Carbamazepine withdrawal: effects of taper rate on seizure frequency.

We prospectively investigated the effects of rate of carbamazepine (CBZ) withdrawal and CBZ level on seizure type and frequency in 12 epilepsy patients withdrawn completely from antiepileptic drugs prior to entering an investigational monotherapy trial. Patients withdrawn from CBZ rapidly (over 4 days) experienced significantly more generalized tonic-clonic seizures (GTCSs) and GTCS clusters than did those withdrawn slowly (over 10 days). Complex partial seizure (CPS) frequency did not differ between the two groups. CPSs preceded GTCSs, with GTCSs occurring in the majority of patients after CBZ had been discontinued, at subtherapeutic or absent CBZ levels. Two of six patients who had been tapered rapidly and all six patients who had been tapered slowly were able to enter the investigational monotherapy trial.

Activation of language cortex with automatic speech tasks.

OBJECTIVE: To identify automatic speech tasks that reliably demonstrate increased regional cerebral blood flow (rCBF) in Broca's and Wernicke's areas of the cortex using PET. BACKGROUND: Localizing language with direct cortical stimulation mapping requires that patients have a stable baseline on tests that engage eloquent cortex. For dysphasic patients or younger children, automatic speech tasks such as counting are often used in lieu of more complex language tests. Evidence from both lesion and neuroimaging studies suggests that these tasks may not adequately engage language cortices. In this study, we examined rCBF during automatic oromotor and speech tasks of varying complexity to identify those eliciting increased CBF in Broca's and Wernicke's areas. METHODS: Eight normal volunteers underwent PET during rest, tongue movements, and three automatic speech tasks: repeating a phoneme sequence, repeating the months of the year, and reciting a memorized prose passage. Images were averaged across subjects and compared across tasks for regional localization and laterality. RESULTS: Whereas all activation tasks produced increased relative CBF in brain regions that correlated with articulation and auditory processing, only the two tasks that used real words (versus phonemes) showed left-lateralized rCBF increases in posterior superior temporal lobe (Wernicke's area), and only the prose repetition task produced left lateralized activity in Broca's area. CONCLUSIONS: Whereas automatic speech typically does not engage language cortex, repeating a memorized prose passage showed unambiguous activation in both Broca's and Wernicke's areas. These results caution against the use of common automatic speech tasks for mapping eloquent cortex and suggest an alternative task for those with poor language abilities or acquired dysphasia who cannot perform standardized language tests reliably.

A direct comparison of PET activation and electrocortical stimulation mapping for language localization.

Mapping eloquent language cortex in presurgical patients typically is accomplished using highly invasive direct cortical stimulation techniques. Functional imaging during language activation using positron emission tomography (PET) is a promising, noninvasive alternative that requires validation. In seven patients undergoing surgical evaluation for intractable epilepsy, we performed both direct cortical stimulation and PET activation mapping of language cortex using identical tasks. MRI, PET, and CT scans were coregistered to directly compare the location of language centers determined by cortical stimulation versus activation PET. We found that cortical regions that showed increased cerebral blood flow during both visual and auditory naming tasks were located in the same regions as subdural electrodes which disrupted language during electrical stimulation. Cortical regions underlying electrodes that did not disrupt language also showed no consistent changes in regional cerebral blood flow during PET activation. Used cautiously, PET activation produces language maps similar to those obtained with direct cortical stimulation, with more complete brain coverage and considerably less invasion.

Repetitive transcranial magnetic stimulation of the dominant hemisphere can disrupt visual naming in temporal lobe epilepsy patients.

We used repetitive transcranial magnetic stimulation (rTMS) to study visual naming in 14 patients with temporal lobe epilepsy. Ten had left hemisphere language by Wada testing and all experienced speech arrest with rTMS of the motor speech area in the left frontal lobe. One left-hander had speech arrest with stimulation of sites on both sides. Subjects were asked to name pictures or read words presented on a computer monitor. rTMS was delivered on half of the trials. Stimulation sites were the motor speech area in the left frontal lobe, the mirror site on the right, and the left and right mid superior and posterior temporal lobes. rTMS at left hemisphere sites caused more naming errors than did right hemisphere rTMS. All individual subjects, except two who had temporal lobe resections and the one with bilateral speech arrest, produced more naming errors with rTMS of left hemisphere sites. There was no significant effect on word reading. rTMS at the left hemisphere and right frontal sites produced reductions in reaction time for picture naming, but not for word reading. This was observed for both correct and incorrect responses. This study shows that left hemisphere rTMS can disrupt visual naming selectively.

