Intraoperative Computed Tomography and Nexframe-Guided Placement of Bilateral Hippocampal-Based Responsive Neurostimulation: Technical Note

Accepted Manuscript Intraoperative CT and Nexframe-guided placement of bilateral hippocampal based responsive neurostimulation: A technical note Kunal Gupta, MBBChir (Cantab), PhD, Jeffrey S. Raskin, MD, MS, Ahmed M. Raslan, MD PII:

S1878-8750(17)30132-8

DOI:

10.1016/j.wneu.2017.01.109

Reference:

WNEU 5206

To appear in:

World Neurosurgery

Received Date: 4 November 2016 Revised Date:

25 January 2017

Accepted Date: 26 January 2017

Please cite this article as: Gupta K, Raskin JS, Raslan AM, Intraoperative CT and Nexframe-guided placement of bilateral hippocampal based responsive neurostimulation: A technical note, World Neurosurgery (2017), doi: 10.1016/j.wneu.2017.01.109. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

Intraoperative CT and Nexframe-guided placement of bilateral hippocampal

RI PT

based responsive neurostimulation: A technical note

Kunal Gupta, MBBChir (Cantab), PhD, Jeffrey S. Raskin, MD, MS, and Ahmed M. Raslan, MD

EP

TE D

Correspondence: Ahmed M. Raslan, MD Department of Neurological Surgery Oregon Health & Science University Mail Code: CH8N 3303 SW Bond Ave. Portland, OR 97239

M AN U

SC

Department of Neurological Surgery, Oregon Health & Science University, Portland, Oregon

AC C

E-mail: [email protected]

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

INTRODUCTION: Surgical resection of a lesion that correlates with seizure onset in patients with epilepsy can dramatically improve seizure burden and quality of life. For bilateral hippocampal lesions, bilateral resection comes with a risk of severe cognitive deficits.

RI PT

Responsive neurostimulation (RNS) devices offers a new modality to treat multifocal lesions in a reversible manner, including bilateral hippocampal stimulation. We describe technical aspects of Nexframe-assisted placement of bilateral NeuroPace mesial temporal electrodes, and case

SC

examples.

METHODS: Retrospective chart review was performed for 4 patients who underwent bilateral

M AN U

mesial temporal RNS placement for medically intractable epilepsy. Operative techniques were assessed and modified. Ambulatory electrocorticographic recordings and a sub-analysis of available data are summarized.

RESULTS: Eight electrodes were placed in 4 patients, who were followed for up to 6 months. 1

TE D

out of 8 electrodes was revised due to vector error greater than 3mm; after surgical technique modification, all subsequent electrodes were reliably placed in a single pass with less than 1.5 mm vector error. Using patients’ seizure diaries, seizure semiologies were correlated with

EP

ambulatory ECoG recording patterns and sub-analyzed; 51.4% were left-sided, 15% right-sided, and 33.6% indeterminate.

AC C

CONCLUSIONS: We report herein the technical nuances of adapting Nexframe to hippocampal based depth electrode RNS system placement. Our group has extensive experience with Nexframe for accurate and safe deep brain stimulation electrode placement. Our preliminary data with bitemporal RNS placement suggests similar accuracy and safety. KEY WORDS responsive neurostimulator/neurostimulation, mesial temporal sclerosis, Nexframe, frameless stereotaxy, epilepsy

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

INTRODUCTION Resective epilepsy surgery can dramatically improve patient quality of life and survival, if patients have a resectable lesion that correlates with seizure focus. In the presence of unilateral

RI PT

hippocampal sclerosis, surgical resection has been shown to result in seizure-free outcomes (Engel class 1) in 50-70% of patients at 1-2 years,1-3 with robust prolonged seizure-free

outcomes in 50% of patients reported at 5 years.4 For bilateral hippocampal sclerosis, bilateral

SC

resection is not possible due to severe cognitive, memory and speech deficits associated with bilateral resection.5 Cukiert et al., report 5 patients with bilateral hippocampal sclerosis, all

