Twice-daily dosing of valproate with divalproex.

Equivalent doses of enteric-coated divalproex were substituted for valproic acid in 15 epileptic patients who were on a three- or four-times-a-day dosing schedule. After 2 wk the dosing regimen was changed to twice-daily divalproex dosing, and plasma levels were determined during the 12-hr period after the morning dose. Peak absorption was reached at 4 hr; extended plateaus were noted thereafter. The mean fluctuation between low and high plasma values for the group was 46%, with a decrease of less than 50% of peak levels in 9 of the 15 patients at a sampling time just before the second dose. Breakthrough seizures did not occur as a result of twice-daily dosing.

Evoked potentials in olivopontocerebellar atrophy.

Pattern-reveral visual evoked potentials, far-field and cortical somatosensory evoked potentials, and auditory brainstem potentials were recorded in two patients with olivopontocerebellar atrophy. In one patient, visual evoked potentials exhibited prolonged latency and interocular latency differences in the absence of clinical visual dysfunction. Median and tibial nerve evoked cortical potentials were severely attenuated in the absence of somatosensory deficit or peripheral nerve slowing. The far-field somatosensory potentials, however, were well preserved. All components of the auditory brain-stem potentials had latencies within normal limits. In the other, more severely afflicted, patient, all visual, somatosensory, and auditory evoked potentials were abnormal.

Auditory brain-stem potentials with unilateral pontine hemorrhage.

Although there have been extensive anatomical and physiological studies in animals, the actual neural sources, or even the laterality, of some components of auditory brain-stem evoked potentials in humans are uncertain. We studied these responses in a 56-year-old patient who had a clearly demarcated pontine hemorrhage on the right side. The patient was somnolent, with dense left hemiplegia and signs of involvement of right cranial nerves V, VI, and VII. Stimulation of the left ear (ie, contralateral to the lesion) evoked a normal series of waves with clearly resolved positive components peaking at 2.0, 3.3, 4.8 (wave IV), and 6.0 ms (wave V). Stimulation of the right ear (ie, ipsilateral to the lesion) evoked only waves I, III, and IV. These results suggest that a pathway ipsilateral to the stimulated ear is necessary and sufficient for generation of auditory wave V and that wave IV is generated in bilateral pathways.

Enhanced auditory postauricular evoked responses after corticobulbar lesions.

The auditory postauricular response is an evoked "microreflex" that is generated in the posterior auricular muscle. This potential, which has two prominent components at latencies of 12 and 16 msec, is best recorded with neck extensor or facial muscles under tension, but it is not recorded if scalp muscles are relaxed. We describe markedly enlarged postauricular responses after upper motor neuron lesions above the foramen magnum.

Effect of gamma-vinyl GABA on human pattern evoked visual potentials.

We studied the effects on human visual evoked potentials of gamma-vinyl GABA, an anticonvulsant drug that increases cerebral levels of gamma-aminobutyric acid. The subjects were 15 epileptic patients undergoing a clinical trial of this drug. Serial recordings were made from each patient periodically for 1 year. The stimulus was a reversing checkerboard pattern with a check size of 50 minutes. Plasma levels of other antiepileptic medications remained constant throughout the study. No changes, other than a normal variation of +/- 5 msec, were observed in peak latency throughout the study. These results indicate that there are no GABA-ergic effects on the P100 component evoked by large checkerboard pattern-reversal.

Cognitive evoked potentials and disorders of recent memory.

P3 potentials evoked by spoken words were recorded from seven patients with disorders of recent memory and seven age-matched controls. Inclusion criteria included a clearly identifiable P3 to pure tone stimuli. Absence or marked decrement of P3 to word stimuli was seen in the patients even though they could perform the word task. The results are consistent with an encoding disorder and failure in elaboration of information processing.

Cholinergic modulation of human P3 event-related potentials.

A paradigm for studying relations between behavioral, neurochemical, and electrophysiologic processes is presented. Scopolamine, a centrally acting anticholinergic agent, abolished the human auditory P3 event-related potential, but had no measurable effect on the pattern reversal visual evoked potential or alpha rhythm. Recent memory was significantly impaired, but assessment of digit span, reaction time, and distant memory showed no impairment. Physostigmine, an anticholinesterase, restored the P3 and reversed the recent memory impairment. These results strongly suggest a cholinergic role in generation of P3 potential and support the concept that P3 generation is related to memory processes.

Valproate tremors.

