8 Analysis of EEG in Medically Intractable Epilepsy
179
the scalp [22]. ESI has been introduced in the late 1990s and anecdotally used in the
presurgical evaluation in a few large epilepsy surgery centers. At that time, simple
spherical head model with a standard MRI template and lower numbers of electrodes
(usually less than 32 channels) were used, and thus it usually provided only a coarse
estimation of source locations in the standard brain models. With advanced computing technology and power, high density electrodes (up to 256), and realistic head
model constructed using patient’s own MRI data can now be used for ESI, and thus
the accuracy and feasibility of ESI have been significantly increased.
In spite of the advances in computational and integrative imaging, ESI has not
become routine everywhere [27]. One of the important problems is that clinical
utility of ESI was largely limited because of the complicated methods without standardization and additionally required workforce and time [30]. A survey conducted
among 25 European epilepsy surgery centers showed that ESI was performed by 12
centers: exclusively with magnetoencephalography (MEG) in 3 centers, exclusively
with EEG in 5 centers, and with both MEG and EEG in 4 centers. Furthermore, a
total of 14 different combinations of inverse methods and volume conduction models
were used: 7 for MEG and 13 for EEG [30]. Therefore, it is apparent that considerable gap between technological advancement and clinical utility exists in the field of
ESI. At present, however, ESI seems to be a promising technique that can positively
contribute to visual EEG analysis for the localization of epileptic spikes, and plays
a role in epilepsy surgery evaluation [22]. ESI can be applied on either interictal
epileptiform discharges (IEDs) or ictal discharges.
8.2.1 Interictal ESI
The IED is transient waves or complexes clearly distinguishable from background
activity, and generally shows a pointed peak with a duration of 20–200 ms usually
followed by a slow wave. The potential should be reflected in physically adjacent
electrodes and perhaps in synaptically linked regions such as the contralateral hemisphere [14]. The IEDs are not accompanied by clinical or subclinical seizure. The
IEDs are generated by the synchronous discharges of a group of neurons in a region
referred to as the epileptic focus. IEDs are so highly correlated with spontaneous
seizures that their presence is used to support the diagnosis of epilepsy and are considered as hallmark of epilepsy [41]. Thus, evaluation of the interictal EEG is an
integral part of the presurgical evaluation.
At least 10–20 cm
2 of synchronous active cortex is usually necessary to produce
a recognizable scalp potential. In other word, cortical areas that are active but small
cannot be detected by the scalp electrodes due to the low conductivity of intervening tissues especially skull between cerebral cortex and scalp. Spikes arising from
deep cortical areas cannot be detectable in scalp EEG until it propagates to more
superficial areas. The site of IEDs represents irritative zone, and it can be extensive,
usually much larger than epileptogenic zone [11]. Patients with focal IEDs included
in surgical resection have good surgical outcomes. The presence of IEDs extending
179
the scalp [22]. ESI has been introduced in the late 1990s and anecdotally used in the
presurgical evaluation in a few large epilepsy surgery centers. At that time, simple
spherical head model with a standard MRI template and lower numbers of electrodes
(usually less than 32 channels) were used, and thus it usually provided only a coarse
estimation of source locations in the standard brain models. With advanced computing technology and power, high density electrodes (up to 256), and realistic head
model constructed using patient’s own MRI data can now be used for ESI, and thus
the accuracy and feasibility of ESI have been significantly increased.
In spite of the advances in computational and integrative imaging, ESI has not
become routine everywhere [27]. One of the important problems is that clinical
utility of ESI was largely limited because of the complicated methods without standardization and additionally required workforce and time [30]. A survey conducted
among 25 European epilepsy surgery centers showed that ESI was performed by 12
centers: exclusively with magnetoencephalography (MEG) in 3 centers, exclusively
with EEG in 5 centers, and with both MEG and EEG in 4 centers. Furthermore, a
total of 14 different combinations of inverse methods and volume conduction models
were used: 7 for MEG and 13 for EEG [30]. Therefore, it is apparent that considerable gap between technological advancement and clinical utility exists in the field of
ESI. At present, however, ESI seems to be a promising technique that can positively
contribute to visual EEG analysis for the localization of epileptic spikes, and plays
a role in epilepsy surgery evaluation [22]. ESI can be applied on either interictal
epileptiform discharges (IEDs) or ictal discharges.
8.2.1 Interictal ESI
The IED is transient waves or complexes clearly distinguishable from background
activity, and generally shows a pointed peak with a duration of 20–200 ms usually
followed by a slow wave. The potential should be reflected in physically adjacent
electrodes and perhaps in synaptically linked regions such as the contralateral hemisphere [14]. The IEDs are not accompanied by clinical or subclinical seizure. The
IEDs are generated by the synchronous discharges of a group of neurons in a region
referred to as the epileptic focus. IEDs are so highly correlated with spontaneous
seizures that their presence is used to support the diagnosis of epilepsy and are considered as hallmark of epilepsy [41]. Thus, evaluation of the interictal EEG is an
integral part of the presurgical evaluation.
At least 10–20 cm
2 of synchronous active cortex is usually necessary to produce
a recognizable scalp potential. In other word, cortical areas that are active but small
cannot be detected by the scalp electrodes due to the low conductivity of intervening tissues especially skull between cerebral cortex and scalp. Spikes arising from
deep cortical areas cannot be detectable in scalp EEG until it propagates to more
superficial areas. The site of IEDs represents irritative zone, and it can be extensive,
usually much larger than epileptogenic zone [11]. Patients with focal IEDs included
in surgical resection have good surgical outcomes. The presence of IEDs extending