M AN U

patients demonstrated unilateral onset and thus received unilateral resection; 4 achieved Engel class 1 seizure-freedom, and 1 achieved Engel class 2 seizure-control.6 In a later report, in a small patient cohort with bilateral hippocampal sclerosis, published by the same group; 2 patients had bilateral ictal onset, however, 80% of activity localized to one side and guided resection, and

TE D

outcomes were similar.7 Of patients who have recurrent seizures after unilateral mesial temporal lobe resection, up to 30% are due to contralateral recurrence; furthermore, equivocal lateralizing ictal activity has been associated with poorer outcomes and complicates surgical treatment for

EP

bilateral hippocampal sclerosis.8

Neurostimulation is a viable treatment option and can offer good seizure control in the

AC C

absence of a resective surgical approach, for example in the presence of multiple lesions or seizure foci in eloquent areas. Open loop stimulation devices include vagal nerve stimulation (VNS); this provides regular afferent abortive stimulation that can be triggered by manual activation of the device. Vagal nerve stimulation (VNS) is typically a salvage procedure for patients with persistent seizures after resection or in the absence of a resectable lesion. NeuroPace (NeuroPace, Mountain View, CA) is a novel closed-loop responsive neurostimulation

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

(RNS) device, which has received indications from the United States Federal Drug Administration (FDA) for medically refractory partial onset epilepsy, in the presence of 1 or 2 seizure foci.9 The device monitors real time electrocorticography (ECoG), and responds to

RI PT

abnormal activity with stimulation. The device is considered modulatory, as opposed to ablative, allowing for use in eloquent cerebral lesions and with bilateral temporal pathology. In addition to a stimulatory function, the device records long-term brain activity via ECoG, which is a novel

SC

benefit, and may enable evaluation of changing seizure patterns over time and guide therapy. Long term RNS monitoring has demonstrated that an average of 40 days of monitoring is

M AN U

required prior to final determination of laterality.10 Long term studies of RNS stimulation have demonstrated 44% seizure reduction at 1 year,11 increasing to 66% over 5 years12, irrespective of prior VNS or resective surgery.13 These studies have focused on mesial temporal pathology, with remaining patients having frontotemporal neocortical pathology.12 Neuropsychological studies

TE D

have demonstrated no worsening in cognitive deficits with bilateral temporal lobe stimulation, and improvements in specific domains, including naming and memory. Adverse effects have rarely been reported, and include a 3.5% infection rate with skin flora primarily, a single

EP

incidence of osteomyelitis in the initial study, and 2.7% rate of hemorrhage.14 The manufacturer (NeuroPace) has not developed a proprietary method of implantation

AC C

(personal communication); therefore it is up to each individual neurosurgeon to develop methodologies for implantation. Prior studies by other neurosurgical groups have reported placement of RNS electrodes primarily by framed based stereotactic systems, and the surgical techniques are not well described.15,16 Our group has extensive experience with placement of deep brain stimulation (DBS) electrodes in the pallidum and thalamus, and we have reported on its accuracy and safety.17 The automatic adaptation of the technical nuances of NexFrame use in

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

DBS placement to placement of deep hippocampal electrodes is not possible without modifications. The purpose of this report is to provide current and/or prospective users of NexFrame with the technical nuances needed for safe use of NexFrame in hippocampal depth

RI PT

electrode placement. At our institute, this is combined with intra-operative computed

tomography (CT; CereTom, NeuroLogica, Danvers, MA) for confirmation of electrode

placement. We then describe follow-up recordings obtained from implanted Neuropace systems.

SC

Two case examples are provided to illustrate the technical nuances.

M AN U

METHODS Patient selection

At our institution patients are selected for bilateral mesial temporal lobe stimulation by the epilepsy multi-disciplinary team. In each case, the presence of bilateral independent temporal

TE D

onset on long-term electroencephalogram (EEG) suggested that unilateral hippocampal resection would be unlikely to provide adequate seizure control. Patients were therefore referred for bilateral mesial temporal electrode placement; the procedures were performed by the senior

EP

author (AMR) using the methodology described below.

