. The mean seizure reduction rate was 87.3%, with 53 patients becoming seizure free. An overall responder rate ([greater than or equal to]50% seizure reduction) of 95% was noted. Thirty-four patients (35%) were able to discontinue 1 or more AEDs and were converted to LEV monotherapy. Clinical impression was good or excellent in 83 patients (85%). Of patients given LEV as first-line monothrapy, a mean seizure reduction of 87.6% was noted, with 32 patients becoming seizure free. The responder rate in the sub-group was 94% with 44 patients with a good or excellent clinical impression. Adverse events were reported in 7 patients overall (7%); 3 required discontinuation for itching/rash (1), dizziness/ataxia (1) or nausea (1). Conclusions: LEV showed excellent efficacy and tolerability in this group of patients when used as monotherapy. Further studies should explore the use of LEV in monotherapy, particularly as a first-line agent. (Supported by UCB Pharma, Inc.) Abstract: .sup.1 Peter Widdess-Walsh , .sup.1 Prakash Kotagal , .sup.2 Guiyun Wu , .sup.1 Jeha Lara , and .sup.1 Richard Burgess (Department of Neurology, Cleveland Clinic, Cleveland, OH ; and Department of Nuclear Medicine, Cleveland Clinic, Cleveland, OH) Rationale: Epileptic auras provide important information about focal epilepsy and its location. Patients with focal epilepsy also report multiple aura types but limited data exists as to the significance of the occurence of multiple auras in the same patient. If multiple auras represent multiple seizure types or ictal onset zones, surgical treatment may not prove successful. Methods: The epilepsy monitoring unit database of the Cleveland Clinic was searched for patients with more than one aura type classified according to the Cleveland Clnic seizure classification system between the years 1989 and 2005. Clinical, EEG, and imaging data was analyzed to determine the location of the epileptogenic lesion, and explore possible mechanisms for the phenomenon of multiple auras. Results: Thirty one patients were identified who experienced multiple auras during a given seizure. 90% of patients with more than one aura had seizures arising from the right/non-dominant hemisphere (n = 31). 100% of patients with at least 3 auras had seizures arising from the right/non-dominant hemisphere (n = 12). In two thirds of cases, the auras occurred as a sequence in the same seizure, and in one third the auras occurred simultaneously. 30/31 patients had seizures with preservation of awareness. EEG seizures remained restricted. Twenty of the patients had resective epilepsy surgery with a good outcome in 59%. Patients who did not have surgery were either not drug-resistant, lost to follow-up, or had tumor-related epilepsy. In one patient with seizures from both temporal lobes, right sided seizures were associated with multiple auras, and left sided seizures were associated with aphasic seizures with loss of awareness. Subdural EEG recordings in 6 patients showed either a 'jacksonian march' of sequential auras, or in one case, several ictal onset zones resulting in separate isolated auras. Ictal SPECT images in 6 patients with right-sided seizures showed a lack of activation of the basal ganglia and brainstem. Conclusions: Most patients who report multiple aura types have localized epilepsy in the non-dominant hemisphere, and are good surgical candidates. The most common mechanism for multiple auras may be a spreading but restricted EEG seizure activating sequential symptomatogenic zones, but without the ictal activation of deeper structures or contralateral spread to cause loss of awareness and amnesia for the auras. Abstract: .sup.1 Ki Hyeong Lee , .sup.1 Yong D. Park , .sup.2 Joseph R. Smith , and .sup.2 Mark R. Lee (Neurology, Medical College of Georgia, Augusta, GA ; and Neurosurgery, Medical College of Georgia, Augusta, GA) Rationale: Functional Hemispherectomy (FH) is performed in patients with intractable epilepsy and disabling hemiparesis due to diffuse hemispheric pathology such as perinatal stroke or diffuse cortical dysplasia (CD). EEG finding in relation to post-op outcome of FH has not been systematically evaluated in the previous reports. Methods: A retrospective analysis included all patients who underwent FH from Sep 91 to June 05 at the Medical College of Georgia. Surgery was performed exclusively by two authors (JRS and MRL) who shared the same surgical technique. Patients were followed at least one yr post-op. Seizure outcome was classified as seizure-free (SF) vs. non seizure-free (NSF). Interictal and ictal EEGs were classified as lateralizing (L) (ipsilateral to the pathology) vs. non-lateralizing (NL) (unlocalizing, contralateral to the pathology, or bilateral-independent onset). Demographic characteristics, etiology, MRI/PET, outcome, and complication were collected from the medical record. Statistical analysis was performed using SPSS 13 package. This study was approved by the institutional review board. Results: Total of 37 patients were included: M 18, F 19. Mean age of seizure onset was 3.4 yrs (0-26 yrs) while mean age of FH was 13.82 yrs (1-40 yrs). Most common etiology was perinatal stroke (40.5%, n = 15) followed by diffuse CD (32.4%, 12), RS (21.6%, 8), head trauma (2.7%, 1), and Sturge-Weber Syndrome (2.7%, 1). Overall surgical outcome was excellent: 81.1% (30/37) SF vs. 18.9% (7/37) NSF. SF outcome was not different between the different etiologies: stroke 73.3% SF (11/5) vs. CD 83.3% SF (10/12) vs. RS 100% SF (8/8). Interictal EEG was nonlateralizing (NL) in 18.9% (7/37) while ictal EEG in 35.1% (13/37). Iinterictal EEG pattern did not predict SF outcome: L 87.5% (21/24) vs. NL 69.2% (9/13), Fisher's Exact Test; p = 0.6. Ictal EEG pattern did not predict SF outcome either: L 87.5% (21/24) vs. 69.2% (9/13), p = 0.18. In addition, within the specific etiologic groups, interictal or ictal EEG did not predict the outcome. Age at the time of surgery tended to be younger in the SF group: 13.1 [+ or -] 11.2 (SF) vs. 16.8 [+ or -] 5.8 (NSF), t-test p = 0.087. Conclusions: Our data showed that interictal or ictal EEG pattern does not predict the surgical outcome. Our study suggests that selection of candidiates for functional hemispherectomy should be based on clinical and radiological evidence, rather than scalp EEG. Abstract: .sup.1 Anthony Marson , .sup.2 David Smith , .sup.1 Catrin Tudur Smith , .sup.1 Paula Williamson , .sup.1 Ann Jacoby , and .sup.1 David Chadwick (Division of Neurological Science, University of Liverpool, Liverpool, Merseyside, United Kingdom ; and Department of Neurology, The Walton Centre for Neurology & Neurosurgery, Liverpool, Merseyside, United Kingdom) Rationale: Carbamazepine is widely accepted as a drug of first choice for patients with partial onset seizures. Several newer drugs possess efficacy against these seizure types but previous RCTs have failed to inform a choice among these drugs as they have failed to examine meaningful longer term outcomes. Methods: SANAD is an NHS HTA sponsored un-blinded randomized controlled trial. Arm A recruited patients for whom carbamazepine was considered to be standard treatment and were randomized to carbamazepine, gabapentin, lamotrigine, oxcarbazepine, topiramate. Choice of drug dose and changes in it were determined by clinicians' usual practice. Outcomes were time to treatment failure, and time to 1-year remission. Sample size calculations were calculated to detect both equivalence and difference. Results: 378, 377, 377, 210, 378 children and adults were randomized to carbamazepine, gabapentin, lamotrigine, oxcarbazepine and topiramate respectively. 87% of patients were classified as symptomatic or cryptogenic partial epilepsy at randomisation. For time to treatment failure lamotrigine was significantly superior to carbamazepine hazard ratio (HR 95% CI) 0.80 (0.64, 0.99), gabapentin 0.65 (0.52, 0.79), oxcarbazepine 0.76 (0.59, 0.99) and topiramate 0.65 (0.53, 0.80). For time to 12 month remission gabapentin was significantly inferior to carbamazepine 0.75 (0.62, 0.89), lamotrigine 0.82 (0.68, 0.98), oxcarbazepine 0.72 (0.58, 0.90). There was no significant difference between carbamazepine and lamotrigine 1.10 (0.92, 1.30). Conclusions: Lamotrigine has similar long term efficacy to, and is better tolerated than, carbamazepine. Lamotrigine should now be considered the treatment of first choice for patients with partial onset seizures. (Supported by NHS HTA programme.) Abstract: .sup.1 H.-J. Meencke , .sup.1 Ch. Dehnicke , .sup.1 M. Merschhemke , and .sup.1 A. Grimmer (Epileptoloy, Epilepsy Center Berlin-Brandenburg, Berlin, Berler, Germany) Rationale: In epilepsy surgery decisions are made by convergent datas from seizure semiology, electroencephalography, neuropsychology and imaging methods. In this study we analysed the distribution and maxima of interictal spiking (epileptogenic zone, EZ) and seizure onset (SZO) in patients with histologically verified hippocampal sclerosis (HS) compared to patients with extrahippocampal pathology (NHS). Methods: We included 182 patients, 81 patients with HS and 101 patients with extrahippocampal pathology. All patients had an EEG work up with closely spaced surface and sphenoidal electrodes (SP 1/2). Results: Interictal: In NHS patients there was a much more widespread involvement of surface electrodes than in HS patients. In HS patients we had a more frequent maximum in SP 1/2 (60/38%) than in NHS patients which was significant (p < 0.004). The sensitivity was 60%, the specificy 62%: Ictal: Ictally we had a much more widespread onset maximum in NHS than in HS. SP 1/2 maximum in HS with 64% was significant higher (p < 0.001) than in NHS (33%). The distribution of EZ and the SZO maximum together resulted in a moderate sensitivity with 74% (p < 0.001). Conclusions: Interictal and ictal involvement of SP 1/2 in HS patients is significantly higher than in NHS patients. But related to the moderate sensitivity the separation of hippocampal (HS) from extrahippocampal pathology is not possible in a single case. Only the mode of distribution of the involved electrodes together with adequate imaging can help to delineate the epileptogenic zone and to separate hippocampal from extrahippocampal cases. Abstract: .sup.1 C.P.J.A. Monte , .sup.2 J.B.A.M. Arends , and .sup.2 I.Y. Tan (Neurology, University Hospital Maastricht, Maastricht, Limburg, Netherlands ; and Epileptology, Kempenhaeghe Epilepsy Centre, Heeze, Brabant, Netherlands) Rationale: Ictal asystole has been hypothesized to be a mechanism for SUDEP. The aim of this study is to analyse the presence of risk factors for SUDEP in a population of patients with a history of ictal asystole. Methods: Risk factors for SUDEP were determined in a submitted systematic review: young age, male sex, early onset, generalized tonic-clonic seizures and the presence of brain lesion. We than performed a literature search (pubmed, embase, cochrane library and references) for cases of ictal asystole and checked them for the presence of these risk factors. Results: We identified 41 cases of ictal asytole. Male sex: n = 39, 71%. Young age n = 33, 66%. Generalized seizures n = 10, 70%. Early onset n = 20, 80%. Structural brainlesion n = 31, 54%. Conclusions: The occurrence of the risk factors is similar in both populations. The statistical significance wil be discussed using a population of epilepsy patients, a population of patients known with ictal bradycardia and a population of patients with a history of asystole. Abstract: .sup.1 Kathryn A. O'Hara , .sup.1 Tom Boro , and .sup.1 John M. Pellock (Neurology, Virginia Commonwealth University, Richmond, VA) Rationale: The emergency outpatient management of seizures varies significantly among EMS providers, depending on established protocols and levels of training/experience. To establish the clinical care provided by a highly trained EMS unit we reviewed records of the Richmond Ambulance Authority (RAA). Methods: The RAA is an all Advanced Life Support service operating in an urban environment. Using a dynamic deployment method matching resources to demand, they field up to 22 units within Richmond. In 2005 31,100 patients were transported including 1,159 seizures. All seizure related calls were reviewed. Of the 1,159 calls there were 539 unique social security numbers. Transports per person ranged from 1 (67%) to 17 (0.2%). Diazepam (IV) is the recommended 1.sup.st line therapy. IV attempts on pediatric (15years) patients that required medication were compared. Results: Priority 1 calls averaged 40 minutes from dispatched time to arrival at the emergency department(ED). The average time from the scene to the ED was 7 minutes. Of the 1,159 seizure calls, 625 patients (54%) were transported to the ED; 85 patients (13.6% of those transported) were under the age of 15. The success rate for starting an IV on seizure patients requiring diazepam was 62% on the 1.sup.st attempt, 16% on the second attempt and 19% on the 3.sup.rd attempt. In 3% of seizing patients an IV could not be established. There was no difference between adult and pediatric IV success rate. Conclusions: Rapid IV access is not always possible in patients who are actively seizing. Alternative forms of anticonvulsant treatment need to be available for EMS. Abstract: .sup.1 Juan G. Ochoa (Neurology, University of Florida, Jacksonville, FL) Rationale: Alteration of extracellular volume, ion concentration and osmolarity may affect the epileptogenic process. Oral hydration with a slight hypotonic solution helps to mantain cerebral perfusion and may prevent abrupt changes in the intersticial electrolyte concentration that may trigger epileptic activity. This therapy may help stabilize the nonsynaptic mechanisms associated with seizure activity. Methods: We treated empirically eight patients with refractory focal or generalized epilepsy using a rice-based oral electrolyte solution therapy at a dose ranging between one to two liters per day, in addition to the standard AEDs. Six other patients fail to comply with therapy or seizure record keeping and were excluded from this report. Concomittant AEDs were either adjusted or discontinued at discretion of the treating physician. Serum electrolytes were obtained both at baseline and during electrolyte therapy. Results: Three children with familial dysautonomia and refractory absence epilepsy became seizure free for one year after therapy with Ceralyte90. All three patients remain on Ceralyte90 therapy for four years. There were occassional absence seizures after AED discontinuation.One patient with juvenile myoclonic epilepsy not able to tolerate traditional AEDs became seizure free on electrolyte therapy for at least one year. One patient with symptomatic generalized refractory epilepsy had over 90% seizure reduction over one year after electrolyte therapy. One child with refractory symptomatic generalized epilesy stpped atypical absence seizures but persist with uncontrolled generalized convultions. One patient with symptomatic generalized epilesy had reduction of cluster seizures preventing hospital admissions after electrolyte therapy through his gastric tube. One patient with symptomatic generalized epilepsy did not improve and discontinued treatment after three weeks of therapy. No significant side effects or complications have been observed during therapy in any of these patients. Conclusions: Rice-based oral electrolyte hydration is a safe novel therapy that may help control seizures in patients with refractory epilepsy. A controlled study is needed to validate this finding. Seizure freedom with empirical treatment in four patients with atypical absences suggests a possible stronger effect in this type of seizures. Abstract: .sup.1 Zhiyi Sha , .sup.1 Manjari Tripathi , .sup.1 Cammy Chicota , .sup.1 Steve Rogers , .sup.1 Jerome Engel , and .sup.1 Susan Bookheimer (Neurology, UCLA Medical Center, Los Angeles, CA) Rationale: Language and memory preferentially develop in the same hemisphere; thus, verbal memory impairment is typically associated with left medial temporal lobe dysfunction while non-verbal memory impairment suggests a right hemisphere focus. However, in patients with left mesial temporal lobe epilepsy, this association may not be well established. Occasionally patients with a clear left hippocampus seizure focus and an apparently normal right hippocampus fail the Wada test with left side injection. We hypothesized that in left mesial temporal lobe epilepsy, a late age of onset may account for these unpredicted Wada failures. Specifically, we reasoned that during the critical years when language develops and becomes established in the left hemisphere, strong connections with the left hippocampus are formed preferentially; in early onset left temporal lobe