Memory and intelligence in lateralized temporal lobe epilepsy and schizophrenia.

Recent neuroimaging studies of patients with schizophrenia have suggested structural and functional abnormalities of mesial temporal lobe structures. We compared the intelligence and memory test performance of 70 patients with schizophrenia and 72 patients with focal, lateralized temporal lobe epilepsy (30 left, 42 right temporal lobe) in order to examine the adequacy of a temporal lobe model of schizophrenic cognitive deficits. The groups did not differ in age, education, or Full Scale IQ. The right temporal lobe group had better overall memory performance than either the left temporal or schizophrenic patients. Unlike the schizophrenic patients, the memory impairment of the left temporal group was most evident with verbal materials and was amplified by delayed testing. Both epilepsy groups had better visual memory than the schizophrenic group. The clear differences in performance pattern between groups suggests that lateralized temporal lobe dysfunction does not by itself provide an adequate model of schizophrenic cognitive impairment.

Regional cerebral blood flow during auditory responsive naming: evidence for cross-modality neural activation.

One issue of continuing debate in language research concerns whether the brain holds separate representations for semantic information through the auditory vs visual modalities. Regardless of whether we hear, see or read meaningful information, our brains automatically activate both auditory and visual semantic associations to the sensory input. The prominent models for how the brain makes these cross-modality associations holds that semantic information conveyed through either sensory input modality is represented in a shared semantic system comprising the traditionally identified language areas in the brain. A few recent case reports as well as activation imaging studies, have challenged this notion by demonstrating category-specific organization within the semantic system in spatially discrete brain regions. Neither view posits a role for primary sensory cortices in semantic processing. We obtained positron emission tomographic (PET) images while subjects performed an auditory responsive naming task, an auditory analog to visual object naming. Subjects heard and responded to descriptions of concrete objects while blindfolded to prevent visual stimulation. Our results showed that, in addition to traditional language centers, auditory language input produced reciprocal activation in primary and secondary visual brain regions, just as if the language stimuli had entered in the visual modality. These findings provide evidence for a distributed semantic system in which sensory-specific semantic modules are mutually interactive, operating directly onto early sensory processing centers.

Use of positron emission tomography for the evaluation of epilepsy.

After a brief introduction to the theoretic aspects of positron emission tomography, four areas of positron emission tomography research are discussed with an emphasis on current concepts and future directions. The use of positron emission tomography as a tool for the localization of the pathologic brain region and as a predictor of surgical outcome in focal epilepsy is reviewed and compared with the sensitivity, specificity, and outcome predicted by other neuroimaging techniques. Research on positron emission tomography measures of regional metabolism, bloodflow, and neuroreceptors is reviewed from the perspective of epileptic pathophysiology with a special emphasis on elucidative integrative neural circuits involved in epileptic spread and termination. A brief review and discussion of the use of positron emission tomography for the understanding of potential neural reorganization of cognitive processes in epilepsy follows. In the final section, an overview of recently developed methods of positron emission tomography data analysis with a focus on application to research questions in epilepsy is presented.

The noninvasive identification of language function. Neuroimaging and rapid transcranial magnetic stimulation.

Neuroimaging techniques that rely on detecting alterations in blood flow may be used to map the cortical localization of cognitive function during task performance. O-15 water positron emission tomography studies have mapped neural networks that subserve language function. These techniques have been adapted to lateralize and localize language function in patients with intractable epilepsy prior to epilepsy surgery. Functional magnetic resonance (fMR) imaging, relying upon fast MR imaging techniques performed during cognitive tasks, allows localization of language areas in individual adults and children and, because there is no radiation exposure, allows for additional or repeat studies in patients. These noninvasive means of language localization may supplant the invasive means of language lateralization (intracarotid amytal procedure) and localization (corticography), and will allow for the continued study of language organization in health and disease.

Retrieval inhibition in anomia.

Reading latencies for anomic temporal lobe epileptics, nonanomic epileptic patient controls, and normal controls were measured in semantic priming paradigms. Both the epileptic controls and the normal controls showed typical semantic facilitation with faster response times following related than unrelated primes. The anomic subjects, on the other hand, were much slower to read targets following presentation of semantically related items than following unrelated primes. This inhibition effect was seen to increase as the number of related primes increased. These patterns were observed both when picture primes (Experiment 1) and word primes (Experiment 2) were used. These findings were interpreted as evidence for a category-specific retrieval inhibition in the anomic epileptic subjects.