AC C

Operative technique

Surgical planning: using a high-resolution 3T magnetic resonance imaging (MRI), an

optimal trajectory of a depth electrode to purchase the entire hippocampus and possibly amygdala, avoiding the ventricle if possible, is planned. Planning should consider the 25-degree coronal angle limit on the Nexframe, which in this case applies to the medial trajectory of the plan. We have found that accommodating the 25-degree limit leads to a more medial entry point

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

on the calvarium. The more medial entry point can prohibit simultaneous mounting of the two Nexframe towers, which could lead to repetition of registration. Therefore a minimum of 7 cm is kept between burr holes, to allow bilateral simultaneous Nexframe mounting.

RI PT

Patient positioning: the patient is placed in a prone position with a DORO radiolucent Mayfield head holder, (pro med instruments, Inc., Cape Coral, FL) taking care to retract the shoulders and allow safe entry of the head into the CereTom portable CT scanner (Figure 1A),

SC

the head is slightly rotated to expose more of the side of the prospective RNS battery implant (Figure 1B). The two pins of the DORO frame are placed contralateral to the side of the RNS

M AN U

battery (Figure 1B). Skull-based fiducial screws are placed into the calvarium, taking into account site of the unsterile registration reference and placement of the calvarial generator (Figure 1C). This can be planned, with regards to the incision and fiducial placement, using the generator template (Figure 1D). An intra-operative thin-cuts CT is performed using CereTom

TE D

and fused to the pre-operative MRI (Figure 1E and F), and the fiducials are registered (Figure 1G). The burr hole sites are identified on the skin using the stereotactic probe and a hand drill used through the scalp after sterile preparation to score the calvarium at the center of each entry

EP

site (Figure 1H). Curvilinear incisions are drawn around the planned entry sites, with a larger incision placed at the planned generator implant site (Figure 1H), cutaneous flaps are reflected

AC C

and burr holes created with an Anspach cranial perforator (Figure 1I). The dura is cauterized, the burr hole edges are burred back and Stimloc base anchoring device is screwed into place (Figure 1J); the locking clips are placed and removed to confirm adequate fixation. The width of the burr hole, over a twist drill hole, increases the degrees of freedom available for targeting with the stereotactic probe. The Nexframe tower is assembled and registration performed with the sterile reference array (Figure 1K). The guidance probe is used to establish the trajectory (Figure 1L)

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

and the microTargeting STar Drive system (FHC, Bowdoin, ME) is then assembled (Figure 1M). The electrode depth stop is placed at the appropriate length for the platform, and using the appropriate distance to target the STar Drive measurement is placed at the appropriate depth, as

RI PT

determined by the Medtronic Framelink targeting system. In our experience this challenged the extreme range of the STar Drive system, which is 115 mm, however remained within its

functioning limits. The cannula was placed and the stylet removed, and the electrode placed until

SC

limited by the depth stop. This was secured, and the cannula sheath withdrawn (Figure 1N). In our initial experience, cannula withdrawal was limited by the targeting system and the electrode

M AN U

could not be accessed and secured proximally; as a result, withdrawal of the cannula resulted in partial withdrawal of the electrode, which therefore required replacement. We then moved to initially passing a short-length cannula with a full-length stylet to guide the electrode: upon withdrawal, the short length cannula allows the electrode to be visualized and secured prior to

TE D

fully removing the cannula sheath. The cannula is withdrawn, the Stimloc locking clip is applied (Figure 1O) and fibrin glue (Tisseel, Baxter, Inc., Deerfield, IL) infused into the burr hole. The electrode stylet is withdrawn and the electrode is withdrawn from the cannula into the Nexframe

EP

tower under direct visualization (Figure 1P). The same process is repeated for the contralateral electrode, which could include re-registration if simultaneous mounting is not possible. The

