Valproate in children with newly diagnosed idiopathic generalized epilepsy

OBJECTIVES

Sparse information on dose-response characteristics for initial antiepileptic drug monotherapy in children with idiopathic generalized epilepsy (IGE) is available. The aim of this study is to characterize the therapeutic dose of valproate in children with newly diagnosed IGE.

MATERIALS AND METHODS

Effect of initial valproate monotherapy and doses associated with seizure freedom were examined in consecutive children with IGE identified from a New Onset Seizure Clinic.

RESULTS

Of 84 patients identified, 48 (57%) became seizure-free on valproate monotherapy and another 10 patients became seizure-free but discontinued VPA because of adverse effects. The mean dose in seizure-free children was 15.7 mg/kg/day and over 95% of IGE patients will respond below 25 mg/kg/day.

CONCLUSIONS

Half of children became seizure-free on valproate monotherapy and did so at modest doses.

Use of topiramate in localization-related epilepsy in children

Topiramate has been shown to be safe and effective in refractory partial epilepsy in children. Pharmacokinetic studies show that the clearance of topiramate is greater in children than in adults; therefore, higher doses may be needed in children than adults. It is generally well tolerated, except for cognitive dysfunction. Weight loss and the risk of renal stones can be significant in some cases. However, when compared with other anticonvulsant medications, topiramate has few serious idiosyncratic reactions such as rash, hematologic reactions, and hepatotoxicity.

Advances in our understanding of early childhood epilepsies: 1999-2000

Epilepsy is frequent in early childhood, but poorly understood. Management of very young children with refractory epilepsy is particularly challenging, but recent advances in classification, medical therapy, and surgical treatment are helping. We summarize work in each area, focusing on developments in the past 2 years. Classification schemes unique to the needs of very young children will likely improve our ability to accurately diagnose and treat those with refractory seizures. Medication studies are now being done that address some of the specific epilepsy syndromes seen in infants and young children, with promising results in some circumstances. Progress is also being made in the identification of good candidates for early, effective epilepsy surgery. For a variety of reasons and incentives, it is likely that research focusing on infants and young children will continue at a brisk pace.

Efficacy, pharmacology, and adverse effects of antiepileptic drugs

This article discusses the factors involved in the appropriate selection of anticonvulsant medications. The clinical use of commonly used traditional antiepileptic drugs and the newly marketed antiepileptic drugs is discussed. This includes the specific indications for use, adverse effects, and dosing of each drug. Drug interactions, mechanisms of action, and pharmacological properties of each drug is also reviewed.

Structure-activity studies of fluoroalkyl-substituted gamma-butyrolactone and gamma-thiobutyrolactone modulators of GABA(A) receptor function

Dihydro-2(3H)-furanones (gamma-butyrolactones) and dihydro-2(3H)-thiophenones (gamma-thiobutyrolactones) containing fluoroalkyl groups at positions C-3, C-4, and C-5 of the heterocyclic rings were prepared. The anticonvulsant/convulsant activities of the compounds were evaluated in mice. Brain concentrations of the compounds were determined and the effects of the compounds on [35S]-tert-butylbicyclophosphorothionate ([35S]TBPS) binding to the picrotoxin site on GABAA receptors were investigated. The effects of the compounds on GABAA receptor function were studied using electrophysiological methods and cultured rat hippocampal neurons. Fluorination at C-3 results in either subtle or pronounced effects on the pharmacological activity of the compounds. When hydrogens are replaced with fluorines at the methylene carbon of an ethyl group, as in 3-(1,1-difluoroethyl)dihydro-3-methyl-2(3H)-furanone (1), the anticonvulsant actions of the compound are not much changed from those found for the corresponding alkyl-substituted analogue. In marked contrast, fluorination at the methyl carbon of the ethyl group, as in dihydro-3-methyl-3-(2,2,2-trifluoroethyl)-2(3H)-furanone (3), produces a compound having convulsant activity. This convulsant activity seems to be due to an increased affinity of the compound for the picrotoxin site on GABAA receptors caused by an interaction that involves the trifluoromethyl group. Results obtained with gamma-butyrolactones containing either a 3-(1-trifluoromethyl)ethyl or a 3-(1-methyl-1-trifluoromethyl)ethyl substituent indicate that the interactions of the trifluoromethyl group with the picrotoxin binding site are subject to both stereochemical and steric constraints. Sulfur for oxygen heteroatom substitution, as in the corresponding gamma-thiobutyrolactones, affects the type (competitive, non-competitive, etc.) of binding interactions that these compounds have with the picrotoxin site in a complex manner. Fluorination of alkyl groups at the C-4 and C-5 positions of gamma-butyrolactones having convulsant activity increases convulsant potency.

