Wednesday 08 September

Wednesday 08 September

Added to your list of favorites
Deleted from your list of favorites
08:00 - 13:00

Satellite course in advanced imaging
Brain Imaging - Postprocessing in Functional Neurosurgery

Moderators: Nadine GIRARD (Marseille, France), Viktor JIRSA (Director INS) (Marseille, France), Ludvic ZRINZO (Professor of Neurosurgery) (London, UK, United Kingdom)
08:00 - 08:30 Basics MR. Jean-Philippe RANJEVA (CNS team leader) (Marseille, France)
08:30 - 09:00 Stereotactic MRI. Ludvic ZRINZO (Professor of Neurosurgery) (London, UK, United Kingdom)
09:00 - 09:30 Statistical approach of the functioning brain. Demian BATTAGLIA (CNRS research scientist) (Marseille, France)
09:30 - 10:00 Morphometry in the brain, VBM …. François MANGIN (France)
10:00 - 10:30 Coffee Break.
10:30 - 11:00 Tractography. Harith AKRAM (Associate Professor) (London, United Kingdom)
11:00 - 11:30 Functionbal connectivity Resting state MRI. Constantin TULEASCA (Lausanne, Switzerland)
11:30 - 12:00 PET. Eric GUEDJ (Directeur DHU Imaging) (Marseille, France)
12:00 - 12:30 Multimodality approaches and ultra-high field MRI. Maxime GUYE (Neurologist) (Marseille, France)
12:30 - 13:00 Discussion.
Espace Vieux-Port

Wednesday 08 September

Added to your list of favorites
Deleted from your list of favorites
08:00 - 14:30

EANS ESSFN European Diploma of Radiosurgery

Moderators: Bodo LIPPITZ (Co-Director) (Hamburg, Germany), Selcuk PEKER (Neurosurgeon) (Istanbul, Turkey)
08:00 - 10:00 I - Arterio Venous Malformations.
08:00 - 08:30 a) Basics, radiobiology predictive factors, angioarchitecture & technical issues. Jean REGIS (PROFESSEUR) (MARSEILLE, France)
08:30 - 09:00 b) Patient selection, natural history, bleeding risk predictors role of imaging modalities, scoring. Piero PICOZZI (Consultant) (Milano, Italy)
09:00 - 09:30 c) Follow up, Outcome, definition of cure, management of failures. Brigitte GATTERBAUER (Gamma Knife) (Vienna, Austria)
09:30 - 10:00 d) Toxicity & complication management and prediction. Bodo LIPPITZ (Co-Director) (Hamburg, Germany)
10:00 - 10:30 Coffee Break.
10:30 - 11:30 II - Fistulas.
10:30 - 10:50 a) Definition indications of the different approaches’ outcome definition of cure. Giorgio SPATOLA (Neurosurgeon) (Brescia, Italy)
10:50 - 11:10 b) Targeting, technic of radiosurgery and outcome (safety efficacy). Selcuk PEKER (Neurosurgeon) (Istanbul, Turkey)
11:10 - 11:30 c) Spinal AVM and FAV indications and results of radiosurgery. Roberto MARTINEZ-ALVAREZ (Neurosurgeon) (Madrid, Spain)
11:30 - 12:30 III - Cavernomas.
11:30 - 11:55 a) Epidemiology, pathology, clinical presentation, natural history and bleeding risk. Mohamed Yassine BELTAIFA (Praticien attaché associé) (Marseille, France)
11:55 - 12:20 b) Role of Radiosurgery and alternatives, efficacy and complication of SRS. Roberto MARTINEZ-ALVAREZ (Neurosurgeon) (Madrid, Spain)
12:20 - 12:30 Discussion & Closure.
12:30 - 13:00 Free Lunch time.
13:00 - 14:00 Manufacturer/demo sessions: How I do with the Gamma Plan.
14:00 - 14:30 Manufacturer/demo sessions: How I do with Element. Selcuk PEKER (Neurosurgeon) (Istanbul, Turkey)
Salle 120

Wednesday 08 September

Added to your list of favorites
Deleted from your list of favorites
14:00 - 18:00

Recent advances in Peripheral Nerve Stimulation (PNS) for intractable pain

Moderators: Denys FONTAINE (Neurosurgeon) (NICE, France), Konstantin V. SLAVIN (professor) (Chicago, USA)
14:00 - 14:30 Mechanisms of action of PNS. Anne BALOSSIER (Dr) (Marseille, France)
14:30 - 15:00 Facial PNS for facial pain. Konstantin V. SLAVIN (professor) (Chicago, USA)
15:00 - 15:30 Occipital Nerve Stimulation for non-primary headache and cervical pain. Sylvie RAOUL (MEDECIN) (NANTES, France)
16:00 - 16:30 Brachial plexus roots PNS for upper limb pain. Denys FONTAINE (Neurosurgeon) (NICE, France)
16:30 - 17:00 Peripheral Nerve Field Stimulation for chronic low back pain. Philippe RIGOARD (Head of Departement Spine-Neurostimulation) (Poitiers, France)
17:00 - 17:30 Dorsal Root Ganglion stimulation. Andrei BRINZEU (MD) (Lyon, France)
17:30 - 18:00 Discussion.
Espace Vieux-Port

Wednesday 08 September

Added to your list of favorites
Deleted from your list of favorites
18:00 - 19:00


Moderators: Antonio GONÇALVES FERREIRA (Head of the Stereotactic and Functional Division) (LISBON, Portugal), Joachim K. KRAUSS (Chairman and Director) (Hannover, Germany), Paul KRACK (Head Center Parkinson and Movement Disorders) (Bern, Switzerland), Jean REGIS (PROFESSEUR) (MARSEILLE, France)
18:00 - 18:15 The DNA of the ESSFN. Marwan HARIZ (Umeå, Sweden)
18:15 - 18:25 Honorary session.
Alim Louis Benabid, Bergman Hagai, Pierre Pollak
18:25 - 19:00 The stunning undersea world of Marseille: portrait of two iconic dwellers in a challenging environment. Jean-Georges HARMELIN (Marseille, France)
Thursday 09 September

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
08:30 - 10:00


Moderators: Patric BLOMSTEDT (Neurosurgeon) (Umeå, Sweden), Antonio GONÇALVES FERREIRA (Head of the Stereotactic and Functional Division) (LISBON, Portugal), Jean REGIS (PROFESSEUR) (MARSEILLE, France)
08:30 - 08:45 Opening Lecture. Antonio GONÇALVES FERREIRA (Head of the Stereotactic and Functional Division) (LISBON, Portugal)
08:45 - 09:15 Stereotactic Navigation: 4 centuries BC. Jean REGIS (PROFESSEUR) (MARSEILLE, France)
09:15 - 10:00 Lesion versus stimulation for OCD & depression. Ludvic ZRINZO (Professor of Neurosurgery) (London, UK, United Kingdom), Veerle VISSER-VANDEWALLE (Head of Dep. of Ster. and Funct. NS) (Cologne, Germany)

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
10:30 - 12:00

Plenary Session 2 - Movement Disorders

Moderators: Jorge GURIDI (Neurosurgery) (Pamplona, Spain), Andres LOZANO (Tasker Chair in Functional Neurosurgery, University of Toronto) (Toronto, Canada), Tatiana WITJAS (neurologist) (Marseille, France)
10:30 - 10:50 PPN stimulation for gait disorders in PD. Stephan CHABARDÈS (head of the unit) (GRENOBLE, France)
10:50 - 11:10 Spinal cord stimulation for gait disorders in PD. Grégoire COURTINE (Prof. Dr. Courtine) (Geneve, Switzerland)
11:10 - 11:30 STN DBS a new treatment of impulse control disorders in PD? Paul KRACK (Head Center Parkinson and Movement Disorders) (Bern, Switzerland)
11:30 - 11:50 Rational past & future of PTT in Parkinson disease. Jorge GURIDI (Neurosurgery) (Pamplona, Spain)
11:50 - 12:00 Discussion.

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
13:30 - 15:00

Plenary Session 3 - LESIONING

Moderators: Kostiantyn KOSTIUK (Neurosurgeon) (KYIV, Ukraine), Ido STRAUSS (Neurosurgeon) (Tel Aviv, Israel), Ludvic ZRINZO (Professor of Neurosurgery) (London, UK, United Kingdom)
13:30 - 13:50 HIFU for Tremor. Jordi RUMIA (Coordinator. Adult and Paediatric Functional Neurosurgery Program) (Barcelona, Spain)
13:50 - 14:10 Is there a future role of LITE in Movement Disorders. Michael SCHULDER (Vice Chair, Neurosurgery) (Lake Success, NY, USA)
14:10 - 14:30 HIFU for OCD. Jin Woo CHANG (Seoul, Korea)
14:30 - 15:00 P&C FUS versus DBS in essential tremor (Vim) & PD (STN) - Pros and cons. Ron ALTERMAN (Chairman) (Boston, USA), Rees COSGROVE (Director, Epilepsy and Functional Neurosurgery) (Boston, USA)

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
15:30 - 16:30

Parallel Session 1
Restorative Surgery

Moderators: Jocelyne BLOCH (Médecin Cadre) (Lausanne, Switzerland), Stéphane PALFI (HEAD) (PARIS, France)
15:30 - 15:45 Restorative transplantation and viral vectors. Nicole DEGLON (Prof.) (Lausanne, Switzerland)
15:45 - 16:00 Restorative transplantation and viral vectors. Stéphane PALFI (HEAD) (PARIS, France)
16:00 - 16:15 #26251 - Efficient representation of hand biomechanics for upper limb neuroprostheses.
Efficient representation of hand biomechanics for upper limb neuroprostheses.


Loss of upper limb function has a major impact on functional independence and quality of life. Neuroprosthetic technologies have been developed to restore function by extracting information from the intact motor cortex to control assistive technologies. However, most current approaches do not take advantage of the latent structure of natural movements.


Hand movements have a large number of possible degrees of freedom, however natural movements tend to be stereotyped combinations of simple movements. The ability to represent complex movements efficiently is valuable. A representation that allows reconstruction of high-dimensional, continuous control signals and classification of movements would allow for more efficient prosthetic systems with increased performance. The ability to interpolate movements beyond the movements used in training, would open the possibility of truly generalizable, naturalistic prosthetic control.



Detailed kinematic data was collected during a series of complex, everyday hand movements.

Linear dimensionality reduction was performed using principal component analysis. The latent structure of hand movements was investigated using the proportion of variance explained by a low-dimensional representation. The quality of movement reconstruction was quantified using the mean squared error between actual and predicted control signals. Classifiers were used to investigate the ability of low-dimensional representations to separate movement classes. Interpolation performance was assessed by measuring the ability to reconstruct movements not contained in the training data.


A deep learning approach was then employed to overcome limitations of these linear approaches. A two-stage autoencoder architecture was developed with recurrent neural networks in both the encoder and decoder. This model was trained using a variety of loss functions in order to tailor the resulting low-dimensional representation to achieve the desired characteristics. The ability of these models to reconstruct, classify and interpolate movements was compared to linear methods.



Naturalistic hand movements can be represented in a low-dimensional space, with >95% of the variance accounted for by 9 dimensions. A linear transformation from this low-dimensional space can accurately reconstruct continuous control signals. Linear classifiers can accurately classify movements within this space, with a softmax regression algorithm achieving accuracy of >80%. New movements can be accurately reconstructed and classified by interpolating their position in latent space, and only a subset of training tasks are required to characterize this space.


A deep learning approach based on a recurrent autoencoder network overcomes many of the limitations of the linear methods developed. Reconstruction error is reduced by >80% using this approach, while customizing the training loss function allows the movement representation to be optimized for movement classification and interpolation, further increasing performance.



Our results indicate that hand biomechanics have a latent structure. This structure can be exploited in order to efficiently represent detailed hand kinematics. This allows for efficient transmission of movement information. Further, this representation allows for the reconstruction and interpolation of movement data, including of previously unseen movements.


This has important implications for neuroprosthetics and brain-computer interfaces. The ability to produce complex movements using low-dimensional data and to generalize accurately to new, unseen movements opens the possibility for complex, naturalistic control of assistive technologies. Further, our results identify a subset of movements required to characterize the space of movement representations, allowing for highly efficient training: by training only the movements required to interpolate generalizable control, the major barrier to translation of excessive training requirements can be overcome.


Overall, our results identify a latent structure in hand biomechanics and a set of hand movement primitives that define this structure. We demonstrate that this structure can be exploited to represent complex hand movements using a low-dimensional representation, and that new hand movements can be interpolated using this latent space model. We show that this approach has potential utility to restoration of naturalistic hand movements in neuroprosthetic systems.

Keogh CONOR (Oxford, United Kingdom), Fitzgerald JAMES

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
15:30 - 19:00

Parallel Session 2
Movement Disorders

Moderators: Juan Antonio BARCIA (Neurosurgeon) (MADRID, Spain), Alexandre EUSEBIO (Professor) (Marseille, France), Niels Anthony VAN DER GAAG (neurosurgeon) (The Hague, The Netherlands)
15:30 - 15:40 #23357 - Long-term evaluation of deep brain stimulation for treatment of Parkinson's disease using a multiple-source, constant- current rechargeable system: 4-year follow-up of a prospective, double-blind RCT.
Long-term evaluation of deep brain stimulation for treatment of Parkinson's disease using a multiple-source, constant- current rechargeable system: 4-year follow-up of a prospective, double-blind RCT.

Objective: The long-term effectiveness of a Deep Brain Stimulation (DBS) device capable of Multiple Independent Current Control (MICC) is assessed in a prospective, sham-controlled, double-blind randomized controlled trial (RCT) where participants are followed for up to 5-years for the treatment of motor symptoms of Parkinson’s disease (PD).

Background: Subthalamic Nucleus (STN) DBS is an established therapeutic option for managing the motor symptoms of PD, and here we report long-term open-label outcomes (up to 4-years) of a double-blind RCT with sham control using an MICC-based DBS device.

Methods: INTREPID ( identifier: NCT01839396) is a multi-center, prospective, double-blinded randomized controlled trial (RCT) sponsored by Boston Scientific. Subjects with advanced PD were implanted bilaterally in the STN with a multiple-source, constant-current DBS system (Vercise, Boston Scientific). Subjects were randomized to either receive active versus control settings for 12 weeks. Upon completion of the 12-week blinded period, subjects received their best therapeutic settings in the open-label phase up to 5 years. During long-term follow-up, motor improvement and quality of life was evaluated using UPRDS, PDQ39, and Schwann and England. Adverse events were also collected.

Results: Analysis of the pre-specified primary endpoint demonstrated a mean difference of 3.03 ± 4.52 hours (p<0.001) between active and control groups in ON time without troublesome dyskinesia, with no increase in antiparkinsonian medication, from post-implant baseline to 12-weeks post-randomization. A 49% (p<0.001) improvement in UPDRS III scores (meds off) at 1-year was previously reported and sustained up to 3-year follow-up (46%, p<0.001). Eighty-nine percent of patients at 3-year follow-up reported high satisfaction with their treatment. At 4-year follow-up, improvement in motor function (41%, UPDRSIII scores) and quality of life was sustained.

Conclusions: Long-term follow-up from the INTREPID RCT demonstrates that the use of a multiple-source, constant-current DBS system is safe and effective with sustained improvement in motor function and quality of life up to 4-years post-implant.

