K. Huh

Comparative cognitive effects of anticonvulsants

We investigated the neuropsychological effects of carbamazepine, phenobarbital, and phenytoin in 15 partial complex epilepsy patients treated with each drug for 3 months, using a randomized double-blind, triple crossover design. Neuropsychological evaluation at the end of each treatment period included Digit Span, Selective Reminding Test, Digit Symbol, Finger Tapping, Grooved Pegboard, Choice Reaction Time, P3 evoked potential, and Profile of Mood States. Employing anticonvulsant blood levels and seizure frequencies as covariates, the only significant difference was for Digit Symbol. Performance with phenobarbital was significantly worse than with the other 2 anticonvulsants despite phenobarbitals having had the lowest overall blood levels. Our data show that patients receiving carbamazepine, phenobarbital, and phenytoin have comparable neuropsychological performance on most measures. The results suggest that the differential cognitive effects of anticonvulsants may be subtle.

Effect of neural transplants on seizure frequency and kindling in immature rats following kainic acid

To study the hypothesis that neural transplantations can alter seizure susceptibility in a chronic animal model of epilepsy 260 immature rats (30- to 32-days-old) were administered a convulsant dosage of kainic acid (KA). Ten days later rats that had severe seizures following KA received either bilateral intracerebroventricular transplants of hippocampal (n = 27), neocortical (n = 29), cerebellar (n = 30), or locus ceruleus (n = 32) tissue, or underwent sham transplantation (n = 66). Spontaneous seizure frequency was assessed for 230 days following which the rats underwent entorhinal kindling. The percentage of rats developing spontaneous recurrent seizures was similar in the 4 transplant groups and the sham-operated controls. Rats receiving hippocampal and locus ceruleus transplants had fewer spontaneous seizures than the sham-operated controls or other transplant groups. However, there were no differences in afterdischarge thresholds or kindling rates in the 5 groups. This study demonstrates that the anticonvulsant effects of neural transplants, using this animal model are mild. Tissue type of the graft appears to be an important variable in the alteration of seizure frequency.

Corpus callosotomy for the treatment of intractable epilepsy in children

Abstract We report the results of anterior two-thirds corpus callosotomies in 20 children undergoing the procedure for intractable seizures. All children recovered from the callosotomy without significant neurological sequelae. Excellent results with regard to seizure frequency were found in 8 of 20 (40%) patients. Following surgery, generalized tonic-clonic seizures, tonic seizures, and atonic seizures were reduced from their preoperative frequency. Although all patients in the postoperative period continued on antiepileptic drugs, there was a significant reduction in the number of drugs used. There was no statistically significant relationship between outcome and age of onset of the seizures, presence or absence of mental retardation, computerized tomography or magnetic resonance imaging findings, presence of hemiplegia, and preoperative EEG abnormalities. Although anterior two-thirds corpus callosotomy may be valuable in reducing seizure frequency in some children, it is currently not possible to predict accurately which children will benefit from the procedure.

Human hippocampal EEg : effects of behavioral activation

We studied spectral components of human hippocampal EEG in relation to behavioral status in 19 patients with intractable complex partial seizures who had depth electrodes implanted into the anterior hippocampi as a part of their preoperative evaluations. Behavioral conditions included: eyes closed resting, eyes open resting, eyes open with a verbal task, and eyes open with a visuospatial task. Hippocampal EEG spectral power uniformly decreased during behavioral activation. EEG activation was quantitatively different among the 3 activated conditions, with the most prominent change occurring during the visuospatial task. The degree of EEG activation corresponded inversely with the site of epileptic focus during the verbal task. The results demonstrate the response of human hippocampal EEG to behavioral activation. The magnitude of EEG change may reflect the degree of functional activation of the given hippocampus.

Attentional mechanisms during the intracarotid amobarbital test

We investigated the integrity of attentional mechanisms following unilateral intracarotid amobarbital injection in 23 patients undergoing preoperative evaluation for epilepsy surgery. Performance for ipsilateral hand-button response to a quasi-random strobe flash was markedly altered following unilateral amobarbital injection as evidenced by decreased correct responses and increased perseverative errors. The increase in perseverations was inversely correlated with subsequent memory performance. The results indicate that unilateral amobarbital injection commonly produces a marked reduction in attention as well as disturbances in strategic control mechanisms.