epilepsy, connections are strengthened with the right hippocampus when the left is dysfunctional. After the critical period of language development (generally accepted as [less than or equal to] 7 years old), we hypothesized that connections from left hemisphere language cortex to right medial temporal lobe would be more difficult to establish, and this connection is necessary for verbal encoding of non-verbal items. Therefore, in patients with late onset ([greater than or equal to] 8 years old) left mesial temporal lobe epilepsy, we predicted that the right hippocampus may not be able to support memory after the left hippocampus loses its function secondary to longstanding seizures. To test this hypothesis we studied the relationship between left hemisphere Wada failures and age of seizure onset. Methods: We reviewed data on 69 patients who were admitted for pre-surgical evaluations for mesial temporal lobe epilepsy. Wada tests were performed as part of the pre-surgical evaluation. Forty-two patients had left mesial temporal lobe epilepsy and 27 patients had right mesial temporal lobe epilepsy. Results: Of the 42 left mesial temporal lobe epilepsy, 18 had an early onset ([less than or equal to] 7 years old) and 24 had a late onset ([greater than or equal to] 8 years old). All 18 early onset patient passed the Wada test with left sided injection. Nineteen of the 24 late onset patient passed the Wada test and 5 of the 24 failed (less than 60% recognition memory). The passing rate of the Wada test in early onset left mesial temporal lobe cases was significantly higher than the late onset group (p < 0.05). In the right mesial temporal lobe epilepsy patients, there was no significant difference between the early and late onset groups (1 out of 10 in the early onset and 1 out of 17 in the late onset did not pass the Wada test). Conclusions: Age of seizure onset may help to predict the Wada failure in left temporal lobe epilepsy. A greater incidence of Wada failures in late onset patients with left mesial temporal epilepsy compared to early onset patients suggests that memory laterality develops in concert with language development. Abstract: .sup.1 David Smith , .sup.2 Anthony Marson , .sup.2 Catrin Tudur Smith , .sup.2 Paula Williamson , .sup.2 Ann Jacoby , and .sup.2 David Chadwick (Department of Neurology, The Walton Centre for Neurology & Neurosurgery, Liverpool, Merseyside, United Kingdom ; and Division of Neurological Science, University of Liverpool, Liverpool, Merseyside, United Kingdom) Rationale: Valproate is widely accepted as a drug of first choice for patients with generalised onset seizures and is recommended for patients with seizures that are difficult to classify because of its broad spectrum of efficacy. It is thought that lamotrigine and topiramate also possess broad spectrum activity. SANAD is the first study to compare the longer term effects of these drugs. Methods: SANAD is an NHS HTA sponsored un-blinded randomized controlled trial. Arm B recruited patients for whom valproate was considered to be standard treatment and were randomized to valproate, lamotrigine or topiramate. Outcomes were time to treatment failure and time to 1-year remission. Results: 716 patients were recruited. At randomization the percentages with idiopathic generalised epilepsy (IGE), unclassified epilepsy, partial epilepsy and other generalised syndromes were 62%, 27%, 7% and 4% respectively. 239, 239, 238 children and adults were randomized to lamotrigine, topiramate and valproate respectively. Overall, for time to treatment failure valproate was significantly superior to topiramate Hazard ratio (HR 95% CI) 0.64 (0.48, 0.84), but there was no significant difference between valproate and lamotrigine 0.80 (0.60, 1.07). For patients with an IGE valproate was significantly superior to both lamotrigine 0.65 (0.45, 0.95) and topiramate 0.53 (0.37, 0.76). For time to 12 month remission valproate was significantly superior to lamotrigine overall 1.31 (1.06, 1.62) and for the subgroup with an IGE 1.48 (1.14, 1.93). But there was no significant difference between valproate and topiramate in either analysis. Conclusions: Valproate is better tolerated than topiramate and more efficacious than lamotrigine. Valproate is the drug of first choice for patients with generalised and unclassified epilepsies. (Supported by NHS HTA Programme.) Abstract: .sup.1 Andrea Sneider , .sup.1 Marianna V. Spanaki-Varelas , .sup.1 Lori A. Schuh , .sup.1 Brien J. Smith , and .sup.2 Lonni Schultz (Neurology, Henry Ford Health System, Detroit, MI ; and Biostatistics and Research Epidemiology, Henry Ford Health System, Detroit, MI) Rationale: There is a delay from the onset of epilepsy to epilepsy surgery of approximately 20 years. The aim of this study was to indentify if a delay in referring patients for epilepsy sugery is attributed to different views that neurologists have for refractory epilepsy. Through a survey, we assessed neurologists' definition of medically refractory epilepsy and their decision-making process on when to refer their patients for epilepsy surgery. Methods: A 10-item survey was developed and mailed to all practicing Michigan neurologists outside our institution. Surveys were anonymous but coded for tracking purposes. A nominal incentive was offered to respondents. Descriptive statistics were used in analysis. Results: Surveys were sent to 415 neurologists, 84 neurologists responded for a response rate of 20.24%. The majority of respondents defined medically refractory epilepsy as failing 3 monotherapy antiepileptic drugs (AEDs) trials and at least 2 polytherapy trials. Nineteen percent said that all approved AEDs had to be failed before considering a patient medically refractory. The vast majority (85%) of respondents did not feel failure of a vagus nerve stimulator (VNS) was required before labeling a patient refractory. However, the majority (62%) of the respondents thought that VNS offers a > 10% chance of seizure-freedom. Our survey showed that 82% of the participants had referred patients for surgery and 73% of respondents felt that epilepsy surgery is under recommended overall. The majority (70%) of neurologists would not consider patients 60 years old as appropriate surgical candidates. The majority stated that their patients should not be considered for surgery unless seizures have been present for a minimum of two years and have a frequency of monthly or greater. Almost 50% percent of respondents were not satisfied with the level of communication from epilepsy centers, and one-third did not have their patients returned to their care in a timely fashion. Another third reported serious complications resulting from surgery. Conclusions: These results provide further insight into the delay seen between onset of epilepsy and time to surgery. The majority of respondents concluded that children less than 10 years of age are inappropriate surgical candidates. In addition, nearly 50% percent of respondents were not satisfied with the level of communication from epilepsy centers, and one-third did not have their patients returned in a timely fashion. Better education of neurologists, as well as better communication from comprehensive epilepsy centers could shorten the time to referral, increase the number of referrals and ultimately improve the lives of our patients. (Supported by Henry Ford Hospital Neurology Department.) Abstract: .sup.1 Arnoldo Soto , and .sup.2 Renata D.C. Van Woensel (Unidad de Neurologia, Hospital Domingo Luciani, El Llanito, Caracas, Venezuela ; and Epilepsy Department, Cyberonics Europe, Zaventem, Belgium) Rationale: The Lennox-Gastaut syndrome (LGS) is widely known as one of the most severe and prognostically unfavourable epileptic conditions. When medications fail, nonpharmacologic treatments should be considered such as the ketogenic diet, epilepsy surgery (corpus callosotomy), and vagus nerve stimulation (VNS) Therapy. Methods: A retrospective analysis of data from Cyberonics' International Patient Registry showed 29 patients (18 male, 11 female) with LGS who were treated with vagus nerve stimulation (VNS) Therapy between May 1994 and September 2001. Patients had a mean age of 16.6 years (range, 4-34). A total of 5 patients were institutionalized, and 7 patients had a history of callosotomy. Twenty-six patients (89.6%) were mentally retarded/developmentally delayed. Patients were taking an average of 2.45 antiepileptic drugs (AEDs) per patient. Mean follow-up was 48.6 months. Data related to seizure reduction, AEDs and quality of life (QoL) were analysed. Results: There was a total median seizure reduction of 60%, and a total mean seizure reduction of 45.9% (p < 0.0001). A total of 41.4% of the patients (n = 12) had [greater than or equal to] 50% seizure reduction. A reduction in the total number of AEDs was also observed (-0.52 AEDs/patient; p = 0.0293). Median stimulation parameters at last follow-up visit were: output current, 1.25 mA; pulse width, 500 [mu]s; frequency, 30 Hz; time 'on', 30 sec; time 'off', 5 minutes. Six patients (20.7%) reported that the magnet was always or most of the time effective in aborting or reducing the intensity and/or duration of their seizures. QoL improvements were more pronounced for alertness (48.3%), post-ictal state (37.9%) and cluster seizures (34.5%). Conclusions: These results show that VNS Therapy allowed a significant reduction in seizure frequency and in total number of AEDs in this patient population with LGS. VNS Therapy was also associated with QoL improvements. Abstract: .sup.1 Michael J. Van Putten , and .sup.1 Maurice H. Van Putten (Neurology and Clinical Neurophysiology, Medisch Spectrum Twente/University of Twente, Enschede, Netherlands; and Massachusetts Institute of Technology-Laser Interferometer Gravitational-Wave Observatory, MIT, NW17-161, Cambridge, MA) Rationale: Increased insight into the ictal phenomena that occur in epilepsy, such as the characterization of dynamic regimes during status epilepticus, adds to the store of knowledge of abnormal brain behavior and may assist in improved patient care, including appropriate seizure termination paradigms. We present a detailed quantitative analysis of ictal phenomena in four patients suffering from a non-convulsive status epilepticus (SE), in whom the EEG shows burst-like patterns of epileptiform discharges (durations 2-30 s) punctuated by low-voltage 'flat periods' with a duration of 1-60 s. Methods: Epileptiform burst-like transients were identified from the power envelope of the EEG signal. Subsequently, correlations between initial phases of all possible combinations of the epileptiform burst-like transients were determined. In addition, histograms of the durations of the ictal events and the interictal periods were created. Quantitative analysis of burst duration and inter-burst intervals was performed by evaluation of the distribution of the burst durations and inter-burst intervals. Given the assumed nonlinear nature of the ictal state and the discrete nature of the bursts and the IBI, potential power law behavior was studied. Results: Histograms of the duration of the epileptiform discharges showed a power-law (n = 2) or bimodal (n = 2) distribution of burst durations, with power-law behavior for the inter-burst intervals, with slopes between -1 and -2, suggesting the presence of type III or on-off intermittency. Analysis of the epileptiform bursts indicates the presence of different, isolated brain states (bursts), that often recur in time (45-99% recurrence rate). Figure 1 illustrates the morphology of various pairs of recurrent epileptiform discharges. Left (A-C): data from patient G and Right (D-F) from patient W. The initial portions of the pairs of traces shown are similar, with divergence of later waveforms (red and blue, respectively). These tracings illustrate qualitatively the sensitive dependence on the initial conditions, - one of the characteristics of nonlinear systems. Conclusions: The presence of intermittency and the power-law distribution of the inter-burst intervals during non-convulsive status epilepticus are indications of nonlinear dynamics. Bursts of epileptiform discharges during a non-convulsive SE are manifestations of intermittency with multiple, recurring, brain states. [figure 1] Abstract: .sup.1 Dominique Parain , and .sup.2 Renata Dias de Carvalho Van Woensel (Service de Neurophysiologie, Hopital Charles-Nicolle, Rouen, France ; and Cyberonics, Europe) Rationale: Epilepsy is very common in tuberous sclerosis complex (TSC) and occurs in 80 to 90% of affected individuals during their lifetime. Although not completely understood, the incidence of epilepsy is thought to relate to the neuropathologic features of the disorder, including cortical tubers and other dysgenetic features. Methods: Analysis of data from Cyberonics' International Patient Registry showed 90 patients (53 male, 37 female) with TSC and medically refractory epilepsy who were treated with adjunctive VNS Therapy. Patients had a mean age of 16.7 ([+ or -]12.00) years. Six (6.7%) patients were institutionalized. Five (5.5%) patients had a prior callosotomy, and 8 (8.9%) patients had a history of lobectomy. Mean follow-up was 32.7 [+ or -] 15.37 months (median: 34.05 months). Epilepsy was localized in 47 (52.2%) cases; 20 patients (22.2%) had generalized seizures, and 14 (15.5%) had Lennox-Gastaut syndrome. Other types of seizures occurred in 22 (24.4%) patients. Data on seizure reductions, magnet use and quality of life (QoL) were collected. Results: There was a median seizure reduction of 59.4% for the entire patient population at the last follow-up visit (p-value < 0.0001; Wilcoxon's paired signed rank test). Seizure reductions [greater than or equal to] 50% occurred in 56.7% of the patients. Magnet use was reported in 67.8% of the cases (n = 61). According to 23 patients (25.5%), the magnet was always or most of the time effective in aborting or reducing the intensity and/or duration of their seizures. Another 26.7% of the patients (n = 24) reported that the magnet had some effectiveness. More than 50% of the patients reported improvements in alertness and post-ictal state. Conclusions: Adjunctive VNS Therapy was effective for the treatment of medically refractory seizures associated with TSC, with a significant median seizure reduction of 59.4%. Seizure reductions [greater than or equal to] 50% were reported in 56.7% of the patients. QoL improvements were most pronounced for alertness and post-ictal state. Abstract: .sup.1 Imran I. Ali , .sup.1 Nabeel A. Herial , .sup.2 Terrance Horrigan , .sup.1 Leah Kellough , and .sup.1 Gretchen E. Tietjen (Neurology, Medical University of Ohio, Toledo, OH ; and Obstetrics and Gynecology, Medical University of Ohio, Toledo, OH) Rationale: There is considerable evidence that older antiepileptic drugs (AEDs) such as phenobarbital and phenytoin adversely affect bone health. However, data regarding newer AEDs such as levetiracetam is lacking. We, therefore, studied the effects of levetirecetam on bone density, vitamin D and osteocalcin levels. Methods: After IRB approval patients on levetiracetam monotherapy for greater than six months were enrolled. Sixteen patients (eight men and eight women) agreed to participate in the study. A detailed questionnaire was used to review medical history. Osteocalcin were measured by chemiluminescent immunoassay and and 1, 25 hydroxy vitamin D levels by a radioimmunoassay. Dual energy X-ray absorptiometry (DEXA) was performed using Hologic System. Results: The mean age of these patients was 37 years (range 20-66 years). Mean duration of therapy with levetirecetam as monotherapy was 27 months (range 6-60 months). The mean serum 1, 25 hydroxy-vitamin D level was 42.5 pg/mL (range 17-100 pg/mL). None of these patients had an abnormally low vitamin D level noted. Mean serum osteocalcin level was 6.4 ng/mL (range 1.8-14.1). Four of the sixteen (25%) of the patients had elevated osteocalcin levels.Two of these had mildly abnormal bone density measurements when compared to normal young adults matched for gender (t score). All four had previous exposure to other AEDs including phenytoin, sodium valproate, carbamazepine and phenobarbital. All sixteen patients underwent DEXA scan. None of the patients had evidence of osteopenia or osteoporosis when compared to age and sex matched controls (z score > -2.0). Borderline abnormal t scores were noted in two patients at lumbar spine level who also had elevated osteocalcin levels. However, these patients had been treated with multiple AEDs prior to being on keppra monotherapy. A significant positive linear correlation was noted between serum 1, 25 hydroxy vitamin D levels and bone mineral density. (r = 0.54, p < 0.05). Conclusions: Levetiracetam does not appear to be associated with a significant decrease in vitamin D levels or bone mineral density. Elevation of osteocalcin levels, implying increased bone turnover was noted in four patients but this finding alone is of unclear clinical significance and requires further study. Slight reduction in BMD in two patients can potentially be explained by previous, prolonged exposure to other AEDs known to cause bone loss. This confounding variable needs to be considered in all cross-sectional studies such as ours in assessing bone mineral density with newer AEDs. Controlled prospective studies in patients with new onset seizures may be a better method of studying effects of newer AEDs on bone mineral density. (Supported by UCB Pharmaceuticals.) Abstract: .sup.1 Carl W. Bazil , .sup.1 Timothy DaGiau , and .sup.1 Elisa A. Salerni (Neurology, Columbia University, New York, NY) Rationale: Sleep disturbances may be more common in epilepsy patients, potentially resulting in decreased memory, attention, and alertness as well as increased seizures. However, sleep problems are frequently overlooked, and the actual prevalence of sleep problems (in general and in epilepsy) is not well understood. This prospective study looked at sleep problems in an epilepsy referral center using a standardized questionairre, and compared scores with subjects accompanying epilepsy patients who did not have epilepsy. Methods: All patients with epilepsy (PWE) seen as inpatients or outpatients at an epilepsy referral center were invited to participate. Questionnaires included basic demographic information, the MOS sleep scale, and questions about sleep hygiene. Companions without epilepsy were also surveyed as controls. PWE had seizure diagnosis verified by treating physicians. The study was approved by the institutional review board. Results: 187 surveys were returned. Average age was 45 years for PWE, 47 years for controls. 