Dual modulation of the gamma-aminobutyric acid type A receptor/ionophore by alkyl-substituted gamma-butyrolactones

Alkyl-substituted gamma-butyrolactones (GBLs) and gamma-thiobutyrolactones exhibit convulsant or anticonvulsant activity, depending on the alkyl substituents. alpha-Substituted lactones with small alkyl substituents are anticonvulsant and potentiate gamma-aminobutyric acid (GABA)-mediated chloride currents, whereas beta-substituted compounds are usually convulsant and block GABAA currents. We have now found that this distinction is not so clear-cut, in that some compounds can both block and augment GABAA currents, but with different time courses. For example, alpha,alpha-diisopropyl-GBL (alpha-DIGBL) potentiates exogenous GABA currents in cultured rat hippocampal neurons but diminishes GABA-mediated inhibitory postsynaptic currents. A more detailed analysis demonstrates a triphasic effect of alpha-DIGBL on GABA currents, with a rapid inhibitory phase, a slower potentiating phase, and then an "off response" when the GABA/alpha-DIGBL perfusion is stopped. Thus, alpha-DIGBL can inhibit and potentiate GABA currents with kinetically different time courses. Inhibition is more rapid, but at steady state potentiation dominates. Using a simplified model of the GABAA receptor/ionophore, we have simulated our experimental observations with alpha-DIGBL. Another lactone, beta-ethyl-beta-methyl-gamma-thiobutyrolactone, also has dual actions, with inhibition predominating at low concentrations and potentiation predominating at high concentrations. We propose two distinct GBL modulatory sites on the GABAA receptor, i.e., an inhibitory "picrotoxin" site and an enhancing "lactone site." New information on the structure of the GABAA receptor/ionophore may allow the molecular dissection of these two sites.

Developmental alteration in GABAA receptor structure and physiological properties in cultured cerebellar granule neurons

Although we now have extensive knowledge about the GABAA receptor subunits determining benzodiazepine modulation of channel function, little is known about subunits influencing other modulatory sites on the GABAA receptor-chloride channel complex. We have identified a developmental change in subunit composition of the GABAA receptor in cultured cerebellar granule neurons that eliminates benzodiazepine-mediated enhancement of GABA responses and alters modulation by a substituted gamma-butyrolactone. Based on data from sequential PCR experiments, we mimicked the functional properties of early and mature receptors with heterologous expression of specific subunit combinations. This report describes one of the most extensive cell- and site-specific developmental changes for an ion channel seen to date.

Alpha-spirocyclopentyl- and alpha-spirocyclopropyl-gamma-butyrolactones: conformationally constrained derivatives of anticonvulsant and convulsant alpha,alpha-disubstituted gamma-butyrolactones

To further study the putative gamma-butyrolactone site of the GABAA/chloride channel complex, constrained derivatives of convulsant and anticonvulsant alpha,alpha-disubstituted gamma-butyrolactones (alpha-spirocyclopropyl- and alpha-spirocyclopentyl-gamma-butyrolactones) were synthesized and evaluated biologically. Most of the spirocyclopropyl agents were anticonvulsants when tested against pentylenetetrazole-induced seizures in mice. These agents effectively displaced 35[S]-tert-butylbicyclophosphorothionate (35[S]-TBPS), a ligand for the picrotoxin binding site of the GABAA/chloride channel, from rat neuronal membranes and affected the GABA-mediated current in hippocampal neurons. The monomethyl-substituted spirocyclopropyl agent with a methyl group cis to the carbonyl (15) potentiates GABA-induced current whereas the trans derivative (16) blocks the current. The only anticonvulsant in the spirocyclopentyl series was the unsubstituted spirocyclopentyl compound 2. All the other substituted spirocyclopentyl targets were inactive in vivo at the highest dose tested except for convulsant 9, which has a trans 2,5-dimethyl-substituted cyclopentyl ring. All the spirocyclopentyl derivatives displaced 35[S]-TBPS from rat neuronal membranes very effectively, and they also all potentiated GABA-induced chloride current except for convulsant 9 which blocked the current. From the data obtained in this investigation, it appears that when the volume occupied above and below the lactone ring is as large as that occupied by spirocyclopentyl agent 9, convulsant activity is observed. Groups with less volume in these areas either are inactive in the behavioral test or have anticonvulsant activity. When bound to the GABAA/chloride channel, the larger molecules may stabilize the closed state of the channel whereas the smaller molecules may stabilize the open state.

