Cerebral activation during speech discrimination in temporal lobe epilepsy

Mapping Language in Epilepsy with Functional Imaging

Surgery is an important therapeutic alternative for patients with uncontrolled epilepsy. Preoperative identification of brain regions important for language is important to reduce the risk of functional impairment after surgery. The Wada test suffers from several technical and clinical disadvantages and provides hemispheric data at best. More invasive methods such as intraoperative or chronic subdural cortical mapping have more limited application. New approaches using neuroimaging methods offer the opportunity to localize, as well as lateralize, language. In addition, normal volunteers can be studied with the same techniques, providing comparative and control data. Although most normal studies have been reported as group data, it is important for individual scans to be available for comparison with patient studies to understand the normal range of interindividual variability. Two techniques, PET with 15O-water-PET and fMRI, have been used. Both detect signal changes associated with increased regional blood flow during neuronal activity. Usually, scans performed during a language task are compared with those obtained during control conditions. It is important to choose activation tasks carefully, to make sure one is imaging activation associated with the particular process of interest. PET has advantages, including a fully diffusible tracer, standardized analytic methods, a more comfortable environment, and less sensitivity to movement artifact. On the other hand, it involves a cyclotron-produced tracer, radiation exposure, and is more difficult to repeat. FMRI over represents the effects of large vascular structures and is very sensitive to movement but uses widely available equipment and has no limitation on the number of studies. For both studies, it is important to understand the potential effects of such factors as attention, fatigue, and familiarity with the material. Several studies comparing 15O-water-PET and fMRI to the Wada test found that the former are at least as accurate for language lateralization. In addition, we compared 15O-water-PET to direct subdural electrode cortical stimulation and found that regions showing increased cerebral blood flow during naming tasks co-registered with subdural electrodes that disrupted language during electrical stimulation. In this and other studies, PET detected more regions than electrical stimulation techniques. The whole brain cannot be covered with electrodes, but some areas participating in a task may not be crucial for it. FMRI is particularly useful for children. We compared cortical activation patterns in normal children, adolescents, and adults. The activation patterns, and laterality of language dominance, in children 8 years and above, were similar to adults, although some differences could reflect maturation and evolving focality of cognitive processes. In children with epilepsy, fMRI successfully identified language laterality and provided data on intrahemispheric localization. Studies also showed the effects of the epileptic focus on normal activation patterns for several tasks. Neuroimaging functional mapping is an important tool, still in the process of development and evolution. Although potentially of great clinical and scientific value, it should be used and interpreted cautiously.

The integration of metabolic imaging in stereotactic procedures including radiosurgery: a review

Object. The authors review their experience with the clinical development and routine use of positron emission tomography (PET) during stereotactic procedures, including the use of PET-guided gamma knife radiosurgery (GKS).

Methods. Techniques have been developed for the routine use of stereotactic PET, and accumulated experience using PET-guided stereotactic procedures over the past 10 years includes more than 150 stereotactic biopsies, 43 neuronavigation procedures, and 34 cases treated with GKS. Positron emission tomography—guided GKS was performed in 24 patients with primary brain tumors (four pilocytic astrocytomas, five low-grade astrocytomas or oligodendrogliomas, seven anaplastic astrocytomas or ependymomas, five glioblastomas, and three neurocytomas), five patients with metastases (single or multiple lesions), and five patients with pituitary adenomas.

Conclusions. Data obtained with PET scanning can be integrated with GKS treatment planning, enabling access to metabolic information with high spatial accuracy. Positron emission tomography data can be successfully combined with magnetic resonance imaging data to provide specific information for defining the target volume for the radiosurgical treatment in patients with recurrent brain tumors, such as glioma, metastasis, and pituitary adenoma. This approach is particularly useful for optimizing target selection for infiltrating or ill-defined brain lesions. The use of PET scanning contributed data in 31 cases (93%) and information that was specifically utilized to adapt the target volume in 25 cases (74%). It would seem that the integration of PET data into GKS treatment planning may represent an important step toward further developments in radiosurgery: this approach provides additional information that may open new perspectives for the optimization of the treatment of brain tumors.

Presurgical contribution of quantitative stereotactic positron emission tomography in temporo limbic epilepsy

BACKGROUND

We quantified the interictal metabolic changes associated with temporal lobe epilepsy by using an accurate stereotactic method.

METHODS

We selected 16 patients who had proven unilateral focal or regional temporal onset defined by SEEG criteria. Each patient underwent stereotactic MRI and stereotactic [18 fluoro] fluorodeoxyglucose positron emission tomography (PET).

RESULTS

Asymmetries (mean, +/- SD) were found in mesio-temporal structures: amygdala (-0.033+/-0.027, p = 0.0002), hippocampus (-0.035+/-0.032, p = 0.0006), and superior temporal gyrus (-0.036+/-0.032, p = 0.0004). Four of the sixteen patients had previously had unlocalized qualitative nonstereotactic PET analysis.

CONCLUSIONS

The quantitative stereotactical PET method allows a higher resolution study of mesio-temporal structures.

Widespread functional deficits in perception-related networks demonstrated by PET in a case with simple visual seizures

PURPOSE

To study benzodiazepine receptor (BZR) density and functional deficits in occipital lobe epilepsy.

METHODS

A 39-year-old man who had simple partial visual seizures after neurosurgical transtentorial extirpation of a pinealoma was studied by EEG, magnetic resonance imaging (MRI), and positron emission tomography (PET) of [18F]2-fluoro-2-deoxy-D-glucose (FDG) at rest and during visual activation task and[11C]flumazenil (FMZ).

RESULTS

Electroencephalographic recordings were nonspecific, and MRI did not reveal any morphologic anomaly in the occipital lobe. Flumazenil-PET demonstrated a small epileptogenic region in the right visual association cortex and FDG-PET showed hypometabolism in a corresponding location and thalamic diaschisis. Stimulation of occipital metabolism by a continuous visual recognition task improved significantly the contrast between the dysfunctional zone and its surround.

CONCLUSIONS

As BZR deficits are restricted to a small region, widespread hypometabolism in networks involved in visual information processing indicates an extensive functional deactivation by the epileptogenic focus.

Future directions for functional mapping

The intracarotid amobarbital procedure and direct cortical stimulation are commonly used techniques for functional mapping in candidates for epilepsy surgery. The role of these two procedures has been challenged recently. Emerging technologies recently applied to functional mapping include optical imaging, functional positron emission tomography studies, transcranial magnetic stimulation, magnetoencephalography, and fast magnetic resonance imaging. Studies have used these new technologies and the potential for these procedures to replace the established but more invasive techniques is being considered.

PET and functional testing in temporal lobe epilepsy

Thirty-seven patients with severe temporal lobe epilepsy were studied interictally with [18F]fluorodeoxyglucose-PET in each of three conditions: resting, during emotional speech, and while performing a visual recognition task. In the resting state, each patient exhibited regional hypometabolism in agreement with his epileptic EEG focus, but that area was typically very large. The zone of maximum dysfunction was significantly better demarcated on activated scans showing an increase in whole-brain metabolism averaging 18%. Concurrently, the midtemporal-focus metabolic contrast was improved by 27%. This effect was more consistently produced by the speech paradigm that also induced significant amygdala recruitment and revealed the individual hemispheric speech dominance. Psychological trait factors played a role only for unspecific global activation.