50% of PWE were women compared with 62% of controls. Nearly all PWE were being treated with anticonvulsant drugs. PWE scored significantly higher on the sleep adequacy subscale of the MOS sleep scale, reflecting worse sleep (53 [+ or -] 2 vs. 43 [+ or -] 3). There were no significant differences in subscales measuring sleep somnolence, optimal sleep, awakening short of breath, or overall sleep problems index although subjects with epilepsy scored worse in all of these. Subjects with epilepsy more commonly reported signs of insomnia (43% vs 34%), restless legs syndrome/periodic limb movements (37% vs 25%), and sleep apnea (43% vs 34%) than control subjects, but controls more commonly reports excessive daytime sleepiness (46% for control vs 37% for PWE). None of these differences were statistically significant. PWE slept significantly longer than controls during the week (6.9 [+ or -] 0.1 vs. 6.2 [+ or -] .22 hrs) but not on weekends. Subjects in both groups had signs of poor sleep hygiene, particularly including use of alcohol shortly before bedtime. Conclusions: In this relatively small sample, sleep adequacy was more impaired in epilepsy patients than in their companions without epilepsy. Other measures suggested increased sleep problems in PWE but did not reach statistical significance. It may be that, with a larger sample, differences between these two groups can be clarified. More objective evaluations, including evaluation by sleep specialists and polysomnography would be needed to idenify the actual frequency of specific sleep disorders. PWE slept longer on average; it was not clear whether this was due to medication, employment status, or other factors. Both groups had signs of poor sleep hygiene, suggesting the need for education regarding sleep in PWE and in the general population. Abstract: .sup.1 Annette M. Chihorek , .sup.1 Bassel W. Abou-Khalil , and .sup.1 Beth A. Malow (Neurology, Vanderbilt University, Nashville, TN) Rationale: Although the incidence of new-onset seizures is much higher in older adults than in any other age group, the etiology remains unknown in about 45% (Epilepsia, 34(3): 453-68, 1993). One possible etiological factor is untreated obstructive sleep apnea (OSA), which can lead to sleep fragmentation and chronic sleep deprivation, potentially facilitating seizures in susceptible individuals (Neurology, 48(5):1389-94, 1997). Methods: To determine whether OSA is more common in individuals with later-onset seizures or worsening of seizure control in later life, we performed a cross-sectional study of adults 50 years or older presenting to a tertiary center epilepsy clinic. Patients with provoked seizures, associated neurological disease, or unreliable seizure histories were excluded. An epilepsy specialist blinded to the study results classified each subject into one of two groups. Group 1: Seizure-onset before age 50 with stable or improved seizure frequency at or after age 50 Group 2: Seizure-onset before age 50 with increased seizure frequency at or after age 50 OR Seizure-onset at age 50 or older Each subject underwent one-night polysomnography. The studies were reviewed by a sleep specialist blinded to the epilepsy histories. The number of apneas and hypopneas per hour of sleep (apnea-hypopnea index; AHI) was calculated. Results: Of 290 patients screened, 135 met inclusion criteria, and 21 consented to participate. Epilepsy syndromes included partial epilepsy of temporal lobe (15) or frontal lobe (3) origin, idiopathic generalized (2), and unspecified (1). The two groups were similar in age, BMI, number of AEDs currently used, and frequency of nocturnal seizures. Group 2 had more men and a higher neck circumference. Group 2 had significantly higher AHIs, as well as higher scores on Epworth Sleepiness Scale (ESS) and Sleep Apnea section of the Sleep Disorders Questionnaire (SA-SDQ). [figure 1] Conclusions: The AHI was higher in the group with either new seizures or increased seizures, suggesting that OSA may play a role in facilitating seizures in this older population. (Supported by Epilepsy Foundation of America Clinical Research and Training Fellowship Program through the Roger F. and Edna F. Evans Fund.) Abstract: .sup.1 Alejandro de Marinis , .sup.2 Francisco Bustamante , .sup.1 Claudia Asmad , .sup.1 Loreto Olate , .sup.2 Francisca Grob , .sup.2 Macarena Parker , .sup.3 Ines Delgado , .sup.1 Vilma Ojeda , .sup.1 Veronica Fernandez , .sup.1 Monica Gonzalez , .sup.1 Ada Chicharro , .sup.1 Jorge Lasso , .sup.1 Claudia Passig , .sup.1 Manuel Lavados , and .sup.4 Andres M. Kanner (Epilepsy Center, Chilean League Against Epilepsy, Santiago, Chile ; Faculty of Medicine, University of the Andes, Santiago, Chile ; Faculty of Medicine, Clinica Alemana - Development University, Santiago, Chile ; and Department of Neurology, Rush Medical Center, Chicago, IL) Rationale: Depression is the most common psychiatric comorbidity in epilepsy, with major impact in quality of life. Recent publications raise concerns of underrecognition of depression in patients with epilepsy. This issues have not been systematically assesed in chilean patients with epilepsy. We evaluate the prevalence and underrecognition of depression in patients with epilepsy attending the epilepsy clinic of the Chilean League Against Epilepsy Methods: We evaluated 200 consecutive adult patients with epilepsy. Depression was diagnosed using a validated CIDI diagnostic interview. The clinical recognition of depression was determined by inspection of medical records. All patients gave their written informed consent. Results: A diagnosis of Major Depression was established in 44/200 patients (22%). In 33/44 (75%) depression was unrecognized. 21/44 (47.7%) have been seizure free for the last year. Conclusions: Depression is highly prevalent and underrecognized in chilean patients with epilepsy. Validated, simple and cost effective screening instruments for the detection of depression would be a major contribution to improve quality of life of patients with epilepsy. Abstract: .sup.1 A. James Fessler , .sup.1 Robert A. Gross , and .sup.1 John T. Langfitt (Neurology, University of Rochester Medical Center, Rochester, NY) Rationale: The SCL-90-R is a self-report inventory that assesses the severity of a number of different types of psychiatric symptoms. We have previously shown that patients with psychogenic, non-epileptic attacks (PNEA) have higher scores on somatization and anxiety indices as measured by this inventory compared to patients with epilepsy. There were no differences between these two groups on measures of paranoia (PA) and psychoticism (PS). These differences may have been underestimated because of the degree of symptom reporting. Patients with epilepsy may have higher scores on these indices, as peri-ictal symptoms may mimic psychiatric disease in select patients. Methods: We retrospectively identified consecutive, adult patients (>17 years of age) who were evaluated at the Strong Epilepsy Center. Two hundred and ninety-two patients completed the SCL-90-R prior to an initial clinic visit between 2001 and 2004 and underwent subsequent video-EEG monitoring. All patients were interviewed by a psychiatrist or psychologist during the admission. One hundred and sixty-six patients were female. One hundred and seventy-six (63.7%) patients were diagnosed with epilepsy and 94 (32.2%) with PNEA. Patients with other non-epileptic events or both PNEA and seizures were excluded. To control for patients who reported multiple somatic symptoms, patients with high positive symptoms total (PST) scores (>65) were excluded from this analysis. Results: One hundred and twenty-four patients with epilepsy and 56 patients with PNEA were identified. Of the psychogenic patients, 6 (10.7%) had an elevated ([greater than or equal to]65) PS score, 3 (5.4%) had an elevated ([greater than or equal to]65) PA score and in 3 (5.4%) patients, both were elevated. Nineteen (15.3%) epilepsy patients had an elevated PS score and 10 (8.1%) had an elevated PA score. In 4 (3.2%) patients, both indices were elevated. All of the patients with an elevated PS score had partial onset seizures with 14/19 (73.7%) having temporal lobe epilepsy, 1 (5.3%) with frontal lobe epilepsy and 4 (21.1%) with hemispheric onset seizures without further localization. Three patients had ictal hallucinations, 2 had fear as a prominent component of the aura and 2 had post-ictal psychiatric symptoms including hallucinations and aggression. Conclusions: No statistically significant differences between patients with epilepsy and PNEA were seen on psychoticism and paranoia indices when patients with high symptoms reporting were excluded. Care must be taken in patients with epilepsy when administering psychiatric batteries, as associated ictal and peri-ictal symptoms may be mistaken for underlying psychiatric disease. Abstract: 1,2 James D. Geyer , .sup.3 Paul R. Carney , and .sup.1 Stephenie C. Dillard (The Sleep Program, Southern Sleep Specialists, Tuscaloosa, AL ; Department of Neurology and Sleep Medicine, The University of Alabama College of Community Health Sciences, Tuscaloosa, AL ; and Department of Pediatrics, Neurology, Neuroscience, and Bioengineering, The University of Florida, Gainesville, FL) Rationale: Restless legs syndrome (RLS) is a common neurological movement disorder occurring in 10-15% of the general population. The prevalence of RLS increases with advancing age and is more common in several neurological disorders such as Parkinson's disease. Epilepsy is also a common neurological disorder with significant associated morbidity and impact on quality of life. Any disorder which adversely affects quality of life may further compound the problems associated with epilepsy. We report the severity and frequency of RLS in patients with localization-related epilepsy. Methods: All epilepsy patients seen in the outpatient clinic over a one month period were screened for movement disorders. Patients with severe language impairment, cognitive impairment and stroke were excluded from this study population. Fifty-one consecutive patients with localization-related (complex partial) epilepsy who met inclusion criteria were seen in the outpatient clinic during this period. Each patient was evaluated with the International Restless Legs Study Group (IRSLSG) questionnaire, NIH RLS diagnostic criteria and comprehensive sleep screening. Restless legs syndrome quality of life questionnaires were also administered to a subset of those patients. Results: Restless legs occurred in 14 of 51 (28%) of the localization-related epilepsy patients. Thirteen of the 14 patients reported at least moderate impact on quality of life parameters related to RLS. An insufficient number of patients with primary generalized epilepsy were screened for adequate statistical analysis during the study period. A number of anti-epileptic agents were used in this study population. Based on patient reported histories, there appears to be a trend toward improvement of RLS symptoms following treatment with carbamazepine, gabapentin, pregabilin, and lamotrigine. Some patients reported worsening of symptoms with valproate and phenytoin. Since a number of anti-epileptic drugs were used in this patient population, there were insufficient numbers of patients treated with each medication for reliable statistical analysis. Conclusions: We identified RLS in patients with epilepsy at a frequency that exceeds the general population. Furthermore, these patients were more likely to have moderate to severe RLS than would typically be seen in patients of similar age, and most reported at least moderate impact on quality of life. It is likely that there are differences in the interaction between RLS and various anti-epileptic medications. More detailed analysis of the effect of medication selection as well as quality of life parameters related to RLS is needed. Abstract: 1,2 Sheryl R. Haut , 1,2 Jonathan H. Masur , .sup.3 Mindy Katz , 1,2 Edith Miller , and 2,3 Richard B. Lipton (Comprehensive Epilepsy Management Center, Montefiore Medical Center, Bronx, NY ; Department of Neurology, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY ; and Department of Epidemiology and Population Health, Albert Einstien College of Medicine, Bronx, NY) Rationale: The age specific incidence of epilepsy appears to be highest during old age. Comorbidities of seizures in the elderly are expected to differ from those in younger patients. To explore these differences, we examined the prevalence of cognitive loss, depression and sleep disorders in elderly patients with epilepsy. Methods: Subjects were recruited from the Epilepsy Management Center at Montefiore Medical Center, and the Einstein Aging Study, an NIH-funded longitudinal study of elderly individuals. Cases were > = 65 years of age with a diagnosis of epilepsy confirmed by an epileptologist, and were paired with age and gender matched controls without a history of seizures. Testing administered to all subjects included the Blessed Information Memory and Concentration (BIMC) test; Prime-MD Depression and Anxiety Scale; and Medical Outcomes Study Sleep Scale. Results: Data were analyzed for 64 subjects. Cases with epilepsy had significantly higher mean BIMC scores (6.3) than controls (1.2) (p < 0.0001). Mean Prime-MD Depression scores were significantly worse for cases (4.2) than controls (.79) (p = 0.006) with six of 32 cases (18%) meeting screening criteria for depression, while anxiety scores were not significantly different. Cases demonstrated significantly poorer sleep scores than controls in the categories of somnolence (p = 0.009) and shortness of breath/headache (p = 0.021) scales. Conclusions: In our population, elderly individuals with epilepsy had reduced cognitive function, a higher prevalence of depression, and poor sleep health when compared with their age mates without epilepsy. This cross-sectional study could not determine the temporal course of epilepsy and co-morbidities, and further investigation will explore these associations prospectively. (Supported by: This study was supported in part by NIH grant K23 NS02192 (SRH), NIH PO1 AG03949 (RBL), and a research grant from Pfizer.) Abstract: 1,2 Hrvoje Hecimovic , .sup.1 Juan M. Santos , .sup.4 Yvette I. Sheline , .sup.5 Mark A. Mintun , .sup.6 Joseph L. Price , .sup.3 Jewell Carter , .sup.3 Victoria Vahle , .sup.4 Abraham Z. Snyder , and .sup.1 Frank Gilliam (The Neurological Institute, Columbia University, New York, NY ; Zagreb Epilepsy Center, University Hospital, Zagreb, Croatia ; Department of Neurology, Washington University, St. Louis, MO ; Department of Psychiatry, Washington University, St. Louis, MO ; Department of Radiology, Washington University, St. Louis, MO ; and Department of Neuroanatomy, Washington University, St. Louis, MO) Rationale: Neurologically normal patients with primary