In today’s practice, epileptologists and neurosurgeons have several options for seizure localization with intracranial electrodes during phase II evaluations. Traditionally, centers in North America have used subdural electrode grids (SDE or SDG) for intracranial seizure localization. However, improvements in technology led to the popularization of stereo-encephalography (SEEG) using depth electrodes. Epilepsy surgery centers highest in volume now offer both SDE and SEEG for seizure localization. This article provides a general guide for considering SEEG versus SDE for intracranial seizure localization based on our experience with both. Several paradigmatic cases are used illustrate the advantages and disadvantages of the different approaches.
Key points
- •
Subdural electrode grids are traditionally used in North American epilepsy centers for intracranial seizure localization; improvements in technology have led to the popularization of stereo-encephalography (SEEG) using depth electrodes.
- •
Epilepsy surgery centers highest in volume now offer both subdural electrode and SEEG as intracranial strategies for seizure localization.
- •
The choice of technique is complex; it depends on the experience of the epilepsy surgical team as well as intrinsic patient characteristics.
Introduction
The stereo-encephalography (SEEG) method was developed in France by Jean Talairach and Jean Bancaud during the 1950s and has been mostly used in France and Italy as the method of choice for invasive localization in refractory focal epilepsy. Its application in the United States has not been widely adopted until the past 5 years. Subdural grids and strips are the most common method for seizure localization used in North America. Large-volume epilepsy surgery centers now offer both subdural electrode (SDE) and SEEG for seizure localization. The choice of technique can be complex, and will depend on the experience and comfort of the epilepsy surgical team as well as patient characteristics. We offer in this article a general guide for considering SEEG along with SDE. These general guidelines have a footing in our experience with both techniques. Table 1 describes some advantages and disadvantages of each strategy. Table 2 describes several paradigmatic epilepsy types and how well SDE and SEEG apply to each. This table illustrates how candidates for SEEG or SDE surgical approach are selected at our institution.
| SDE | SEEG | |
|---|---|---|
| Advantages | Accurate anatomic electrical/functional mapping of covered brain surfaces. Monitoring and resection within one hospital stay. | Enhanced targeting capability for deeper targets. Improved bihemispheric monitoring as well as mapping of functional networks. More facile placement in the reoperated patient. Given smaller access to implant leads, lessened wound healing morbidity. |
| Limitations | Difficulty in coverage of intrasulcal, deep brain and interhemispheric targets. Multilobar or bilateral sampling challenges. Morbidity associated with surgery: infectious consequences, hemorrhage, cerebral edema, if following previous surgery potential for adhesions and difficult dissection. Inaccuracy encountered when SDG used with additional depth electrodes. | Functional mapping restricted. If hemorrhage associated with placement of leads, can be large scale with significant consequences. |
| Ideal surgical candidates | The patient with a possible cortical target/lesion within eloquent cortex, a virgin surgical resection and/or goal of surgery to perform cortical mapping. | The patient with nonlesional MRI, deep lesions or EZ; and/or the previously operated patient. Also, the patient in whom bilateral exploration is required. |
| SEEG | SEEG or SDE | SDE |
|---|---|---|
| Suspected bitemporal onset Insular onset Nodular heterotopia or tuberous sclerosis with multiple lesions Post craniotomy | Temporal/temporal plus (unilateral) SMA/Midline Occipital Lesion negative near motor cortex | Cortical lesion, near motor cortex Onset is suspected near the language cortex |
The first principle we use in choosing an implantation strategy is to determine whether or not a lesion is present, and if this lesion is thought to correlate with the ictal onset area. If such a lesion is located adjacent to eloquent motor, visual, or speech cortex, we generally favor a grid electrode investigation, perhaps augmented with selected depth electrodes. If seizure semiology includes dyscognitive features that imply early involvement or primary onset in hippocampal-cingulate-insular areas, even with an associated lesion, we mostly prefer to start with stereo-electroencephalogram investigations.
A second guiding principle is concern for bilaterality. For suspected frontal or fronto-parietal onset locations in which we believe bilateral sampling is important, we favor stereo EEG. If onset is clearly unilateral, either stereo EEG or grid recordings can be used. At our institution, if we believe good para-falcine sampling is important, we prefer stereo EEG, although other centers use multiple custom-designed grids or strips. For bilateral temporal lobe sampling, stereo EEG is preferable to burr hole strip electrocortigography (EcoG) investigations, we believe. Posttraumatic epilepsy often falls into this category, although individual examples may differ. Recording effectively from the insula requires stereo EEG.