Philip STARR (san francisco, USA), Roshini JAIN, Lilly CHEN, Alexander TRÖSTER, Lauren SCHROCK, Paul HOUSE, Monique GIROUX, Adam HEBB, Sierra FARRIS, Donald WHITING, Timothy LEICHLITER, Jill OSTREM, Marta SAN LUCIANO, Nicholas GALIFIANAKIS, Leo VERHAGEN METMAN, Sepehr SANI, Jessica KARL, Mustafa SIDDIQUI, Stephen TATTER, Ihtsham UL HAQ, Andre MACHADO, Michal GOSTKOWSKI, Michele TAGLIATI, Adam MAMELAK, Michael OKUN, Kelly FOOTE, Guillermo MOGUEL-COBOS, Francisco PONCE, Rajesh PAHWA, Jules NAZZARRO, Cathrin BUETEFISCH, Robert GROSS, Corneliu LUCA, Jonathan JAGID, Gonzalo REVUELTA, Istvan TAKACS, Michael POURFAR, Alon MOGILNER, Andrew DUKER, George MANDYBUR, Joshua ROSENOW, Scott COOPER, Michael PARK, Suketu KHANDHAR, Mark SEDRAK, Fenna PHIBBS, Julie PILITSIS, Ryan UITTI, Jerrold VITEK
15:40 - 15:50 #23378 - Stereotactic venture-intermediate nucleus for essential tremor tremor normalizes aberrant dynamic functional connectivity of extrastriate visual system: a resting-state functional MRI study.
Stereotactic venture-intermediate nucleus for essential tremor tremor normalizes aberrant dynamic functional connectivity of extrastriate visual system: a resting-state functional MRI study.


Tremor circuitry has been commonly hypothesized as driven by one or multiple pacemakers within the cerebello-thalamo-cortical pathway, including the cerebellum, contralateral motor thalamus and M1. However, previous studies, using multiple methodologies, have advocated that tremor could be influenced by visual feedback. Furthermore, visual feedback itself would increase tremor and would be accompanied by abnormal changes, spreading outside the cerebello-thalamo-cortical pathway, in the visual system. 


The study included 42 participants: 12 HC (group 1), 15 patients with essential tremor (ET) (group 2; right-sided, drug-resistant) before ventro-intermediate ncuelus (Vim) radiosurgery (RS) and the same 15 (group 3) one year after left unilateral Vim RS. Imaging was done on a head-only 3T magnetic resonance imaging (MRI) scanner, SIEMENS SKYRA (Munich, Germany, 32-channel receive-only phased-array head coil). We used blood-oxygenation-level-dependent (BOLD) fMRI during resting state to characterize dynamical interactions of the extrastriate cortex, and compare healthy controls (HC) against ET patients before and 1 year after Vim RS. In particular, we applied the co-activation patterns (CAPs) methodology to extract whole-brain spatial patterns of brain activity that occur dynamically over time. 


We found that three different patterns are equally occurring in HC and ET and were reminiscent for the “cerebello-visuo-motor” (1), “thalamo-visuo-motor”(including the targeted thalamus, 2), and “basal ganglia and extrastriate” (3) networks. The occurrence of the first one was decreased in pretherapeutic ET as compared to HC, while the other two showed increased occurrences. This suggests a misbalance between the more prominent cerebellar circuitry and the thalamo-visuo-motor and basal ganglia networks. Multiple regression analysis showed that pretherapeutic standard tremor scores negatively correlated with the increased occurrence of the thalamo-visuo-motor network, suggesting a compensatory pathophysiological trait. The clinical improvement after thalamotomy was related to changes of occurrences of the basal ganglia and extrastriate cortex circuitry, which came back to HC values after the intervention, thus suggesting a role of dynamics of the extrastriate cortex in tremor generation and further arrest after the intervention. 


In sum, we found that resting-state dynamic functional connectivity of the extrastriate cortex can be characterized by three CAPs. These patterns corresponded to (1) a “cerebello-visuo-motor”, (2) a “thalamo-visuo-motor” and (3) a “basal ganglia and extrastriate” network. The first one is decreased in occurrence in pretherapeutic ET as compared to HCs, while the following two rather display increased occurrences. This would suggest a balance between the cerebellar circuitry, and the thalamo-visuo-motor and further basal ganglia ones. The pretherapeutic tremor scores correlated with the abnormal increase in occurrence of the thalamo-visuo-motor network, suggesting a compensatory pathophysiological trait. The improvement in tremor scores after Vim RS is more related to changes within the basal ganglia and extrastriate cortex. All these circuitries aligned to healthy controls after thalamotomy, suggesting a prominent role of the extrastriate cortex and its dynamic connectivity in tremor generation and further arrest after interventional procedures. This opens the discussion for a potential new target for tremor, as we previously advocated in earlier studies, aiming at the extrastriate cortex. These findings support the idea that the visual system plays a prominent role in tremor generation and further arrest after interventional procedures, such as Vim RS.

Constantin TULEASCA (Lausanne, Switzerland), Thomas BOLTON, Jean RÉGIS, Elena NAJDENOVSKA, Tatiana WITJAS, Nadine GIRARD, Mohamed FAOUZI, Jean-Philippe THIRAN, Meritxell BACH CUADRA, Marc LEVIVIER, Dimitri VAN DE VILLE
15:50 - 16:00 #23786 - Asleep deep brain stimulation for essential tremor using a machine-learning approach for targeting: preliminary results of a phase-2 clinical trial (OPTI-VIM).
Asleep deep brain stimulation for essential tremor using a machine-learning approach for targeting: preliminary results of a phase-2 clinical trial (OPTI-VIM).


DBS of the VIM nucleus is an efficacious treatment for refractory essential tremor, although targeting the intra-thalamic nuclei remains challenging. In a previous work, using machine-learning algorithms, we were able to predict a clinical target for DBS in essential tremor. The learning database consisted in clinical and radiological features of patients previously operated on with optimal outcomes. The OPTI-VIM trial (NCT03760406) is now ongoing to validate this approach.

Patients and Methods:

In this prospective bi-centric (Lyon and Bordeaux), non-comparative, phase-2 clinical trial, we planned to include 22 patients with severe essential tremor despite optimal medical management, aged between 18 and 75 years, with normal MRI, without cognitive impairment (MDRS score ≥ 130) or depression (BDI scale < 20).

The primary endpoint is the efficacy of the procedure on tremor as assessed by the improvement on the Fahn-Tolosa-Marin (FTM) scale between the pre- and post-operative assessments 3 months after surgery. Secondary endpoints are (1) the efficacy of the procedure on tremor as assessed by accelerometry recordings; (2) complications related to surgery and neurostimulation-related side effects, mainly dysarthria and ataxia assessed by the SARA scale (scale for assessment and rating of ataxia); (3) improvement in quality of life’s mPDQ-39 scale between the pre-operative and post-operative assessments; and (4) the stereotactic accuracy was evaluated by calculation of the Euclidian distance between the target and the electrode by co-registration between the marked MRI with the target and the postoperative CT scan.

The target was planned with the “Optim-DBS” software we developed, on a 3D T1 MRI at 1.5 or 3 Tesla. DBS surgery was performed under general anaesthesia, without intra-operative clinical and electrophysiological testing.

Preliminary results:

Seven patients underwent surgery under general anaesthesia between June 2019 and January 2020 (4M/3F, mean age 63 years old).

The (pre-operative / post-operative) FTM scale means were 55/23.3. The mean improvement of the tremor was 55% on the FTM scale and 76% for the subscore of upper limb tremor. These scores were confirmed by accelerometry.

The (pre-operative / post-operative) SARA means were 5.9/4.2. Two patients worsened their SARA, the first one significantly worsened his ataxia with a score which increased from 4.5 to 10.5. The second one increased his score non-significantly from 3 to 4.

The (pre-operative / post-operative) PDQ39 means were 42.8/18.5, respectively. The mean improvement of quality of life were 58% on the PDQ39.

There were no surgical or device-related complications during the 3 months of follow-up.

The mean distance between the target and the electrode surface was 0.98mm (min 0, max 2.3).


Asleep DBS for essential tremor using our machine-learning model for targeting may be a safe, efficient procedure leading to outcomes comparable to those published in the literature for standard awake DBS surgery. The final conclusions will be drawn once the study has been completed.

Julien ENGELHARDT, Emile SIMON, Dominique GUEHL, Stephane THOBOIS, Nathalie DAMON-PERRIERE, Teodor DANAILA, Louis NADAL, Pierre BRIAU, Olivier BRANCHARD, Nicolas AUZOU, Marie BONNET, Wassilios MEISSNER, Antoine BENARD, Pierre BURBAUD, Polo GUSTAVO, Patrick MERTENS, Emmanuel CUNY (Bordeaux)
16:00 - 16:10 #23788 - Study of MRgFUS thalamotomy lesions in essential tremor and Parkinson disease in a large cohort of patients.
Study of MRgFUS thalamotomy lesions in essential tremor and Parkinson disease in a large cohort of patients.


To assess the clinical and radiological outcome of 134 patients with esential tremor (ET) and tremor dominant Parkinson’s disease (PD) at 6 month follow up treated by magnetic resonance guided focused ultrasound (MRgFUS).


The efficacy of unilateral Vim (Ventralis Intermedius) thalamotomies by MRgFUS in tremor has been reported. Some authors consider this nucleus as part of the VLp (Ventral Lateralis posterior). It receives projections from the dentato-rubro-thalamic tract (DRT).

MRgFUS is an image-guided, incision-less procedure based on the focalization of high intensity ultrasound beams in a precise area of the brain where it produces a thermal ablation.

The resultant lesion usually presents 3 different areas visible in magnetic resonance (MR): a central necrosis (i) surrounded by cytotoxic (ii) and vasogenic (iii) edema.


Treatment was performed using MRgFUS equipment (Ex Ablate 4000, InSightec) coupled with a high field MR (3T Skyra, Siemens). Skull density ratio (SDR) was calculated in all patients before treatment. Parameters like intensity and duration of sonications were modulated during the procedure to achieve the target temperature.

Pre-treatment and immediately after treatment 3T MR studies were acquired in all patients. Classic coordinates were used for targeting of the Vim. Probabilistic tractography of the DRT was performed placing the regions of interest (ROIs) in the thalamus, premotor cortex and dentate nucleus (SyngoVia, Siemens).

Volumetry (areas i, ii and iii) and diameter of the lesions were analyzed by manual delineation on 3D T2 SPACE MR sequences with the Iplannet planification software (Brainlab). The volumes were exported as 3D objects and were superimposed onto individualized probabilistic atlas of the thalamic nuclei. Overlapping regions were studied.

Tremor severity was assessed at baseline, 1 (1m), 3 (3m), and 6(6m) month (follow-up using the Clinical Rating Scale for Tremor (CRST) in ET patients and the tremor items of the Unified PD Rating Scale (UPDRS) in PD patients.

Side effects were classified as mild (slight disturbances), moderate (partial impairment of daily activities), or severe (established neurological deficit with any level of disability).

For the statistical analysis, Mann Whitney U and Pearson correlation tests were applied with the STATA statistical tool.


134 patients (87 ET, 47 PD; mean age 72.9 y.o; male 77%) underwent Vim thalamotomy contralateral to the patient´s hand tremor from october 2018 to december 2019. In patients with ET, the contralateral CRST score improved 80,98% (1m, 83 patients), 74% (3m,58 pt), 78% (6m, 34 pt). In PD patients, tremor improvement was 84,78% (1m, 37 pt), 78,84% (3m, 24 pt) and 65,35% (6m,16 pt). 

The most common side effects were gait instability and dysmetria, followed by dysarthria and finger or lip/tongue paresthesias. At 1m, 64% of the patients presented adverse events; 27,5% at 3m and 12,5% at 6m. More than 80-90% were mild, 10-15% were moderate and 0-4% were severe along the 6m period.  

Average volume of the lesions was 11,5 (i); 157,6(ii) and 376,3 (iii) cm3. The average outer volume of the lesions was 404,4 cm3. Mean diameter: 8,9 mm. Larger volumes were directly correlated with a greater number and severity of side effects at 1m, 3 m and 6m (moderate correlation; p<0,05).

Overlapping of individual atlas segmentation and lesions was performed in 47 patients. The mean overlapping percentage of the atlas VLp and the lesion was 70.44% (i), 58,37(ii) and 47,08 (ii). The overlapping region of the necrosis was inferior in those patients whose tremor responded worse to treatment (p<0,01)

Pre-treatment DRT tracts were used as an adjuvant tool during treatment to adjust the targeting coordinates when initial clinical response was poor.

Mean temperature (t) was 58,22º. SDR values were directly correlated with t (r=0,43; p<0,001); i and ii volumes (r=0,28 and 0,37; p<0,001); and inversely correlated with intensity (r=-0,64; p<0,01) and sonication duration( r=- 0,53; p<0,01).


MRgFUS thalamotomy for ET and PD is associated with a great improvement of tremor at 6m follow-up. Although side effects are frequent, the majority are mild. Larger lesions have an impact on its occurrence. SDR values play a role on the treatment outcome.

16:10 - 16:20 #23801 - Structure-Function Relationship of the Posterior Subthalamic Area with Directional Deep Brain Stimulation for Essential Tremor.
Structure-Function Relationship of the Posterior Subthalamic Area with Directional Deep Brain Stimulation for Essential Tremor.

Objective Deep Brain Stimulation (DBS) of the Posterior Subthalamic Area (PSA) is an emergent target for the treatment of Essential Tremor (ET). Due to the hetereogenous and complex anatomy of the PSA  it remains unclear which specific structures mediate tremor suppression and different sorts of side effects. The objective of the current work was to yield a better understanding of what anatomical structures mediate different clinical effects observed during directional DBS of the PSA.

Methods We analysed a consecutive series of 12 ET patients. Imaging analysis and systematic clinical testing performed 4-6 months postoperatively yielded location, clinical efficacy and corresponding therapeutic windows for 160 directional contacts. Overlap ratios between individual stimulation activation volumes (VTA) and neighbouring thalamic and subthalamic nuclei as well as individual fiber tracts were calculated. Further, we generated stimulation heatmaps to assess the area of activity and structures stimulated during tremor suppression and occurrence of side effects.

Results Stimulation of the Dentato-Rubrothalamic Tract (DRTT) and the Zona incerta (Zi) was most consistently correlated with tremor suppression. Both individual and group analysis demonstrated a similar pattern of activation for tremor suppression and different sorts of side-effects. Unlike current clinical concepts, induction of spasms and paresthesia were not correlated with stimulation of the corticospinal tract and the medial lemniscus.  Furthermore, we noticed a significant difference in the therapeutic window between the best and worst directional contacts while the best directional contacts did not provide significantly larger windows than omnidirectional stimulation at the same level.

Conclusion PSA DBS is effective in suppressing all aspects of ET but can be associated with concomitant side effects limiting the therapeutic window. Activation patterns for tremor suppression and side effects were similar and predominantly involved the DRTT and the Zi. We found no different activation patterns between different types of side effects and no clear correlation between structure and function. Future studies with use of more sophisticated VTA modelling taking into account fiber heterogeneity and orientation may eventually better delineate these different clusters, which may allow for a refined targeting and programming within this area.

Andreas NOWACKI (Bern, Switzerland), Jean-Philippe LÉVY, Anh Khoa NGUYEN, Lennard LACHENMAYER, Ines DEBOVE, Gerd TINKHAUSER, Katrin PETERMANN, Joan MICHELIS, Michael SCHÜPBACH, Claudio POLLO
16:20 - 16:30 #23885 - Long-term local field potential-guided directional deep brain stimulation in Parkinson’s disease.
Long-term local field potential-guided directional deep brain stimulation in Parkinson’s disease.


High-frequency deep brain stimulation of the subthalamic nucleus is the preferred surgical treatment for advanced Parkinson's disease. A recent innovation in this technique is the advent of directional leads allowing for current steering, which can increase the stimulation threshold for adverse effects and widen the therapeutic window. However, selecting programming settings is time consuming as it entails a thorough monopolar clinical review. To overcome this, directional stimulation programming may be guided by intraoperatively recording local field potential beta oscillations (13-35Hz) from the implanted leads.