depression have hippocampal volumes smaller than controls. Temporal lobe epilepsy (TLE) patients have a wider range of abnormalities in the mesial temporal structures, which allows investigation of the association of hippocampal volume with severity of depressive symptoms. Based on the hyperexcitable hippocampal efferents to limbic structures, we hypothesized that TLE patients will express increased symptoms of depression. In addition, we speculated that severe hippocampal atrophy will not support the hyperexcitable network required to maintain clinical depression. Methods: We evaluated 28 consecutive adults with TLE using clinical variables, long-term video/EEG monitoring, high resolution MRI and tests for depression, including the Beck Depression Inventory (BDI). On the MRI we manually delineated regions of interest in the mesial temporal (hippocampus, amygdala) and prefrontal (gyrus rectus, pregenual cortex, subgenual and subcallosal cortex) structures, based on prior human and animal studies of mood dysfunction. Distribution of the data was tested using the Kolmogorov-Smirnov test. Student's t-test was used to determine group differences, with a significance level set at p < 0.05. Data were analyzed using the SPSS statistical software. Results: Based on a total BDI score greater than 15, 42.8% of the group were experiencing significant symptoms of depression. None of the subjects in the quartile with the smallest left hippocampal volume had a BDI score greater than 15, whereas 57.1% of the subjects in the upper three quartiles had increased symptoms of depression, p < 0.008. No other brain structure volume had an association with depressive symptoms. Conclusions: Patients with hippocampal volume loss that is consistent with severe hippocampal sclerosis had no evidence of significant symptoms of depression. This finding suggests that there is a critical number of hyperexcitable hippocampal neurons required to maintain depression in temporal lobe epilepsy. [figure 1] (Supported by NIH grants NS40808 and NS047551 and Epilepsy Foundation fellowship grant.) Abstract: .sup.1 Jennifer L. Hopp , .sup.1 Helga Matausch , .sup.1 Jingkun Zhu , and .sup.1 Allan Krumholz (Neurology, University of Maryland Medical Center, Baltimore, MD) Rationale: Mood and anxiety disorders are common in patients with epilepsy and many hypotheses have been proposed to understand the relationship between emotional dysfunction and seizures. Despite the recognition of this comorbidity, psychiatric disease remains under-recognized in neurology patients. This study was designed to identify the prevalence of depression and anxiety, as well as the co-existence of these symptoms, in patients with epilepsy and examine their impact and relative contribution on health related quality of life (QOL). Methods: 110 consecutive patients in the University of Maryland Medical Center Epilepsy Outpatient Center and Inpatient Epilepsy Monitoring Unit (EMU) completed Beck Depression Inventory-II.sup.[R] (BDI-II.sup.[R]), Beck Anxiety Inventory.sup.[R] (BAI.sup.[R]) and Quality of Life in Epilepsy-31 (QOLIE-31) measures between December 1, 2005 and May 25, 2006. Patients who completed these questionnaires in the inpatient EMU did so on the day of admission before their medications were changed and prior to any seizure activity. Scores were calculated for each inventory and means were compared using standard t-test calculations. Results: 109 patients completed the BAI and anxiety was noted to be severe in 9% of patients, moderate in 9%, mild in 26%, and minimal in 56% of patients. 98 patients completed the BDI and symptoms suggestive of depression were severe in 7% of patients, moderate in 12%, mild in 16% and minimal in 65%. 88 patients had complete BAI, BDI, and QOLIE-31 data for analysis. When anxiety and depression scores were independently compared with quality of life scores in 88 patients, it was found that lower levels of anxiety and depression were correlated with higher quality of life indices. The most significant reduction in quality of life was seen when comparing BAI and BDI scores between patient groups with minimal versus mild levels of depression and anxiety (table1) (p < 0.0001, 95% confidence interval). all values represent mean QOLIE-31 scores for each category of symptom severity Conclusions: These findings are supportive of prior data which have noted increased prevalence of anxiety and depression in patients with epilepsy. The presence of even mild symptoms of anxiety and depression have a significant impact on quality of life. When the levels of anxiety and depression were simultaneously compared to QOLIE-31 scores, there was a more significant impact on QOL by depression levels than anxiety. These data also suggest the possibility that although more severe symptoms of depression and anxiety are associated with reduced QOL, that there may be a low threshold which is important with regard to the effect of depression and anxiety on quality of life in patients with epilepsy. Abstract: .sup.1 Ruben Kuzniecky , .sup.1 Charles K. Vorkas , .sup.1 Kenneth Alper , .sup.1 Chad Carlson , .sup.1 William Barr , and .sup.1 Orrin Devinsky (NYU Comprehensive Epilepsy Center, NYU School of Medicine, New York, NY) Rationale: Postictal psychosis (PIP) is an important behavioral complication of epilepsy and its early recognition is essential for appropriate management. Yet, the risk factors for PIP remain controversial. The majority of studies to date have been relatively small and have largely focused on patients with temporal lobe epilepsy (TLE). We sought to verify previous findings that correlate PIP with bilateral epileptiform interictal/ictal activity, clustering of seizures, and mesial temporal sclerosis (MTS). Methods: We retrospectively reviewed a series of 140 patients with complex partial epilepsy undergoing video-EEG (VEEG) monitoring as candidates for epilepsy surgery. These patients were evaluated by structured neurologic and psychiatric interviews by a team of epileptologists and a single neuropsychiatrist (KA). Each patient had at least one MRI prior to surgery. No patient had psychotic symptoms at baseline. 55 of these patients were diagnosed with PIP using the criteria presented in Devinsky et al. [1], which were adapted from the definition provided by Logsdail and Toone [2]. Results: Age, sex, age at onset of epilepsy, and history of febrile seizures were similar in the PIP and control groups. Patients with PIP experienced more generalized slowing (19.6% v 7.0% p 2 seizures in 24 hours) during VEEG monitoring (45.0% v 28.2% p < 0.045). Patients with PIP had more frequent clinical secondary generalization (64.7% v 43.5% p < 0.017), bilateral interictal epileptiform discharges (52.9% v 32.9% p < 0.021), and non-localized seizures (54.9% v 23.8% p < 0.001) during monitoring. MRI lesion lateralization was similar in both groups, though patients in the control group had more bilateral focal abnormalities than in the PIP group (12.3% v 4.4 p = ns). More patients in the PIP group had a nonfocal abnormality (11.8% v 2.7% p = ns). Imaging diagnosis of MTS was similar in the PIP and control groups (58.3% v 55.6% p = ns). Conclusions: This retrospective study of PIP is the largest to date. PIP was more frequent in patients with electrophysiologic evidence of bilateral cortical hyperexcitability. No significant correlation was found between PIP and structural abnormalities on MRI. Although the correlation between secondary generalization and seizure clusters suggest an association with "epilepsy severity," the presence of independent bilateral seizure onsets was not significant and overall baseline seizure frequency could not be formally evaluated in the population. These findings suggest PIP correlates more with neurophysiologic function than structural anatomy. Devinsky O, Abramson H, Alper K, et al: Postictal psychosis; a case control series of 20 patients and 150 controls. Epilepsy Research 1995; 20(3): 247-253. Logsdail SJ, Toone BK: Postictal psychoses: a clinical and phenomenological description. Br J Psychiatry 1988; 152:246-252. Abstract: 1,3 Eun-Jeong Lee , .sup.3 Brian Bell , .sup.2 Christian Dow , .sup.3 Jana E. Jones , .sup.3 Paul A. Rutecki , .sup.2 Michael Seidenberg , and .sup.3 Bruce P. Hermann (Rehabilitation Psychology & Special Education, University of Wisconsin-Madison, Madison, WI ; Psychology, Rosalind Franklin University of Medicine and Science, North Chicago, IL ; and Neurology, University of Wisconsin-Madison, Madison, WI) Rationale: Depression is a major comorbidity of epilepsy that contributes to unemployment, underemployment, medical expenses, and reduced self-reported quality of life (Bishop et al., 2002). The etiology of depression remains controversial and rarely have causal modeling techniques been used in this literature. The purpose of this study is to test the applicability of Lewinsohn et al.'s (1985) integrative model of depression to epilepsy. This model views depression as the end result of environmentally initiated changes in behavior, affect, and cognition as the primary triggers to episodes of depression. Methods: 100 individuals with temporal lobe epilepsy underwent structured interviews to obtain sociodemographic and epilepsy related information, and completed self-report measures of related handicap (Subjective Handicap of Epilepsy scale), Quality of Life (Quality of Life in Epilepsy-89), and depression (Beck Depression Inventory). Path analysis was used to (a) evaluate the general compatibility (i.e., goodness of fit) of the model with the data, and (b) examine relationships between epilepsy related variables, disruptions in daily activities, self-perception, stress, social support, employment status, and depression among individuals with epilepsy. Results: In path analysis, acceptable fit is indicated by a nonsignificant chi-square value (i.e., p > .05; Bollen, 1989), and by comparative fit index (CFI) and normed fit index (NFI) values great than .90 (Tabachnik & Fidell, 2001). The model tested revealed a non-significant chi-square, [chi].sup.2(16, n = 100) = 19.398, p > .05, suggesting a good-fit. CFI and NFI were .98 and .90, respectively, both of which are greater than the cutoff of .90. The root mean square error of approximation (REMSA) was .46, which represents "a close fit" recommended by Quintana and Maxwell (1999). Conclusions: Overall, the findings provide preliminary support for the applicability of the integrative model of depression proposed by Lewinson et al (1985) to epilepsy. The three main fit indices indicated that the model fits the data well, suggesting that epilepsy related negative events, self-perception of epilepsy, and stress play an important role in triggering depression. In contrast, employment and especially social support are found to provide buffering effects for depression. The strength of the model is the utilization of a more adequate and comprehensive representation of vulnerabilities as well as immunities that have been shown in the literature to be relevant to the process of depression in the general and epilepsy literatures. (Supported by NIH NS R01-44351, F32 MH649882, MO1 RR03186 (GCRC), and Epilepsy Foundation J.W.G. Memorial Fellowship.) Abstract: .sup.1 Lee A. Lineberry , .sup.2 Braxton B. Wannamaker , .sup.1 Pamela L. Ferguson , .sup.1 Anbesaw W. Selassie , and .sup.3 Gigi M. Smith (Biostatistics, Bioinformatics and Epidemiology, Medical Universty of South Carolina, Charleston, SC ; Clinical Neuroscience, Medical University of South Carolina, Charleston, SC ; and Pediatric Neurology, Medical University of South Carolina, Charleston, SC) Rationale: The many studies of sudden unexplained death in epilepsy (SUDEP) have yet to identify consistent risk factors for its occurrence. Further, no single mechanism has been elucidated. Mortality associated with SUDEP is population dependent. It is difficult to analyze SUDEP in that it occurs relatively infrequently. In this study, SUDEP cases were identified in a clinical practice of 30 years. Generally, all patients are known personally to the investigator (BBW) and the clinical and therapeutic information was similar. Methods: 3096 patient records were accumulated in a single practice between 1974 and 2004. Abstractions were performed on 711 cases including 26 SUDEP cases and 87 other death cases; the remainder are alive. Specific demographic, clinical, seizure, treatment and laboratory information was obtained using a developed abstracted tool. SUDEP cases were matched on gender and 5-year age intervals with 3 controls that have epilepsy and are alive. McNemar tests were performed using SAS. Results: The rate of SUDEP was 4.2 per 1000 person-years. A history of generalized tonic clonic (GTC) seizures was seen more often in the SUDEP group (RR = 5.0, 95% CI 1.90, 13.13). There was a trend towards persons who died of SUDEP being more likely to have ever experienced status epilepticus (p = 0.08). Also, persons with SUDEP were more likely to have more than 1 type of seizure (p < 0.001), be taking more than 2 antiepileptic drugs (AEDs) at last visit (p < 0.001), to have ever taken 3+ AEDs (p < 0.001), and to have had a medication change in the 3 months prior to the last visit (p < 0.001). There was also a trend towards childhood onset epilepsy in the SUDEP cohort (p = 0.07). No difference was observed in having a primary generalized epilepsy. Conclusions: The rate of SUDEP is consistent with other published findings from tertiary referral epilepsy clinics. This study included persons who attended a tertiary referral practice, and therefore likely had more severe epilepsy than general neurological populations. Results have also confirmed that cases of SUDEP are more likely to be taking more AEDs currently and in the past, have more than 1 type of seizure, and to have had a medication change within three months of death. It has been documented that type of seizure, mainly GTC, is associated with SUDEP; however, less documented is the type of epilepsy. Specifically, primary generalized epilepsy as JME was found equally in both SUDEP and matched cohorts. Trends towards ever having status epilepticus and childhood onset epilepsy should be explored further when more data is available. (Supported by Epilepsy Services & Research, Inc.) Abstract: 1,3 Amelie Lothe , .sup.1 Adrien Didelot , .sup.2 Nicolas Costes , .sup.3 Mohamed Saoud , and 1,2 Philippe Ryvlin (Unite 280, INSERM, Lyon, France ; CERMEP, Bron, France ; and EA3092, UCBL, Lyon, France) Rationale: Depression and epilepsy are comorbid diseases that seem likely to involve common pathophysiological pathways, including serotonin (5-HT) and 5-HT.sub.1A receptors. Two studies have investigated 5-HT.sub.1A receptors in patients with temporal lobe epilepsy (TLE) as a function of depressive symptoms and have shown a negative correlation between the binding potential (BP) of [11.sup.C]-WAY100635 (WAY) or [18F]-FCWAY and varying depression scales, though in different brain regions. The aim of our work was to study the relationships between 5-HT.sub.1A receptor density and the Beck Depression Inventory (BDI) scale in a homogeneous group of patients suffering from TLE with hippocampal sclerosis (HS), using a PET ligand, [18.sup.F]MPPF, which affinity differs from WAY and allows displacement by endogenous serotonin. Methods: 21 patients suffering from TLE with HS underwent a [18.sup.F]MPPF-PET. Depressive symptoms were evaluated on the day of the PET study using the BDI scale. A simplified model was used to generate parametric images of 5-HT.sub.1A receptors binding potential values. Images from patients with a left HS were flipped so that HS were on the same side (right) for all patients. To investigate the association between BP values and BDI scores, we carried out SPM analyses using an ANCOVA design, considering the BDI scores as a covariable of interest, with masks that either included all limbic and paralimbic regions, or the brainstem. In addition, we placed regions of interest (ROIs) over the same brain regions. Results: SPM analyses disclosed a significant positive association between the MPPF BP values and the BDI scores in the insula contralateral to the seizure focus as well as in the raphe nuclei (p < 0.001 and p < 0.01 respectively, at the cluster level corrected for multiple comparisons). These results were confirmed by ROI analysis. Conclusions: In contrast with previous WAY-PET studies, we have shown a positive, rather than negative, association between 5-HT.sub.1A receptors BP values and the BDI scores, in regions known to be involved in depression (insula, raphe) that also differ from those reported in the previous studies (anterior cingulate, hippocampus). These differences could be partly explained by the lower affinity of