Alkyl-substituted gamma-butyrolactones act at a distinct site allosterically linked to the TBPS/picrotoxinin site on the GABAA receptor complex

Effects of alkyl-substituted gamma-butyrolactones and gamma-thiobutyrolactones on [35S]t-butylbicyclophosphorothionate (35S-TBPS) dissociation from the picrotoxinin receptor were studied. Unlike picrotoxinin, these lactones accelerated the dissociation rate of 35S-TBPS. Thus, previous reports that these lactones change the Kd but not the Bmax of 35S-TBPS in equilibrium binding experiments is explained not by competitive inhibition, but by an allosteric interaction with the 35S-TBPS binding site. These results indicate that modulatory effects of alkyl-substituted gamma-butyrolactones may result from their action at a distinct site on the gamma-aminobutyric acid (GABA)A receptor.

Relative anticonvulsant effects of GABAmimetic and GABA modulatory agents

Anticonvulsant properties of compounds that enhance GABA-mediated inhibition through modulatory sites on the GABAA receptor [phenobarbital (PB), clonazepam (CZP), alpha-ethyl-alpha-methyl-gamma-thiobutyrolactone (alpha-EMTBL)] were compared with anticonvulsant effects of compounds believed to be antagonists at these modulatory sites (Ro15-1788 and alpha-isopropyl-alpha-methyl-gamma-butyrolactone gamma-IMGBL)] and to 4,5,6,7-tetrahydroisoxazolo-[4,5-c]-pyridin-3-ol (THIP, GABAA receptor agonist), (+/-) baclofen (GABAB receptor agonist), and gamma-vinyl GABA, a compound that increases endogenous GABA. The compounds were tested for their ability to block experimental seizures caused by maximal electroshock, pentylenetetrazol, picrotoxin, methyl-6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate (DMCM), bicuculline (BIC), aminophylline, strychnine, and t-butyl-bicyclophosphorothionate (TBPS) in mice. CZP blocked all but strychnine seizures. PB was also highly effective, blocking all but TBPS seizures. alpha-EMTBL, representing a new class of experimental anticonvulsant drugs, prevented all seizures except strychnine (STR)- and aminophylline-induced seizures. The antagonists are effective only against one convulsant stimulus. Ro15-1788 and alpha-IMGBL prevented only DMCM- and pentylenetetrazol (PTZ)-induced seizures, respectively. THIP and gamma-vinyl GABA both blocked only BIC and picrotoxin seizures. Baclofen had no anticonvulsant activity. These data demonstrate that compounds that increase neuronal inhibition by potentiating the action of GABA have a broader spectrum of anticonvulsant action than either antagonists or GABAmimetic agents or compounds that increase endogenous GABA.

Synthesis and structure-activity studies of alkyl-substituted gamma-butyrolactones and gamma-thiobutyrolactones: ligands for the picrotoxin receptor

A series of gamma-butyrolactones and gamma-thiobutyrolactones possessing a variety of alkyl groups and alkyl-substitution patterns was prepared and evaluated for anticonvulsant and convulsant activity. Behavioral studies performed on these compounds suggest that maximal anticonvulsant activity (against maximal electroshock and pentylenetetrazol) results when three or four carbon atoms are present at the alpha-position. For convulsant potency, a similar dependence on the size of the alkyl chain at the beta-position was observed. Additional gamma-dimethyl groups were found to increase the convulsant potency of a beta-substituted compound and to cause an alpha-substituted anticonvulsant to become a convulsant. In general, sulfur for oxygen heteroatom substitution in the alpha-substituted lactones resulted in improved anticonvulsant potency and spectrum of activity. Binding of these compounds to the picrotoxin site of the GABA receptor complex was demonstrated with a [35S]-tert-butylbicyclophosphorothionate radioligand binding assay. Measurements of brain concentrations for selected compounds supports a hypothesis that correlates binding to the picrotoxin site with the pharmacological effects of these compounds.