Objective of this study is first, to evaluate whether the power of beta oscillations intraoperatively recorded from directional local field potentials can predict the clinically most effective contacts; and second, to assess the clinical impact of beta-based programming of directional deep brain stimulation at long term follow-up.


We conducted a non-randomized, open-label, prospective study with 24 Parkinson`s Disease patients (mean age 55.8 years, 66.7% male), divided in two groups. In the group A (14 patients, 2016-2018), we investigated whether beta activity in the directional contacts correlated with clinical efficacy (defined by clinical monopolar review, blinded to beta activity). Stimulating parameters were selected according to the clinical monopolar review, assessed with a mean follow-up of 27 months (15-38 months). In the group B (10 patients, 2018-2019), stimulating parameters were selected according to beta activity and assessed with a mean follow-up of 14 months (6-21 months).


Neurophysiological results showed that the strongest correlation between beta activity and clinical efficacy was obtained with the low-beta sub-band (13-20Hz) rather than with total beta activity (13-35 Hz). Contacts with highest beta peaks increased the threshold window by 25%. Selecting the two contacts with highest beta peaks provided 82% probability of selecting the best clinical contact. Clinical results showed similar improvement of motor features with reduced need of dopaminergic medication in both groups.  In group A (stimulation parameters based on clinical monopolar review), showed a 78% UDPRS-III reduction and 60% LEDD reduction. In group B (stimulation parameters based on beta peak power), showed a 75% UDPRS-III reduction and 62% LEDD reduction. Importantly, this clinical improvement was maintained at long-term follow up.


Our results demonstrate the clinical efficacy of directional stimulation over 3 years of follow up and validate the use of intraoperative local field potentials beta oscillations to guide the initial programming of long-term directional deep brain stimulation in Parkinson`s disease. Local field potential-guided programming provides a unique opportunity to adjust the stimulation for each individual patient, and the combination of physiological guidance with directionality will allow us to shape activation volumes tailored to patient needs  

16:30 - 16:40 #23910 - Effects of deep brain stimulation on depressive symptoms and cerebral glucose metabolism in parkinson’s disease.
Effects of deep brain stimulation on depressive symptoms and cerebral glucose metabolism in parkinson’s disease.

Background:Subthalamic nucleus (STN) deep brain stimulation (DBS) can ameliorate motor symptoms in Parkinson’s disease (PD), however, its utility as well as mechanism in reducing comorbid major depression remained uncertain. Here, we shed light on the neural mechanism of STN-DBS in managing comorbid treatment-resistant depression (TRD) in PD.

Objective:To assess motor symptoms, depressive symptoms, quality of life, and regional cerebral glucose metabolism before and after STN-DBS in PD patients with TRD.

Method:We prospectively studied 18 PD patients with TRD (d-PD) who underwent bilateral STN-DBS and an age- and sex-matched group containing 16 health controls. Clinical assessments were performed before and after STN-DBSat approximately 1-year follow-up. Cerebral regional glucose metabolism was assessed by 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) before and after STN-DBS.

Results:After STN-DBS, the d-PD patients presented a substantial reductionon both motor and depressive symptoms. Improvement was also observed in their daily functioning and quality of life, whereas cognitive functioning was not significantly modified. Compared to controls, d-PD patients had widespread abnormalities in cerebral regional glucose metabolism before surgery  which were partially restored after STN-DBS. Moreover, improvement in depressive symptoms were associated with metabolic patternrestoration in brain regions implicated in emotional regulation (i.e., right inferior frontal gyrus, right orbitofrontal cortex and right lingual gyrus.).

Conclusion:STN-DBS may offer additional benefits in d-PD patients. The antidepressant effects appears to be associated with the restoration of abnormal glucose metabolism in widely distributed networks involved in emotion, motivation, and attention.

Xiaoxiao ZHANG (Shanghai, China), Bomin SUN, Chencheng ZHANG
16:40 - 16:50 #23914 - MRgFUS thalamotomy sensory side effects follow the thalamic structural and functional homonculus.
MRgFUS thalamotomy sensory side effects follow the thalamic structural and functional homonculus.

Introduction: Magnetic Resonance-guided focused ultrasound thalamotomy (MRgFUS) is an effective treatment for tremor, however, side effects may occur. The purpose of the present study was to investigate the spatial relationship between thalamotomies and specific sensory side effects as well as their functional connectivity with somatosensory cortex and relationship to the medial lemniscus (ML).

Methods: Of 103 patients treated with MRgFUS for tremor, 17 developed sensory side effects after thalamotomy persisting 3 months after the procedure. These side effects were categorized into four groups based on the location of the disturbance: face/mouth/tongue numbness/paresthesia, hand-only paresthesia, hemi-body/limb paresthesia, and dysgeusia. Then, areas of significant risk (ASR) for each category were defined using voxel-wise mass univariate analysis and overlaid on corresponding odds ratio maps. The ASR area associated with the maximum risk was used as a region-of-interest in a normative functional connectome to determine side-effect specific functional connectivity. Finally, each ASR was overlaid on the medial lemniscus derived from normative template.

Results: Thalamotomies producing sensory side effects extended posteriorly into the principle sensory nucleus of the thalamus. Thalamic regions associated with significant risk of the specific sensory side effects were distributed according to the somatotopy of the sensory thalamus, with the dysguesia ASR being located the most medial. ML fibers touched by ASRs were found to be organized according to the known somatotopic organization of the ML below the thalamus and at the level of the midbrain. Positive functional connectivity was found between each of the sensory-specific thalamic seeds and the primary somatosensory cortex and insular cortex.

Discussion: Distinct regions in the sensory thalamus may give rise to specific side effects when included in a thalamotomy lesion. These findings demonstrate the relationship between the sensory thalamus, ML, and bilateral sensory cortex. The functional connectivity patterns found between the sensory-specific thalamic seed regions and the insular cortex support the role of the insula in primary processing of gustatory information and also in multi-sensory integration.

Michelle PAFF (Orange, CA, Canada), Alexandre BOUTET, Jurgen GERMANN, Gavin ELIAS, Clement CHOW, Aaron LOH, Walter KUCHARCZYK, Alfonso FASANO, Michael SCHWARTZ, Andres LOZANO
16:50 - 17:00 #23927 - Surgery-related 30-day morbidity of functional stereotactic neurosurgery in a large cohort of 600 operations.
Surgery-related 30-day morbidity of functional stereotactic neurosurgery in a large cohort of 600 operations.


In this retrospective study we analyzed the surgery and hardwarerelated morbidity of deep brain stimulation within 30 days after surgery.


600 functional stereotactic operations (DBS electrode implantation or radiofrequency lesioning) were performed from 1997 to 2018. All procedures were performed or supervised by the senior neurosurgeon in three different centers using the same technique. The target was determined with CT-stereotactic surgery supplemented by MR imaging and approached via a guiding cannula. MER was performed in 2/3 of the cases via a single channel technique, supplemented by additional trajectories if decided necessary. Surgery was performed while patient was awake in 531 instances. Postoperative CT scans obtained within 24 hours after surgery were searched for haemorrhage of any size at any site. Hardware- or other surgery-related complications within 30 days after surgery were documented.


A total of 251 women and 349 men with a median age of 55 years were operated. The majority of patients underwent DBS (580), while a subset had radiofrequency lesioning procedures (20). Overall in 19 (3.25 %) procedures an intracranial haemorrhage was detected, which was asymptomatic in all patients except in 1 patient (0,16 %), who had a persistent mild hemiparesis on the right side. Early infections were noticed in 3 patients (0,53 %), 1 at the site of the IPG, 2 at the site of the cranial skin incision. Four patient (0,7 %)  had an intraoperative seizure, and 2 (0,3 %) had a seizure one day after surgery. Three patients (0,53 %) had clinically relevant intraoperative air embolism, and in 1 patient (0,16 %) pulmonary embolism occurred. One patient (0,16 %) had an acute coronary syndrome during IPG implantation, in four patients (0,7 %) surgery terminated because of cardiac coronary syndromes were suspected, which however, weren’t confirmed later.


Stereotactic functional surgery is a generally safe procedure with low morbidity and no mortality when performed by an experienced team. Intraoperative haemorrhage constitutes the highest surgically related complication. Awake surgery is tolerated without relevant problems by the majority of patients. Infections within 30 days after surgery are rare.

Joachim RUNGE (Hannover, Germany), Assel SARYYEVA, Marc WOLF, Christian BLAHAK, Christoph SCHRADER, Holger H. CAPELLE, Hansjörg BÄZNER, Mahmoud ABDALLAT, Joachim K. KRAUSS
17:00 - 17:10 #23933 - Globus pallidus stimulation as treatment for camptocormia in Parkinson’s disease.
Globus pallidus stimulation as treatment for camptocormia in Parkinson’s disease.

Objective: Camptocormia is a common and often debilitating postural deformity in Parkinson's disease (PD). Few treatments are currently effective. Deep brain stimulation (DBS) of the globus pallidus internus (GPi) shows potential in treating camptocormia, but evidence remian limited to case reports. We here report the effect of GPi-DBS in the treatment of camptocormia in a retrospective cohort of PD patients.


Methods: Video recordings of patients who recieved GPi-DBS were retrospectively reviewed. The total and upper camptocormia angles (TCC angle and UCC angle), were used to compare camptocormia alterations. The Unified Parkinson’s Disease RatingAssessment Scale Part III (UPDRS-III) was used to assess the patients’ motor symptoms.


Results: Thirty-six consecutive patients with advanced PD who underwent GPi-DBS were reviewed. Twelve patients manifested per-operative camptocormia: eight had lower camptocormia (TCC >30°; TCC-camptocormia), four had upper camptocormia (UCC >45°; UCC-camptocormia). Mean follow-up time was 7.3 ± 3.3 months. GPi-DBS improved TCC-camptocormia by 40.4% (angle from 39.1° ± 10.1° to 23.3° ± 8.1°, p=0.017) and UCC-camptocormia by 22.8% (angle from 50.5° ± 2.6° to 39.0° ± 6.7°, p=0.012) (Fig.1).Larger improvement was seen in patients with a larger pre-surgical TCC angle. No significant outliers among the leads were identified through the location analysis (Fig.2).


Conclusion: Our study demonstrates the potential effectiveness of GPi-DBS for treating camptocormia in PD patients. Controlled studies with larger numbers of unselected camptocormia patients should extend our findings.

Figure Legends

Fig. 1. Pre- and post-surgical total- (A) and upper (B) camptocormia angles (TCC/UCC angles) in patients with lower camptocormia (blue symbols and lines), upper camptocormia (red symbols and lines) and without camptocormia (green symbols and lines). *p < 0.05; ns: not significant. 

Fig. 2. The DBS leads of all patients with TCC-camptocormia (A) and UCC-camptorcormia (B) were merged on the T2-weighted (upper) and T1-weighted (lower) Montreal Neurological Institute (MNI) templates, with active contacts marked in red. Masses with yellow described the location of the STN, red for the red nucleus, green for the GPi, blue for the globus pallidus externus (GPe).

Yijie LAI, Yunhai SONG (Shangahi, China), Daoqin SU, Linbin WANG, Chencheng ZHANG, Jorik NONNEKES, Bastiaan BLOEM, Dianyou LI
17:10 - 17:20 #23943 - Subthalamic nucleus and EEG signatures underlyingleg force modulation in patients with Parkinson's disease.
Subthalamic nucleus and EEG signatures underlyingleg force modulation in patients with Parkinson's disease.

Impairments of gait and balance are amongst the most disabling and least well-understood symptoms of Parkinson's disease (PD). Well-established neuromodulation therapies for PD are highly effective for the treatment of upper-limb motor symptoms such as tremor or bradykinesia. However, these approaches exhibit modest results for alleviating locomotor deficits, presumably because the control of gait involves additional pathways governed by different neural dynamics.

A key limitation holding back the design of novel therapies is the lack of mechanistic readouts that may help understand and capture pathological neural activity patterns related to leg dysfunction. Here, we sought to identify the cortical and subcortical neural signatures underlying leg movements in PD. We recorded the local field potentials from deep brain stimulation electrodes implanted in the subthalamic nucleus (STN) in conjunction with 64-channel cortical electroencephalographic (EEG) in patients with advanced Parkinson's disease, as they performed a well-controlled leg motor task requiring single-joint modulations of force (high and low force).

We found distinct modulations in STN and EEG frequency bands that strongly correlate with leg motor effort, predominantly in the low beta (~13-20Hz), high beta (~20-35Hz) and gamma (35-90Hz) bands. In particular, high beta desynchronization was time-locked to leg muscle recruitment and maintained throughout force production with amplitudes that correlated with effort, both in STN and EEG (predominantly central sensorimotor electrodes) signals. Low beta and low gamma (35-50Hz) bands also exhibited modulations with leg movements, but with behaviors that differed across patients and between EEG and STN. Interestingly, force-related correlates were identified in STN and EEG regardless of the mobilized leg joint (ankle or knee) or of the direction of movement (flexion or extension), although power differences across force levels varied for each condition.

Our results highlight clear neural patterns underlying effort-related leg movements, both in STN and EEG signals, which hold promises to help clarify the role of such networks during adaptations in gait, initiation and turning, or during episodes of shuffling steps or freezing. Our results additionally emphasize differences across patients in specific frequency bands, which may be related to the pathology of each subject, and which may critically need to be accounted for when developing therapies that address leg-related disorders in PD.

Yohann THENAISIE, Andrea GALVEZ, Kyuhwa LEE, Charlotte MOERMAN, Mayte CASTRO-JIMENEZ, Elvira PIRONDINI, Grégoire COURTINE, Jocelyne BLOCH, Eduardo MARTIN MORAUD (Lausanne, Switzerland)
17:20 - 17:30 #23944 - Subthalamic nucleus activity patterns correlate with modulations in leg muscle synergies during locomotion in Parkinson’s patients.
Subthalamic nucleus activity patterns correlate with modulations in leg muscle synergies during locomotion in Parkinson’s patients.

Most patients with advanced Parkinson's disease (PD) suffer from disturbances of gait and balance, which severely affect their everyday mobility, independence and quality of life. Unlike upper-limb motor symptoms, these deficits respond poorly to commonly available therapies. A key limitation holding back the design of better focused, evidence-based therapies stems from the lack of biomarkers that correlate pathological neural activity patterns and leg dysfunction during gait. However, this identification is contingent on technologies, concepts and methodologies allowing to simultaneously record and link brain states to whole-body biomechanical features representative of gait deficits.

Here we employed a whole-body gait monitoring platform, operating wirelessly and in real-time, to map the activity of the subthalamic nucleus onto kinematic and muscle activity patterns while patients executed a range of locomotor activities. We contrasted the neural correlates underlying basic walking, turning and precision locomotion in order to uncover the STN involvement in locomotor tasks requiring different levels of muscle recruitment and effort.

Similarly to our previous observations during a single-joint leg motor effort task, we found that STN activity patterns exhibit distinct modulations in low beta (~13-20 Hz), high beta (~20-35 Hz) and low-gamma (~35-50 Hz) power during gait, which were aligned to the recruitment of well-defined leg muscle groups (synergies) during each phase of gait, initiation and turning. Although STN patterns displayed significant differences across patients, they all exhibited clear monotonic modulations with effort that correlated in time with increases in muscle synergy activations related to propulsion. We additionally identified specific modulations in patients suffering from freezing of gait.

