Robotic-Guided Stereoelectroencephalography for Refractory Epilepsy: Technique and Nuances
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.319246
Source of Support: None, Conflict of Interest: None
Keywords: Amygdalectomy, epilepsy, MRI negative, robotic guidance, SEEGKey Message: Robotic SEEG placement is a widely accepted and well documented technique globally. It is a safe and efficacious method of SEEG placement with comparable accuracy compared to the frame based technique.
Localization of the epileptogenic zone is complex and not straightforward in a significant proportion of patients with refractory epilepsy. Invasive intracranial monitoring in the form of stereoelectroencephalography (SEEG) and subdural grids are used in the absence of concordance between the clinical, imaging, and the video electroencephalography (VEEG) data. The other indication being magnetic resonance imaging (MRI) negative epilepsy, otherwise known as substrate negative/nonlesional epilepsy. Subdural grid better evaluates the convexity cortex for epileptogenicity, while SEEG is best for the hidden cortex in the depths of the sulcus as well as deep-seated structures.,
To describe the technique and operative nuances of SEEG electrode placement using robotic guidance.
A 28-year-old right-handed lady presented with seizure onset at the age of 15 years with a frequency of 2–3/day initially to 5–6/week with three appropriately dosed antiepileptics during the course of treatment. She had two types of seizures: one was associated with an aura of chest discomfort, palpitations along with oral and bilateral automatisms. There was associated speech and behavioral arrest along with ictal urinary incontinence. The second type had head turning to the right with secondary generalization lasting up to 1 min. The seizures were more during the day time. VEEG revealed bilateral seizure onsets; MRI brain was normal; and positron emission tomography (PET) and single-photon emission computerized tomography (SPECT) were suggestive of left anteromesial temporal and left posterobasal temporal focus, respectively. Magnetoencephalography (MEG) localized to bilateral temporal lobes (left >> right), while neuropsychology localizing to predominantly to the right temporal lobes. In view of lack of a clear hypothesis regarding localization, the patient was planned for SEEG placement.
The plan for placement of the SEEG electrodes is based on the final hypothesis formulated at the comprehensive epilepsy surgery meeting (CSEM), involving the epilepsy surgery team. At our institution, we routinely perform robotic-guided SEEG placement (ROSA, Zimmer Biomet, Warsaw, Indiana). The planning of the trajectories is made prior to the surgery on the robotic platform.
We follow a standardized MRI protocol for robotic surgeries at our institution:
The head is fixed with either a Mayfield clamp or a Leksell frame depending on the site and the number of SEEG electrodes to be placed. A Leksell frame is ergonomically better compared to a Mayfield clamp, as the operative space is completely unhindered for the movement of the robotic arm [Figure 1] and [Figure 2].
>Video Link: https://youtu.be/Hoi3ByS9k7o
Video timeline with audio transcript (Minutes)
0:00–0:15 – This video demonstrates the technique of robotic SEEG electrode implantation and our approach to MRI negative epilepsy
0:15–0:43 – A 28-year-young lady presented with a history of epilepsy for a duration of 13 years. She had two types of semiology: one was associated with an aura of chest discomfort and palpitations. There was a history of automatisms involving bilateral hands followed by speech arrest. The other type of semiology began with head turn to the right side followed by secondary generalization.
0:43–1:17 – VEEG revealed bilateral temporal seizure onset with right-sided predominance
1:19–1:38 – MRI brain revealed no focal lesions and was reported as normal
1:39–2:78 – PET and SPECT localized to left anterior and posterior basitemporal regions, respectively
2:07–2:18 – Neuropsychology assessment showed impaired visual memory more than the verbal suggestive of predominant right temporal involvement
2:21–2:48 – CSEM hypothesis bilateral temporal with clinical semiology, PET/SPECT, and MEG showing predominant right-sided localization, while VEEG and neuropsychology predominantly localized to the right temporal lobe.
2:49–3:09 – A total of 12 SEEG electrodes were planned in bilateral temporal to include the lateral and the mesial structures in addition to left basitemporal, bilateral cingulate, and insula in view of ictal urinary incontinence.
3:14–3:25 – The patient is positioned supine with the head fixed with a Mayfield clamp and attached to the robotic arm.
3:25–3:49 – Alternatively, Laksell frame B may use, which allows free space without hindering the robotic arm in cases of multiple bilateral implantations. There is scope to use three-point fixation for smooth registration, when laser forehead scanning is done.