AC C

Nexframe towers are removed, the Stimloc caps placed and the leads protected until they are secured to the implanted generator (Figure 1Q). The contralateral lead is tunneled under the galea to the incision containing the generator. A full-thickness craniotomy to the shape of the NeuroPace generator is then performed (Figure 1R), the generator secured to the calvarium (Figure 1S), and the leads affixed to this (Figure 1T). A CereTom CT is performed and merged with the pre-operative MRI and surgical plan, to confirm placement of the electrodes along the

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

planned trajectory (Figure 2). Recordings are made to confirm detection of ECoG and the incisions are irrigated and closed.

RI PT

RESULTS NeuroPace RNS placement

Four patients underwent bilateral hippocampal NeuroPace electrode placement. Patient

SC

data are presented in Table 1. Three were male and 1 was female. The median age was 44.5 ± 6.94 years. Over a series of 4 cases including 8 electrodes placed using this method, there was a

M AN U

single electrode replacement due to difficulties encountered with electrode depth and the electrode insertion cannula. All other electrodes were accepted within 2 mm of target and did not require revision or replacement. We have previously evaluated the NexFrame targeting system for vector error in thalamic and pallidal DBS electrode placement and found trajectory and

TE D

vector errors to be 1.24 ± 0.87 mm and 1.59 ± 1.11 mm respectively;17 similar studies by other groups utilizing Nexframe for deep brain stimulation have demonstrated vector errors of 2.78 ± 0.25 mm,18 and 2.8 ± 1.3 mm.19 The deviation encountered in the present study for hippocampal

EP

lead placement is similar to these previously reported values (median 1.65 ± 0.7 mm). The initial case duration was 314 minutes, the subsequent cases’ operative duration ranged from 195 to 226

AC C

minutes (median 225.5 ± 45 minutes). The mean intensive care unit stay duration was 2 ± 1.30 days, with total duration of stay 4.5 ± 2.38 days. At 6 months follow-up, one patient reported the same seizure frequency as pre-op however much reduced duration and intensity of ictal events, one patient reported 50% reduction in frequency and duration, one patient reported no change, one patient is deceased due to events unrelated to surgery. Data are presented as median ± standard deviation.

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

CASE EXAMPLES CASE #1 A 44-year-old male, who developed seizures 2 years prior to surgery, with no prior

RI PT

history of seizures. At the time of his original presentation (2 years prior to surgery), he

developed non-specific symptoms of low-grade fever, dizziness and difficulty sleeping; he was seen in urgent care and was prescribed azithromycin. A few days later he experienced worsening

SC

confusion and had a generalized tonic-clonic seizure. He was transferred to our institution for care; at that point he had continued subclinical electrographic seizures despite multiple

M AN U

concurrent anti-epileptic drugs (AEDs) including phenytoin, levetiracetam, topiramate and benzodiazepines. Cerebrospinal fluid (CSF) was unremarkable. Extensive infectious and autoimmune evaluation was negative, and a high dose solumedrol trial had no clear benefit, however his seizures gradually abated, and after 18 days he was discharged. After discharge, he

TE D

continued to have daily seizures of multiple semiologies, often triggered by stress. He therefore underwent out-patient long-term video-EEG monitoring, which revealed bilateral independent temporal sharp waves and multiple seizures of bilateral independent onset. Brain MRI

EP

demonstrated bilateral mesial temporal sclerosis, and cerebral angiogram demonstrated chronic proximal left middle cerebral artery occlusion with reconstitution by lentriculostriate collateral

AC C

vessels. He is verbal and neurologically non-focal on clinical examination, though he reports episodic memory difficulties.