These results reinforce the link between STN activity patterns and the modulation of vigor in locomotor networks, and may help predict the signatures that underlie propulsion-related deficits of gait and balance in PD, such as asymmetry, shuffling steps or eventually freezing of gait. Our results additionally emphasize particularities in individuals that exhibit freezing, which may critically need to be accounted for when developing therapies that address leg-related disorders in PD.

Yohann THENAISIE, Charlotte MOERMAN, Kyuhwa LEE, Flavio RASCHELLÀ, Andrea GALVEZ, Mayte CASTRO-JIMENEZ, Elvira PIRONDINI, Grégoire COURTINE, Jocelyne BLOCH, Eduardo MARTIN MORAUD (Lausanne, Switzerland)
17:30 - 17:40 #23978 - Bilateral Subthalamic Nucleus Theta Frequency Stimulation Improves Episodic Verbal Fluency in Patients with Parkinson's Disease: A Double-Blinded Randomized Crossover Trial.
Bilateral Subthalamic Nucleus Theta Frequency Stimulation Improves Episodic Verbal Fluency in Patients with Parkinson's Disease: A Double-Blinded Randomized Crossover Trial.

Cognitive outcomes following deep brain stimulation (DBS) in Parkinson’s disease (PD) have gained much attention with particular concern for verbal fluency. Current stimulation paradigms utilize high (gamma) frequency stimulation for optimal motor benefits; however, little work has been done to optimize stimulation parameters for cognition. Recent evidence implicates a role of subthalamic nucleus (STN) low (theta) frequency oscillations in executive function and suggests that theta frequency stimulation could improve executive function in PD. The aim of this study was to evaluate the acute on/off effects of bilateral gamma frequency STN stimulation on verbal fluency and executive function in PD and compare with the effects of theta frequency stimulation.

Twelve patients (all males, mean age 60.8) with bilateral STN DBS for PD underwent a double-blinded, randomized neuropsychological evaluation during stimulation at (1) 130-135Hz (gamma, baseline), (2) 10Hz (theta) and (3) off. Processing speed, verbal fluency, and executive functions were evaluated using a verbal fluency task (phonemic, episodic, nonepisodic, and switching), color-word interference task, and random number generation task. Performance at each stimulation frequency was compared within subjects. Preoperative (mean 4.7 months) scores were compared with postoperative (mean 11 months) scores.

Theta frequency stimulation significantly improved  compared to gamma frequency stimulation (p=0.02), but not compared to off (p>0.05). There were no significant differences between stimulation frequencies in phonemic or nonepisodic verbal fluency, color-word interference or random number generation (p>0.05). Comparing preoperative to postoperative scores, there were no differences in verbal fluency or color-word interference (p>.05).

Our study is the first to show improved episodic verbal fluency during theta versus gamma frequency stimulation, corroborating a role of theta oscillations in cognition. Further work is needed to explore if theta frequency stimulation can be interleaved with gamma frequency stimulation to concomitantly improve both motor and cognitive outcomes.

Jordan LAM, Justin LEE, Brian LEE, Darrin LEE (Los Angeles, USA)
17:40 - 17:50 #23990 - Does dystonia severity impact deep brain stimulation surgery and related complications?
Does dystonia severity impact deep brain stimulation surgery and related complications?

Background/ AimsDeep brain stimulation (DBS) has demonstrated pronounced benefit in treating dystonic conditions, however adverse events (AE) related to the procedure might impact the outcome. DBS-related AEs have been reported in literature; nevertheless, one unanswered question relates to whether the rate of complications is higher in patients with severe dystonia when compared to patients with less severe forms. 

The aim of this study was to first, determine which features explain the best dystonia severity at baseline and second, whether the frequency of DBS AEs was related to symptom severity.

Methods: A total of 80 patients treated with DBS for dystonia in a single center over 20 years were evaluated. We considered the occurrence of dystonic storm (DS) prior to DBS as a criterion for dystonia severity. We performed a case-control study in order to compare 40 patients who developed at least one DS episode (group 0) with a reference group who did not (group 1), matched for sex, age at first DBS and clinical phenotype. 

Related to DBS complications, we considered infection, hemorrhage, lead fracture, extension cable fracture or tension requesting surgical replacement and skin erosion. All patients were treated for dystonia with initial bilateral pallidal stimulation. 

First, we aimed to explain dystonia severity as expressed by the Burke Fahn Marsden dystonia rating scale (motor section) at baseline previous to DBS administration, based on the following variables: gender, family history, age at symptom onset, type of movement disorders, other neurologic or systemic manifestations, etiology, age at worsening with functional disability, number of SD episodes pre-DBS, disease duration pre-DBS, symptom distribution at onset, abnormal movements and postures at onset, duration of ICU stay pre-DBS, distribution of MRI abnormalities. 

Analysis was performed for n=37 subjects from group 0 and n=32 from group 1, for whom all the variables were available. Several machine learning algorithms have been used such as linear regression as well as models based on decision trees (Random Forest and Xgboost). We tested parameters and hyperparameters to obtain the combination providing the best accuracy. To validate model accuracy, mean absolute error (MAE) and root mean square error (RMSE) have been applied and cross-validation for each algorithm and for each group. 

Second, we assessed complications following DBS surgery for n=35 subjects from group 0, for whom 64 features were available and studied. Machine learning algorithms have been applied with use of Random Forest Classifier and XGBoost Classifier. 

Results: Generalized symptom distribution at disease onset was the most frequent presentation among group 0 (42.5%), vs. focal symptom distribution in group 1 (43.6%), p<0.005. Among group 0, 75% patients presented disability at disease onset (gait disturbance, hand movement or sitting abnormalities, dysarthria, dysphagia or dyspnea) vs. 20.5% of patients in group 1, p<0.005. An independent-sample t-test was conducted to compare the preoperative M-BFMDRS scores for both groups. There was a significant difference in scores for DS-patients (mean 85.15 ± 19.92) and non-DS patients (mean 57.80 ± 19.03), p<0.005.

Surgical revision was performed in 37,5% of subjects from group 0 vs. 25% from group 1 (p>0,05). Infections occurred in 20% of the subjects from group 0 vs. 11,1% from group 1. Lead fracture occurred in 20% vs. 17,1% in groups 0 and 1, respectively. No bleeding occurred in both groups. Revision of lead position was more frequent in group 1.

Based on the correlation matrix and models applied, the main features explaining the dystonia severity scores at baseline for group 0 were the following: age at worsening with functional disability and age at symptom onset, symptom distribution at onset, etiology, disease duration before DBS and MRI abnormalities of cortical distribution and for group 1, disease duration pre-DBS, symptom distribution at onset and age at worsening with functional disability.

The main features explaining hardware related complications were etiology, age and symptom distribution at dystonia onset.

Conclusion: DS prior DBS, generalized symptom distribution and disability at disease onset seem to be surrogates for dystonia severity. DBS-related AEs appear to be higher in severe dystonia, which could be due to the fact that these patients have a poorer clinical condition and hence are more prone to surgical complications.


Joana MONTEIRO (Lisboa, Portugal), Emilie CHAN SENG, Gaëtan POULEN, Philippe COUBES, Laura CIF
17:50 - 18:00 #26026 - European Multicentre Probabilistic Stimulation Map for Essential Tremor.
European Multicentre Probabilistic Stimulation Map for Essential Tremor.

Background: Deep Brain Stimulation (DBS) is an established therapy for medication-refractory Essential Tremor (ET). However, predictors of outcome and the optimal stimulation site remain a matter of controversy.

Objective: The objective of this study was to collect clinical and neuroimaging data of a large cohort of patients with DBS for ET from different European centers to identify predictors of outcome and to construct a probabilistic stimulation map and identify an optimal stimulation site.

Methods: This study enrolled 119 ET patients treated chronically with unilateral or bilateral DBS operated at five different European DBS centres for retrospective analysis of data on baseline covariates, pre- and postoperative clinical tremor scores (12-month) as well as individual imaging data to obtain individual electrode positions and stimulation volumes.  We calculated a stimulation map using voxel-wise statistical analysis. We used multiple regression analysis to estimate predictors of tremor reduction.

Results: The mean tremor reduction per patient across all centers was 59.3 % (55.7 – 62.9%, 95% CI). Preoperative tremor severity was the only baseline clinical characteristic that was significantly associated with outcome (r2 = 0.05, p= 0.03). We identified an area of optimal stimulation that extended from the posterior part of the PSA to the Vim and coincided with the area of highest likelihood to contain the DRTT. The anatomical location of stimulation was the overall best predictor of outcome.

Conclusions: Our multicentre ET probabilistic stimulation map identified an area of optimal stimulation along the course of the DRTT. This target may be used to guide surgical planning and for computer-assisted planning and programming of deep brain stimulation to in the future.

Andreas NOWACKI (Bern, Switzerland), Sabry BARLATEY, Bassam AL-FATLY, Till DEMBEK, Maarten BOT, Alexander GREEN, Alba SEGURA-AMIL, Anh Khoa NGUYEN, Lennard LACHENMAYER, Ines DEBOVE, Verle VISSER-VANDEWALLE, Andreas HORN, Richaard SCHUURMAN, Michael BARBE, Tipu AZIZ, Andrea KÜHN, Claudio POLLO
18:10 - 18:20 #24013 - A randomized controlled trial of awake versus asleep microelectrode guided frame-based deep brain stimulation for Parkinson’s disease.
A randomized controlled trial of awake versus asleep microelectrode guided frame-based deep brain stimulation for Parkinson’s disease.

Introduction, objective

It is unknown if there is a difference in outcome in asleep versus awake deep brain stimulation (DBS) of the subthalamic nucleus for advanced Parinson’s disease.

Objective: We hypothesized that there would be a difference in adverse effects concerning cognition, mood and behavior between awake and asleep DBS favoring the asleep arm of the study. We further hypothesized that the motor outcomes would be identical between the groups.



The study was a single-center prospective randomized open-label blinded end-point trial. There were 187 persons with Parkinson’s disease referred for DBS between May 2015 to March 2019 and of those 110 were eligible for participation in the study. The primary outcome follow-up visit was conducted six months following DBS.

Bilateral subthalamic nucleus DBS was performed asleep (under general anesthesia) or awake without sedation. Both arms of the study used a frame-based intraoperative microelectrode recording technique for placement of the DBS leads.

The primary outcome variable was the between group difference in cognitive, mood and behavioral adverse effects as measured by a composite score. The secondary outcomes included the Movement Disorders Society Unified Parkinson Disease Rating Scale (MDS-UPDRS), the patient assessment of surgical burden and operative time.



The participants were randomized to awake (local anesthesia) (N=56, mean age 60.0 years (SD7.4), 40 male (71%)) or to asleep (general anesthesia) (N=54, mean age 61.3 years (SD7.9), 38 male (70%)) DBS surgery. The six months follow-up visit was completed by 103 participants. The proportion of patients with adverse cognitive, mood and behavioral effects on the composite score was 15 of 52 (29%) after awake and 11 of 51 (22%) after asleep DBS (odds ratio 0.7 (95% CI 0·3-1·7)). There was no difference in improvement in the off-medication MDS-UPDRS Motor Examination scores between groups (local anesthesia -27·3±17·5 points, general anesthesia -25·3±14·3 points, mean difference -2·0 (95% CI -8·1 to 4·2)). Asleep surgery was experienced as less burdensome by the patients and was 26 minutes shorter than awake surgery. 



There was no difference in the primary outcome of asleep versus awake DBS. Future large randomized studies should examine some of the newer asleep based DBS technologies as this study was limited to frame-based microelectrode guided procedures.

Rozemarije HOLEWIJN (Amsterdam, The Netherlands), Dagmar VERBAAN, Pepijn MUNCKHOF, VAN DEN, Maarten BOT, Gert GEURTSEN, Rob BIE, DE, Rick SCHUURMAN
18:20 - 18:30 #24082 - Probabilistic tractography aids lead placement in DBS for essential tremor.
Probabilistic tractography aids lead placement in DBS for essential tremor.


Since the first implantation of a deep brain stimulation system carried out by Pollack and Benabid in essential tremor, it became the gold standard treatment in the 30 years in movement disorders. Today, accurate planning and electrophysiological mapping of the target region is still essential to achieve successful results. Since 2017, the introduction of diffusion tensor imaging, tractography and their routine use in our department, the technique provided a valuable tool to non-invasively map and pinpoint the stereotactic target in 42 Parkinson’s disease, 8 dystonia, and 19 tremor patients. This trend steers us further away from awake surgeries even in tremor DBS, where anatomical boundaries of the ventral intermediate nucleus, mainly the closeness of the sensory thalamic nuclei can result in unwanted sensations under stimulation. 


Materials and methods

16 patients were enrolled in this study. Each patient underwent a preoperative MRI session under general anesthesia, T1, T2, contrast enhanced T1, and DTI images have been acquired in two different centers (10 patients - Philips Achieva - 32 directions, b0=800; 7 patients - Siemens Magnetom Verio 64 directions, b0=1000). Preoperative DTI analysis and fiber tracking has been carried out using tools available in FSL 6.0 (FMRIB) and Freesurfer 6.0 (Martinos Center for Biomedical Imaging). Reconstruction of diffusion tensors, probabilistic fiber tracking, and visualization of the results has been carried out on a Titan XP GPU based system. Connectivity segmentation of thalamus, to visualize the sensory nuclei and the Vim, the traversing lemniscal and dentate-rubro-thalamic pathways were done preoperatively, results were integrated into a Medtronic Cranial Planning Station. Implantation of DBS leads has been carried out using a Leksell stereotactic G frame and a ROSA stereotactic robot system. Intraoperative electrophysiological mapping and macrosimulation were done to control the imaging results.



Of the 32 implanted electrodes patients 29 were within a 1 mm close range of the DRT, 65,63% was placed within the tract. The remaining 3 electrodes were within a 2 mm range. Sufficient tremor control was achieved in all of the implanted patients, no sensory side effect has been observed under implantation or postoperative programming, chronic stimulation. Optimal tremor control was found near the thalamic entry of the DRT. 



Visualization of the DRT, the sensory nuclei, and the lemniscal pathways provide a patient specific implementation of DBS surgery. Relying only on standard stereotactic coordinates might result in unwanted side effects, sensory activation during stimulation. The described method is not only sufficient to achieve tremor control but can also minimize the possibility of suboptimal lead placement, thus might reduce the requirement of intraoperative electrophysiology and awake surgery.

Loránd ERŐSS (Solymar, Hungary), László HALÁSZ
18:30 - 18:40 #24083 - Pedunculopontine Nucleus Deep Brain Stimulation for Parkinsonian disorders: A Case Series.
Pedunculopontine Nucleus Deep Brain Stimulation for Parkinsonian disorders: A Case Series.



Deep brain stimulation (DBS) of the Pedunculopontine nucleus (PPN) has been investigated for the treatment of levodopa-refractory gait dysfunction in Parkinsonian disorders, with equivocal results so far.



To summarise the clinical outcomes of PPN-DBS treated patients at our centre and elicit any patterns that may guide future research.


Materials and Methods:

Pre- and post-operative objective overall motor and gait subsection scores as well as patient-reported outcomes were recorded for six PPN-DBS treated patients; three with Parkinson’s disease (PD) and three with Progressive supranuclear palsy (PSP). Electrodes were implanted unilaterally in the first three patients and bilaterally in the latter three, using an MRI-guided MRI-verified technique. Stimulation was initiated at 20-30Hz and optimised in an iterative manner.



Unilaterally treated patients did not demonstrate significant improvements in gait questionnaires, UPDRS-III or PSPRS scores or their respective gait subsections. This contrasted with at least an initial response in bilaterally treated patients. Diurnal cycling of stimulation in a PD patient with habituation to the initial benefit reproduced substantial improvements in FOG 3 years post-operatively. Among the PSP patients, one with a Parkinsonian subtype had a sustained improvement in FOG while another with Richardson syndrome (PSP-RS) did not benefit.