Effects of neural transplantation on seizures in the immature genetically epilepsy-prone rat

To study the hypothesis that neural transplantations can alter seizure susceptibility in a genetic animal model of epilepsy, 93 pubescent genetically epilepsy-prone rats with stage 9 seizures received either bilateral inferior colliculi (N = 21) or lateral ventricle (N = 42) transplants or sham transplants (N = 30). The grafts consisted of embryonic locus ceruleus, neocortical, or cerebellar tissue. Starting 2 days after the transplantation the rats were subjected to audiogenic stimulations every other day for 61 days. Latency to the running and tonic phase, seizure severity score, and duration of the tonic and clonic phase were compared in the neural transplant and sham-operated controls. Rats that received transplants had a longer latency to the tonic phase and a shorter duration of the clonic phase than the controls. At age 110 days the rats had electrodes implanted bilaterally into the angular bundle and were kindled. No difference in kindling rate was found between the rats that received neural grafts and the sham-operated controls. Cerebrospinal fluid concentration of norepinephrine was not altered by the transplants. This study demonstrates that the anticonvulsant effects of neural transplants, using the genetically epilepsy-prone model of epilepsy, are mild.

Spectral Analysis of Sphenoidal Evoked Potentials Predicts Epileptic Focus

Summary: Sphenoidal evoked potentials (SpEPs) were recorded in 14 patients with intractable temporal lobe epilepsy. Twelve patients had electrographically proven unilateral seizure onset (five left, seven right). Two patients had partial complex seizures and unilateral mesial temporal lobe lesions shown by magnetic resonance imaging (MRI). Thirteen patients subsequently underwent temporal lobectomy. SpEPs were recorded using the P3 tonal oddball paradigm from sphenoidal electrodes referenced to scalp electrodes (i.e., T3, T4, C3, C4, Cz) and were subjected to spectral analysis for whole band (1–12.6 Hz) power. A significant relationship was found for side of seizure focus and reduced spectral power of the sphenoidal‐temporal target SpEPs. In 10 of 14 patients, SpEP spectral power was less from the “epileptogenic” than from the “nonepileptogenic” temporal lobe. Of the remaining 4 patients, 2 had bilaterally equal spectral power, and 2 had lower SpEP power from the nonepilep‐togenic side. Postoperatively, all 13 patients are seizure‐free except for the 2 patients with reduced SpEP power opposite to the ictal onset and 1 patient with fluctuations in SpEP asymmetries. Spectral analysis of SpEPs appears to have utility in the assessment of temporal lobe dysfunction and thus in the preoperative evaluation for epilepsy surgery.

Analysis of a Four-Year Experience with Depth Electrodes and a Two- Year Experience with Subdural Electrodes in the Evaluation of Ablative Seizure Surgery Candidates

Chronically implanted depth and subdural electrodes have both been shown to be satisfactory means of localizing epileptogenic foci. Utilizing bilateral mesial temporal depth electrodes, we have localized mesial temporal foci in a large percentage of patients. Depth electrode investigation of a more limited number of patients with suspected extramesial temporal foci has not been as reliable in giving localizing information. We have more recently used subdural electrodes to investigate this latter category of patients, and preliminary findings suggest that this technique may be of localizing value in several subcategories.

Long-Latency Evoked Potentials During Aura of Temporal Lobe Origin

Long-latency evoked potentials recorded from scalp or scalp/sphenoidal electrodes have been shown to have diminished amplitude and power during interictal recordings on the side ipsilateral to an unilateral mesial temporal lobe (MTL) focus. We now report recordings of long-latency evoked potentials both during a prolonged aura and also on two seizure-free days. P3 amplitude and sphenoidal evoked potential (SpEP) spectral power in the scalp/sphenoidal channels were less, ipsilateral to the MTL focus. During the aura, the SpEPs had reduced power and reduced amplitude. However, the amplitude of the N2 component for the target averages in the CZ-A1A2 channel was markedly enhanced during the aura.

Spectral analysis of human hippocampal EEG: Behavioral activation

Hippocampal EEG was recorded from eight patients with intractable temporal lobe epilepsy who had depth electrodes implanted as part of their presurgical evaluation. The patients all had a mesial temporal seizure onset (5L, 3R) demonstrated by multiple ictal EEG recordings. A bipolar montage consisting of ascending pairs of contacts on the hippocampal electrodes was employed. EEG was recorded during two states: eyes-closed resting and behavioral activation (i.e., eyes-open, paired associate, verbal memory task). Hippocampal EEG power uniformly decreased during behavioral activation. The percent decrease in spectral power was greater for the nonepileptogenic hippocampi than the epileptogenic hippocampi in six of eight patients. The results demonstrate response of the human hippocampal EEG to behavioral activation. Similar techniques may have utility in the preoperative evaluation for epilepsy surgery.