Functional mapping of human memory using PET: comparisons of conceptual and perceptual tasks.

An experiment is reported in which regional cerebral blood flow (rCBF) as measured using positron emission tomography (PET) as participants performed conceptual and perceptual memory tasks. Blood flow during two conceptual tests of semantic cued recall and semantic association was compared to a control condition in which participants made semantic associations to nonstudied words. Analogously, rCBF during two perceptual tasks of word fragment cued recall and word fragment completion was compared to a word fragment nonstudied control condition. A direct comparison of conceptual and perceptual tasks showed that conceptual tasks activated medial and lateral left hemisphere in frontal and temporal regions as well as the lateral aspect of bilateral inferior parietal lobule. Perceptual tasks, in contrast, produced relatively greater activation in right frontal and temporal cortex as well as bilateral activation in more posterior regions. Comparisons of the memory tasks with their control conditions revealed memory-specific deactivations in left medial and superior temporal cortex as well as left frontal cortex for both conceptual tasks. In contrast, memory-specific deactivations for both perceptual fragment completion tests were localized in posterior regions including occipital cortex. Results from this and other functional imaging experiments provide evidence that conceptual and perceptual memory processes are subserved, at least in part, by different neurological structures in the human brain.

Dissociations among memory measures in memory-impaired subjects: evidence for a processing account of memory.

Deficits in conceptual transfer on both implicit and explicit memory tests were obtained for memory-impaired temporal lobe epileptic (TLE) subjects in three studies. In Experiment 1, in which a generate-read paradigm was employed, memory-impaired TLEs failed to show normal generation effects on conceptually driven tests of semantic cued recall and general knowledge questions, although their data-driven memory as measured by word-fragment completion and graphemic cued recall tasks was normal. In Experiment 2, memory-impaired patients having left temporal lobe seizure foci were tested on these four tasks and compared with nonimpaired TLEs having right temporal foci. The left TLEs showed deficits on conceptually driven tasks and normal memory for data-driven tests. These findings were extended in Experiment 3, in which left TLE patients failed to show any benefit from blocked study, as compared with random study, on category production and semantic cued-recall tests, although right TLEs and normal controls showed blocking effects on both tasks. These findings may be accommodated by a processing framework of memory in which memory-impaired patients are characterized as having deficits in conceptual, but not in data-driven, processing capabilities.

The role of the temporal lobes in recognizing visuospatial materials: remembering versus knowing.

Recognition memory for abstract visuospatial designs was assessed in unilateral temporal lobe epilepsy (TLE) patients and normal controls using a remember/know recognition paradigm. Subjects assigned "remember" judgments to recognized items for which they could recall the study presentation, and "know" judgments to items recognized on the basis of familiarity without conscious recollection of the study episode. In Experiment 1 normal controls and left TLE patients gave more "know" than "remember" recognition judgments for visuospatial materials. Right TLE subjects, however, showed the opposite response pattern. Experiment 1a demonstrated that this dissociation between left and right temporal patients occurred in both presurgery and postsurgery patients. In Experiment 2 recognition was assessed following encoding conditions in which subjects answered questions about either the number of lines in the designs or the appropriateness of verbal labels for presented stimuli. The previous pattern of "know" and "remember" responses was replicated for all groups in the line count condition, but was reversed for normal controls in the label condition. These results are interpreted within a theoretical framework in which "remember" responses are based on the contribution of distinctiveness of individual items to recognition whereas "know" judgements reflect perceptual fluency.

Developmental differences in implicit and explicit memory performance.

Performance of preschool, elementary school, and college students was compared on a series of perceptual and conceptual implicit and explicit memory tasks that followed perceptual or conceptual processing during study. As expected, performance on the conceptual explicit memory task improved across age groups. In contrast, performance on the perceptual explicit memory task as well as that on both types of implicit memory tasks showed no developmental change. Also, perceptual processing during study led to better memory performance than conceptual processing for both the perceptual implicit and perceptual explicit tasks and conceptual processing during study led to better memory performance on the conceptual explicit memory task. Performance on the conceptual implicit memory task, in contrast, was affected equally by both types of study processing. The results are discussed in terms of transfer-appropriate processing (Roediger & Blaxton, 1987b) and unitization and grouping processes (Graf & Schacter, 1989).

Relationship of carbamazepine reduction rate to seizure frequency during inpatient telemetry.