PPN-DBS remains an investigational treatment for levodopa-refractory FOG. This series corroborates some previously reported findings: bilateral stimulation may be more effective than unilateral stimulation, the response in PSP patients may depend on the disease subtype, and diurnal cycling of stimulation to overcome habituation merits further investigation.


Viswas DAYAL, Ali RAJABIAN (London Queen Square, United Kingdom), Marjan JAHANSHAHI, Iciar AVILES-OLMOS, Dorothy COWIE, Amy PETERS, Brian DAY, Jonathan HYAM, Harith AKRAM, Patricia LIMOUSIN, Marwan HARIZ, Ludvic ZRINZO, Thomas FOLTYNIE
18:40 - 18:50 #25588 - Stereotactic Targeting in Europe: The European Society of Stereotactic and Functional Neurosurgery Survey.
Stereotactic Targeting in Europe: The European Society of Stereotactic and Functional Neurosurgery Survey.

Deep brain stimulation (DBS) of the subthalamus nucleus (STN) is a highly effective surgical technique for providing relief to patients suffering from Parkinson’s disease (PD). The ventral intermediate nucleus of the thalamus (VIM) target is also very effective for treating essential tremor (ET) with DBS or lesion. Nevertheless, targeting procedures remain controversial and variable. Two main techniques are used to identify the target: (i) the indirect technique, which transforms or wraps atlas or statistical information in a patient-specific space; (ii) the direct technique, which creates direct visualisation of the target on pre-operative magnetic resonance imaging (MR) with the multiplication of complex MR sequences.

With the help of the ESSFN, we wanted to interview European centres for functional and stereotactic neurosurgery to assess the reality of heterogeneous targeting and the difficulties encountered by surgeons.


Two surveys on STN and VIM targeting were sent to ESSFN members and the French centres of functional and stereotactic neurosurgery.

The first part of the survey focused on targeting, and the second part on difficulties encountered by centres during targeting.

Results: Forty-seven centres responded to the survey (about 40% of the 120 European centres of stereotactic and functional neurosurgery) with 1,548 targeting procedures having been conducted (STN 1000, VIM 548) in 18 countries in 2019.

All STN targeting was performed for DBS. 39% of centres targeted STN for diseases other than PD but these represent only 0.7% of STN targeting.

VIM targeting was performed by centres which practice DBS only (46%), DBS and lesion (31%) or lesion only (23%). VIM targeting was performed for tremors other than ET by 59% of centres, accounting for 18% of VIM targeting procedures.


Centres perform a mean of 26 STN targeting procedures per year with a median of 15, and a mean of 13 VIM targeting procedures per year with a median of 8. A significant number of centres carry out less than 10 VIM or STN targeting procedures per year.

Direct STN targeting was performed by 15 centres for 32% of STN targeting procedures without using statistical coordinates. Only 6 centres conducting 17% of targeting procedures did not use per-operative evaluations. Direct VIM targeting was used by 3 centres for 4% of VIM targeting procedures without using statistical coordinates and 10 centres (35% of targeting) did not use per-operative evaluations.

The mean targeting duration was 1 h 40 for STN (median of 2 h) and 1 h 25 for VIM (median of 1 h 00). Targeting is always performed by a neurosurgeon who is sometimes assisted by a neurologist or radiologist.

The two main problems noted by neurosurgeons are the complexity and heterogeneity of practices, followed by the duration of targeting and intra-operative evaluation, and the invasive nature of the latter.




More VIM targeting procedures are carried out than expected. Per-operative evaluation remains the reference method. Some lesion centres have significant VIM activities. STN targeting for indications other than PD is anecdotal, and the complexity and lack of homogeneity of practices are the main problems encountered by neurosurgeons.

Emmanuel CUNY (Bordeaux), Jean REGIS
18:50 - 19:00 #23993 - Paediatric Deep Brain Stimulation: the challenge of secondary dystonia.
Paediatric Deep Brain Stimulation: the challenge of secondary dystonia.

Unlike primary dystonias where the results of deep brain stimulation (DBS) can be responsible for an improvement of 70 to 90%, the benefit in secondary dystonias is less significant, on average 30%.
We present our cohort of implanted secondary dystonias, comparing our results to the literature data in terms of efficacy, complications and prognostic factors.

Patients & Methods
Among all the dystonic children implanted in our centre since 2015 (n=42), 18 patients (8-16 years old) with secondary dystonia (8 cerebral palsy, 4 PKAN, 1 glutaric aciduria, 5 without etiology), were implanted in the internal pallidum (GPi) bilateral implantation; One patient received a multi-targeted implantation (4 electrodes > GPI+STN).
All implantations were performed under general anaesthesia with electrophysiological control. Validation of the position of the electrodes was carried out intraoperatively by scanner and then again by scanner at 10 days postoperatively. The target chosen was the ventral posterolateral part of the GPi corrected according to the anatomical variability associated with the pallidal lesions induced by the pathology.
All indications were validated in RCP or Multidisciplinary Consultation for paediatric movement disorders. The expected benefits of DBS ranged from a reduction in painful attacks to functional improvement, depending on the initial pathology.

The results are very heterogeneous, ranging from a spectacular improvement for a progressive, unlabelled disease to a lack of efficacy for 2 children with Cerebral Palsy. There were no additional postoperative complications compared to primary dystonias, including infectious dystonias (2/2 cases).

We currently have few prognostic means to evaluate good candidates. Precise electrode placement in the GPi is crucial despite the possible lesions of the GP. Other targets than GPi or multi targets schemes can be a alernative surgical option. Early implantation would ensure better results. 
Patient selection and definition of expectations regarding the benefit to be expected from DBS are particularly important in this population.
DBS may be useful in secondary dystonias, particularly in dyskinetic, painful, and malignant forms. It should be discussed on a case-by-case basis with clear goals to be achieved, as functional gain may not always be the primary goal of DBS in this population.

Vincent D'HARDEMARE (Paris), Nathalie DORISON, Diane DOUMMAR, Marie HULLY, Florence RENALDO, Domitille GRAS, Emmanuel ROZE, Cecilia ALTUZZARA, Aude CHAROLLAIS, Véronique DARMENCY, Alice GIGNOUX, Georg DORFMULLER, Bernard PIDOUX, Julie BONHEUR, Laurent GOETZ
Salle Major

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
15:30 - 17:00

Parallel Session 3
Radiosurgery - Other

Moderators: Roman LISCAK (head) (PRAGUE, Czech Republic), Roberto MARTINEZ-ALVAREZ (Neurosurgeon) (Madrid, Spain), Piero PICOZZI (Consultant) (Milano, Italy)
15:30 - 15:40 #23675 - Long-term Cognitive Outcome after Radiosurgery in Epileptic Hypothalamic Hamartomas.
Long-term Cognitive Outcome after Radiosurgery in Epileptic Hypothalamic Hamartomas.

Introduction: Epileptic patients with Hypothalamic Hamartoma (HH) are frequently presenting with cognitive impairment. Surgical techniques aiming at the HH can be very efficient for epilepsy relief and cognitive improvement but are also demonstrated to carry a significant risk of additional reduction in memory function in these already disabled patients. Gamma Knife Radiosurgery (GKS) offered efficient non-invasive procedure. We evaluated the effect of stereotactic radiosurgery on cognitive outcome.

Methods: Thirty-nine epileptic patients (median age 17, range 3-50) with HH underwent preoperative and postoperative comprehensive neuropsychological standard testing. All patients were prospectively evaluated and underwent complete pre-surgical and post-surgical clinical, electrophysiological, endocrinal and visual assessment. In all patients the post-operative assessment was performed at least 3 years after radiosurgery. We explored what variables correlate with cognitive outcome. Literature review was done for other surgical techniques and their risks for cognitive complications after surgery. The median dose received by Mammillarybodies (MB) and Fornix (Fx) was respectively 16,5Gy and 8 Gy.

Results: No memory or cognitive decline was observed after GKS. We observed significant improvement (>1SD inz-score) especially in working memory index (46%) and processing speed index (35%)but also inintelligence quotient full scale (24%), verbal comprehension index (11%), perceptual organization index (21%), verbal leaning (20%), visual learning (33%). Before GKS, the probability of seizure cessation was higher in patients with higher cognitive performances. After GKS, the cognitive improvement was significantly higher in the seizure free patients compared to the non-seizure-free.

Conclusion: We found clear cognitive improvement in a high percentage of patients but importantly no significant cognitive worsening in long-term comprehensive neuropsychological assessment and specially memory three years after GKS. GKS is comparing very favorably to the other surgical techniques in term of cognitive outcome with similar seizure freedom.

Hussein HAMDI (Marseille), Faisal ALBADER, Virginie LAGUITTON, Agnes TREBUCHON, Fabrice BARTOLOMEI, Jean REGIS
15:40 - 15:50 #23934 - Gamma Knife Radiosurgery in the treatment of Glomus jugulare tumours - the Vienna series.
Gamma Knife Radiosurgery in the treatment of Glomus jugulare tumours - the Vienna series.


A Glomus jugulare tumour (GJT) is considered a slowly growing, benign lesion located in the skull base. The tumour is frequently highly vascular and surgical removal is rarely radical. Consequently, radiosurgery became a relevant role in the treatment of these tumours.



A retrospective analysis identified 42 patients with GJT treated with Gamma Knife Radiosurgery (GKRS) at the Department of Neurosurgery, Medical University Vienna. Nineteen out of 42 patients underwent surgery before GKRS. Twenty-three patients had GKRS as primary treatment.



Five patients were lost to follow-up (5/42, 12%). The median total follow-up time was 49months (range: 11-212months). The median dose to the tumour margin was 13Gy (range 9-16Gy). The median tumour size was 7.4ccm (1.2-74.0ccm).

In MRI controls 13 tumours decreased (35%), 22 remained stable (60%) and two (5%) showed a progression and were managed conservatively. Treatment failures received a marginal tumour dose of 13 and 15Gy, respectively.



GKRS is an effective treatment option for GJTs even after prior surgical resection and provides a tumour control of nearly 95%.

Brigitte GATTERBAUER (Vienna, Austria), Josa M. FRISCHER
15:50 - 16:00 #23969 - Safety of Gamma Knife ICON Hypo-fractionated treatment for benign peri-optic lesions: preliminary results from a single centre.
Safety of Gamma Knife ICON Hypo-fractionated treatment for benign peri-optic lesions: preliminary results from a single centre.


Traditionally Gamma Knife Radiosurgery (GKRS) has been identified as the gold standard therapy for single-fraction high-dose irradiation of intracranial lesions. The Leksell Gamma Knife Icon (GK Icon) system can utilize cone-beam computed tomography (CBCT) to evaluate motion error and it is currently used in some centers for hypo fractionated GKRS treatment (fGKRS).

In this study, we analyzed the safety of GK Icon hypo-fractionated radiosurgery for the treatment of benign lesions close to the optic apparatus.


The study was conducted with the approval of our institutional review board. We retrospectively analyzed a prospectively maintained database of patients treated with hypo-fractionated GKRS between September 2017 and December 2019 using the GK Icon system. We included patients harboring benign perioptic lesion, including skull base meningiomas, pituitary adenomas and craniopharyngiomas. Patients had at least 12 months of clinical ophthalmological and radiological follow up.


We collected a total of 41 patients (29 female and 12 male) with a mean age of 59,8 years (38-85). The most frequent lesion was meningioma, except for nine patients (two craniopharingiomas and seven pituitary adenomas). Tumors were most commonly located in the cavernous sinus, 3 lesions originated from the optic nerve sheath. The average lesion volume at fGKRS was 5,136 cm3 (range 0.608-16.441 median 3.240 cm3). 19 patients manifested neurological symptoms at the time of fGKRS, including diplopia, trigeminal paresthesia, visual disturbance or exopthalmus. Sixteen patients had previous surgery (four adenomas, one craniopharyngioma, two patients with WHO grade II meningioma and nine with WHO grade I meningioma). Three patients had prior radiotherapy or radiosurgery. All patients were treated using the Icon mask system; most patients were treated in 5 fractions, while 2 patients were treated in 3 fractions. The majority (92%) of patients received 25 Gy at a median isodose of 50% with a median conformality index of 1.0. When comparing the CBCT and preoperative CT images, the inter-fractional movement was less than 0.5 mm. Daily difference in intra-fractional movement was 0.1 mm with a maximum error of 0.5 mm in HDMM system. The median follow-up period was 14.3 months (range 6-36).

Among the symptomatic patients, three experienced symptom improvement after fGKRS, and eleven remain stable. 2 patients, presenting large lesion volume (16.473 and 10.632 cm3) complained of subjective diplopia. None of asymptomatic patients became symptomatic. All visual fields and visual acuity of asymptomatic patients remained normal, none of the treated patients experienced a worsening in visual field or visual acuity.


Despite the short follow-up doesn’t allow us to assert the results on tumor control, this study reports encouraging preliminary results on the safety of hypo-fractionation with Gamma Knife Icon in peri-optic lesions.

Giorgio SPATOLA (Brescia, Italy), Lodoviga GIUDICE, Karol MIGLIORATI, Bassetti CHIARA, Cesare GIORGI, Oscar VIVALDI, Mario BIGNARDI, Alberto Bernardo FRANZIN
16:00 - 16:10 #24048 - Comparison of dorsal root entry zone and trigeminal nerve fibers adjacent to trigeminal ganglion in targeting of gamma knife radiosurgery for trigeminal neuralgia.
Comparison of dorsal root entry zone and trigeminal nerve fibers adjacent to trigeminal ganglion in targeting of gamma knife radiosurgery for trigeminal neuralgia.

Background: The dorsal root entry zone (DREZ) and trigeminal nerve fibers that are adjacent to the trigeminal ganglion (TFatTG) as a target in Gamma Knife radiosurgery (GKRS) for the treatment of medically refractory TN were compared and evaluated. Special interest was given to the complication rates and effectiveness of these targets. The aim was to determine whether (TFatTG) targeting could be an alternative method to DREZ targeting in medically refractory TN patients with pain in V2 and/or V3 distribution. These selective targeting technique could minimize possible side effects like numbness, especially in V1 distribution. 


Methods: Twenty-six medically refractory TN patients were enrolled in this study. The patients were divided into two groups. The DREZ and the TFatTG were selected as a target for GKRS. TN patients with pain in V2 and/or V3 distribution especially with good visualization of the Meckel cavum, ganglia and trigeminal fibers in MRI studies were selected for targeting of this area. Selective targeting of V2 and/or V3 nerve fibers was performed with the help of multiplanar reformatting of high resolution CISS MRI studies. DREZ targeting technique was performed as reported in the relevant literature. The irradiation time, pain control, and treatment related complications like numbness were evaluated. 


Results: Chi-square test of homogeneity was used for the evaluation of statistical differences of BNI numbness scores after treatment between groups. Three patients (23.1%) in the TFatTG targeting group and  8 patients in the DREZ targeting group complaints from numbness. Numbness after GKRS was statistically higher in the DREZ targeting group in comparison to TFatTG targeting group (p<0.05). Irradiation time for each targeting group were normally distributed, as assessed by Shapiro-Wilk's test (p > .05). An independent-sample t-test was run to determine if there were differences in irradiation time between DREZ and TFatTG. Irradiation time was significantly longer in DREZ targeting (86.9±19.2 min) in comparison to TFatTG targeting (65.0±2.5 min) (P<0.01).