A third simple principle we adopt is to use stereo EEG in patients who have undergone a previous craniotomy. The morbidity of placing subdural grids in a scarred subarachnoid space, wound and cerebrospinal fluid (CSF) leak issues after grid placement, and the limitations of the previous incision all lead us to prefer stereo EEG in this situation. Grid electrodes after a previous craniotomy would be for unusual circumstances.
Other, more specific patient situations are described later in this article. Different details of semiology and surface EEG patterns in suspected temporal lobe epilepsy present the richest variety and most nuanced decision making, which we attempt to recapitulate through a series of case examples.
Introduction
The stereo-encephalography (SEEG) method was developed in France by Jean Talairach and Jean Bancaud during the 1950s and has been mostly used in France and Italy as the method of choice for invasive localization in refractory focal epilepsy. Its application in the United States has not been widely adopted until the past 5 years. Subdural grids and strips are the most common method for seizure localization used in North America. Large-volume epilepsy surgery centers now offer both subdural electrode (SDE) and SEEG for seizure localization. The choice of technique can be complex, and will depend on the experience and comfort of the epilepsy surgical team as well as patient characteristics. We offer in this article a general guide for considering SEEG along with SDE. These general guidelines have a footing in our experience with both techniques. Table 1 describes some advantages and disadvantages of each strategy. Table 2 describes several paradigmatic epilepsy types and how well SDE and SEEG apply to each. This table illustrates how candidates for SEEG or SDE surgical approach are selected at our institution.
| SDE | SEEG | |
|---|---|---|
| Advantages | Accurate anatomic electrical/functional mapping of covered brain surfaces. Monitoring and resection within one hospital stay. | Enhanced targeting capability for deeper targets. Improved bihemispheric monitoring as well as mapping of functional networks. More facile placement in the reoperated patient. Given smaller access to implant leads, lessened wound healing morbidity. |
| Limitations | Difficulty in coverage of intrasulcal, deep brain and interhemispheric targets. Multilobar or bilateral sampling challenges. Morbidity associated with surgery: infectious consequences, hemorrhage, cerebral edema, if following previous surgery potential for adhesions and difficult dissection. Inaccuracy encountered when SDG used with additional depth electrodes. | Functional mapping restricted. If hemorrhage associated with placement of leads, can be large scale with significant consequences. |
| Ideal surgical candidates | The patient with a possible cortical target/lesion within eloquent cortex, a virgin surgical resection and/or goal of surgery to perform cortical mapping. | The patient with nonlesional MRI, deep lesions or EZ; and/or the previously operated patient. Also, the patient in whom bilateral exploration is required. |
| SEEG | SEEG or SDE | SDE |
|---|---|---|
| Suspected bitemporal onset Insular onset Nodular heterotopia or tuberous sclerosis with multiple lesions Post craniotomy | Temporal/temporal plus (unilateral) SMA/Midline Occipital Lesion negative near motor cortex | Cortical lesion, near motor cortex Onset is suspected near the language cortex |
The first principle we use in choosing an implantation strategy is to determine whether or not a lesion is present, and if this lesion is thought to correlate with the ictal onset area. If such a lesion is located adjacent to eloquent motor, visual, or speech cortex, we generally favor a grid electrode investigation, perhaps augmented with selected depth electrodes. If seizure semiology includes dyscognitive features that imply early involvement or primary onset in hippocampal-cingulate-insular areas, even with an associated lesion, we mostly prefer to start with stereo-electroencephalogram investigations.
A second guiding principle is concern for bilaterality. For suspected frontal or fronto-parietal onset locations in which we believe bilateral sampling is important, we favor stereo EEG. If onset is clearly unilateral, either stereo EEG or grid recordings can be used. At our institution, if we believe good para-falcine sampling is important, we prefer stereo EEG, although other centers use multiple custom-designed grids or strips. For bilateral temporal lobe sampling, stereo EEG is preferable to burr hole strip electrocortigography (EcoG) investigations, we believe. Posttraumatic epilepsy often falls into this category, although individual examples may differ. Recording effectively from the insula requires stereo EEG.
A third simple principle we adopt is to use stereo EEG in patients who have undergone a previous craniotomy. The morbidity of placing subdural grids in a scarred subarachnoid space, wound and cerebrospinal fluid (CSF) leak issues after grid placement, and the limitations of the previous incision all lead us to prefer stereo EEG in this situation. Grid electrodes after a previous craniotomy would be for unusual circumstances.