Gamma-butyrolactone antagonism of the picrotoxin receptor: comparison of a pure antagonist and a mixed antagonist/inverse agonist

Multiple receptors modulate the ion channel gated by the inhibitory neurotransmitter gamma-aminobutyric acid (GABA). gamma-Butyrolactones and gamma-thiobutyrolactones are compounds that act at the picrotoxin recognition site on the GABA receptor complex as either agonists or inverse agonists, depending on the nature of the alkyl substituents. Here we have compared the effects of two gamma-butyrolactones, alpha-ethyl-alpha-methyl-gamma-butyrolactone (alpha EMGBL) and alpha-isopropyl-alpha-methyl-gamma-butryolactone (alpha IMGBL), on GABA currents and inhibitory postsynaptic currents (IPSCs) in cultured, voltage-clamped, rat hippocampal neurons. alpha EMGBL also decreased the rate of IPSC decay without altering IPSC peak amplitude. At higher GABA concentrations (30 microM), alpha EMGBL has already been shown to block picrotoxin receptor agonists and inverse agonists. Thus, alpha EMGBL is a mixed antagonist/inverse agonist. In contrast to alpha EMGBL, alpha IMGBL had no effect on responses to either 0.5 or 30 microM GABA or on IPSCs, but it was able to block the effects of picrotoxin receptor agonists and inverse agonists. Therefore, alpha IMGBL is the first pure antagonist to be described for the picrotoxin receptor. The main conductance state of the GABA-gated channel probably has two or more open states, brief openings associated with binding of a single GABA molecule and longer openings due to the binding of two GABA molecules. We were able to simulate the results obtained with alpha EMGBL, using a computer model, by assuming that alpha EMGBL altered only the opening and closing rate constants for the monoliganded open channel of the GABA receptor. In addition to having site-selective actions, these results suggest that drugs modulating the GABA-linked chloride ionophore may be specific for the kinetic state of the GABA-gated channel.

Binding interactions of convulsant and anticonvulsant gamma-butyrolactones and gamma-thiobutyrolactones with the picrotoxin receptor

Alkyl-substituted gamma-butyrolactones (GBLs) and gamma-thiobutyrolactones (TBLs) are neuroactive chemicals. beta-Substituted compounds are convulsant, whereas alpha-alkyl substituted GBLs and TBLs are anticonvulsant. The structural similarities between beta-alkyl GBLs and the convulsant picrotoxinin suggested that alkyl substituted GBLs and TBLs act at the picrotoxin receptor. To test this hypothesis we examined the interactions of convulsant and anticonvulsant GBLs and TBLs with the picrotoxin, benzodiazepine and gamma-aminobutyric acid (GABA) binding sites of the GABA receptor complex. All of these convulsants and anticonvulsants studied competitively displaced 35S-t-butylbicyclophosphorothionate (35S-TBPS), a ligand that binds to the picrotoxin receptor. This inhibition of 35S-TBPS binding was not blocked by the GABA antagonist bicuculline methobromide. The convulsant GBLs and TBLs also partially inhibited [3H]muscimol binding to the GABA site and [3H]flunitrazepam binding to the benzodiazepine site, but they did so at concentrations substantially greater than those that inhibited 35S-TBPS binding. The anticonvulsant GBLs and TBLs had no effect on either [3H]muscimol or [3H]flunitrazepam binding. In contrast to the GBLs and TBLs, pentobarbital inhibited TBPS binding in a manner that was blocked by bicuculline methobromide, and it enhanced both [3H]flunitrazepam and [3H]muscimol binding. Both ethosuximide and tetramethylsuccinimide, neuroactive compounds structurally similar to GBLs, competitively displaced 35S-TBPS from the picrotoxin receptor and both compounds were weak inhibitors of [3H] muscimol binding. In addition, ethosuximide also partially diminished [3H]flunitrazepam binding. These data demonstrate that the site of action of alkyl-substituted GBLs and TBLs is different from that of GABA, barbiturates and benzodiazepines. We suggest that the GBLs and TBLs act at the picrotoxin receptor.