Conclusion: TFatTG is a reasonable target in patients with a distribution of pain to V2 and/or V3 TN, with less collateral effects over brain stem and less side effects like numbness especially in V1 distribution. Hence, shorter irradiation time and equal pain control compared with the DREZ target, the TFatTG may be an effective and functional target in selected patients with pain in V2 and/or V3 distribution. Randomized prospective study evaluating DREZ, TFatTG and cistern targeting in patients with medically refractory TN is in progress. 


Mesut Emre YAMAN (Ankara, Turkey), Burak KARAASLAN, Munibe Busra ERDEM, Gökberk EROL, Beste GÜLSUNA, Şükrü AYKOL
16:10 - 16:20 #26276 - Long-term hearing outcome after radiosurgery for vestibular schwannomas: a systematic review and meta-analysis.
Long-term hearing outcome after radiosurgery for vestibular schwannomas: a systematic review and meta-analysis.

INTRODUCTION: Stereotactic radiosurgery (SRS) has emerged over the last thirty years as the main treatment option in the management of small to medium size vestibular schwannomas (VS), due to high tumor control rate and low cranial nerves morbidity. In most series, hearing preservation after SRS is generally reported over 60% at 3-year follow-up. However, series reporting long-term hearing outcome are scarce.

OBJECTIVE: We performed a systematic review of the literature and meta-analysis, with the aim of focusing on the long-term hearing preservation after SRS and related prognostic factors.

METHODS: Using PRISMA guidelines, we reviewed manuscripts published between January 1990 and October 2020 and referenced in PubMed®. Inclusion criteria required that each article be a peer-reviewed clinical study or a case series of VS treated with SRS (single dose), irrespective of the technique, reporting hearing outcome after SRS with a median or mean audiometric follow-up of at least 5 years. Hearing preservation, cranial nerves outcomes, and tumor control were evaluated with separate meta-analyses.

RESULTS: Twenty-three studies were included. Hearing preservation at last follow-up was reported in 58.9% of cases, with a median follow-up of 6.7 years (range: 0.2-23 years; 1,409 patients). The main favorable prognosis factors were young age, good hearing status at SRS (Gardner-Robertson IA), early treatment after diagnosis, small tumor volume, low marginal irradiation dose, and low maximal dose to the cochlea. Tumor control was achieved in 96.1% of patients. Persistent facial palsy and trigeminal neuropathy were reported in 1.1% and 2.6% of patients, respectively. The rates of facial palsy and trigeminal neuropathy were significantly higher in Linear Accelearor (LINAC) series (p<0.05), respectively 0.8% vs 6.4% and 1.7% and 7.1%.

CONCLUSIONS: Long-term hearing preservation remains one of the main issues after SRS, with major impact on health-related quality of life. Our meta-analysis suggests that hearing preservation can be achieved in almost 60% of patients after a median follow-up of 6.7 years, irrespective of the technique. However, Linac series present higher risks of trigeminal and facial nerve neuropathy.

Anne BALOSSIER (Marseille), Constantin TULEASCA, Christine DELSANTI, Lucas TROUDE, Jean-Marc THOMASSIN, Pierre-Hugues ROCHE, Jean RÉGIS
16:30 - 16:40 #23795 - Opening of blood brain barrier (BBB) in the hippocampus and entorhinal cortex in early Alzheimer’s disease with focused ultrasound (FUS).
Opening of blood brain barrier (BBB) in the hippocampus and entorhinal cortex in early Alzheimer’s disease with focused ultrasound (FUS).


There is an urgent need for innovative research addressing Alzheimer’s disease (AD) treatment as no effective therapy exists. Animal models have demonstrated that magnetic resonance (MR)-guided low intensity focused ultrasound (FUS) can reversibly open the blood-brain barrier (BBB), resulting in a reduction of amyloid-beta plaque, improvement of memory, and allowing for targeted drug and stem-cell delivery. We report initial clinical and PET analysis of phase II clinical trial targeting the hippocampus and entorhinal cortex (EC) with FUS in patients with AD.



Patients with early AD and positive amyloid-beta PET underwent MRI-guided FUS sonication of the hippocampus and EC (220 kHz, ExAblate Neuro Type 2 system) with concomitant IV microbubble (Definity®) injection. Treatment areas were selected based upon individual anatomy and amyloid burden. FUS treatment occurred on three separate sessions, each two weeks apart, at the same target sites. Outcome assessments included MRI evaluating BBB opening and closure, safety, changes in amyloid PET scans, and cognitive/behavioral evaluations.



Six subjects (5 female and 1 male: ages 55-72 years) were enrolled at two trial participating institutions. All six Subjects completed three separate treatment sessions, each two weeks apart for a total of 18 FUS treatments. FUS sonication of up to five targets of the right (n=2) or left (n=4) hippocampus/EC was performed. The follow-up ranged from 2 to 18 months. All 18 treatment sessions were tolerated well and there were no overt hemorrhages.  Post-FUS contrast MRI in all six subjects revealed immediate and sizable hippocampal parenchymal enhancement indicating BBB opening, followed by BBB closure within 24 hours (Figure). The average opening was 95% of the FUS target volume, corresponding to 29% of the overall hippocampus volume. To date, no neurological adverse events have been noted in any of the 6 subjects. Cognitive and behavioral testing demonstrated no meaningful changes at the 3 months (n=5) follow up since the last treatment. Follow up post-FUS treatment PET analysis (approximately 60-days time interval from the baseline pretreatment PET scan; n=5) showed a decrease in amyloid-beta plaque in all participants in the treated hippocampus (range: 2.7% – 6.2%).



This study is the first to demonstrate precise, safe, non-invasive, reversible, and substantial opening of the BBB with FUS in the human hippocampus/EC, with the suggestion of reduction of amyloid-beta plaque, providing a unique translational opportunity to investigate novel therapeutics in AD.

Ali REZAI (Morgantown, USA), Manish RANJAN, Pierre-Francois D’HAESE, Mark HAUT, Jeffrey CARPENTER, Umer NAJIB, Rashi MEHTA, Alexander SONG, Daniel CLAASSEN, J. Levi CHAZEN, Michael LIN, Zion ZIBLY, Gary MARANO, Mor DAYAN, Nathaniel KELM, Sally HODDER, Michael KAPLITT
16:40 - 16:50 #26280 - Deep Brain Stimulation in Disorders of Consciousness: A 10-year institutional experience.
Deep Brain Stimulation in Disorders of Consciousness: A 10-year institutional experience.

Aims: Consciousness disorders, namely vegetative state (VS) and minimally conscious state (MCS), represent serious conditions with mayor consequences for patients and their families. Several studies have described regaining of consciousness in such patients with the use of deep brain stimulation (DBS) of brain or brainstem nuclei. The aim of our study is to present  10 years’ experience results in using DBS as a therapy for VS/ MCS patients.

Methods: Ovrall, our study included 63 patients; entry criteria included neurophysiological and neurological evaluation, as well as neuroimaging examination. DBS system was implanted in 26 patients; 20 patients were in VS and 6 in MCS. The implantation and stimulation target was centromedian-parafascicular complex in the left hemisphere in HI-BI or the one better preserved in TBI patients. The span of the follow up period was 10 to 111 months. 

Results: Level of consciousness was improved in six patients. Two MCS patients substantially improved, regaining the ability to walk, to speak fluently and without need for everyday life assistance. Four patients showed improvements in consciousness in a way of regaining the possibility to communicate non-verbally, but are still dependent on the care of their guardians.

Conclusion: In patients with consciousness disorders, spontaneous recovery to the level of consciousness without assistance is rare. Thus, for such patients who meet neurological, neurophysiological and neuroimaging criteria, DBS of certain thalamic nuclei could be recommended as a treatment option, especially in earlier phases when irreversible changes of musculoskeletal system are still not so drastic.


16:50 - 16:55 #23854 - Awakening the brain with DBS of the intralaminar thalamus after severe brain injury.
Awakening the brain with DBS of the intralaminar thalamus after severe brain injury.

Rationale: Severe brain injury can result in prolonged disorders of consciousness and other hypo-responsive behavioral syndromes associated with severely diminished motivation, such as akinetic mutism. Hypoactivation of intralaminar thalamic outflow tracts to the (pre)frontal cortex has been suggested to be the main cause of such behavioral deficits. DBS of the intralaminar thalamus (centromedian-parafascicular complex, CM-Pf) may increase arousal and restore purposeful behavior in these severely injured patients. 

Objective: To explore the efficacy of CM-Pf DBS in restoring consciousness and purposeful behavior in patients with disorders of consciousness and diminished motivation.

Methods: Six patients with a minimally conscious state (MCS) without improvement >24 months after traumatic brain injury and 1-2 patients with severe akinetic mutism will undergo DBS targeted to the CM-Pf complex of the thalamus. The primary outcome measure for post-TBI MCS patients is the change in the Coma Recovery Scale-Revised (CRS-R) score. Secondary endpoints are changes on a subset of additional behavioral scales for patients recovering from coma (such as the PALOC-S). The primary outcome for akinetic mutism patients is the change in the Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment (MoCA) tool. Also, neurophysiological changes will be measured using both EEG and MEG.

Results/conclusion: Preliminary results show that CM-Pf DBS improves arousal in one MCS patient and restores the ability to walk and talk in a previously wheelchair-bounded patient with akinetic mutism. These results will be shown in pre-DBS and post-DBS videos. 

Hisse ARNTS (Amsterdam, The Netherlands), Willemijn VAN ERP, Berno OVERBEEK, Rick SCHUURMAN, Pepijn VAN DEN MUNCKHOF
16:55 - 17:00 #26301 - Entraining human wakefulness: selective, closed-loop stimulation of brainstem arousal circuits.
Entraining human wakefulness: selective, closed-loop stimulation of brainstem arousal circuits.

Background: Deep sleep features slow wave activity (SWA) in cortical areas crucial for decision-making and vigilance. Disorders of Consciousness also share this electrophysiological marker –while remaining challenging to treat. In addition, hypersomnia is a feature of advanced neurodegeneration as well as part of sleep disorders (some, like narcolepsy, also potentially treatment-resistant). Therefore paradigms of targeted SWA modulation, as well as enhancement of cortical rhythms positively associated with attention and cognition (such as cortical gamma oscillations), may help open new treatment avenues of these conditions. The pedunculopontine nucleus of the brainstem (PPN) delivers activating cholinergic afferents, closely linked to wakefulness. Its role in locomotion and autonomic control has made it a target for movement disorders therefore accessible for surgical neuromodulation.

Methods: Four patients underwent bilateral electrode implantation in the PPN (Fig.1C). We allowed for a period of recovery, then applied two closed-loop stimulation protocols over multiple nights. We compared their ability to reduce SWA, entrain cortical gamma activity and increase arousal. We also compared their efficacy against a control -sham stimulation trials, delivered at the same level of pre-stimulation SWA. Sham trials (no stimulation) were 'delivered' on separate nights, not mixed with real trials, to avoid any effect of real stimulation washout.

Both protocols reliably entrained fast cortical rhythms when compared to sham trials. Overall, stimulation increased gamma power and reduced delta activity (p=0.0000, one-way ANOVA with Bonferroni correction) as measured on the EEG (Fig.1 A,B). When examined separately, both protocols led to statistically significant gamma entrainment compared to sham (P=0.0000 for both protocols, ANOVA with Bonferroni correction). However, one of the two protocols was significantly more efficient at SWA reduction (P= 0.0006, CI: [0.8043 4,8094]).


In a human study, we demonstrated selective modulation of arousal state through deep brain stimulation of brainstem arousal pathways, against a control (sham) paradigm. We show that different stimulation parameters can offer different levels of SWA control –therefore offering us the option of staged, even more tailored interventions featuring gamma entrainment. These findings highlight the importance of investigating brainstem DBS in treatment-refractory hypersomnia and disorders of consciousness.

Alceste DELI, Alceste DELI (Oxford, United Kingdom), Shenghong HE, Yongzhi HUANG, Sean MARTIN, Tipu AZIZ, Timothy DENISON, Alexander GREEN
Espace Vieux-Port

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
15:30 - 17:00

Parallel Session 4

Moderators: Claire HAEGELEN (Neurosurgeon) (Lyon, France), Dirk VAN ROOST (Consultant) (Ghent, Belgium)
15:30 - 15:40 #23939 - Network substrates of centromedian DBS in generalized pharmacoresistant epilepsy.
Network substrates of centromedian DBS in generalized pharmacoresistant epilepsy.

Objective: Deep brain stimulation (DBS) of the centromedian (CM) nucleus of the thalamus in

patients with pharmacoresistant epilepsy has yielded first promising results. However, the

structural network substrates of its therapeutic effect remain to be identified. A proper

characterization of the targeted networks could improve DBS efficiency and reduced stimulation

related side-effects. We aim at evaluating the targeted network with the description of structural

connectivity patterns derived from volumes of tissue activated (VTA) and its relationship with

CM-DBS outcomes.

Methods: We retrospectively analyzed magnetic resonance imaging (MRI) and diffusion tensor

imaging (DTI) data from patients with generalized pharmacoresistant epilepsy and CM-DBS (N

= 10, mean age at surgery = 30.8±5.9 years, 4 female). We modelled VTAs according to the

individual stimulation parameters and included these as seeds to reconstruct the targeted network,

as derived from deterministic tractography, for the clinical outcome of CM-DBS.

Results: Out of the 10 patients, nine significantly improved (> 50%) with CM-DBS at 6 months

after implantation and later on. The structural connectivity analysis revealed high connectivity

between VTAs and sensorimotor, premotor, brainstem and cerebellar regions, as well as fiber

projections to frontal and temporal cortices.

Interpretation: The outcome of CM-DBS in pharmacoresistant epilepsy is highly dependent

from individual connectivity profile of the volume-of tissue activation at the implantation site and

targeted network encompassing frontal and central regions, brainstem and cerebellum. The

proposed framework could be implemented in future studies to refine stereotactic implantation or

the parameters of the targeted neuromodulation.

Cristina TORRES (Madrid, Spain), Gabriel GONZALEZ-ESCAMILLA, Dumitru CIOLAC, Marta NAVAS, Paloma PULIDO, Jesus PASTOR, Vega-Zelaya LORENA, Rafael G. SOLA, Sergiu GROPPA
15:40 - 15:50 #25980 - Anterior Nucleus of the Thalamus DBS versus best medical therapy including VNS for pharmaco-resistant epilepsy: results of the FRANCE STUDY, a randomized, open label trial.
Anterior Nucleus of the Thalamus DBS versus best medical therapy including VNS for pharmaco-resistant epilepsy: results of the FRANCE STUDY, a randomized, open label trial.

Deep Brain Stimulation(DBS) is an alternative treatment to treat patients with severe pharmaco- resistant epilepsy. However, there is no randomized control study assessing the effect of ANT-DBS in patients who have failed medical treatment and vagus nerve stimulation.  We aimed to compare the efficacy of ANT-DBS in pharmacoresistant epileptic patients who have failed VNS, to the efficacy of best medical therapy.


Mathods. We conducted a randomized controlled trial in 12 french expert centers for epilepsy.  We recruited patients who failed medical and VNS treatment and who experienced at least 4 severe seizures (according to the Chalfont Scale) per months for 3 months during the screening period. 

Enrolled patients were randomised in either a neurostimulation group (DBS group) or a best medical therapy group ( BMT group). VNS therapy was maintained ON in patients in whom it was not stopped at the inclusion.  Patients assigned into the DBS group were stimulated bilaterally using different type of stimulation during 12 months.  

The primary endpoint was the difference in the mean number of severe seizures that occurred during 3 months before the randomization and that occurred during the last 3 months at 1 year follow up, as assessed by a Khi-deux test. We also assessed the side effects occuring in all patients. The trial is registered  with NCT02076698.