NeuroPace RNS Recordings The ability to record ECoG was confirmed intra-operatively, and recording abilities were activated immediately. Stimulation parameters were held until the first post-operative visit with

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

his neurologist at 4 weeks. A novel and previously unprecedented ability of an implanted closed loop system is the ability to perform long-term ECoG in the patient’s home environment. Coupled with a physical seizure diary maintained by the patient, this enables neurologists to

RI PT

obtain real-time ECoG and correlate it to patient seizure semiology. The patient described 5 different seizure semiologies: 1) simple partial seizures, 2) complex partial seizures, 3) brief episodes he described as “jolts”, 4) brief jolts accompanied by a remote visual aura, and 5)

SC

feeling of being dissociated from reality, which he described as being in the “third person”. As this patient experienced these events he was able to trigger the device, recording ECoG traces for

M AN U

each of these seizure semiologies for review by his neurologist (Figure 3a-f). In the initial 3 weeks after placement, the device recorded 107 events triggered manually by the patient’s magnet, he endorsed at the clinic visit that he was able to swipe his device for almost all events. Laterality of events was identified by the presence of rhythmic high amplitude activity, of which

TE D

51.4% were left-sided, 15% were right-sided and 33.6% were indeterminate (Figure 4a). These data suggest that a considerable number of his events are below detection threshold, and may not have been captured by conventional staged ECoG. Furthermore, just over 51% were left-sided,

EP

well below the 80% threshold recommended by aforementioned methodology describing unilateral temporal resection in bilateral disease.6 Upon evaluation of the patient’s seizure diary,

AC C

he was able to record semiology and date/time, allowing his events to be correlated with magnettriggered recordings made by the device. He clearly identified a single complex partial seizure, 10 simple partial seizures, 9 dissociative or “3rd person” type events, 5 auras, and 3 “jolts”. The device data suggest a bilateral component to the majority of his seizure types, and highlight the difficulty in accurately determining laterality for his auras and jolts (Figure 4b). It is likely that these will be better defined with prolonged recording time.

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

Post-RNS outcome The patient tolerated the procedure well, and the electrodes were placed using Nexframe

RI PT

and intra-operative CT methodology with high accuracy. The incisions healed well, and reliable ECoG tracings were obtained. His device was activated at low stimulation parameters, and his neurologist continues to follow him closely to determine his optimal stimulation pattern for

SC

seizure control.

M AN U

CASE #2

A 45-year-old female with intractable bilateral temporal lobe epilepsy secondary to limbic encephalitis. At age 26 years she was diagnosed with temporal lobe epilepsy from simple partial seizures. She then failed medication trials including carbamazepine, oxcarbazepine,

TE D

lamotrigine, and tiagabine for lack of efficacy or unwanted side effects. In 2007, she suffered non-paraneoplastic anti-glutamic acid decarboxylase antibody (GAD-Abs) positive limbic encephalitis (LE) diagnosed by serum and CSF anti-GAD Abs with

EP

treatments including pulse cyclophosphamide, plasmapheresis, rituximab, prednisone, and azathioprine. She suffered from refractory complex partial seizures numbering 12 per day,

AC C

beginning with an aura of a deja vu and an odd dreamy/fearful feeling, followed by diminished responsiveness, facial tightening, vacant staring, aimless arm movements, and automatic verbalization.

At this time video EEG and bilateral foramen ovale electrodes captured 50 seizures and

45 of the seizures were of clear right temporal onset with ictal discharge and variable spread over the right temporal surface electrodes. Five of the seizures were subclinical seizures of left

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

temporal onset. A brain MRI with and without contrast showed no abnormal enhancement, but bilateral amygdala and hippocampal increased T2 signal of FLAIR sequence with the left hippocampus appearing atrophic and the right hippocampus appearing edematous. A vagal nerve

RI PT

stimulator (VNS) was placed in 2012 as palliative therapy, with encouraging responses including shorter durations of seizures when using the magnet during seizures, and quicker recovery thereafter. She reported some mild dysphonia but otherwise tolerated the treatment well.

SC

She was determined to not be a neurosurgical candidate, because of the immunologic etiology and bilateral temporal lobe involvement. When evaluated for RNS placement she

M AN U

suffered three disabling seizures per week and 3-5 seizures per day despite VNS therapy, antiepileptic therapy including lacosamide, levetiracetam, pregabalin and immunotherapy including azathioprine.