We made accelerometric recordings of the tremor induced by valproic acid. The tremor was similar to essential tremor and appeared within a month of starting therapy. It was present at rest and exacerbated by action or antigravity positioning. There was no close correlation of tremor severity and plasma valproate level, but the tremor usually appeared at dosages greater than 750 mg per day. This tremor has appeared in 20 of 25 patients recently studied. In some patients the tremor is markedly active; however, others note only minimal tremor activity.

Valproic acid and plasma levels of phenobarbital.

During concurrent administration of phenobarbital and valproic acid, phenobarbital plasma concentrations often increase. This often requires a reduction of phenobarbital dosage. In normal cats and patients with epilepsy, we found no evidence of decreased renal excretion of phenobarbital. Metabolic studies in four patients revealed a decrease in the conversion of phenobarbital to hydroxyphenylphenobarbital and decreased urinary ratios of hydroxyphenylphenobarbital to phenobarbital. These data suggest that phenobarbital metabolism is inhibited by therapeutic plasma levels of valproic acid.

Treatment of valproate tremors.

Chronic valproate therapy induces symptomatic tremor in about 10% of patients. We studied the effects of propranolol, amantadine, diphenhydramine, benztropine, and cyproheptadine on these tremors in 19 patients by using serial accelerometric recordings. Propranolol was clearly the most therapeutic. Amantadine was moderately effective, but cyproheptadine, diphenhydramine, and benztropine gave little or no relief.

Effect of valproic acid on spike and wave discharges in patients with absence seizures.

Twenty-five patients with absence seizures were treated with valproic acid in doses from 17 to 62.5 mg per kilogram per day. Nineteen patients experienced reduction of spike and wave discharges; in 11 it was greater than 75 percent. Twenty-one patients had a reduction of the total time of spike and wave discharge. Four patients had increase of spike and wave discharges. Nineteen patients had fewer absence seizures. There was no correlation between plasma concentration of valproic acid and EEG change, but clinical improvement occurred when plasma levels of valproic acid reached 50 to 60 microgram per milliliter.

Comparison of valproic acid and phenytoin in newly diagnosed tonic-clonic seizures.

Sixty-one newly diagnosed epileptic patients with generalized tonic-clonic, clonic, or tonic seizures were randomly allocated to treatment with valproate (VPA) and phenytoin (PHT). After 6 months, both drugs had been effective in preventing recurrence of seizures. Seventy-three percent of patients receiving VPA and 47% of patients receiving PHT had no recurrences. Side effects of either drug were mild. Laboratory abnormalities were similar for both drugs. Except for one PHT patient with toxic hepatitis, therapy was not discontinued.

Treatment of epilepsy by stimulation of the vagus nerve.

We treated 14 patients with medically refractory partial seizures by stimulation of the vagus nerve in two single-blind pilot studies. Patients received stimulation through an implantable, programmable NeuroCybernetic Prosthesis, consisting of a pulse generator and a lead-electrode assembly. The mean reduction in seizure frequency after 14 to 35 months of vagal stimulation was 46.6%. Of the 14 patients, five (35.7%) had a 50% or greater reduction in seizure frequency. Two patients, one of whom had had 10 to 100 seizures per day before stimulation, have been seizure-free for over 1 year. Adverse events were primarily limited to initial hoarseness and a tingling sensation at the electrode site in the neck when the device was activated. Most patients tolerated the device and stimulation well. There were no permanent adverse events. Some cases of medically refractory partial seizures are improved by vagal stimulation.

Effects of gamma-vinyl-GABA on the human electroencephalogram.

Gamma-vinyl-GABA (GVG) is a new anticonvulsant drug that significantly raises the level of the gamma-aminobutyric acid in the brain (GABA). The effects of gamma-vinyl-GABA on the human electroencephalogram were studied to assess the role of GABAergic mechanisms on electrocortical activity. Serial EEGs were recorded in 15 epileptic patients undergoing a controlled clinical trial of gamma-vinyl-GABA. The effects of gamma-vinyl-GABA on alpha, beta, or theta activity, sleep spindles and epileptiform activity were studied. No changes could be detected in any of the intrinsic brain rhythms. Three patients showed a mild amelioration of epileptiform activity; no increase in epileptiform activity was seen.

Prolonged survival of cultured keratinocyte allografts in the nonimmunosuppressed mouse.