Physiological regulation of the picrotoxin receptor by gamma-butyrolactones and gamma-thiobutyrolactones in cultured hippocampal neurons

We examined the effects of alkyl-substituted gamma-butyrolactones (GBLs), and gamma-thiobutyrolactones (TBLs) on GABA currents in cultured, voltage-clamped rat hippocampal neurons. Convulsant GBLs and TBLs reversiby diminished GABA responses in a concentration-dependent manner. beta-Ethyl-beta-methyl GBL (beta-EMGBL) completely abolished GABA responses at 3 mM (IC(50)390 microM), while TBL and beta-ethyl-beta-methyl TBL (beta-EMTBL)-induced inhibition of GABA currents was incomplete, saturating at about 50% of control at 300 microM and 10 mM for beta-EMTBL and TBL, respectively. beta-EMGBL and beta-EMTBL both increased the rate of decay of inhibitory post-synaptic currents (IPSCs) and beta-EMGBL also decreased IPSC peak amplitude. In contrast, the anticonvulsant alpha-ethyl-alpha-methyl TBL (alpha-EMTBL) potentiated GABA currents at all GABA concentrations tested; maximal potentiation was 190% of control at 1 mM alpha-EMTBL (EC50 102 microM). Another anticonvulsant alpha-ethyl-alpha-methyl GBL (alpha-EMGBL), potentiated responses to low (0.5 microM) but not high (greater than or equal to 10 microM) GABA. It also blocked the inhibitory effects of picrotoxin and beta-EMGBL and the facilitative effect of alpha-EMTBL on responses to 30 microM GABA. alpha-EMGBL did not interfere with other agents which augment GABA currents. Both alpha-EMTBL and alpha-EMGBL decreased the rate of IPSC decay without altering IPSC peak amplitude. None of these compounds had any direct membrane effects. We propose that beta-alkyl GBLs diminish GABA currents, and therefore, we hypothesize that these compounds are picrotoxin receptor agonists. beta-Alkyl TBLs partially diminish GABA currents and may be partial agonists.(ABSTRACT TRUNCATED AT 250 WORDS)

Physiological modulation of the GABA receptor by convulsant and anticonvulsant barbiturates in cultured rat hippocampal neurons

The actions of convulsant and sedative barbiturates on responses to gamma-aminobutyric acid (GABA) application and on inhibitory postsynaptic currents were compared using voltage-clamp techniques in cultured rat hippocampal neurons. The convulsant barbiturates, 5-ethyl-5-(3-methylbut-2-enyl) barbituric acid (3M2B), and (+)-5-ethyl-5-(1,3-dimethylbutyl) barbituric acid [+)-DMBB), and the sedative barbiturate, 5-ethyl-5-(3-methylbutyl) barbituric acid (3MB), all potentiated GABA-mediated chloride currents. In addition, these compounds prolonged the duration of GABAergic inhibitory postsynaptic currents. The similarity between the action of convulsant and sedative barbiturates suggests that the convulsant activity of 3M2B and (+)-DMBB are not mediated by their actions at GABAergic synapses.

Convulsant and anticonvulsant cyclopentanones and cyclohexanones

The convulsant and/or anticonvulsant activity of unsubstituted and mono-alkyl-substituted cyclopentanones and cyclohexanones were examined by testing the ability of these compounds to produce seizures or to inhibit seizures induced by pentylenetetrazol and maximal electroshock in CF-1 mice. In addition, these compounds were tested for their ability to bind to the picrotoxin receptor. The unsubstituted compounds, cyclopentanone and cyclohexanone, prevented both pentylnetetrazol- and maximal electroshock-induced seizures. Cyclopentanones and cyclohexanones with small (less than 3 carbon atoms) alkyl substituents in the 2-position were also anticonvulsant; all of these compounds, except 2-ethylcyclohexanone, blocked both pentylenetrazol- and maximal electroshock-induced seizures. 2-Ethylcyclohexanone was very effective against pentylenetetrazol seizures but did not prevent maximal electroshock seizures. Cyclohexanones with larger alkyl substituents in the 2-position, 2-propylcyclohexanone and 2-t-butylcyclohexanone, caused clonic seizures following injection into mice. Of the cyclopentanones and cyclohexanones with alkyl substitutions in the 3-position that were studied, one was an anticonvulsant (3-methylcyclopentanone), one was a mixed convulsant/anticonvulsant (3-ethylcyclohexanone), and the other two (3-ethylcyclopentanone and 3-t-butylcyclohexanone) were convulsants. Finally, two cyclohexanones with alkyl substituents in the 4-position were studied. Both 4-ethylcyclohexanone and 4-t-butylcyclohexanone produced convulsions when injected into mice. All the neuroactive cyclopentanones and cyclohexanones competitively displaced [35S]t-butylbicyclophosphorothionate, a ligand specific for the picrotoxin receptor, from rat brain membranes. The convulsant compounds were generally more potent than the anticonvulsants. The cyclohexanones were more potent than their corresponding cyclopentanones and the binding potency of both increased as the size of the alkyl substituent increased. These results suggest that cyclopentanone, cyclohexanone, and their alkyl-substituted derivatives act at the picrotoxin receptor to increase or decrease neuronal activity. Thus, they appear to have sites and mechanisms of action similar to those of the neuroactive gamma-butyrolactones and gamma-thiobutyrolactones.

