Faculty Bibliography

We examined the brain organization for language and auditory functions in five high-functioning autistic and five normal adults, using [15O]-water positron emission tomography (PET). Cerebral blood flow was studied for rest, listening to tones, and listening to, repeating, and generating sentences. The autism group (compared to the control group) showed (a) reversed hemispheric dominance during verbal auditory stimulation; (b) a trend towards reduced activation of auditory cortex during acoustic stimulation; and (c) reduced cerebellar activation during nonverbal auditory perception and possibly expressive language. These results are compatible with findings of cerebellar anomalies and may suggest a tendency towards atypical dominance for language in autism.

PURPOSE/OBJECTIVE:Intermittent photic stimulation (IPS) is an activation procedure used during electroencephalogram (EEG) recording to elicit paroxysmal discharges in individuals with photosensitivity. Specific responses on EEG recording may be provoked by IPS at different frequencies of flickering in normal individuals and in patients with photosensitive epilepsy. METHODS:Changes in regional cerebral blood flow (rCBF) were studied during IPS in two groups of subjects by using [15O]-water positron emission tomography (PET): a control group consisting of eight healthy adults with photic driving response during IPS on EEG recording (mean age, 25 +/- 10.5 years) without history of neurologic or psychiatric abnormalities and a patient group consisting of four adults (mean age, 33 +/- 7.5 years) with history of photosensitive epilepsy. [15O]-water PET scanning with concomitant EEG monitoring was performed during baseline and IPS at 4-, 14-, and 30-Hz frequencies. RESULTS:The control group showed photic driving response at 14-Hz IPS frequency. The patient group showed photoparoxysmal response at 14 and 30 Hz, but not at 4 Hz. Changes in rCBF were determined by means of statistical parametric mapping. Increases in rCBF in occipital cortex (Brodmann's areas 17, 18, and 19) were observed in both groups. In addition, during photic driving responses, the control group showed rCBF increases in the insula and in the thalamus, on the right side. The patient group showed a significant rCBF increase in the hypothalamic region inferior to the left caudate nucleus during photoparoxysmal response. This activation was not found in the control group. Increased rCBF also was observed in the patient group in the head of the left caudate nucleus, in the left hippocampus, and in left insula during IPS without photoparoxysmal response. No activations in these regions were observed during photoparoxysmal response. CONCLUSIONS:These data may indicate involvement of the hypothalamus in photosensitive epilepsy and may suggest a modulatory function of the caudate nucleus, which might be associated with an inhibition of epileptic discharges during IPS in patients with photosensitive epilepsy.

Although functional imaging with positron emission tomography (PET) and single-photon emission computed tomography are useful in the clinical evaluation of intractable epilepsy, these techniques have not been widely applied to understanding the basic mechanisms of the epilepsies. Among patients with infantile spasms, PET studies with 2-deoxy-2[18F]fluoro-D-glucose (FDG) suggest that the spasms are the result of secondary generalization from cortical foci and that maturational factors result in the recruitment of basal ganglia and brainstem serotonin mechanisms that lead to secondary generalization and the unique semiology of the spasms. Attempts to develop an animal model of infantile spasms have not been successful. Glucose utilization studies in the Lennox-Gastaut syndrome also indicate cortical lesions and further suggest that the electroencephalographic pattern of 1 to 2.5 Hz spike-wave activity (slow spike-wave pattern) is an interictal phenomenon. There is a remarkable consistency between 14C-2-deoxyglucose autoradiographic findings and PET observations of glucose utilization performed for patients in the ictal, interictal, and postictal states. Although three patterns of ictal glucose hypermetabolism have been described, hypermetabolism also can be seen in the postictal and interictal clinical states and in various animal models. Preliminary studies of benzodiazepine receptor binding with PET have found that the cortical epileptic region of decreased binding is smaller than the region of hypometabolism on glucose utilization studies, but detailed electrophysiologic comparisons have not been made. Development of new PET methods for the study of presynaptic and postsynaptic neurotransmitter functions will offer unique opportunities in the study of epileptic mechanisms.

Several reports have indicated that cortical resection is effective in alleviating intractable epilepsy in children with tuberous sclerosis complex (TSC). Because of the multitude of cortical lesions, however, identifying the epileptogenic tuber(s) is difficult and often requires invasive intracranial electroencephalographic (EEG) monitoring. As increased concentrations of serotonin and serotonin-immunoreactive processes have been reported in resected human epileptic cortex, we used alpha-[11C]methyl-L-tryptophan ([11C]AMT) positron emission tomography (PET) to test the hypothesis that serotonin synthesis is increased interictally in epileptogenic tubers in patients with TSC. Nine children with TSC and epilepsy, aged 1 to 9 years (mean, 4 years 1 month), were studied. All children underwent scalp video-EEG monitoring, PET scans of glucose metabolism and serotonin synthesis, and EEG monitoring during both PET studies. [11C]AMT scans were coregistered with magnetic resonance imaging and with glucose metabolism scans. Whereas glucose metabolism PET showed multifocal cortical hypometabolism corresponding to the locations of tubers in all 9 children, [11C]AMT uptake was increased in one tuber (n=3), two tubers (n=3), three tubers (n=1), and four tubers (n=1) in 8 of the 9 children. All other tubers showed decreased [11C]AMT uptake. Ictal EEG data available in 8 children showed seizure onset corresponding to foci of increased [11C]AMT uptake in 4 children (including 2 with intracranial EEG recordings). In 2 children, ictal EEG was nonlocalizing, and in 1 child there was discordance between the region of increased [11C]AMT uptake and the region of ictal onset on EEG. The only child whose [11C]AMT scan showed no regions of increased uptake had a left frontal seizure focus on EEG; however, at the time of his [11C]AMT PET scan, his seizures had come under control. [11C]AMT PET may be a powerful tool in differentiating between epileptogenic and nonepileptogenic tubers in patients with TSC.