- Top of page
- MATERIALS AND METHODS
Summary: Purpose: To characterize the epileptogenic condition of patients with mesial temporal lobe epilepsy, the interictal patterns of glucose metabolism, perfusion, and magnetic field in the temporal lobe were evaluated by using [18F]fluorodeoxyglucose–positron emission tomography, [99mTc]-ethylcysteinate dimer–single photon emission computed tomography, and magnetoencephalography (MEG).
Methods: Twenty-one patients with mesial temporal lobe epilepsy related to hippocampal sclerosis were studied. The ictal-onset area was located by continuous video-EEG monitoring. Quantitative analysis of glucose metabolism and perfusion in the temporal lobe was performed, and the cerebral magnetic field was evaluated to measure the equivalent current dipole (MEG-ECD).
Results: Although hypometabolism and hypoperfusion in the temporal lobe were lateralized with the ictal-onset area in 16 (76.2%) and in 11 (52.4%) respectively, they were localized in diverse ways without any coupling. MEG-ECD was distributed in diverse ways unrelated to the ictal-onset area: ipsilateral medial temporal origin in five (23.8%), ipsilateral lateral temporal origin in two (9.5%), ipsilateral mixed (medial and lateral) temporal origin in six (28.6%), bilateral temporal origin in four (19.0%), and contralateral temporal origin in two (9.5%).
Conclusions: MEG-ECD was distributed in varied ways with the disorder and uncoupling of glucose metabolism and perfusion in the temporal lobe. These results may help resolve the clinical controversy over the possibility that the cortical irritative area generating the interictal epileptic discharge is distinct from the ictal-onset area, and also may have some functional implications in identifying different brain compartments in the generation of metabolic signals.
For the presurgical evaluation of patients with intractable epilepsy, it is important to identify the area that may be generating the seizures and to understand how it is functionally related to cortical structures. When patients with mesial temporal lobe epilepsy are evaluated, neurophysiologic and neuroimaging studies are performed based on semiology. The diagnostic methods of [18F]fluorodeoxyglucose–positron emission tomography (FDG-PET) or [99mTc]-ethylcysteinate dimer–single-photon emission computed tomography (TcECD-SPECT) have been found to be useful for assessing the glucose metabolism or perfusion in intractable epilepsy (1–11). Furthermore, magnetoencephalography (MEG), detecting cerebral magnetic fields induced by epileptic discharges, has recently been introduced as an aid to the identification of the cortical irritative area and has some great advantages over scalp electroencephalography (EEG) (12–14). MEG, with ≤100-plus channels, can record the cerebral magnetic fields in both hemispheres simultaneously, resulting in a major breakthrough in the evaluation of intractable epilepsy (15–18). However, the interictal functional properties of the epileptogenic zone, consisting of the ictal-onset area that initiates seizures and the cortical irritative area, remain unclear. The precise mechanism by which mesial temporal lobe epilepsy is produced is still not fully understood, nor is its relation to interictal neural activity.
Here we addressed the question of whether functional imaging studies with FDG-PET, TcECD-SPECT, and MEG can characterize the interictal functional properties of the epileptogenic zone. The present study involved 21 patients with mesial temporal lobe epilepsy related to hippocampal sclerosis. The ictal-onset area was located by continuous video-EEG monitoring. The interictal functional properties of the epileptogenic zone were analyzed comprehensively in terms of neural activity of glucose metabolism, perfusion, and magnetic field.
- Top of page
- MATERIALS AND METHODS
In this study, the interictal patterns of cerebral glucose metabolism, blood perfusion, and magnetic field in patients with mesial temporal lobe epilepsy were analyzed comprehensively with functional imaging studies by using FDG-PET, TcECD-SPECT, and MEG to characterize the epileptogenic zone relative to cerebral neural activity.