TE D

Post-RNS outcome

The patient tolerated the implantation procedure well although she experienced an immediate increase in seizure burden, which tapered over postoperative week one. Her incisions

EP

healed well. Her short-term memory is significantly impaired, and she lives in an adult foster home, staying with her mother and daughter 3 days per week. She underwent monitoring for 1

AC C

month after bilateral mesial RNS electrode placement, followed by initiation of stimulation. At 3-month follow-up she had incurred a 50% reduction in her seizure burden.

DISCUSSION

The Nexframe and intraoperative CT method for delivery of DBS electrodes has been previously described by our group.17 One advantage of Nexframe is that it replaces a

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

conventional stereotactic frame, which would allow performance of this surgery at institutions worldwide that for fiscal and/or other reasons do not possess such infrastructure. This in turn allows more patients, who may benefit from the procedure, access to therapy. The absence of a

RI PT

large frame also allows access to the calvarium for generator placement, without requiring

removal of a frame and risking electrode dislodgement. At our institution, we also perform intraoperative CT to confirm electrode placement prior to closure, reducing our requirement for post-

SC

hoc lead revision.

In our operative experience, the methodology for DBS electrode placement using

M AN U

Nexframe was not easily transferred to RNS temporal lobe electrode placement. The differences could be summarized in three main issues and three further subsidiary issues: 1- Hippocampal electrode depth (105-115 mm), which is much greater than conventional DBS depth (75-95 mm), which could challenge the maximum range of the STar Drive delivery system of 115. That could

TE D

be mitigated by adjusting the target to second contact with an offset equal to the distance between second contact and first contact. 2- The need to use of shorter cannula (short of target) fitted with longer stylet (at target) to allow retrieval of the cannula and visualization of electrode

EP

to be able to apply the stimloc clip. This is also caused by the larger electrode depth hence a smaller distance from the skull to the end of the StarDrive where the cannula would be retrieved

AC C

to. 3- The ideal hippocampal depth trajectories may not allowed a 7 cm minimum distance between the two NexFrame towers, which could create the need for sequential placement of towers and dual sterile registration. 4- Finally, the conventional need for 7 cm distance between the two NexFrame towers may not be necessary due to skull geometry that makes occipital placement of NexFrame towers not on the same horizontal plane therefore a smaller linear distance is possible without collision, so we trial the placement of the towers even through the

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

linear measured distance may be less than 7 cm. 5- The procedure is best done prone in DoroFrame if intraoperative CT is contemplated and there is enough clearance using the Ceretom for

placement is possible and allows for easier access to parietal skull.

RI PT

the head and NexFrame towers placed. 6- Slight head tilt to elevate the side chosen for battery

As noted above, the surgical method required extensive adaption from the method used for conventional DBS and is worth reporting to prospective users. The limitations are induced by

SC

depth, skull and trajectory limitations. This technical report provides key details and nuances that permit a stereotactic neurosurgeon to replicate this procedure in their practice, and broaden

M AN U

patient access to these cutting edge technologies.

We have followed up our small series of patients and obtained high fidelity electrocorticographic recordings for epilepsy management. The large amount of data that can be obtained by a closed loop system heralds a new modality of treatment, one that allows rare

TE D

seizures to be captured and analysis of patterns for individual seizure semiologies. In the longterm, this may permit individual stimulation parameters for each seizure pattern and allow truly tailored epilepsy management for individual patients.

EP

We acknowledge the limitations to this study, this is not an outcome study, it is purely a technical note to aid surgeons in their endeavor. We also acknowledge that conclusions regarding

AC C

safety of this method needs to be interpreted in lieu of the small number of subjects therefore is considered a feasibility report as well. We report herein our experience at a single center; evaluation of the long-term efficacy of this technique would benefit from application in multiple centers with analysis of technical and patient outcome.