MPPF as compared to that of WAY, and reflect decreased level of endogenous 5-HT in the insula and raphe in the more depressed patients. (Supported by Programme Hospitalier de Recherche Clinique.) Abstract: .sup.1 Bernardo Moreira , .sup.1 Silvia Vincentiis , .sup.1 Lia Fiore , .sup.1 Renato Marchetti , and .sup.1 Kette Valente (Clinical Neurophysiology - Psychiatry, USP, Sao Paulo, SP, Brazil) Rationale: Analysis of the semiology of psychogenic non-epileptic seizures (PNES) is challenging. Use of the same terminology as that used for epileptic phenomema may be inappropriate for PNES. Groppel et al. suggested a classification based on the analysis of clusters of symptoms and signs. In this study we applied this classification to the analysis in order to analyze its application and possible limitations Methods: We studied the semiology of 27 patients with PNES documented by VEEG. We observed and characterized motor phenomena, classifying them as events with positive motor phenomena (clonic, hypermotor, tremors, pelvic thrusting, lateralized movements, versive, erratic and anomalous movements simulating automatisms, hypertonic posture, retropulsion with or without opistotonus) or as events with negative motor phenomena (loss of tonus, motor arrest, staring, etc). We also evaluated responsiveness, hyperventilation, sound emission and aura. Postictal phenomena and event duration were likewise evaluated. Results: Eight patients were classified as Cluster I (hypermotor and clonic movements of extremities, pelvic thrusting, head movements, tonic head posture - motor PNES); 4, as Cluster II (limb tremor - minor motor PNES) and 4 as Cluster III ("atonic PNES"). Six patients had more than one event, with distinct semiology, being classified in two groups. Fifteen NES could not be classified according to these criteria. Among those presenting positive motor phenomena, 4 had Cluster I phenomena associated to tremors (Cluster II), and 3 had Cluster I phenomena associated to automatisms, 2 patients had only automatic movements and 1, had subtle lateralization movements. Not all patients with negative motor phenomena had atonia; 2 had speech arrest, 1 had irresponsiveness, 1 had visual hallucinations, and 2 had irresponsiveness with eye flutter. Ten patients had hyperventilation, 12 with whimpering, and less often, wailing. Loss of responsiveness was observed in 16 patients from this group. Four patients reported aura preceding the event. Postictal confusion with agitation was observed in 6 patients. Twelve patients presented events with prolonged duration. Conclusions: Our study suggested that although the attempt of a categorical classification of PNES, such as that carried out in epileptic seizures and syndromes, may seem quite interesting, it seems to be of little practical use because of the heterogeneity of PNES. PNES do not always present as stereotyped seizures in the same patient and frequently undergoes changes in semiology during its course. Moreover, in the literature there is a tendency to overestimate motor phenomena, when other non-motor features may be just as important for diagnosis of PNES and remain undervalued. Groppel G, et al. Cluster analysis of clinical seizure semiology of psychogenic nonepileptic seizures. Epilepsia 2000;41(5):610-4. Abstract: .sup.1 Laura M. Moreno , .sup.1 Gina Mapes Jetter , .sup.1 Susan Rogers , .sup.1 Elizabeth Clark , .sup.1 Jan M. Bruder , and .sup.1 Jose E. Cavazos (Medicine (Neurology) and Pharmacology, University of Texas Health Science Center, San Antonio, TX) Rationale: Recent studies have shown that some enzyme-inducing antiepileptic drugs (EIAEDs) are associated with osteoporosis and an excess number of fractures. 3A4 is the main CYP450 isoenzyme involved in the degradation of active Vitamin D (Vit D) to its inactive form. It is thought that the primary mechanism of this problem is due to an AED induction of the metabolism of Vit D through the CYP450 system. The potential development of osteoporosis might be prevented by the use of Vit D and calcium (Ca) supplements. It is uncertain as to how to treat the Vit D deficiency in epilepsy patients, who often have been taking AEDs for many years. Although maintenance therapy guidelines for those at high risk for developing osteoporosis exist, a more rapid protocol might be useful to normalize Vit D levels. The objective of this study was to determine if replenishment of 25-hydroxy Vitamin D.sub.3 (25-OH- D.sub.3) levels by a 12-week course of weekly high-dose Ergocalciferol, vitamin D.sub.2, will normalize the 25-OH- D.sub.3 levels. Methods: The protocol took male patients on AEDs and looked at Vitamin D3, PTH and calcium levels. If the results showed the patient to be deficient in Vitamin D3 ( Results: Analysis showed a difference in Vit D, iPTH, and Ca levels before and after treatment with Ergo. However, only the change in Vit D levels was significant. We further divided into subgroups of EIAEDs, NEIAEDs, and valproic acid (VPA) examining Vit D only. Conclusions: Patients that took a 12-week course of Ergo had a significant increase in 25-OH- D.sub.3 levels. Although iPTH trended down and Ca levels increased after Ergo treatment, there was no significant effect seen between the before and after treatment group. Long term follow-up of these labs in this study group is needed to correlate whether the increase in Vit D will decrease the rate of osteoporosis. Abstract: .sup.1 Mary Jo V. Pugh , .sup.2 Jan E. Knoefel , .sup.3 Joyce A. Cramer , and .sup.4 Dan R. Berlowitz (VERDICT, South Texas Veterans Healthcare System, San Antonio, TX ; Neurology, Veterans Healthcare System New Mexico, Albuquerque, NM ; Psychiatry, Yale University, New Haven, CT ; and CHQOER, Bedford VA Hospital, Bedford, MA) Rationale: The incidence of epilepsy is most common in the elderly, and treatment choices may be complicated by physical and psychiatric comorbidities. Studies have not systematically examined the patterns of comorbidities in older patients with new onset epilepsy. Methods: We identified new onset epilepsy patients using national diagnostic and administrative data from the Veterans Health Administration and Medicare data files and national VA pharmacy data. We first identified individuals who were 65 years or older in each year (2000-2004). Individuals with a first diagnosis (ICD-9 codes) of epilepsy (345) or convulsion (780.39) with a concomitant first prescription for an antiepileptic drug (AED) were classified as having new onset epilepsy. Of those older veterans with both a diagnosis of epilepsy and a concomitant AED (n = 69,494), 9,234 met the criteria for new onset epilepsy. We then identified the common comorbid conditions that may affect treatment patterns. Results: The cohort of new onset epilepsy patients was primarily male (98%), White (78%; 16% Black, 5% Hispanic) with a mean age of 75.1 years. On average, these individuals had 11 physical comorbidities and slightly less than one psychiatric comorbidity. Approximately 30% had a prior stroke and 16% had a history of transient ischemic attacks. This group was also substantially affected by metabolic syndrome: 92% hypertension, 52% type 2 diabetes, and 68% hypercholesterolemia. Psychiatric comorbidities included depression (22%), psychosis (18%) and anxiety (13%). On average these patients received 17 (SD = 8.8) unique classes of drugs in addition to their AEDs, including many that pose potential interactions with older AEDs (e.g. warfarin, statins). Conclusions: The comorbidity profile of this populations indicates that they have profound comorbid disease. Diseases such as hypercholesterolemia and coagulopathy are commonly treated with drugs that have been found interact with older enzyme inducing AEDs. We know that over 80% of these patients received these older, enzyme inducing AEDs (65% phenytoin; 5% phenobarbital, 10% carbamazepine) Further research is needed to determine the extent to which drug-drug interactions occurred, and the impact of these interactions. It is possible that treatment with older, less expensive AEDs is offset by hidden costs that occur as a result of drug-drug interactions (e.g. higher mortality due to subsequent stroke or MI for patients treated with statins and who receive little benefit due to the drug-drug interaction) or hip fractures that may occur due to osteoporosis that has been associated with chronic use of enzyme inducing AEDs. Much research is needed to explore these hypotheses. (Supported by Department of Veterans Affairs, Health Services Research and Development Service.) Abstract: .sup.1 Dalin T. Pulsipher , .sup.2 Jana Jones , .sup.1 Christian Dow , .sup.2 Bruce Hermann , .sup.1 Michael Seidenberg , and .sup.2 Bell Brian (Psychology, Rosalind Franklin University of Medicine & Science, North Chicago, IL ; and Neurology, University of Wisconsin Hospital, Madison, Madison, WI) Rationale: There is an extensive literature examining quality of life (QOL) among individuals with epilepsy. Some clinical seizure variables (eg, seizure frequency, age of onset, duration of epilepsy) have been shown to influence QOL. Psychiatric comorbidities (e.g., depression) are reliably and strongly associated with decreased QOL. Recent studies have documented high rates of medical comorbidities (e.g., hypertension, diabetes, cancer) among people with epilepsy compared to healthy controls, but their impact on QOL has yet to be examined. The purpose of this study was to determine the relationship between comorbidities (medical and psychiatric) and QOL in a sample of adults with temporal lobe epilepsy (TLE). Methods: 93 adult patients with TLE were administered the QOLIE-89. Medical history was obtained through structured patient/collateral interviews and chart review. All subjects underwent the Structured Clinical Interview for DSM-IV which provided a current DSM-IV Axis I diagnosis. Correlation and regression analyses were performed to determine the contribution of clinical seizure variables and comorbid medical and psychiatric conditions to QOL. Results: 49.5% (n = 46) of the TLE patients reported the presence of a medical comorbidity, psychiatric comorbidity, or both. 19 (20%) subjects had only a medical comorbidity, 16 (17%) subjects had only a current Axis I psychiatric comorbidity, and 12 (13%) subjects had both. An increased total number of comorbid conditions were significantly associated with poorer QOL. Regression analyses demonstrated that the total number of comorbidities ([sz]= 3.79, t = 4.06, p < .01) was the strongest predictor of overall QOL, followed by education ([sz]= .95, t = 2.35, p = .02). Clinical seizure variables did not significantly predict QOL. In addition, a stepwise regression analysis showed that both psychiatric ([sz]= 4.46, t = 2.68, t < .01) and medical comorbidities ([sz]= 2.97, t = 2.24, p = .03) were distinct and unique predictors of overall QOL. Conclusions: Consistent with previous reports, we found that comorbid medical and psychiatric conditions were common in patients with TLE. The total number of combined medical and psychiatric comorbid conditions significantly predicted QOL after accounting for basic demographic and clinical seizure variables. This is the first study to date to examine the impact of medical comorbidities on QOL in TLE. The high rates of medical comorbidity observed in this and other studies should alert clinicians to the potential significance of their impact on QOL. (Supported by: This project was funded by the Epilepsy Foundation.) Abstract: .sup.1 Ross F. Liebman , .sup.1 Sarah Shalev , and .sup.1 Alcibiades J. Rodriguez (Department of Neurology, NYU Comprehensive Epilepsy Center, New York University School of Medicine, New York, NY) Rationale: Several studies have shown that sleep complaints are more common in epilepsy patients when compared to normal controls from the general population. The specific sleep complaints of an epilepsy patient may be different than those of a person without epilepsy seen in a sleep disorder center. This difference would have significant implications to the approach of treating epilepsy patients with a sleep disorder. Comparing the sleep complaints of epilepsy and non-epilepsy patients seen in consultation at a sleep disorders center could show this difference. Methods: A retrospective chart review was conducted on patients seen for initial consultation at the New York Sleep Institute over a 5 month period. All patients over 18 years of age were included. A total of 112 patients met these criteria, 39 in the epilepsy group and 73 in the control group. For each patient, information was collected from the initial consultation report written by a Board Certified Sleep Specialist. The data included basic demographic information, reason for referral, other sleep complaints, Epworth Sleepiness Scale (ESS) score, co-morbidities (including if they have been diagnosed with epilepsy based on the International League Against Epilepsy criteria), medication history, family history, social history and Body Mass Index (BMI). The data was analyzed using Pearson chi-square analysis for categorical data and t-test for continuous variables. Results: The chief complaint/reason for referral to the sleep institute was noted to be statistically different between the epilepsy group and the controls, via Pearson's chi-square analysis (p = 0.0035). In the epilepsy group, 48.7% had a chief complaint of insomnia, compared to 26.0% in the control group (p = 0.0157). Snoring was more common in the control group (34% vs. 18%) and subjective EDS was more common in epilepsy group (31% vs. 22%). From the interview, all sleep disturbances that were afflicting each patient were recorded. Of these sleep disturbances, only difficulty falling asleep (Insomnia subtype) was significantly different between the two groups (56% of the epilepsy patients, 25% of controls, p = 0.0012). Subjective EDS, snoring, un-refresing sleep, restlessness in legs and difficulty staying asleep were statistically equally frequent between the groups. Conclusions: Insomnia is the prevailing chief complaint in patients with epilepsy that present to a sleep disorder center. Over all complaints, difficulty failing asleep was most identifying with the epilepsy patients. Both of these problems were significantly different from the ones seen in patients without epilepsy. Abstract: .sup.1 Vivianne P. Rosa , .sup.1 Gerardo Maria de AraujoFilho , .sup.1 Katia Lin , .sup.1 Luis Otavio Sales Ferreira Caboclo , and .sup.1 Elza Marcia T. Yacubian (Department of Neurology, Universidade Federal de Sao Paulo/UNIFESP, Sao Paulo, Sao Paulo, Brazil) Rationale: The complex relationship between temporal lobe epilepsy (TLE) and psychiatric disorders (PD) has been matter of interest, and important studies have emphasized this association. PD compromise patients' quality of life. Few studies, however, studied the frequency of PD in patients presenting the same etiology of TLE. We evaluated the frequency of PD in a homogenous series of patients with refractory TLE related to mesial temporal sclerosis (MTS) aiming to determine the frequency of PD and possible correlations with clinical variables as well as laterality and degree of atrophy in volumetric MRI measures. Methods: Data from 106 patients with refractory TLE related to MTS accompanied at the outpatient epilepsy clinic of Universidade Federal de Sao Paulo, Brazil, from March 2005 to May 2006 were reviewed. The psychiatric evaluation was based on axis I DSM-IV criteria and on Bear and Fedio Inventory (BFI). For statistical analysis, the X.sup.2 test and Student's T test were used. P value considered significant was < 0.05. Results: Sixty-one females and 45 males were evaluated. The mean age and mean duration of the disease were 39.5 [+ or -] 13.4 and 24 [+ or -] 11.3 years, respectively. The mean age at epilepsy onset was 19.0 [+ or -] 12.8 years. MTS occurred more frequently on left side (72 cases; 68%), followed by right side (30 cases; 28.3%) and by 4 cases of bilateral MTS (3.7%). The hippocampus were significantly smaller (p < 0.05) in the side of the epileptogenic zone. Carbamazepine and phenytoin were the most used single drugs, and their association to a benzodiazepine, mostly clobazam, the most common association. PD were found in 65 patients (61.3%). Humor disorders were the most frequent (32 cases; 30%), followed by interictal (15 cases; 14%) and postictal (10 cases; 9.4%) psychosis. Twelve patients performed criteria for 2 axis I disorders, and the most frequently found association was major depression and postictal psychosis (7 cases), related to left MTS. Postictal and interictal psychosis were significantly associated with left side (p < 0.05), while PD in general and humor disorders were not associated to any side. Conclusions: There was a high