Other, more specific patient situations are described later in this article. Different details of semiology and surface EEG patterns in suspected temporal lobe epilepsy present the richest variety and most nuanced decision making, which we attempt to recapitulate through a series of case examples.
Brief description of our institutional stereo-encephalography methodology and protocol
SEEG electrode implantation is planned according to a hypothesis for localized seizure onset combining surface EEG with careful review of seizure semiology and radiographic findings. In our practice with SEEG, electrode trajectories are calculated using neuronavigation software with calculations rendered for frame-based or frameless stereotaxy. Preoperative magnetic resonance or computed tomography (CT) angiography is performed to ensure the absence of vascular structures along the anticipated electrode trajectory. Electrode implantation is performed under general anesthesia. A robotic assistant device (such as the robotic system ROSA; Medtech, Montpellier, France) facilitates increased accuracy and reduced operative time, although frame-based or frameless strategies are also used. Electrodes traverse the skull through small twist-drill access sites and are fixed in place for the duration of recording by bolts seated within the bone of the cranial vault ( Fig. 1 ). Electrodes may be inserted using orthogonal or oblique orientation. Intraoperatively, fluoroscopy is obtained to confirm the general accuracy of implanted electrode trajectories in real time. A postimplantation volumetric CT of the brain without contrast is obtained from the post anesthesia care unit. From the CT data set fused with the preoperative MRI, final verification of electrode placement is achieved. With assurance of targeting fidelity, initiation of recording in the monitored epilepsy ward commences. At the completion of recording, SEEG electrodes are removed in the operating room under local anesthesia with sedation as needed. Based on the intracranial SEEG recording, resective surgery is scheduled 2 to 3 months following SEEG electrode removal. This article is organized according to different seizure-onset hypotheses and explains our reasoning for choosing one approach or another.
Stereo-encephalography as the preferred method of invasive evaluation
Bitemporal Lobe Epilepsy
We perform bilateral SEEG explorations for MRI lesion-negative cases with bilateral interictal activity, variable surface EEG patterns, or with semiology that does not match the surface EEG pattern observed. In cases with bilateral radiographic changes, such as bilateral mesial temporal sclerosis, we use SEEG to confirm the laterality of seizures and to help guide placement of Neuropace (Mountain View, California, USA) electrodes or other palliative options if resection cannot be contemplated.
Signs of mesial temporal sclerosis on MRI do not exclude the possibility of neocortical temporal onset, extratemporal, or dual pathology (discussed later in this article). For instance, seizure onset from the temporal lobe opposite to subdural grid (SDG) placement is a known etiology of SDG-guided resection failure. Seizures from the nondominant mesial temporal lobe may be clinically silent or not recognized by the patient or care providers. We feel standard practice should be to place at least one electrode into each hippocampus, or hippocampus and entorhinal/perirhinal/anterior fusiform cortex in lesion-negative cases without obvious semiological lateralization. However, other combinations of contralateral SEEG placement have been described, such as the hippocampus and insula or hippocampus and frontal operculum. This position is supported by data from implanted Neuropace devices. These data, in some patients, confirms that mesial temporal lobe seizures may have “cyclic” patterns. In cyclic mesial temporal seizures, epileptiform discharges are seen during a prolonged period of time (up to several months) in one temporal lobe but may periodically be recorded from the opposite temporal lobe during the next prolonged period of time. Hence, we prefer bilateral temporal lobe sampling in most cases of suspected temporal lobe seizure onset.
In brief, the typical temporal SEEG exploration should include all temporal lobe cortices covering the hemisphere of the most probable seizure origin, and at least one electrode always should be placed into the opposite mesial temporal structures ( Fig. 2 ). If a patient’s clinical data strongly imply bilateral temporal seizure onset, mirror limbic system exploration should be performed bilaterally (broad symmetric bilateral temporal coverage; Fig. 3 ). Bilateral coverage by one electrode is possible for anterior cingulate and orbitofrontal areas ( Fig. 4 ).
Related posts:
Hippocampal Transections for Epilepsy
Minimally Invasive Neurosurgery for Epilepsy Using Stereotactic MRI Guidance
The Role of Stereotactic Laser Amygdalohippocampotomy in Mesial Temporal Lobe Epilepsy
The Stereo-Electroencephalography Methodology
Responsive Direct Brain Stimulation for Epilepsy
Neuromodulation for Epilepsy
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
