182
K.-Y. Jung
who were younger than 18 years old [36]. EEG data were recorded with electrodes
less than 32 channels (conventional 10–20 system plus 4–10 extra-channels over the
suspected epileptogenic region). ESI was done with the realistic subject-dependent
head models with a boundary element model (BEM) using a single rotating dipole
for each spike or sharp wave. Accuracy of ESI was evaluated whether the single
dipole was localized within surgical resection cavity after at least 1 year of followup. Epileptogenic focus in 65% of patients could be localized by single rotating dipole
with low-resolution EEG, which was comparable to ictal SPECT (68%) and superior
to PET (55.5%). Correct localization was significantly higher in case of temporal lobe
epilepsy compared to extratemporal lobe epilepsy (84.6% vs. 48.0%). MRI-negative
patients showed tendency to be more frequently localized within surgical resection
cavity compared to MRI-positive cases (78.6% vs. 60.9%). Sensitivity and specificity
of ESI in relation to 2-year surgical outcome were 60.6 and 50.0%, respectively.
Although these figures are somewhat lower than those of previous reports, it is
acceptable and reasonable considering lower number of electrodes adopted in this
study, compared to high density (from 64 to 256 channels) EEG. Nonetheless, lowresolution electrode ESI was superior to PET or ictal SPECT images in the present
study. Therefore, this study is encouraging that using only low-resolution EEG can be
helpful in presurgical evaluation of intractable focal epilepsy at least with pediatric
patients.
In summary, interictal ESI can be a relatively inexpensive, noninvasive modality
for the evaluation of focal intractable epilepsy, and should be encouraged to use in
routine presurgical evaluation in clinical practice. However, standardization of ESI
processing steps and determination of IEDs should be provided for use in routine
clinical practices. More strong data such as data from double-blinded randomized
controlled study with large sample size are also needed.
8.2.2 Ictal ESI
Ictal EEG activity is characterized as a run of EEG rhythm that changes the EEG
background and evolves over time in waveform frequency, amplitude, and morphology [28]. Ictal EEG is usually accompanied by clinical seizure. Rarely, no seizure
can be associated with apparent ictal discharges, so-called subclinical EEG seizure,
which occurs in cases when very small cortical areas are synchronized, which is only
noted during electrocorticography rather than scalp EEG, and in cases of simple partial seizures that do not evolve complex partial or generalized seizure.
Ictal onset zones are areas of cortex where seizures are generated including areas
of early propagation under certain circumstances, and can be identified by the source
localization at ictal EEG onset time. Ictal onset zone can be either within the distribution of irritative zone or distinct from irritative zone. In other words, ictal onset
zone is not necessarily the same as irritative zone. Nevertheless, the ictal ESI results
are largely in agreement with interictal ESI results in temporal lobe epilepsy [34].
K.-Y. Jung
who were younger than 18 years old [36]. EEG data were recorded with electrodes
less than 32 channels (conventional 10–20 system plus 4–10 extra-channels over the
suspected epileptogenic region). ESI was done with the realistic subject-dependent
head models with a boundary element model (BEM) using a single rotating dipole
for each spike or sharp wave. Accuracy of ESI was evaluated whether the single
dipole was localized within surgical resection cavity after at least 1 year of followup. Epileptogenic focus in 65% of patients could be localized by single rotating dipole
with low-resolution EEG, which was comparable to ictal SPECT (68%) and superior
to PET (55.5%). Correct localization was significantly higher in case of temporal lobe
epilepsy compared to extratemporal lobe epilepsy (84.6% vs. 48.0%). MRI-negative
patients showed tendency to be more frequently localized within surgical resection
cavity compared to MRI-positive cases (78.6% vs. 60.9%). Sensitivity and specificity
of ESI in relation to 2-year surgical outcome were 60.6 and 50.0%, respectively.
Although these figures are somewhat lower than those of previous reports, it is
acceptable and reasonable considering lower number of electrodes adopted in this
study, compared to high density (from 64 to 256 channels) EEG. Nonetheless, lowresolution electrode ESI was superior to PET or ictal SPECT images in the present
study. Therefore, this study is encouraging that using only low-resolution EEG can be
helpful in presurgical evaluation of intractable focal epilepsy at least with pediatric
patients.
In summary, interictal ESI can be a relatively inexpensive, noninvasive modality
for the evaluation of focal intractable epilepsy, and should be encouraged to use in
routine presurgical evaluation in clinical practice. However, standardization of ESI
processing steps and determination of IEDs should be provided for use in routine
clinical practices. More strong data such as data from double-blinded randomized
controlled study with large sample size are also needed.
8.2.2 Ictal ESI
Ictal EEG activity is characterized as a run of EEG rhythm that changes the EEG
background and evolves over time in waveform frequency, amplitude, and morphology [28]. Ictal EEG is usually accompanied by clinical seizure. Rarely, no seizure
can be associated with apparent ictal discharges, so-called subclinical EEG seizure,
which occurs in cases when very small cortical areas are synchronized, which is only
noted during electrocorticography rather than scalp EEG, and in cases of simple partial seizures that do not evolve complex partial or generalized seizure.
Ictal onset zones are areas of cortex where seizures are generated including areas
of early propagation under certain circumstances, and can be identified by the source
localization at ictal EEG onset time. Ictal onset zone can be either within the distribution of irritative zone or distinct from irritative zone. In other words, ictal onset
zone is not necessarily the same as irritative zone. Nevertheless, the ictal ESI results
are largely in agreement with interictal ESI results in temporal lobe epilepsy [34].