prevalence of PD in patients with refractory TLE associated to MTS. The most common were humor and psychotic disorders. Psychosis was significantly associated to left side. These findings are concordant with data in current literature relating the prevalence of PD in TLE, confirming the existence of anatomic alterations, and also a possible left laterality effect in the mesial temporal lobe structures in patients with epileptic psychosis. (Supported by CAPES and FAPESP from Brazil.) Abstract: .sup.1 Nicole A. Seminario , .sup.1 Sarah T. Farias , .sup.1 Julie Jorgensen , and .sup.1 Masud Seyal (Neurology, University of California, Davis, Sacramento, CA) Rationale: Depression in epilepsy is common and a powerful indicator of the quality of life. Yet neurologists do not routinely screen epilepsy patients for depression in the clinic setting. In a busy neurology practice, where the major focus is seizure control, it is likely that symptoms of depression may be missed. The PHQ-9 is a brief, self-administered questionnaire based on current DSM-IV criteria that has been validated in several patient populations. PHQ-9 scores of 10 or higher have high sensitivity and specificity for diagnosing major depression. Utilizing the PHQ-9, we sought to determine the frequency of depression in patients with epilepsy and to correlate patient characteristics with the degree of depression. Methods: 263 patients seen in a tertiary referral epilepsy clinic over a 6-month period were asked to complete the questionnaire. Depression scores were assessed in relation to seizure frequency; epilepsy diagnosis; antidepressant and antiepileptic drug usage. Differences across patients were compared using ANOVA. Results: Nearly one-third of patients (29.3%) had scores consistent with major depression. Seizure-free patients had lower depression scores (mean score 5.1, s.d. 5.4; median score 3) than those with persistent seizures (mean score 8.1, s.d. 6.2; median score 7) (p = 0.003). Patients with non-epileptic psychogenic seizures had higher scores than those with localization related epilepsy, idiopathic generalized epilepsy or with symptomatic generalized epilepsy (Difference across groups significant p = 0.044). Depression scores were not related to the number or type of anti-epileptic drugs used. Patients currently taking antidepressant drugs had higher scores than those not on antidepressant drugs (p < 0.001). Just over one-half of patients with major depression were not on antidepressant medication. Conclusions: In patients with epilepsy, major depression was detected at a rate similar to that reported previously using "gold-standard" instruments. The PHQ-9 is a sensitive measure of depression in this population. The brevity of this instrument is conducive to routine screening of patients with epilepsy. Depression is under-recognized in this group of patients. Over half the patients with major depression were not on antidepressant drugs. Depression is associated with persistent seizures rather than the type of epilepsy or antiepileptic drugs. [figure 1] (Supported by Department of Neurology, University of California, Davis.) Abstract: .sup.1 Sadat A. Shamim , .sup.2 Gregor Hasler , .sup.1 Kathleen Kelley , and .sup.1 William H. Theodore (Clinical Epilepsy, NINDS, NIH, Bethesda, MD ; and NIMH, NIH, Bethesda, MD) Rationale: To evaluate the relationship between hippocampal volume loss, depression, and epilepsy. Background: There is a significantly increased incidence of depression and suicide in patients with epilepsy. Both epilepsy and depression are associated with reduced hippocampal volumes, but it is uncertain whether patients with both conditions have greater atrophy than those with epilepsy alone. Previous studies used depression measures strongly weighted to current state, and did not necessarily assess the influence of chronic major depressive disorder ("trait"), which could have a greater impact on hippocampal volume. Methods: Sixty-one epilepsy patients with complex partial seizures (CPS) confirmed by ictal video-EEG had 3D spoiled gradient recalled acquisition (spgr) with 1.5 mm slice MRI Scans. These were loaded into a linux based system, hippocampi were manually traced on each slice, and then assembled into 3 dimensional volumes. The anterior head of the hippocampus was separated from the amygdala by the appearance of the inferior limb of the lateral ventricle. The hippocampus was traced superiorly around the choroidal fissure, curving laterally along the medial border of the temporal horn, and medially along the gray matter of the hippocampus up to its junction with the parahippocampal gyrus. Depression screening was performed with Beck Depression Inventory (BDI, 56 patients) and with the structured clinical inventory for DSM-IV (SCID, 36 patients). For the BDI, a score above 10 was considered mild to moderate, above 20 moderate to severe, and above 30 severe depression. MRI and clinical analysis were performed blinded to other data. Statistical analysis was performed with Systat (Point Richmond CA) using Student's T-test and analysis of variance. Results: SCID evaluation revealed 20 patients with and 16 without depression. BDI revealed 24 and 32, respectively. There was no relationship between the right or left hippocampal volumes and depression for either measure. There was a clear and expected relationship between hippocampal volume and the side of the seizure focus (p = 0.001). However, this was not affected by the presence of depression based on SCID or BDI. Conclusions: We found no significant effect of depression, using measures of train as well as state, on hippocampal volume in patients with CPS. The high rate of comorbidity may be better evaluated with measures of cellular function rather than by anatomic imaging of the hippocampus. (Supported by National Institutes of Health.) Abstract: .sup.1 Elizabeth S. Stroup , .sup.1 Erica L. Coady , .sup.1 Mark D. Holmes , .sup.1 Daniel L. Drane , .sup.1 Naomi S. Chaytor , and .sup.1 John W. Miller (University of Washington Regional Epilepsy Center, Harborview Medical Center, Seattle, WA) Rationale: Comorbidity (CO) of both epileptic (ES) and psychogenic nonepileptic (PNES) seizures in the same patient is well documented. Less well studied is the effort required to reach this diagnosis, and the yield from additional days of monitoring. Methods: We identified 36 adult patients (64% female) with video-EEG confirmed ES and PNES events, from all long term monitorings performed at the UW Regional Epilepsy Center from mid-1999 to 2006. ES was defined as having definite ictal EEG abnormalities, while PNES was defined as episodes of unresponsiveness or abnormal motor activity in the absence of an electrographic seizure. For a diagnosis of CO, both types of events had to be captured during one or more video-EEG monitorings. Mean age at time of final diagnosis was 33 years (SD = 11). Results: The mean number of monitorings required to diagnose CO was 1.6 (SD = 0.9; median = 1.0; range = 1-5). Average number of required monitoring days (across one or more admissions) was 6.0 (SD = 4.0; range = 2-21). The mean number of discrete unequivocal events recorded before reaching a diagnosis of CO was 6.2 (SD = 6.9; median = 4.0; range = 2-35). Elapsed time between a patient's first recorded event and a diagnosis of CO averaged 3.8 days (SD = 4.0; range = 0-20). For the 50% of patients whose initial event was a PNES spell, elapsed time to an epileptic seizure averaged 2.6 days of additional monitoring (SD = 1.7; range = 0-5). In contrast, patients who initially experienced an epileptic seizure required nearly twice this time to convert from an exclusive diagnosis of ES to CO (M = 5.1 days; SD = 5.2; range = 1-20). Twelve of the 36 patients underwent surgery for intractable ES, with 42% experiencing their first PNES event after surgery. Conclusions: Most patients with verified comorbid diagnoses of ES and PNES events require relatively few monitorings to document both. In the current data, a comorbid diagnosis was reached in 90% of the sample in two or fewer monitoring admissions cumulatively totaling eight days or less. A full 64% of the sample was diagnosed with CO in only one monitoring admission. Perhaps of greatest clinical utility is the observation that 83% of patients who initially presented with a PNES event went on to have an unequivocal epileptic seizure within four monitoring days of their initial spell. Thus, when a diagnosis of epilepsy is strongly suspected in a patient who initially presents with a PNES event, evidence suggests that allowing several more monitoring days in selected patients can yield clinically useful data critical to appropriate patient care. Additionally, there was a subgroup of patients who developed PNES following surgical resection of an epileptic focus. Prediction of which patients warrant additional admissions or monitoring days remains to be determined, but is critical to efforts to reduce length of stay, without compromising quality of care. Abstract: .sup.1 Joanne M. Wrench , .sup.1 Sarah J. Wilson , .sup.2 Peter F. Bladin , and .sup.3 David C. Reutens (School of Behavioural Science, University of Melbourne, Melbourne, Vic, Australia ; Comprehensive Epilepsy Program, Austin Health, Melbourne, Vic, Australia ; and Department of Medicine, Monash University, Melbourne, Vic, Australia) Rationale: Patients undergoing temporal lobectomy are at risk of developing de novo depression after surgery despite good seizure outcome. Neurobiological predictors of de novo depression largely remain unknown. Previous research has indicated that patients with Major Depression have smaller hippocampi than controls. We explored whether this finding was also true for epilepsy surgery patients by examining the relationships between pre-operative hippocampal volumes, seizure outcome, and de novo depression after epilepsy surgery. Methods: The sample comprised 42 patients undergoing epilepsy surgery in the Comprehensive Epilepsy Program, Austin Hospital (26 mesial temporal lobe resections and 16 non-mesial temporal resections). Patients were prospectively assessed pre-operatively, and at discharge, one, three, six, and 12 months post-operatively. Assessment included a semi-structured clinical interview (Austin CEP Interview) and the Beck Depression Inventory (BDI-II). Hippocampal volumetry was conducted on patients' preoperative T(1)-weighted MRI scans and on 41 neurologically normal controls following stereotactic normalisation of the images. Results: Post-operatively, 42% of mesial temporal patients and 19% of non-mesial patients were depressed within the first 12 months after surgery. The occurrence of depression was associated with significantly smaller hippocampi contralateral to the resection in the mesial temporal group (p = 0.003). This effect was also seen in mesial temporal patients who developed de novo depression after surgery (p = 0.027). Contralateral hippocampi were particularly small in patients suffering from both depression and seizures post-operatively (p = 0.031). Furthermore, there were significant negative correlations between contralateral hippocampal volumes and BDI-II scores at one and three months post-operatively (r =-0.46 and r =-0.42 respectively, p < 0.05). There were no significant findings between depression, seizure outcome, and hippocampal volumes in the non-mesial group. Conclusions: Our findings suggest that the hippocampus is a structural marker of vulnerability to post-operative depression, particularly in the presence of post-operative seizures. They also indicate that a smaller hippocampus predates the onset of depression. Abstract: .sup.1 Miya R. Asato , .sup.1 Patricia K. Crumrine , .sup.1 Enami Yasui , and .sup.1 Beatriz Luna (Pediatrics, Division of Child Neurology, Children's Hospital of Pittsburgh, Pittsburgh, PA; and Psychiatry, Western Psychiatric Institutes and Clinics, Pittsburgh, PA) Rationale: While the elevated risk for psychopathology in pediatric epilepsy has long been recognized, the relationship between brain function and the risk for psychiatric comorbidity has not been well studied. These initial results are from an ongoing prospective study of different seizure subtypes of well-characterized medically treated pediatric epilepsy patients and matched controls. These children performed a battery of neurocognitive tests, diagnostic psychiatric interviewing, and tasks of cognitive control of behavior using oculomotor tasks. Methods: We report results from 12 pairs of pediatric epilepsy patients and age, gender, and IQ matched control subjects. From the 12 pairs studied up to the present, the average age of the children was 11.92 years (SD 3.2), and gender was split evenly between males and females. Mean IQ of the patient group was 96.5 (SD 11.9) and duration of disease was 3.5 years. Epilepsy type in the patient group was classified as complex partial (n = 8), primary generalized (n = 2), and benign focal epilepsy (n = 2) based on EEG data and clinical semiology. All patients except one were being treated with anticonvulsants. Control subjects were free of psychiatric diagnoses, and had no history of neurological or psychiatric disorders. Participants completed cognitive control oculomotor tasks including a test of sensorimotor control (visually guided saccade task) and a test of response inhibition (antisaccade). Oculomotor tasks are objective, non-verbal measures of executive function which are easily performed by children, and have been successfully used to study cognitive and brain systems in children with typical development and neurodevelopmental disorders. Comprehensive psychiatric assessment including a diagnostic psychiatric interview (KSADs -PL) and the Behavior Rating Inventory of Executive Function (BRIEF). Results: The rate of psychiatric diagnoses in the epilepsy patient group was elevated at 50% including attention deficit disorder (n = 5) and depression (n = 1). There were no psychiatric diagnoses in the control population. Preliminary results of the oculomotor tasks demonstrate that all 24 children performed well on the control task, indicating intact basic sensorimotor function. Children with epilepsy commited more response inhibition errors on the antisaccade task compared to controls (p = 0.06), indicating a deficit in response inhibition. Results from the BRIEF, a parent and teacher report of executive functioning indicate a significant correlation between poorer functioning and having psychiatric comorbidity (p < 0.001). Conclusions: As part of an ongoing study to evaluate cognitive processes at risk due to epilepsy mediated effects, these results indicate a neurobiologically linked relationship between executive functioning and psychiatric comorbidity. (Supported by NINDS K23 NS052234.) Abstract: .sup.1 Janice M. Buelow , .sup.1 Joan K. Austin , and .sup.2 Susan M. Perkins (School of Nursing, Indiana University, Indianapolis, IN ; and Division of Biostatistics, Indiana University School of Medicine, Indianapolis, IN) Rationale: Adaptive functioning refers to the physical, social, and communication skills necessary for functioning in daily life. We could find no study that described adaptive functioning in children with both epilepsy and mild intellectual disability (ID). Some authors suggest that measuring adaptive functioning is more important than measuring intelligence because adaptive functioning reflects actual behavior in the context of daily life and may be amenable to intervention. The purpose of this study was to describe the relationship of adaptive functioning to behavior problems in a population of children with both epilepsy and mild developmental disability. Methods: To measure adaptive functioning, parents completed structured interviews using the Vineland Adaptive Functioning Scale (VABS). To measure behavior problems, parents completed the Child Behavior Checklist (CBCL). Results: The sample was 49 children (24 girls and 25 boys) ages 9-16 years (M = 12.3) who had both epilepsy and mild ID. The mean estimated IQ was 73.5 (SD = 17.8). The mean age of seizure onset was 5.3 years and the mean duration was 7.0 years. About half (53%) of the children had two seizure types and 46% were receiving more than one AED. The CBCL scores showed high rates of behavior problems; total (M = 66.36, SD = 8.86), internalizing (M = 64.21, SD = 11.26), and externalizing (M = 60.19, SD = 10.78). The mean VABS scores were approximately 2 standard deviations below the norm scores for the general population; total (M = 60.69, SD = 11.12), communication (M = 60.12, SD = 12.94), daily living (M = 68.67, SD = 16.41), and socialization (M = 68.42, SD = 15.40). There was a trend for the VABS total adaptive functioning score to be negatively correlated with CBCL total