The development of neuroimaging techniques such as PET and SPECT has had a great impact on the functional investigation of patients with intractable epilepsy. Interictal PET has proven to be a reliable imaging method for defining the area of hypometabolism in the brain. Several studies have demonstrated the efficacy of interictal PET to define the lateralization of the epileptogenic zone, consisting of the ictal-onset area and the cortical irritative area. Interictal SPECT is another useful neuroimaging tool to provide information about areas of hypoperfusion in brain that might be related to the epileptogenic zone. In mesial temporal lobe epilepsy, the epileptogenic zone is usually characterized by hypometabolism of glucose or hypoperfusion, with rates reportedly from 60 to 90% (3–10,28) and from 50 to 70% (24,29–31), respectively, not notably different from the rate estimated in the present study. The area of hypometabolism or hypoperfusion is usually defined as more extensive than that of the ictal-onset area, and often involves the entire temporal lobe or extends beyond the temporal lobe in mesial temporal lobe epilepsy (1,2,22,28,32–34). In the present study, hypometabolism of glucose and hypoperfusion also were observed not only in the medial temporal lobe but also in the neocortex of the lateral temporal lobe without any other structural abnormalities except for the hippocampal sclerosis. It is thought that most of the observed focal metabolic and hemodynamic disturbances might be due to functional rather than structural changes (34,42). Lateral temporal hypometabolism or hypoperfusion in mesial temporal lobe epilepsy may be due to disordered functional pathways between medial and lateral temporal cortex caused by cell loss in medial temporal lobe (32,42). The area and degree of hypometabolism were not completely coupled with those of hypoperfusion in the present study. Previous studies also reported uncoupling of glucose metabolism and blood flow in the epileptogenic zone in intractable epilepsy (35–40). Differential behavior of metabolism and hemodynamics may occur in the epileptogenic zone. Furthermore, it remains unclear how hypometabolism or hypoperfusion correlates with epileptic phenomena. Engel et al. (41) demonstrated that no quantitative relation could be seen between the degree of focal hypometabolism and the frequency of interictal EEG spikes in 50 patients with complex partial seizures. Guillon et al. (42) analyzed the relation between interictal spikes and interictal hypoperfusion in 20 patients with mesial temporal lobe epilepsy and suggested that the correlation between those parameters does not necessarily indicate a casual relation and may reflect different aspects of the cerebral dysfunction of the epileptic brain. Although some possibility exists that the hypometabolism determined by PET may be related to subclinical ictal discharges or interictal spikes occurring during PET scanning, as well as to permanent structural or functional abnormalities (43), metabolic and electrophysiologic techniques measure different aspects of cerebral functions in seizure disorders. These controversies may be attributed to the lack of understanding of interictal patterns of brain activity in patients with intractable epilepsy.
Although MEG has been proposed to identify the cortical irritative area with a noninvasive technique, the traditional MEG system has the technical limitation that it is difficult to detect MEG-ECD from deep areas in the brain such as the medial structure of the temporal lobe (13–17,44). The MEG used in the present study has high enough sensitivity to detect such signals from deep areas in the brain by using a pickup coil with a longer baseline and a wider-area sensor array (25). The distribution of the interictal MEG-ECD was thought to indicate the area generating interictal epileptic discharges. In the present study, only five of 21 patients revealed interictal MEG-ECD originating in the medial temporal lobe, and other patients revealed lateral, mixed, bilateral, contralateral, or no localization of the origin of MEG-ECD. The pattern of MEG-ECD distribution was seen to be nonuniform, regardless of typical symptoms of mesial temporal lobe epilepsy, suggesting diversity and a complex mechanism of interictal epileptic phenomena in mesial temporal lobe epilepsy. Some clinical controversies are found concerning the relation between interictal spikes and seizures. Gotman (45) investigated the presurgical long-term monitoring of focal epilepsies and suggested that no systematic changes occur in spiking activity in the minutes or hours preceding seizures and increases in spikes after many partial seizures. Janszky et al. (46) investigated the 18 patients with intractable temporal lobe epilepsy with continuous video-EEG monitoring and suggested that the localization of interictal spikes may depend on the brain areas involved by preceding seizures. Colder et al. (47) examined autocorrelation and the interval-distribution characteristics of neuronal discharge patterns in patients with complex partial seizures and suggested that possible mechanisms of decreased burst discharge in epileptogenic regions include selective loss of burst-discharging neurons and increased recurrent inhibition. In basic studies of epilepsy, Jansen and Yaari (48) demonstrated that focal seizures arise independent of interictal spikes in their in vitro model of acute focal epilepsy, and determined that the underlying cellular mechanisms of interictal spikes and seizure generation are different: the former required chemical synaptic excitation, whereas the latter could be initiated and maintained by nonsynaptic neuronal interactions. Barbarosie and Avoli (49) showed that ictal discharge generation recorded in the entorhinal cortex after Schaffer collateral cut is prevented by mimicking CA3 neuronal activity through rhythmic electrical stimulation of the CA1 hippocampal output region in their combined mouse hippocampal–entorhinal cortex slice model. They speculated that the interictal activity controls rather than promotes ictal events, and functional integrity of CA3 constitutes a critical control mechanism in temporal lobe epilepsy. These clinical and basic researches suggested that interictal spikes and seizures are two quite distinct phenomena.
The epileptogenic zone is the region of the brain that is necessary and sufficient for seizure generation and consists of the ictal-onset area and the cortical irritative area. The size of these areas and the boundaries between them may be subject to continuous dynamic changes that reflect the underlying modulation of neuronal synchronization (44,50). In the present study, MEG-ECD distribution was found to be diverse, with no particular pattern of association with the ictal-onset area. Such a diversity of MEG-ECD may reflect the fact that interictal spikes and ictal discharges are generated by different populations of neurons through different cellular and network mechanisms. It is crucial to understand the cellular mechanisms that generate epileptic phenomena and the functional implications of different brain compartments in the generation of metabolic signals.