CONCLUSION

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

We describe in detail the technical nuances and feasibility of adapting the Medtronic Nexframe frameless stereotactic system for the placement of bilateral mesial temporal RNS electrodes. Our group has extensive experience in the use of Nexframe for DBS electrode

RI PT

placement, and we sought to apply the same methodology to RNS. We noted that this required extensive adaptation and knowledge of intricate surgical nuances to perform this with similar accuracy and reliability. We therefore report in detail surgical techniques and nuances that would

SC

allow other practicing neurosurgeon to replicate this procedure. The authors hope this technical report will be of value to neurosurgeons who practice functional stereotactic operative techniques

M AN U

for epilepsy and allows more adult patients who have not been controlled with two or more AEDs access to this relatively new technology.

REFERENCES

Berkovic SF, McIntosh AM, Kalnins RM, et al. Preoperative MRI predicts outcome of

TE D

1.

temporal lobectomy: an actuarial analysis. Neurology 1995;45:1358-1363 2.

Gilliam F, Faught E, Martin R, et al. Predictive value of MRI-identified mesial temporal

EP

sclerosis for surgical outcome in temporal lobe epilepsy: an intent-to-treat analysis. Epilepsia 2000;41:963-966

Kumlien E, Doss RC, Gates JR. Treatment outcome in patients with mesial temporal

AC C

3.

sclerosis. Seizure 2002;11:413-417

4.

Wiebe S, Blume WT, Girvin JP, et al. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311-318

5.

Corkin S. What's new with the amnesic patient H.M.? Nat Rev Neurosci 2002;3:153-160

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

6.

Cukiert A, Sousa A, Machado E, et al. Results of surgery in patients with bilateral independent temporal lobe spiking (BITLS) with normal MRI or bilateral mesial temporal sclerosis (MTS) investigated with bilateral subdural grids. Arq Neuropsiquiatr

7.

RI PT

2000;58:1009-1013

Cukiert A, Cukiert CM, Argentoni M, et al. Outcome after cortico-amygdalo-

invasive recording. Seizure 2009;18:515-518 8.

SC

hippocampectomy in patients with severe bilateral mesial temporal sclerosis submitted to

Harroud A, Bouthillier A, Weil AG, Nguyen DK. Temporal lobe epilepsy surgery

9.

RNS® System - P100026. 2013;

M AN U

failures: a review. Epilepsy Res Treat 2012;2012:201651

http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/DeviceApprovalsan dClearances/Recently-ApprovedDevices/ucm376685.htm. Accessed 1 July, 2016. King-Stephens D, Mirro E, Weber PB, et al. Lateralization of mesial temporal lobe

TE D

10.

epilepsy with chronic ambulatory electrocorticography. Epilepsia 2015;56:959-967 11.

Heck CN, King-Stephens D, Massey AD, et al. Two-year seizure reduction in adults with

EP

medically intractable partial onset epilepsy treated with responsive neurostimulation: final results of the RNS System Pivotal trial. Epilepsia 2014;55:432-441 Bergey GK, Morrell MJ, Mizrahi EM, et al. Long-term treatment with responsive brain

AC C

12.

stimulation in adults with refractory partial seizures. Neurology 2015;84:810-817

13.

Morrell MJ. Responsive cortical stimulation for the treatment of medically intractable

partial epilepsy. Neurology 2011;77:1295-1304

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

14.

Loring DW, Kapur R, Meador KJ, Morrell MJ. Differential neuropsychological outcomes following targeted responsive neurostimulation for partial-onset epilepsy. Epilepsia 2015;56:1836-1844 Lee B, Zubair MN, Marquez YD, et al. A Single-Center Experience with the NeuroPace

RI PT

15.

RNS System: A Review of Techniques and Potential Problems. World Neurosurg 2015;84:719-726

Sillay KA, Rutecki P, Cicora K, et al. Long-term measurement of impedance in

SC

16.

chronically implanted depth and subdural electrodes during responsive neurostimulation

17.