behavior problems score (p = .09). The VABS daily living skills score was negatively correlated with CBCL total behavior problems score (p = .05), and CBCL externalizing score (p = .02). VABS socialization and communication scores were not significantly correlated with CBCL total scores. Conclusions: This pilot study showed an association between more behavior problems and poorer adaptive functioning in children with epilepsy and mild intellectual disability. Specifically, total behavior scores were correlated with both total adaptive functioning and daily living skills. Our findings are similar to those in other studies of children with mild ID and chronic disorders other than epilepsy. Our pilot study provided foundational information regarding overall adaptive functioning and behavior in children with both epilepsy and mild ID. (Supported by NR 04536.) Abstract: .sup.1 Tara Clarke , .sup.1 Bhavna Bali , .sup.1 Lisa J. Strug , .sup.1 Peregrine L. Murphy , and .sup.1 Deb K. Pal (Epidemiology, Columbia University Medical Center, New York, NY) Rationale: An association between Rolandic epilepsy and migraine headache has been suggested in previous case series (Bladin, 1987), and controlled studies are conflicting (Santucci, 1985;Giroud, 1989). It has further been suggested that the two diagnoses share etiological background. We therefore tested the hypothesis that migraine headaches are over-represented in children with Rolandic epilepsy. Methods: We performed an unmatched case-control study. We collected 44 cases of strictly defined Rolandic epilepsy with oro-facial seizures, median age 9 years, sex ratio 3:2, M:F from pediatric neurology clinics. We collected 92 control children lacking a primary brain diagnosis, median age 9 years, from other subspecialty clinics at the same hospitals. We made a clinical diagnosis of migraine using features in the history that include those required for the International Headache Society criteria, but with modifications for childhood as proposed by Maytal et al 1997. We calculated odds ratios of association, adjusted for age and sex, using multiple logistic regression. Results: 17% of cases and 17% of controls had a clinical diagnosis of migraine without aura. No subject was diagnosed with other migraine variants. When adjusted for age and sex, the odds ratio was 0.94 (95%CI: 0.57-1.52). Conclusions: The prevalence of migraine in controls was comparable to that reported in meta-analysis estimates (Scher, 1999). The prevalence of migraine in Rolandic epilepsy cases is no different than in controls without a primary brain diagnosis. Our finding of no association is in agreement with a previous controlled study (Santucci, 1985). In contrast, previous uncontrolled reports have suggested that migraine occurs in 62-80% of children with Rolandic epilepsy. Our results suggest that previous reports may have overestimated the prevalence of migraine in Rolandic epilepsy patients, and hence concluded an association. Although our findings do not support the idea that migraine is over-represented in Rolandic epilepsy, this does not exclude the possibility of an association with other focal epilepsies, or the possibility that shared etiological mechanisms pertain for migraine and epilepsy. (Supported by Epilepsy Foundation.; Partnership for Pediatric Epilepsy.; National Institutes of Health NS 047530.) Abstract: .sup.1 Gary A. Stobbe , .sup.1 Kathy Eileen , .sup.2 Julie M. Davies , and .sup.1 Susan Malmquist (Research, Autism Spectrum Treatment and Research Center, Seattle, WA ; and Department of Psychology, Seattle Pacific University, Seattle, WA) Rationale: Previous research has recognized the frequent association of epilepsy and autism (7%-42%; Tuchman, 2000). Debate has ensued whether a causal relationship exists between EEG epileptiform abnormalities and autistic regression in the absence of clinical seizures. AED use with autistic children has suggested benefits in areas of hyperactivity, impulsivity, aggression, and affective instability associated with autism (Hollander, 2001; Rugino, 2002). The purpose of this study was to investigate whether autistic children with EEG epileptiform abnormalities treated with levetiracetam would show improvement in certain core features of autism. Methods: Participants were 9 boys and 2 girls, ages 3 to 11 years (M= 6.43) with a diagnosis of autism spectrum disorder and with epileptiform discharges on EEG. Multiple baseline and outcomes measures evaluating core features in autism were employed, including daily behavioral diaries assessing social interaction, emotional response, awareness, expressive language, receptive language, play/imagination, body movement, rigid/habitual behaviors, and other behaviors. Study design was a 24-week, open label, non-forced titration of levetiracetam with initial daily dose of 10 mg/kg/d and target dose of 60 mg/kg/d. Baseline concommitant medications were not considered reason for exclusion. Results: Ten (91%) children had a formal diagnosis of autism; one of Aspergers. Incidence of regressive sybtype was overrepresented (64%) compared to expected (30-39%). Early termination occured in 6 of 11, primarily due to aggression and emotionality seen at higher doses. Curvilinear analysis suggested the maximum benefit was at lower doses (5-10/mg/kg) in 7 of 11 children. No serious adverse events were observed and all side effects reversed with discontinuance of medication. Preliminary results from daily behavioral diaries indicate positive trends in awareness (10 of 11). Differential results for regressive subtype suggest benefit in the domain of play/imagination (6 of 7). Participants with EEG patterns of left focal epileptiform discharges and bifrontal sharps evidenced positive trends in the area of expressive language (5 of 6, and 2 of 2, respectively). Children with right focal discharges experienced negative effects in expressive language (3 of 3). Conclusions: Preliminary results suggest levetiracetam is safe in children with ASD and EEG epileptiform abnormalities, and that certain behavioral domains of autism may benefit from low doses. Increasing dose did not correspond to increased performance, and resulted in a higher incidence of early termination from the study. Epileptiform discharges involving the dominant hemisphere may predict benefit in the domain of expressive language. These results warrant further investigation targeting concepts of EEG abnormaility and language effects with levetiracetam. (Supported by UCB Pharma, Inc.) Abstract: .sup.1 Jana E. Jones , .sup.2 Christian Dow , .sup.1 Raj Sheth , .sup.3 Monica Koehn , .sup.1 Ryann Watson , .sup.3 Rochelle Caplan , .sup.2 Michael Seidenberg , and .sup.1 Bruce P. Hermann (Neurology, University of Wisconsin, Madison, WI ; Psychology, Rosalind Franklin University of Medicine and Science, North Chicago, IL ; Psychiatry, UCLA, Los Angeles, CA ; and Neurology, Marshfield Clinic, Marshfield, WI) Rationale: Among children with recent onset idiopathic epilepsies (age 8-18), a subset have been found to exhibit learning and education problems prior to the first recognized seizure. These children exhibit greater abnormalities in neuropsychological status as well as left hemisphere MRI volumetrics compared to children with recent onset epilepsy without problematic educational histories and healthy controls. These findings are consistent with the hypothesis that a subset of children with new onset epilepsy sustained an antecedent neurobiolgical insult affecting brain function and structure. The degree to which comorbid mental health problems may also be overrepresented in this at risk cohort remains to be determined and represents the focus of this investigation. Methods: Subjects were children age 8-18 with recent onset (diagnosed within the past 12 months) epilepsy of idiopathic etiology (n = 51). Participants and their parents underwent a standardized psychiatric interview to characterize DSM-IV diagnoses (K-SADS). Structured interview identified 35 children with no history of academic problems (AP-) and 16 children with prior (pre-epilepsy onset) academic problems (AP+) that included either special education, special assistance in reading and math, mandatory summer school, tutors, or grade retention. Results: AP+ children exhibited significantly elevated rates of lifetime-to-date Axis I disorders of any type compared to AP- children (93.8% vs. 60.0%, p < .01). There were higher rates of externalizing disorders (62.5% vs. 11.4%, p < .001), including ADHD (26.4% vs. 10%) and conduct disorders (12.5% vs. 0.0%, p < .05). A majority of the AP+ children (62.5%) exhibited psychiatric comorbidity prior to the first recognized seizure. Internalizing disorders (depression, anxiety) were not overrepresented in AD+ children. Conclusions: Children with academic problems (AP+) antedating epilepsy onset also exhibit an increased rate of psychiatric comorbidity, especially externalizing disorders which includes ADHD and conduct disorders. This emerging pattern of pre-epilepsy onset learning and psychiatric disorders as well as MR volumetric abnormalities is consistent with a presumed antecedent neurobiological insult independent of seizures, epilepsy syndrome, and medication treatment. Evidence of early learning problems and psychiatric comorbidity needs to be integrated in models of the development and course of epilepsy in children and adolescents. (Supported by NIH NS R0144351, M01RR03186 (GCRC), and F32 MH649882.) Abstract: .sup.1 Susan Koh , .sup.2 Brenda Bursch , .sup.2 Erin Lanphier , and .sup.2 Rochelle Caplan (Department of Pediatric Neurology, University of California, Los Angeles, CA ; and Department of Psychiatry, University of California, Los Angeles, CA) Rationale: This study examined the rate, types, and demographic features of non-epileptic events (NEE) in a large pediatric epilepsy center over 5 years. The single pediatric study conducted to date found NEE in 15.2% of the children (Kotagal 2002). These included stereotyped and paroxysmal movements, hypnic jerks, parasomnias, Sandifer syndrome, inattention and conversion disorder. Epilepsy was found in 19-25% of the middle childhood and adolescent cases. Methods: A chart review was conducted on all children under 18 years with NEE during long term videotelemetry at UCLA between 2000 and 2005. Patients with auras were excluded from this study since it is unclear whether these spells were NEE or not. Information gathered included ethnicity, gender, age at time of telemetry admission, history of psychiatric illness, type of medications, history of epilepsy, history of developmental delay, EEG ictal data (if the patient had a seizure), and type of NEE. Type of NEE was categorized as arousal/sleep myoclonus, abdominal complaints (reflux), apnea, psychiatric disorders (psychosis, impulse control disorder, conversion disorder), movement disorders (tics, dystonia, tremors, myoclonus) and staring. Results: 469 of the 1509 telemetry admissions had NEE. Among these 469 patients, 150 had movement disorders, 112 sleep related issues, 107 staring, 16 gastrointestinal related issues, 16 apnea (mainly breath holding spells), and 8 headaches. Many had different types of NEE noted in the same procedure. 128 patients had a psychiatric diagnosis with a conversion disorder in 31 and impulse control disorders psychosis, masturbation or enuresis in the remaining 97 children. Of those with conversion disorders, 13 were female and 18 male ranging in age from 3-18 years (mean 12.87 yrs). The ethnic distribution was similar to that of the greater Los Angeles community. 14 children had a past history of epilepsy and 3 had documented epileptic seizures during the same telemetry. Six patients had no medication treatment for epilepsy or psychiatric condition prior to the telemetry, 3 had treatment for a psychiatric condition and 25 had treatment for seizures. 12 patients had a prior psychiatric history of depression (2), bipolar disorder (1), psychosis (5), attention deficit disorder (2), and autism (2). 9 patients had a history of mild mental retardation or learning disability. Conclusions: Unlike Kotagal et al. (2002), 31% of our telemetry patients had NEE. 27% had comorbid psychiatric disorders with conversion disorder in 7% of all telemetry cases. There also was a higher rate of previous epilepsy (45%) and past psychiatric history (39%) in the NEE cases associated with conversion disorder. The high rate and marked morbidity of this disorder highlights the importance of further studies to identify neurocognitive and socioeconomic factors that increase the risk of children with epilepsy for this disease. Abstract: .sup.1 Jay A. Salpekar , .sup.1 Marian J. Kolodgie , .sup.1 Audrey Scully , .sup.1 Leia Foster , .sup.1 Andrew Renuart , and .sup.1 Joan A. Conry (Center for Neuroscience and Behavioral Medicine, Children's National Medical Center, Washington, DC) Rationale: Pediatric epilepsy patients with intellectual and developmental disabilities (IDD) are challenging both in terms of epilepsy treatment and behavioral comorbidity. Behavioral and psychiatric characteristics in this population are also difficult to isolate given the potential masking of psychiatric symptoms by intrinsic characteristics of IDD. IDD patients have been perceived to have limited ability to swallow pills and thus have restricted treatment options. Our goal was to characterize behavioral problems in a study population with epilepsy and moderate-severe IDD and to assess behavioral outcomes through the clinical trial. Methods: Pediatric epilepsy patients with IDD were recruited to participate in a clinical trial involving transition from Depakote.sup.[R] sprinkle medication to extended release tablets. A clinical evaluation, CBCL, and adaptive behavior rating scales were completed initially and at the conclusion of the eight week trial. Patients underwent a protocol to instruct them on swallowing pills. Parents completed CBCLs and category scores were deemed clinically significant for t scores of 65 or greater. Results: Ten patients with pediatric epilepsy and IDD were recruited (average age 11.1; 6 male) for the dosage conversion study. Adaptive functioning scales and/or documented IQ levels revealed all participants to be well within the moderately mentally retarded range (IQ approx. 40-60). Two patients were nonverbal, but could accomplish many activities of daily living with assistance. At the initial visit, 8/10 subjects had at least one CBCL narrow band category score in the clinically significant range. Seven subjects had completed behavioral measures and one subject had dropped out at the time of this analysis. Of the seven completers, five had at least one category score that improved, and three had at least one category score that worsened. Three patients had improvement in thought problems, two had improved attention, and none of the sample worsened in these two categories. Two patients had worse scores in somatic concerns. Conclusions: We did not expect significant behavior change in this study that focused upon medication dosage form conversion; however, some specific aspects of behavior improved with the transition to once daily Depakote.sup.