To establish guidelines for medication reduction during inpatient telemetry, the records of 18 children and young adults with refractory partial seizures undergoing carbamazepine (CBZ) reductions during continuous video/EEG telemetry were reviewed. Six patients were receiving CBZ monotherapy, and 12 patients were treated with an additional antiepileptic drug (AED) maintained at baseline dosage during CBZ taper. Despite relatively rapid mean reductions in dosage of 44% by day 2 of taper, no patients experienced frequent repetitive seizures or status epilepticus (SE). Seizure rate during the entire CBZ reduction period correlated significantly with rate of drug reduction. Linear regression analysis showed drug reduction rate to be a good predictor of seizure rate. Fourteen patients experienced at least three seizures during CBZ taper. On the average, the third seizure occurred on day 5 of taper at a percentage of dose reduction of 79%. In 8 patients, CBZ concentrations were measured both before taper and < or = 24 h after the third seizure. For these patients, seizure rate also correlated significantly with reduction in CBZ level. We conclude that manipulation of CBZ dose reduction rate is important in maximizing seizure frequency during telemetry and, in our patients, a relatively rapid rate of dose reduction was safe and effective in promoting seizure recordings.

Hematoma-related seizures detected during subdural electrode monitoring.

A 28-year-old woman with epilepsy developed the new onset of paroxysmal tongue tingling during subdural electrode monitoring. Her symptoms coincided with electrographic seizure activity arising from an area of the perisylvian region that had not previously been involved in her habitual seizures. At electrode removal, a focal 2 x 2 cm hematoma was detected and evacuated from beneath these electrode contacts. Unexpected episodic events may represent nonhabitual seizure activity related to the surgical procedure.

(15)O water positron emission tomography in language localization: a study comparing positron emission tomography visual and computerized region of interest analysis with the Wada test.

We compared (15)O water positron emission tomography (PET) auditory and visual confrontational naming activation with an intracarotid amobarbital (Amytal) injection procedure (IAP) for language lateralization in 12 patients with intractable epilepsy. PET scans were evaluated by three raters experienced in functional imaging as well as by a region of interest (ROI) approach. Compared with IAP, raters' positive predictive value for language lateralization ranged from 88 to 91%. ROI analysis had a positive predictive value of 80%. Six patients had surgery; 1 with right-sided IAP language dominance but left-sided PET activation had dysphasia for 6 months after left temporal lobectomy.

Cortical stimulation elicits regional distinctions in auditory and visual naming.

We used electrical stimulation mapping to compare performance on auditory and visual naming tasks in inferotemporal, lateral temporal, frontal, and parietal cortex in 8 temporal lobe epilepsy (TLE) patients with subdural electrodes placed for preoperative language localization. Performance on auditory responsive naming (ARN) and visual confrontation naming (VCN) was best during stimulation of parietal cortex and was equally impaired during stimulation of inferotemporal and frontal cortex. In contrast, ARN performance was significantly poorer than VCN performance during stimulation of anterior and posterior lateral temporal cortex. In most patients, stimulation of inferotemporal cortex at relatively low stimulus intensities (< or = 5 mA) during either ARN or VCN elicited reproducible errors in which patients could describe, gesture, spell, or draw, but not name, in response to auditory or visual cues. Inferotemporal and frontal cortex appear to be multimodality language regions distinct from lateral temporal cortex.

Functional mapping of human learning: a positron emission tomography activation study of eyeblink conditioning.

Regional cerebral blood flow (rCBF) was measured using positron emission tomography during eyeblink conditioning in young adults. Subjects were scanned in three experimental conditions: delay conditioning, in which binaural tones preceded air puffs to the right eye by 400 msec; pseudoconditioning, in which presentations of tone and air puff stimuli were not correlated in time; and fixation rest, which served as a baseline control. Compared with fixation, pseudoconditioning produced rCBF increases in frontal and temporal cortex, basal ganglia, left hippocampal formation, and pons. Learning-specific activations were observed in conditioning as compared with pseudoconditioning in bilateral frontal cortex, left thalamus, right medial hippocampal formation, left lingual gyrus, pons, and bilateral cerebellum; decreases in rCBF were observed for bilateral temporal cortex, and in the right hemisphere in putamen, cerebellum, and the lateral aspect of hippocampal formation. Blood flow increased as the level of learning increased in the left hemisphere in caudate, hippocampal formation, fusiform gyrus, and cerebellum, and in right temporal cortex and pons. In contrast, activation in left frontal cortex decreased as learning increased. These functional imaging results implicate many of the same structures identified by previous lesion and recording studies of eyeblink conditioning in animals and humans and suggest that the same brain regions in animals and humans mediate multiple forms of associative learning that give meaning to a previously neutral stimulus.