Comparison of the anticonvulsant activities of ethosuximide, valproate, and a new anticonvulsant, thiobutyrolactone

Anticonvulsant properties of alpha-ethyl-alpha-methyl-gamma-thiobutyrolactone (alpha-EMTBL) were compared with those of the antiepileptic drugs ethosuximide (ESM) and valproate (VPA) by testing their ability to block seizures in mice caused by maximal electroshock (MES), pentylenetetrazol (PTZ), picrotoxin (PICRO), bicuculline (BIC), methyl-6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate (DMCM), N-methyl-D,L-aspartate (NMDA), aminophylline (AMPH), strychnine (STR), beta-ethyl-beta-methyl-gamma-thiobutyrolactone (beta-EMTBL), and t-butylbicyclophosphorothionate (TBPS). ESM was able to prevent PTZ-, PICRO-, DMCM-, and beta-EMTBL-induced seizures. In contrast, VPA and alpha-EMTBL blocked all of these plus MESTBPS-, and BIC-induced convulsions. Only VPA prevented AMPH-induced seizures. None of the anticonvulsants blocked STR or NMDA seizures. Rotorod testing for acute neurotoxicity demonstrated that ESM was the least toxic and alpha-EMTBL and VPA were equivalent. Animals treated daily with high doses of alpha-EMTBL for a 2-week period appeared healthier and had a higher survival rate than animals treated with VPA in the same manner. After a single intraperitoneal (i.p.) injection, the duration of anticonvulsant action of alpha-EMTBL was 1.3 and 4 times longer than that of ESM and VPA, respectively. These results indicate that alpha-EMTBL has a wide spectrum of anticonvulsant action like VPA but may be less toxic and longer acting. We suggest that alpha-EMTBL is a compound worthy of further testing and development as an antiepileptic drug (AED).

Identification of a median thalamic system regulating seizures and arousal

This study better defines the way in which the thalamus controls expression of experimental generalized seizures. The effects of small intrathalamic injections of the direct GABA agonist muscimol on the thresholds of pentylenetetrazol (PTZ)-induced seizures and on spontaneous behavior were determined in the rat and compared with the effects of injections of gamma-vinyl-GABA (GVG), an irreversible inhibitor of GABA transaminase. Muscimol injections produced neuronal inhibition in a relatively small area of thalamus, whereas GVG injections produced inhibition in a much larger area. Muscimol injections in the midline thalamus in the vicinity of the paraventricular, paratenial, interanteromedial, intermediodorsal, and central medial nuclei facilitated PTZ myoclonic and clonic seizures and also produced sedation. These effects on seizure thresholds were attributable both to a lower PTZ threshold dose for initiation of electroencephalographic (EEG) seizure activity and to an increased probability of this EEG activity being expressed as behavioral seizures. Midline injections located more posteriorly in the thalamus also inhibited tonic seizures. Muscimol injections placed laterally, dorsally, or ventrally to this midline thalamic region had much less effect on behavior or seizures. In contrast, GVG injections in the anterior medial thalamus elevated the threshold for all PTZ seizure types and for associated EEG seizure activity but had little effect on spontaneous behavior. These findings demonstrate the existence of an important seizure regulatory system in the midline of the thalamus and a direct anatomic link between the mechanisms for regulating arousal and seizure production which may help explain the association between sleep and seizure facilitation in humans.(ABSTRACT TRUNCATED AT 250 WORDS)