Results: We enrolled 67 patients and 61 were randomized, 30 in the neurostimulation group (DBS group) and 31 in the best medical therapy group ( BMT group). For the DBS group, 29 received bilateral implantation of the ANT using the 3389 lead from Medtronic. 59 patients completed the final assessment at 12 months.

In the DBS group, 37,93% of patients achieved 50 % of reduction of all seizures  versus 16.67 % in the BMT group ( p= 0.066), and  51.72 % of patients achieved 40 % of severe seizure reduction, compared to 30 % in the BMT group (p= 0.08). In the DBS group, the mean number of severe seizures per month was reduced by 38 %, and was increased by 4 % in the BMT group.  The Beck depression inventory scored was improved in 88.46% of the patients in the DBS group, versus 55.56 % in the BMT group ( p=0.014). 2 patients died due to SUDEP, one in each group, before having received DBS. 

Conclusion: the result of this randomized study shows that ANT -DBS tend to have an effect on the number of severe seizures and on the total number of seizures  in a population of patients suffering from severe epilepsy and who failed medications and VNS, compared to a control group who received the best medical treatment including VNS. 

Stephan CHABARDÈS (GRENOBLE), Haegelen CLAIRE, Fabrice BARTOLOMEI, Sylvain RHEIMS, Emmanuel CUNY, Sophie COLNAT COLBOIS, Louis MAILLARD, Philippe KAHANE, Stephane CLEMENCEAU, Bertrand DEVAUX, Marc Guenot GUENOT, Guillaume PENCHET, Denys FONTAINE, Anca ANCA, Lorella MINOTTI, Study Group FRANCE, Sandra DAVID-TCHOUDA, Jean REGIS
15:50 - 16:00 #26112 - Surgery for multifocal epilepsy.
Surgery for multifocal epilepsy.


        Introduction. Multifocal epilepsy predominantly occurs in childhood and characterises by high incidence of intractable, severe and traumatic seizures. High rates of developmental delay, mental retardation and psyhoemotional disturbances are noted in children with multifocal epilepsy. Patients with such epilepsy are often considered unsuitable for epilepsy surgery. However some surgical procedures, such as multilobar resections, functional hemispherotomy, disconnections, neurostimulation or ablative interventions can stop seizures or significantly reduce their frequency and improve patient’s quality of life.  The purpose of the report is to demonstrate our experience of surgical treatment of multifocal epilepsy.

Material and methods. 48 patients with multifocal epilepsy were enrolled in study, among them there were 37 (77%) children and 11 (23%) adults. Patient’s age ranged from 2 to 44 years (mean – 13 years). Most patients had severe epilepsy, refractory to medications. 14 (29%) patients had repeated epileptic statuses and 18 (38%) had epilepsia partialis continua. Epileptic encephalopathy noted in 15 (31%) cases. Mean duration of epilepsy before surgery was 8.9 years.

Patients underwent the following surgical interventions: microsurgical callosotomy - 15 (31%); stereotactic anterior callosotomy - 12 (25%); multilobar resections – 6 (13%); callosotomy in combination with resective interventions - 3 (6%); peri-insular functional hemispherectomy - 12 (25%). Postoperative long-term follow-up ranged from 6 months to  11 years (mean – 5.8 years).

Results. In 25 (52%) cases multiple epileptic focuses or diffuse structure and electrophysiologic changes were found within one hemisphere and in 19 (40%) patients structure abnormalities were revealed in both hemispheres, 4 (8%) patients had MRI -negative epilepsy with bilateral epileptiform discharges. Causes of epilepsy in our group were:  malformation of cortical development - 12 (25%), Rasmussen encephalitis - 8 (16%), intracerebral hemorrhage - 7 (15%), hypoxic-ischemic encephalopathy - 5 (10%), meningoencephalitis - 5 (10%), neoplastic lesions - 3 (6%), Sturge-Weber syndrome - 2 (4%) and microencephaly - 2 (4%). Cause of epilepsy was unknown in 4 (8%) cases. 

After treatment 23 (48%) patients became seizure-free (Engel 1), 8 (17%) patients had rare short auras (Engel 2), in 11 (23%) cases seizure frequency reduced over 75%, in 6 (13%) cases seizure frequency reduced less then 75% or did not change significantly. Best seizure  control achieved in patients who underwent functional hemispherectomy and multilobar resections - 90%, while after stereotactic callosotomy such result achieved only in 25%. However after both types of callosotomy drop-attacks stoped in 21 from 26 (81%) patients who had it before treatment.

After hemispherectomy one child died because of brain hypoxia. Post-hemispherectomy hydrocephalus occurred in one case and was needed repeated shunt interventions. Overall postoperative mortality was 2%, morbidity was 2%.  

Discussion. Our results may to conclude that multilobar resections, combined resective surgery with callosotomy and functional hemispherectomy can be the valuable treatment approach for multifocal epilepsy because of their effectiveness and safeness. Functional disconnecting interventions, such as callosotomy aimed to blocking epileptiform discharges and are reserved for patients, suffering from severe epilepsy who are not good candidates for resective surgery. Sufficient seizure control, normalisation of brain electrical activity, prevention of secondary epileptogenesis and reduction of adverse effects of antiepileptic drugs allow to stop progressive cognitive decline that can be seen in paediatric epilepsy patients.

Kostiantyn KOSTIUK (KYIV, Ukraine), Varelii CHEBURAKHIN, Maxim SHEVELOV, Yuri MEDVEDEV, Andriy POPOV, Sergii DICHKO, Oleksiy KANAYKIN, Vladyslav BUNYAKIN, David TEVZADZE
16:00 - 16:10 #26153 - Towards refined targeting the anterior nucleus of thalamus in DBS for epilepsy.
Towards refined targeting the anterior nucleus of thalamus in DBS for epilepsy.


In Deep Brain Stimulation (DBS) for epilepsy addressing the Anterior Nucleus of the Thalamus (ANT) region, the analysis of pre- and post-operative MRI and CT images shows that the active stimulation electrode position is not the sole discriminator for therapy outcome. Electrodes which have been placed accurately (as evaluated by postop imaging) in the ANT nucleus but also the ones which clearly are outside the intended target region (but still in the ANT region) show both good and bad therapy efficacy outcomes.  This may be due to different brain networks passing through the ANT region, which are important in different types of epilepsies.  This research is tailored to finding anatomical hotspots that correlate to good therapy outcome for different types of focal onset epilepsy.


A total of 179 European MORE study epilepsy patients with bilateral ANT deep brain stimulation were analyzed. For 109 patients active stimulation electrode contact location could be linked to stimulation parameters and therapy outcome, at 2 year follow up.  Patients were grouped as Responders (≥ 50% Seizure Reduction (SR)); Improvers, 50%<SR<0% and No Benefit Patients; SR increase. The ANT connectivity description combined a diffusion weighted imaging data from Human Connectome Project (n = 109) with histological data in a synthetic pathway reconstruction process.


Our non-rigid registration procedures, aims at accurate transfer of patient images into MNI space. We present extensive analytic and probabilistic mapping data of a large cohort of MORE study patients These data were analyzed to define the optimal stimulation volume and to define parameters yielding optimal results.

The analysis of the outcomes included:

1. ANT optimized automatic registration and reconstruction of the electrodes in the MNI space 

2. automatic neuroanatomical characterization of each lead contact and volumes of tissue activated (VTA) computation for the clinically used stimulation electrode contacts

3. probabilistic mapping of stimulation induced effects were constructed by aggregating individual electrode locations and their VTAs.

4. anatomical fingerprint mapping of stimulation outcome defined VTA groups

5. evaluation of the stimulation focus across the MORE patients with a construction of an ANT-DBS stimulation atlas (DSA)

6. multinomial logistic regression analysis to identify and optimize a set of covariates consisting of neuroanatomical, seizure related features. 


Our analysis revealed hotspots for all three outcome groups (Responders; Improvers and No Benefit Patients) in the ANT region. The volumes of the hotspots showed considerable overlap preventing clear outcome discrimination based on these hotspots. 

The volume of the responders was located inferiorly to that of improvers and No Benefit Patients. The anatomical fingerprinting of the atlas revealed that all three outcome clusters have similar anatomical characteristics (Figure 1). 

The multinomial logistic regression of MORE data identified one anatomical area ventral anterior nucleus (VAM) that significantly contributed to the separation of the responder and No Benefit patient groups. 

The stimulation outcome score maps suggest a different hotspot location for the responder group in the US SANTE study (analysis in progress). The average number of contacts in ANT (MORE left=1.72, right=1.67 ) was comparable between the studies. 

The connectivity of the ANT region was characterized by 10 reconstructed pathways. These pathways constitute main connection sources of the ANT region. 


The considerable overlap of the aggregated maps (stimulation outcome score maps) does not allow for clear definition of discriminative stimulation targets. However, the identified various areas and associate fiber relations modulate the probability of being responder. The analysis shows that even if the VTAs are large small fields in the ANT region can nevertheless be segregated in terms of positive and negative improvement.

Milan MAJTANIK (Düsseldorf, Germany), Juergen MAI, Frans GIELEN, Kai LEHTIMÄKI, Volker COENEN, António José GONÇALVES FERREIRA, Antonio GIL-NAGEL, Jukka PELTOLA, Philippe RYVLIN, Abdallah ABOUIHIA, Thomas BRIONNE, Paul BOON
16:10 - 16:15 #26042 - Cost-Effectiveness Analysis of Responsive Neurostimulation for Drug-Resistant Focal Onset Epilepsy.
Cost-Effectiveness Analysis of Responsive Neurostimulation for Drug-Resistant Focal Onset Epilepsy.

Objective: We evaluated the incremental cost-effectiveness of responsive neurostimulation (RNS System) therapy for management of medically refractory focal onset seizures compared to pharmacotherapy alone. 


Methods: We created and analyzed a decision model for treatment with RNS therapy versus pharmacotherapy using a semi-Markov process. We adopted a public payer perspective and used the maximum duration of 9 years in the RNS long-term follow-up study as the time horizon. We used seizure frequency data to model changes in quality of life and estimated the impact of RNS therapy on the annual direct costs of epilepsy care. The model also included expected mortality, adverse events, and costs related to system implantation, programming, and replacement. We interpreted our results against societal willingness-to-pay thresholds of $50,000, $100,000, and $200,000 per quality-adjusted life year (QALY). 


Results: Based on 3 different calculated utility value estimates, the incremental cost-effectiveness ratio (ICER) for RNS therapy (with continued pharmacotherapy) compared to pharmacotherapy alone ranged between $34,867-$56,253. Multiple sensitivity analyses yielded ICERs often below $50,000 per QALY but predominantly below $100,000 and consistently below $200,000/QALY.


Significance: Modeling based on 9 years of available data demonstrates that RNS therapy for medically refractory epilepsy very likely falls within the range of cost-effectiveness depending on method of utility estimation, variability in model inputs, and willingness-to-pay threshold. Several factors favor improved cost-effectiveness in the future. Given the increasing focus on delivering cost-effective care, we hope that this analysis will help inform clinical decision-making for this surgical option for refractory epilepsy.

Brett YOUNGERMAN, Timothy DYSTER, Shraddha SRINIVASAN, Casey HALPERN, Guy MCKHANN, Sameer SHETH (Houston, USA)
16:15 - 16:20 #26046 - A role for modulation of the mamillothalamic tract in seizure control?
A role for modulation of the mamillothalamic tract in seizure control?

Frédéric L.W.V.J. Schaper, Birgit R. Plantinga, Albert J. Colon, G.
Louis Wagner, Paul Boon, Nadia Blom, Erik Gommer, Govert
Hoogland, Linda Ackermans, Rob P.W. Rouhl, and  
Yasin Temel

Background: Deep brain stimulation of the anterior nucleus of the thalamus (ANT-DBS) can improve seizure control for patients with DRE. Yet, one cannot overlook the high discrepancy in efficacy among patients, possibly resulting from differences in stimulation site. 

Objective: We tested the hypothesis that stimulation at the junction of the ANT and mammillothalamic tract (ANT-MTT junction) increases seizure control.

Methods:The relationship between seizure control and the location of the active contacts to the ANT-MTT junction was investigated in 20 patients treated with ANT-DBS for DRE. Coordinates and Euclidean distance of the active contact relative to the ANT-MTT junction were calculated and related to seizure control. Stimulation sites were mapped by modelling the volume of tissue activation (VTA) and generating stimulation heat-maps. 

Results: After 1 year of stimulation, patients had a median 46%-reduction in total seizure frequency with a 50% responder rate and 20% of patients were seizure free. The Euclidean distance of the active contacts to the ANT-MTT junction correlates to change in seizure frequency (r = 0.39, p = 0.01) and is ~30% smaller (p = 0,015) in responders than in non-responders. VTA models and stimulation heat maps indicate a hot-spot at the ANT-MTT junction for responders whereas non-responders had no evident hot-spot. 

Conclusion: ANT-MTT junction stimulation correlates to increased seizure control. Our findings suggest a relationship between stimulation site and therapy response in DBS for DRE with a potential role for the mammillothalamic tract. DBS directed at white matter merits further exploration for the treatment of epilepsy.

Yasin TEMEL (Maastricht, The Netherlands)
16:20 - 16:25 #26129 - Thalamic deep brain stimulation in refractory epilepsy. Surgical practices and outcomes.
Thalamic deep brain stimulation in refractory epilepsy. Surgical practices and outcomes.

Background: Deep brain stimulation (DBS) of the anterior nucleus of thalamus (ANT) is a treatment option for drug  resistant focal epilepsy. ANT-DBS has been available in European countries after receiving CE-mark in 2010 based on data from a randomized controlled trial (SANTE) (Fisher et al., 2010) and received FDA approval in 2018. Medtronic Registry for Epilepsy (MORE) is an observational open label registry aiming to collect clinical data about DBS implantation practices in Europe.


Objective: The aim of the present study is to report surgical implantation practices in centers participating in MORE registry. Secondly, we studied the potential correlation between surgical technique, active contact localization and outcome using advanced image processing tools to co-register individual patient brain and MNI brain for group analysis.

Patients and methods: A total of 191 patients were enrolled between February 21st 2012 and April 30th 2017 by 25 investigational sites in 13 countries. Implant data was available from 179 patients and full analysis set (FAS) was available from 157 patients that completed two year follow up. DBS surgery was performed by neurosurgeons according to local treatment practices aiming to implant lead contacts bilaterally to ANT. Importantly, surgical method or target definition was not controlled by the study group, but education regarding surgical technique used in the US SANTE trial (transventricular) was provided for participating centers. Lead positions were verified postoperatively with magnetic resonance imaging (MRI) and/or computed tomography (CT). Patient outcome was classified as follows: Responders (≥ 50% Seizure Reduction (SR); Improvers, (50%<SR<0%) and No benefit patients (SR increase). At least two-month baseline seizure diary data was required for participation in the study. Each patient’s imaging data was co-registered to MNI brain and contact locations and volume of tissue activation (VTA) were calculated. Statistical outcome score (SoS) maps were calculated for outcome groups by surgical trajectories.


Results: In 129 (76%) of patients Medtronic 3389 leads and in 39 (23%) patients Medtronic 3387 leads were implanted bilaterally (one patient had mixed lead types). One hundred patients (59%) had leads implanted using transventricular trajectory bilaterally and 60 (35%) had leads implanted using extraventricular approach. Ten patients (6%) had mixed trajectories. The statistical analysis resulted in a significant difference in the average number of contacts in ANT between transventricular and extraventricular approach (F(1,334)=49.38, p<0.01), where transventricular approach resulted in an average of twice as many ANT contacts compared to the extraventricular approach. The VTAs in transventricularly implanted leads covered ANT in its slightly anterior and superior aspect, while VTAs in extraventricularly implanted leads were distributed slightly more inferior, posterior and lateral (Figure 1). Interestingly, the distribution of VTAs was almost identical between responders and no benefit patient both in transventricular and extraventricular approaches (Figure 2).