M AN U

in humans. Brain Stimul 2013;6:718-726

Burchiel KJ, McCartney S, Lee A, Raslan AM. Accuracy of deep brain stimulation electrode placement using intraoperative computed tomography without microelectrode recording. Journal of Neurosurgery 2013;119:301-306

Kelman C, Ramakrishnan V, Davies A, Holloway K. Analysis of stereotactic accuracy of

TE D

18.

the cosman-robert-wells frame and nexframe frameless systems in deep brain stimulation surgery. Stereotact Funct Neurosurg 2010;88:288-295

EP

Bot M, van den Munckhof P, Bakay R, et al. Analysis of Stereotactic Accuracy in Patients Undergoing Deep Brain Stimulation Using Nexframe and the Leksell Frame. Stereotact Funct Neurosurg 2015;93:316-325

AC C

19.

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

FIGURE LEGENDS Figure 1. Intra-operative images recording key aspects of NexFrame assisted frameless

RI PT

stereotactic placement of medial temporal RNS electrodes; a) the patient is placed in a radiolucent Doro head holder, b) bilateral occipital incisions are planned with a larger incision at the generator site, c, d) skull-based fiducials are placed, e, f) intra-operative CT is performed and

SC

the CT merged with the pre-operative MRI, g) fiducials are registered with the intra-operative CT with the non-sterile reference frame, h) occipital entry points are planned, i) the bilateral

M AN U

curvilinear occipital incisions are opened and burr holes placed at the planned entry sites, j, k) the NexFrame towers are assembled and frameless registration performed, l) the trajectories are planned using the targeting system, m, n) electrodes are placed with the assistance of Microdrive targeting towers, and o, p) secured to the skull, q, r) the craniotomy is planned for generator

TE D

placement, s, t) the generator is placed and the wires are connected.

Figure 2. Merged image of the post-placement CT and the pre-operative MRI, which

EP

demonstrates placement of the electrode and each contact in the mesial temporal lobe.

AC C

Figure 3. ECoG recordings from the NeuroPace device, correlated to seizure semiologies; a) baseline, b) aura, c) jolt, d) simple partial, e) complex partial, and f) 3rd person.

Figure 4. Laterality of a) total recorded events, and b) events by correlated seizure semiology from patient 1’s seizure diary. These demonstrate the variation in laterality of ictal onset between differing seizure semiologies.

ACCEPTED MANUSCRIPT

Electrode deviation (mm) 1.9 1.65 1.55 1.95

TE D

Intensive care stay (days) 1 3 1 4

SC

Operating time (minutes) 314 195 225 226

M AN U

RNS location Hippocampus Hippocampus Hippocampus Hippocampus

EP

M F M M

Age at surgery (years) 44 45 48 30

AC C

Gender

RI PT

Table 1. Patient demographics, case details, and outcome for bilateral hippocampal NeuroPace electrode placement.

Overall in-patient stay (days) 2 8 3 6

Patient outcome Engel 2 Engel 3 Deceased Engel 4

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

Highlights RNS offers a reversible therapy in bilateral mesial temporal sclerosis

•

Conventional frames can limit surgical exposure for device implantation

•

The authors describe adaption of Nexframe for bitemporal RNS placement

•

RNS provides a high volume of clinically relevant ambulatory electrocorticography

•

These technologies herald a new era for treatment of medically refractory epilepsy

AC C

EP

TE D

M AN U

SC

RI PT

•

ACCEPTED MANUSCRIPT CT Nexframe-guided RNS

Abbreviations

AEDs – anti-epileptic drugs

CT – computed tomography DBS – deep brain stimulation ECoG – electrocorticography EEG – electroencephalogram

LE – limbic encephalitis MRI – magnetic resonance imaging RNS – responsive neurostimulation

AC C

EP

TE D

VNS – vagal nerve stimulation

M AN U

GAD-Abs – glutamic acid decarboxylase antibody

SC

RI PT

CSF – cerebrospinal fluid