The effect of in vitro culture on the survival of allografts of epidermal keratinocytes has been examined using a mouse model. Female BALB/c tail epidermal cells were cultured from single cell suspensions to form confluent sheets that were grafted onto male CBA recipients using a transplantation technique that ensured separation of donor graft from host skin. Animals were killed at defined intervals, and the status of the grafts determined histologically. Full thickness skin allografts rejected at 13-15 days. Allografts of epidermis (obtained by enzymatic cleavage at the dermoepidermal junction) rejected at 19-20 days. Cultured keratinocyte allografts were not rejected at least within 70 days and had a histological appearance identical to syngeneic controls. The expression of MHC class I and class II determinants and the leukocyte common (Ly5) surface marker on the donor cells before and after culture were examined using indirect immunofluorescence and monoclonal antibodies. These and other cytochemical studies showed that freshly dissociated tail epidermal cells contained 0.3% of cells that expressed membrane-bound ATP-ase activity, Ia antigens, and the Ly5 surface antigen. These are the Langerhans' cells of the epidermis. In culture, these cells decrease so that by day 8 of culture, no such cells can be detected. At confluence, there are no Ia positive cells, but all cells express MHC class I antigens and stain with an antikeratin antibody. The loss of Ia expression on culture correlates with a decreased stimulation of the class II H-2d-restricted T cell clone D7.1 by cultured keratinocytes compared with freshly dispersed epidermal cells. Furthermore, cultured keratinocytes fail to prime allogeneic mice as determined by the survival of whole thickness skin grafts, whereas freshly dispersed cells induce an accelerated rejection. The results suggest that the survival of cultured keratinocyte allografts is due to the elimination of cells expressing Ia antigens and supports the passenger leukocyte theory of graft rejection.

Homocysteine induced convulsions: enhancement by vitamin B6 and inhibition by hydrazine.

The effects of pyridoxal phosphate, pyridoxine and hydrazine were studied on homocysteine-induced seizures in mice. Both of the B6 vitamers significantly decreased the latency and increased the severity, lethality and duration of seizures induced by homocysteine. The B6 inhibitor hydrazine sulfate, which is normally a convulsant, prevented the tonic component of the convulsions and increased the latency to the clonic component. This experiment indicates that a vitamin B6 dependent step is critically involved in the metabolic changes which precede homocysteine seizures.

In vivo uptake of valproic acid into brain.

The pharmacokinetics of valproic acid (VPA) penetration into the central nervous system of cats were studied. VPA levels in cortical gray matter and plasma were measured at timed intervals after rapid intravenous drug infusion. Brain uptake of the drug was maximal at 1 min postinfusion and decayed rapidly with a mean elimination half-life of 41 min. After a rapid distribution phase, plasma VPA levels remained stable for 90 min. The brain:plasma ratio was maximal at 1 min and also declined rapidly. The volume of distribution was 0.125 1/kg. The small volume of distribution, low brain:plasma ratios and rapid clearance from brain indicate that VPA is not significantly bound in cerebral cortex after a single dose.

Neurochemical effects of vagus nerve stimulation in humans.

An implanted stimulating device chronically stimulated the left cervical vagus nerve in epileptic patients. Cerebrospinal fluid concentrations of free and total gamma-aminobutyric acid, homovanillic acid, 5-hydroxyindoleacetic acid, aspartate, glutamate, asparagine, serine, glutamine, glycine, phosphoethanolamine, taurine, alanine, tyrosine, ethanolamine, valine, phenylalanine, isoleucine, vasoactive intestinal peptide, beta-endorphin, and somatostatin were measured before and after 2 months of chronic stimulation in six patients. Significant increases were seen in homovanillic acid and 5-hydroxyindoleacetic acid in three patients, and significant decreases in aspartate were seen in five patients. These changes were associated with a decrease in seizure frequency.

Absence of cortical white matter changes in three patients undergoing long-term vigabatrin therapy.

Chronic administration of the experimental antiepileptic drug vigabatrin (gamma-vinyl GABA) to animals has been shown to cause dose-dependent neuropathological changes characterized by a microvacuolation in specific white matter tracts. This finding has led to some concern as to whether similar pathologic changes might occur in patients taking this medication. Here we report on analysis of tissue specimens taken during neurosurgery from three patients undergoing chronic vigabatrin therapy (4 g/day). The first patient, a 34-year-old woman, had taken vigabatrin for 2 years prior to surgery, the second, a 50-year-old man, had taken the drug for 1 year, and a 34-year-old man had taken the drug for 5.3 years. For comparison, similar specimens were taken from three other patients not taking vigabatrin who were undergoing surgery for intractable epilepsy. Specimens from each subject were prepared in an identical manner and examined with light and electron microscopy. All specimens were examined in a blinded fashion. There was some minor nonspecific myelinic splitting seen in both controls and vigabatrin-treated patients but there was no evidence for any drug-induced lesions similar to that seen in experimental animals.