3:50–4:43 – We routinely use six-point facial skin-based laser registration. This is followed by the mesh registration covering the nose and the forehead, followed by zig-zag laser scanning of the forehead and then manually scanning bilateral scanning of both the temple regions. Once the registration is complete the accuracy check is done for all the six facial points and to correct any errors more than 1 mm by matching these points manually. If the error of >1 mm persists, reregistration is performed.
4:44–5:00 – The robotic arm is driven to the desired trajectory and the entry point is marked as per the laser pointer.
5:01–5:49 – A twist drill craniostomy of 2.1 mm diameter is performed using a pneumatic-powered handheld drill. The robotic adaptor is closely applied to the scalp while drilling and a stopper is tightened at appropriate length accounting for the scalp and bone thickness. This is followed by dural coagulation using a monopolar coagulation.
5:52–7:13 – Anchor bolt is inserted and the distance to the target (A) and distance B (from the top of the adaptor to the top of the anchor bolt is calculated). This is followed by the calculation of the length of the electrode by negating B + 3 mm from A.
7:14–7:55 – Various types of electrodes are available in the market ranging from for simple recording purpose to multimodal functions (recording, stimulation, and lesioning) simultaneously. The desired electrode is inserted with cap fixation over the bolt after removing the stylet.
8: 00–13 – This is followed by O arm confirmation of the accuracy of electrode placement
8:15–8:42 – Accuracy check is performed after placing all the electrodes with a thin-slice postoperative CT brain
8:43–8:56 – The anterior, middle, and posterior basi-temporal along with the amygdalar contacts showed seizure activity; however, no such activity was seen on hippocampal contacts [Figure 3].
9:02–9:32 – Hippocampal sparing lateral temporal neocortical and amygdala resection was planned. Preresection electrocorticography (ECOG) showed grade V discharges, which after anterior temporal resection reduced to grade-IV. Further resection of posterior and basitemporal region achieved a grade III ECOG. Finally, on resecting the amygdala grade II, readings were recorded, while recording with the strip electrode placed on the hippocampus did not reveal any activity. Hence, it was spared. Postoperative course was uneventful and CT brain showed a good resection cavity. Histopathology confirmed the diagnosis of focal cortical dysplasia Type Ib in both the neocortex and the amygdalar specimen. The patient remains to be seizure free at 1-year follow up.
The postoperative course was uneventful and patient is seizure free (ILAE Class 1) at 1-year follow up. The resected specimen of temporal neocortex and amygdala revealed focal cortical dysplasia type-1b on histopathology.
Pearls and Pitfalls
Intracranial EEG is a prerequisite in the evaluation of complex drug refractory epilepsy, not localized despite maximal multimodal evaluation., SEEG has gained universal acceptance over subdural grids in recent times, paralleling the advances in the field of stereotactic neurosurgery. Approximately, two thirds of the gray matter remain buried in the sulcal depths of the brain, which is otherwise inaccessible for recording with the conventional subdural grid. Hence, SEEG electrodes are advantageous in sampling these areas apart from the deep-seated lesions like heterotopias and hamartomas., MRI negative, discordance between MRI, and electrical and clinical findings, as well as ancillary investigations, represent definitive indications for SEEG implantation.
SEEG electrodes implantation can be accomplished with either frame-based or frameless technique. Frameless stereotactic methods, especially the robotic-guided technique is less laborious as well as faster with a minimal scope for human error compared to the frame-based methods. Efficacy and accuracy of the robotic method is comparable to the conventional frame-based technique as demonstrated in various studies. Stereotactic procedures demanding extreme accuracy like deep brain stimulation, lesioning of hypothalamic hamartomas, psychosurgery, robotic thermocoagulative hemispherotomy and even ventriculoperitoneal shunts in slit like ventricles have been performed robotically.,,,,,,,
Robotic-guided SEEG placement is a safe and elegant technique which is faster and equally efficacious compared to the frame-based technique.
Declaration of patient consent
A full and detailed consent from the patient/guardian has been taken. The patient's identity has been adequately anonymized. If anything related to the patient's identity is shown, adequate consent has been taken from the patient/relative/guardian. The journal will not be responsible for any medico-legal issues arising out of issues related to patient's identity or any other issues arising from the public display of the video.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3]