[R] extended release tablets. Thought processing and attention were improved in some subjects and did not worsen for any subject. Somatic concerns increased in two subjects and may be related to preoccupation with pill swallowing. It is notable that in this severely impaired population, once daily medication was successfully used and may have benefits for behavior function in addition to facilitating easier adherence to treatment. (Supported by Abbott Laboratories, Inc. provided unrestricted research support.) Abstract: .sup.1 Elisabeth M.S. Sherman , .sup.2 Daniel J. Slick , .sup.3 Mary B. Connolly , and .sup.3 Kim L. Eyrl (Neurosciences, Alberta Children's Hospital, Calgary, AB, Canada ; Community NeuroRehab Services, Calgary, AB, Canada ; and Neurosciences, BC Children's Hospital, Vancouver, BC, Canada) Rationale: ADHD is a frequent comorbid condition in pediatric epilepsy, especially in tertiary care centers. In a large sample of children with epilepsy, we aimed to determine (1) the prevalence of ADHD symptoms, (2) neurological and epilepsy-related risk factors for ADHD, 3) the influence of intellectual deficits on the expression of ADHD in epilepsy, and 4) the extent to which ADHD symptoms pose a risk for poor health-related quality of life (HRQOL) and restrictions in daily activities in children with epilepsy. We also explored whether there was an epilepsy-specific presentation of ADHD in children with epilepsy. Methods: Data from the ADHD Rating Scale-IV (ADHD-IV) from 203 cases (mean age = 11.8, SD = 3.8) from a tertiary centre serving children with severe epilepsy were reviewed. The majority of children had focal symptomatic epilepsy, with a median seizure frequency of 3 seizures per month and median age at onset of 2.5 years. Results: Clinically significant problems with Inattention were more frequent than problems with Hyperactivity-Impulsivity (40% of children vs. 18%). About 2/3 of the sample met screening criteria for either ADHD-I or ADHD-C, with equal representation of both subtypes. Age of onset, epilepsy duration, and intractability did not predict ADHD symptoms or subtype, but ADHD-I was more prevalent in localization-related epilepsies, and there was a trend for higher medication load in this subgroup. Severity of ADHD symptoms was inversely related to HRQOL. Overall, ADHD-C children were younger, had low functional levels, had the poorest HRQOL and the most restrictions in daily activities. No gender difference in the prevalence of ADHD subtypes was seen. Conclusions: Children seen at tertiary care centres for severe epilepsy are at high risk for attention problems and ADHD. The clinical characteristics of ADHD occurring in the context of epilepsy differ significantly from the known characteristics of children with primary ADHD, raising questions about the validity of using the DSM-IV-based criteria and the notion of comorbidity to describe what may be an epilepsy-specific disturbance in children with severe epilepsy. (Supported by BC Medical Services Foundation/Vancouver Foundation.) Abstract: .sup.1 Anastasia C. Sullwold , .sup.1 Gretchen E. Weatherly , and 1,2 Frank J. Ritter (Epilepsy, Minnesota Epilepsy Group, P.A. of United Hospital and Children's Hospitals and Clinics - St. Paul, St. Paul, MN ; and Department of Neurology, University of Minnesota, Minneapolis, MN) Rationale: Pediatric epilepsy patients are at risk for cognitive and mental health problems. Previously, we found suicidal ideation (SI) was more prevalent in hospitalized pediatric patients with psychogenic nonepileptic seizures (PNES) or both PNES and epileptic seizures (MIX) than those with epileptic seizures alone (ES). Currently, our objective is to describe the occurrence of SI as it relates to cognitive functioning, age, and seizure diagnosis in a sample of pediatric epilepsy patients. Methods: Records of 162 children admitted to an inpatient epilepsy unit from 9/99-4/06 for video-EEG monitoring were reviewed. Inclusion criteria included evaluation by a psychologist during the hospitalization, diagnosis of ES, PNES or MIX by a neurologist, and IQ data obtained from within 5 years of the hospitalization. Those with IQ below 40 and those with physiologic nonepileptic events were excluded. All epileptic seizure types were grouped together. Data were analyzed using T-Tests for Independent Measures, ANOVAs, Chi-Squares, and Partial Correlations. Results: 47% of the sample was male. 125 had ES, 20 had PNESs, and 17 had MIX. Gender was equally distributed across seizure groups. ANOVA indicated those with ES were younger (mean = 12 yrs; p < .01) than those with MIX (mean = 14 yrs) and IQs were similar across the three seizure groups (ES mean = 79, PNES mean = 89, and MIX mean = 80). 20% of the sample reported SI. This included 18% of ES (n = 23), 24% of MIX (n = 4), and 30% of PNES (n = 6). Chi-square analyses showed a non-significant but clear trend toward higher SI in those with PNES/MIX than ES. T-Tests showed those with SI were older (mean = 13 yrs) and with higher IQ (mean = 87) than those without SI (mean age = 12 yrs, p < .05; mean IQ = 79, p < .01). Correlations confirmed the relationship between SI and IQ (p < .01) and between SI and age (p < .05). However, age and seizure group were also significantly correlated (p < .01). A partial correlation holding the seizure group constant eliminated significance from the age-SI relationship. Conclusions: In this selective sample, IQ was related to SI, with IQs closer to the average range showing higher occurrence of SI. A trend of higher SI in PNES/MIX was notable, similar to our previous study that indicated a robust relationship between SI and PNES/MIX in a larger sample. Taken together, these results suggest the presence of PNES and an average IQ may increase risk for SI in pediatric epilepsy patients. This study is limited by patient selection factors. Nonetheless, this report represents progress towards determining key factors in identifying pediatric epilepsy patients at highest risk for SI. Further study of this important issue is clearly warranted. Abstract: .sup.1 Sigride Thome-Souza , .sup.1 Evelyn Kuczynski , .sup.2 Patricia Rzezak , .sup.2 Daniel Fuentes , and 1,3 Kette Valente (Clinical Neurophysiology - Psychiatry, USP, Sao Paulo, SP, Brazil ; Neuropsychology - Psychiatry, USP, Sao Paulo, SP, Brazil ; and LIM 21- Psychiatry, USP, Sao Paulo, SP, Brazil) Rationale: Recent studies have demonstrated a high prevalence of psychiatric disorders (PD) in children and adolescents with epilepsy. Due to particularities of epilepsy in this age group, especially as to disease duration, it is of interest that this group be studied separately. Most studies with children approach heterogeneous groups with focal and generalized epilepsy, both symptomatic and idiopathic. We aimed to evaluate the frequency and type of PD that occurs in children and adolescents with temporal lobe epilepsy (TLE) and to analyze possible predictive factors, concerning the epilepsy, for the occurrence of PD in these patients. Methods: A multidisciplinary team evaluated patients taking part in this study. Psychiatric interviews were carried out with KIDDIE-SADS and patients were classified according to DSM IV and CID 10. Twenty-six children and adolescents with mean age of 12 years were studied. Mean age of epilepsy onset was 3.6 years (SD 3.2) and mean duration of 8.6 years (SD 4.5). As to etiology, nineteen patients had mesial temporal sclerosis (MTS) and seven, lateral temporal epilepsy (LTE). Fourteen patients were on polytherapy and fourteen were considered refractory to clinical treatment. Results: From 26 patients with TLE, 22 (84.6%) presented PD, of which 10, more than one. The diagnosed PD were: affective disorders in 13 patients (59%), behavioral disorders in 6 (27.3%); hyperkinetic disorder, mental deficiency and dissociative disorder in 4 (18.2%); and oppositional defiant disorder and bipolar disorder in 1 patient (4.5%). Of the 13 patients with affective disorders, depressive disorder was diagnosed in 9 patients (mild in 4 and moderate to severe in 5) and anxiety disorder in 4 patients. Nine families (40.9%) reported history of PD preceding the onset of epileptic seizures. Mean age of epilepsy onset in patients with PD was 3.5 years, and in those without PD of 4.3 years. As to duration of epilepsy, patients with PD had a longer duration (8.5 years) when compared to patients without PD (7.5 years). As to etiology, out of 19 patients with mesial structure lesions, 19 (100%) presented PD, and out of seven patients with lateral lesions, three (42.9%) presented PD, a statistically significant difference. Family history for PD was reported in 13 out of 22 patients (59%). Conclusions: The study of children with epilepsy enables us to study in an unusual manner the relationship between epilepsy and PD, without interference of important variables observed in adult series such as prolonged disease duration. In this series the prevalence of PD in children with TLE was as high as that observed in adults, mainly affective disorders. Patients with PD present an earlier age-onset of epilepsy and longer duration, although these differences were not significant. In this study presence of MTS was the most relevant predictive factor for PD. (Supported by FAPESP.) Abstract: .sup.1 Alcy R. Torres , .sup.2 Jane E. Whitney , .sup.2 Sneha N. Rao , .sup.2 Rachel M. Lobel , .sup.2 Claire A. Tilley , and .sup.2 Joseph M. Gonzalez-Heydrich (Neurology, Children's Hospital/Harvard Medical School, Boston, MA ; and Psychiatry, Children's Hospital/Harvard Medical School, Boston, MA) Rationale: Twelve to 39% of children with epilepsy have Attention Deficit/Hyperactivity Disorder (ADHD).sup.1, yet only methylphenidate has controlled data in this population. Methods: We reviewed medical records of pediatric patients with ADHD plus epilepsy given atomoxetine (ATM) abstracting diagnoses, clinical course, adverse events, and prospectively entered Clinical Global Impressions scores (CGI). Response at last visit is a CGI-Improvement score of 1 or 2 plus a drop of at least 1 point in CGI-Severity score compared to baseline, and continuation of ATM. Results: Twenty-seven patients (10.1 [+ or -] 4.2 years, 63% male) were treated with ATM for median 26, range 4-141 weeks. The average daily dose was 35 [+ or -] 24 mg (median 25, range 10-100mg). For the 22 patients with weights available the dose was 0.9 [+ or -] 0.4 (range 0.1-1.6) mg/kg/day. Twenty-four (90%) had been unsuccessfully treated with stimulants prior to ATM. ADHD was combined subtype in 21 (78%) patients. Twenty-one patients (88%) were diagnosed with a psychiatric disorder in addition to ADHD including mood disorders in 16, anxiety in 8, and psychosis in 6. Six patients had a concomitant antipsychotic when they began treatment with ATM and 11 patients (41%) were treated with an antipsychotic at the last visit on ATM. Average seizure frequency at baseline was 9 [+ or -] 37 per month (median 0, range 0 to 90). Epilepsy etiology was idiopathic (n = 18, 67%), cryptogenic (n = 8, 30%) or symptomatic (n = 1, 4%). Seizure onset was strictly partial (n = 4, 15%), generalized (n = 9, 33%), or mixed partial and generalized (n = 13, 48%). Twenty-three patients (85%) were on antiepileptic medications (AED) at baseline, and 21 (78%) at the last visit on ATM. Four patients increased and two patients decreased their AED regimens during ATM treatment. There were no serious adverse events and no patients discontinued ATM due to an increase or worsening of seizures. Seventeen of 27 patients (63%) discontinued ATM. The reasons were: inadequate response (n = 7, 23%), increased irritability/activation (n = 8, 30%), appetite decrease and tremor (n = 1, 4%), or noncompliance (n = 1, 4%). Eight patients (30%) met criteria for response. ATM dose, epilepsy type, and comorbid psychiatric disorders did not predict response or discontinuation. Conclusions: The modest response rate (30%) and high discontinuation rate (63%) on ATM could be due to the high percentage of stimulant non-responders in this sample. The absence of serious adverse events or discontinuations for seizure exacerbation indicates that prospective studies should be planned. Limitations of this study include its small sample size, retrospective nature, and lack of randomization and placebo control. Reference Dunn, DW et al. Dev Med Child Neurol 2003 Volume: 45 Issue: 1 50-54 Abstract: .sup.1 Monique J. Veendrick-Meekes , .sup.1 Boudewijn B. Gunning , and .sup.1 Loek L. Steenbeek (Childrens Department, Epilepsy Eenter Kempenhaeghe, Heeze, Netherlands) Rationale: To assess signs and symptoms related to seizures and to the use of antiepileptic drugs (AED) inducing Lower Respiratory Tract Affections (LRTA) in children with Catastrophic Epilepsies (CE). Methods: This study is a 2 years retrospective case study including 28 children, age 0;8-7;4 mean 2;9), males 20(71%) and females 8(29%) with intractable seizures. Seizures were observed by trained nurses. All children underwent excessive diagnostic procedures to define epilepsy syndrome and etiology. Pulmonary symptoms were diagnosed on clinical observations, physical examinations, X ray and measurement of O2 saturation. Results: LRTA occurred in 12 children (43%)(LTRA), 16 (57%) children had no LRTA. (n-LRTA. Daily seizures occurred in 9(75%) infants in LRTA and in 6 (37%) in n-LRTA. Seizure related LTRA inducing factors were seen 8 (67%) in LRTA, 4 (25%) in n-LRTA. Vomiting 4(33%) in LRTA, 2(12%)in n-LRTA. Salivation 5(42%) in LRTA, 0 (0%) in n-LRTA. Apnea 1(8%) in LRTA and 0(0%) in n-LRTA, and somnolence 1(8%) in LRTA, 2 (12%) in n-LRTA. LRTA inducing adverse events of antiepileptic drugs (AED) were seen11(92%) and 5 (31%) in n-LRTA. Salivation was present 4 (33%) in LRTA and 0(0%) in n-LRTA, somnolence 4 (33%) in LRTA, 3(19%)in n-LTRA, vomiting 2 (16%) in LRTA, 0(0%) in n-LRTA and apnea 1(8%) in LRTA and 0(0%) in n-LRTA .1(8%)child died in LRTA, 2 (12%) in n-LRTA. Conclusions: LRTA in infants with CE can be life threatening and have negative impact on quality of life. Daily seizures, salivation and vomiting might induce LRTAThere is a further need of investigations to improve prevention and treatment options. Abstract: .sup.1 Elaine C. Wirrell , and .sup.1 Judy Wong (Pediatrics and Clinical Neurosciences, University of Calgary, Calgary, AB, Canada) Rationale: To determine (a) if children and teens with epilepsy participate in less physical activity and have higher body mass index (BMI) percentiles for age than their non-epileptic siblings and (b) what epilepsy-specific factors limit their participation. Methods: Patients 5-17 years, with a [greater than or equal to]3 month history of epilepsy, a development quotient [greater than or equal to]80, no major motor or sensory impairments, and at least one non-epileptic sibling in a similar age range, were identified from the neurology clinic database or at the time of clinic visit. Parents completed a questionnaire regarding sedentary activities and group, individual and total sports activities. Children aged 11-15 years also completed the physical activity portion of the Health Behavior in School Aged Children questionnaire. Clinic charts were reviewed for seizure type, etiology, frequency, duration of epilepsy and number of anti-epileptic drugs (AEDs) ever taken. Results: Teens with epilepsy participated in less group and total sports activities than controls and were more likely to be potentially overweight or overweight. Receiving three or more AEDs in the past showed a significant negative correlation with sports participation. While there was a trend for those with higher seizure frequency to be less active, no other epilepsy specific factors nor prior seizures or seizure-related injury during a sports event correlated with participation in physical activity. Conclusions: Programs that promote exercise in adolescents with epilepsy should be encouraged to improve their physical, psychological, and social well-being. Abstract: .sup.1 Jerry J. Shih , and .sup.1 David R. Chabolla (Department of Neurology, Mayo Clinic, Jacksonville, FL) Rationale: Ictal asystole or bradyarrythmias are recognized cardiac processes associated with certain epilepsies and are the subject of case reports and case series. Less is known about the incidence of patients originally thought to have epilepsy who ultimately are diagnosed with asystole or bradyarrythmias not associated with any seizure manifestation. Methods: All patients admitted to epilepsy monitoring units from two hospitals from 2000-2005 were retrospectively evaluated for clinically significant bradyarrythmias which explained their paroxysmal episodes of alteration of consciousness or "convulsions". 550 subjects ranging from age 5 months to 81 years-old were included in the analysis. All subjects analyzed had one or more events documented with concomitant video, electroencephalographic, and electrocardiographic data. Total recording time ranged from 8 hours to 8 days. Patients with known epilepsy admitted for ictal single-photon emission computed tomography or intracranial EEG monitoring were excluded from analysis. Results: Three patients were diagnosed with clinically significant asystole or bradyarrythmias. One patient (age 5 months) died in the hospital during an event in which she could not be resuscitated. The other two (ages 49 and 76) patients had episodes for 39 and 8 years, respectively and carried the diagnosis of medically refractory epilepsy. Both of these patients had an implantable cardiac defibrillator placed and are event-free. Conclusions: The incidence of patients admitted for video-EEG monitoring who are ultimately diagnosed with asystole and bradyarrythmias not associated with epilepsy is rare. However, the lack of timely diagnosis can be fatal, and epileptologists should recognize the possibility of significant cardiac arrythmias being misinterpreted as seizures. (Supported by Mayo Clinic Foundation grant to JJS NIH 3 M01 RR00997-20S3 to JJS.) Author Affiliation: (1)Department of Psychiatry, Field of Social and Behavioral Medicine, Kagoshima University Graduate School of Medical and Dental, Kagoshima, Kagoshima, Japan (2)Neurosurgery, Tanaka Neurosurgical Clinic, Kagoshima, Kagoshima, Japan (3)Department of Neurosurgery, Kagoshima University Graduate School of Medical and Dental, Kagoshima, Kagoshima, Japan