Discussion: Due to location of ANT in thalamus bordering to CSF from anterior, superior and medial aspects, two distinct surgical approaches (transventricular and extraventricular) were selected by neurosurgeons. VTAs were distributed slightly differently depending on surgical techniques selected, but no obvious difference in VTA coverage between responders and no benefit patients was observed neither in transventricular trajectory nor extraventricular trajectory groups. More consistent lead contact placement in ANT seems to favor transventricular approach, but no superiority of one surgical approach in terms of efficacy could be demonstrated in this open label multicenter dataset.


Conclusions: Structural gray matter anatomy and lead contact locations do not clearly explain differences in therapy outcomes. Yet unidentified patient and/or epilepsy related factors or more complex white matter network effects may predict therapy response more reliably and remain to be explored in future studies.

Kai LEHTIMÄKI (Tampere, Finland), Yasin TEMEL, António José GONÇALVES FERREIRA, Volker COENEN, Juergen MAI, Milan MAJTANIK, Antonio GIL-NAGEL, Jukka PELTOLA, Philippe RYVLIN, Abdallah ABOUIHIA, Thomas BRIONNE, Frans GIELEN, Paul BOON
16:25 - 16:30 #27684 - Deep brain stimulation of anterior nucleus of thalamus for intractable epilepsy: analysis of contacts position.
Deep brain stimulation of anterior nucleus of thalamus for intractable epilepsy: analysis of contacts position.

Anterior nucleus of the thalamus as a target of deep brain stimulation (ANT-DBS) is one of the well-tolerated' safe and promising procedures for treatment of epilepsy based on the data from both the experimental studies and limited clinical trials. These preliminary evidences were encouraging enough to design more comprehensive anatomical and functional analysis to improve the outcome.
Our aim was to create mapping analysis of all contacts potentially stimulated and consequently involved in therapeutic and adverse effects with further analysis of "hot spot" and "cold spot" within the target.
Methods through a prospective randomized multicenter trial in 12 French expert centers, 48 patients have been implanted by Medtronic 3389 lead (348 contacts) and were analysed based on high resolution neuroimaging data of brain anatomy and leads implanted. We did patient-specific comprehensive 3D visualization of lead location and further contact-specific anatomical characterization within the detailed thalamic parcellation and surrounding structures invaded in the all three planes on each side using automatic algorithm and StatMaps working pipeline of MRX-Brain platform in MNI space. Group level and individual level analysis of the coordinates of contacts were conducted in correlation with the clinical results and adverse effects if present.
Better understanding and mapping of ANT based on the functional clinical effect in correlation and detailed precise contact characterisations and location, could help for making decisions better on how to implant, program, and fine tune ANT-DBS therapy.

Hussein HAMDI (Marseille), Milan MAJTANIK, Claire HAEGELEN, Juergen MAI, Stephan CHABARDES, Jean REGIS
16:30 - 16:35 #26730 - A tractography study after bilateral anterior thalamic stimulation in drug-resistant epilepsy.
A tractography study after bilateral anterior thalamic stimulation in drug-resistant epilepsy.

Despite optimal medical treatments, many epilepsy patients have drug-resistant seizures. Even after surgery, only 58% of the patients are free of seizure against 8% of the patients without surgery and with best medical treatments (1). In France, a national protocol called FRANCE has included prospectively 61 patients with refractory partial-onset epilepsy and failure of vagus nerve stimulation to be treated with a bilateral chronic stimulation of the anterior thalamic nucleus (ATN). Previous studies have shown depression and memory impairment as the main stimulation-related adverse effects. The prefontal cortex being involved in depression and memory, the object of the study was to assess the patterns of connectivity between the prefrontal cortex and the ATN in epileptic patients with ATN stimulation.

   In FRANCE, all the patients were assessed preoperatively and at one year postoperatively using psychiatric, neuropsychologic and quality of live assessments. Among the 61 patients, twelve of them had, more than the morphological MRI, a diffusion MRI based on the CUSP (Cube and Sphere) method. We used first the PyDBS template (2) to delineate the ATN and 10 Brodmann’s areas (BA) involved in depression and memory impairments. Our regions of interest were 8 prefrontal areas (BA n°8, 9, 10, 11, 44, 45, 46, 47) and 2 areas from the anterior cingulate gyrus (BA n°24, 32). We used the BRAINResample from 3D Slicer to better segment the areas and we corrected manually the segmentation of the ATN if necessary. We used a deterministic method from the SlicerDMRI project to study the tracts between the ATN and our 10 regions of interest. To compare our patients, we used the data from 12 healthy subjects from the Human Connectome Project and analysed them with the same tools. We analysed the correlations between the fractional anisotropy and the mean diffusivity with the clinical data in the patients with Spearman’s rank test. P values < 0.05 were deemed to be significant.

    In all patients and the healthy subjects, the ATN was connected bilaterally with the BA 11 and 47 (Fig. 1). Only 4 patients had connections between the ATN and the BA 9, 10, 45 and 46. In the patients, the mean diffusivity of the right BA 47 and BA 11 was significantly correlated with the verbal and working memories, and with the trail making test difference. The mean diffusivity of the left BA 47 and BA 11 was significantly correlated with the working memory.  Only the fractional anisotropy of the left BA 11 was significantly correlated with trail making test difference. A significant difference was found between the pre- and the postoperative conditions for the depression score (Beck), which was reduced after ATN stimulation.

   We showed that the ATN seems to have strong connectivity with the areas BA 11 and BA 47, perhaps with a tendency for a right hemisphere dominance. BA 11 belongs to the orbitofrontal cortex involved in the depression. BA 47 is the pars orbitaris of the inferior frontal gyrus involved in the emotional control and the inhibition. These two areas were not impaired in these epileptic patients. Even if our study has several limitations, deterministic tractography appears as a valuable non-invasive tool for the exploration of the thalamocortical connections in epileptic patients.  



1. Wiebe S, Blume WT, Girvin JP, Eliasziw M (2001) A randomized, controlled trial of surgery for temporal-lobe epilepsy. New England J Med 345(5), 311318.

2. Haegelen C, Coupé P, Fonov V, Guizard N, Jannin P, Morandi X, Collins DL (2013) Automated segmentation of basal ganglia and deep brain structures in MRI of Parkinson’s disease. Int J Comp Assist Radiol Surg 8(1), 99110.

Claire HAEGELEN (Lyon), Mathilde GAUDIAN, Alfonso ESTUDILLO, John BAXTER, Elise BANNIER, Stephan CHABARDES, Anca NICA, Pierre JANNIN
16:40 - 16:45 #26195 - Developing prediction models of SEEG electrode implantation accuracy in epileptic patients.
Developing prediction models of SEEG electrode implantation accuracy in epileptic patients.


Resection of the epileptogenic zone (EZ) is a curative treatment option for selected patients with drug-resistant focal epilepsy. Stereoelectroencephalography (SEEG) aims to localize the EZ. Accuracy of the implanted intracerebral depth electrodes is of critical importance because it improves the delineation of the EZ leading to more precise resective surgery and higher treatment-gain on the one hand, and it lowers the risk of SEEG-associated complications, such as intracranial hemorrhage, on the other hand. Assessment of the implantation accuracy is measured by calculating entry - and target point localization errors (EPLE and TPLE). Many factors appear to contribute to EPLE and TPLE, which may be considered as predictors of accuracy.


To identify potential predictors of depth electrode implantation accuracy, in order to develop prediction models for EPLE and TPLE. 


Retrospective data retrieval and analysis of 75 patients with focal drug-resistant epilepsy (DRE) in whom S-EEG depth electrode implantation was performed between September 2008 and February 2020. A list of 21 potential accuracy predictors was composed and analyzed retrospectively for all patients. Due to the missing data being scattered across electrode cases, the data set was imputed. The potential predictors were identified based on objectiveness, external validity, and relevance, and subsequently investigated for multicollinearity using bivariate correlations (> 0.9 = eliminated) and Variance Inflation Factor (VIF) (VIF > 10 = eliminated). Additionally, univariable multilevel analysis was performed, followed by multivariable multilevel analysis using stepwise backward elimination, to obtain a prediction model for EPLE and TPLE for all implanted electrodes and specified per implantation direction (orthogonal or oblique). Finally, to check the accuracy of the prediction models, the root mean square error (RMSE) was calculated for all prediction models, which is measured in millimeters (mm). 


1725 electrodes were analyzed. No variables were excluded based on multicollinearity. 6 prediction models were obtained. The most accurate prediction model was aimed at predicting the EPLE for the electrode direction orthogonal and had an RMSE of 1.311mm. Moreover, the prediction models aimed at predicting the remaining outcome variables were the following: total EPLE = RMSE 2.102 mm, EPLE oblique = RMSE 2.358 mm, total TPLE = RMSE 2.765 mm, TPLE oblique = 22.818 mm and TPLE orthogonal = RMSE 2.276 mm. 


The present study was accomplished to develop prediction models aimed at SEEG electrode implantation accuracy. This is the first study ever where prediction models have been obtained at predicting SEEG implantation accuracy. Although it is mentioned that EPLE orthogonal has the most accurate prediction model, it is still considered clinically inaccurate due to the relatively large RMSE value. Therefore, all the 6 models may not be useful in clinical practice; however, it may be a good starting point for neurosurgeons and neuroscientists. The limitations of the study include planning bias of the electrodes, specifically regarding electrode direction. Some valuable information was not analyzed due to external validity, but may affect the predicted outcome variables.


Intracerebral depth electrodes implantation accuracy can be predicted based on 21 predictors, of which the outcome variable EPLE, for the subgroep orthogonally implanted electrodes, could be predicted most accurately. 

Cyan KORT (Maastricht, The Netherlands), Janvi KAKADIA, Pieter KUBBEN, Olaf SCHIJNS, Louis WAGNER, Lars VAN DER LOO, Govert HOOGLAND, Jim DINGS, Kim RIJKERS, Sander VAN KUIJK
Salle 120

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
16:30 - 19:00

Parallel Session 5
Pain Surgery

Moderators: Konstantin V. SLAVIN (professor) (Chicago, USA), Pawel SOKAL (head of department) (Bydgoszcz, Poland)
16:30 - 16:50 Limbic or sensory disciminative DBS in Neuropathic pain. Joachim K. KRAUSS (Chairman and Director) (Hannover, Germany)
16:50 - 17:10 Renaissance of Peripheral Nerve Stimulation in Treatment of Pain" with possible subheading "Does it really work? Konstantin V. SLAVIN (professor) (Chicago, USA)
17:10 - 17:30 Dorsal root ganglion stimulation. Dirk DE RIDDER (Belgium)
17:30 - 17:50 DBS for Pain. Tipu AZIZ (Professor) (Oxford, United Kingdom)
17:50 - 18:00 Mesencephalotomy: anatomo-clinical correlations based on a brainstem normalized coordinate system. Laurent GOETZ (Chercheur) (Paris, France)
18:00 - 18:10 #23493 - Effectiveness of Stereotactic Thalamotomy in Patients with Intractable Pain.
Effectiveness of Stereotactic Thalamotomy in Patients with Intractable Pain.

OBJECTIVE Ablative procedures are still useful in the treatment of intractable pain despite the proliferation of neuromodulation techniques. In this study, the authors present the results of stereotactic thermolesion (ST) of the centromedian-parafascicularis (CM/Pf) nucleus in various pain syndromes.

METHODS Between 1999 and 2018, unilateral ST of the CM was performed in 62 patients suffering from various severe pain syndromes, in whom conservative treatment had failed. There were 37 women and 25 men in the study population, with a mean age of 63 years (range 30–85 years). The pain syndromes consisted of 22 patients with treatment-resistant trigeminal neuralgia (TN), 10 with postherpetic TN, 1 with TN related to multiple sclerosis, 12 with thalamic pain, 3 with phantom pain, 11 with facial pain related to surgery and 2 related to trauma. The median follow-up period was 23.5 months (range 1–124 months). Other invasive procedures for pain release preceded in 61.2% of patients. The Leksell stereotactic frame, SurgiPlan software, and T1- and T2-weighted sequences acquired at 1.5 T were used for localization of the targeted centromedian-parafascicularis (CM/Pf) nucleus. A stereotactic procedure with a frontal transdermal approach was performed under local anesthesia. The thermolesion was created by a unipolar 16 gauge electrode and Diros generator. A maximum temperature in the range of 75-80 ° C for 60 seconds was applied. Pain relief before and after ST was evaluated. The effect of the procedure was evaluated in percentage scale of remaining pain.

RESULTS The complete data were analyzed in 46 of 62 patients (74%). Overall pain relief was achieved in 19 (41.3%) patients and the pain intensity decreased to 39.5% on average. In 63% of these patients, the effect was recorded on the first follow-up 3 months after the procedure. Pain recurred only in one patient (4.5%) 5 months after the procedure. In one patient, cortical abscess was observed (1.6%) and treated conservatively, no other adverse events occurred.

CONCLUSIONS Our results suggest that stereotactic thermolesion of the CM/Pf complex in patients suffering from severe pain syndromes is a relatively effective and safe method, which can be used even in severely-affected patients. This work was supported by MH CZ – DRO (NHH, 00023884), IG191201.

Jaromir HANUSKA (Prague, Czech Republic), Dusan URGOSIK, Roman LISCAK
18:10 - 18:20 #23800 - DBS for chronic cluster headache: a clinical and image-based meta-analysis of individual patient data.
DBS for chronic cluster headache: a clinical and image-based meta-analysis of individual patient data.

Background: Deep brain stimulation (DBS) is a treatment alternative for refractory chronic Cluster headache (CCH). Despite several recent prospective case series reporting on good outcome, the effectiveness and the optimal stimulation target of DBS for CCH remain unclear. We aimed to obtain precise estimates and predictors of long-term pain relief in an individual patient data meta-analysis. Further, we aimed to construct a probabilistic stimulation map of effective DBS.

Methods:  We invited investigators of published cohorts of patients undergoing DBS for CCH, identified by a systematic review of MEDLINE from inception to Febuary 15, 2019, to provide individual patient data on baseline covariates, pre- and postoperative headache scores at medium (12-month) and long-term follow-up as well as individual imaging data to obtain individual electrode positions.  We calculated a stimulation map using voxel-wise statistical analysis. We used multiple regression analysis to estimate predictors of pain-relief.

Results: Among 40 patients from four different cohorts representing about 50% of all previously published cases, we found a significant 77% mean reduction in headache attack frequency over a mean follow-up of 44 months (corresponding to 75% responder rate). Positive outcome was not associated with baseline covariates. We identified two hotspots of stimulation covering the ventral tegmental area and the retrorubral midbrain tegmentum.

Conclusion: This study supports that DBS provides long-term pain relief for the majority of CCH patients. Our stimulation map of the region of influence of therapeutic DBS identified an optimal anatomical target site that can help surgeons guide their surgical planning in the future.

Andreas NOWACKI (Bern, Switzerland), Martin SCHOBER, Lydia NADER, Assel SARYYEVA, Anh Khoa NGUYEN, Alexander GREEN, Claudio POLLO, Joachim KRAUSS, Denys FONTAINE
18:20 - 18:30 #23896 - Molecular inflammatory phenotyping in neuropathic pain patients under unilateral L4-dorsal root ganglion stimulation.
Molecular inflammatory phenotyping in neuropathic pain patients under unilateral L4-dorsal root ganglion stimulation.

Introduction: Complex regional pain syndrome (CRPS) has been associated with a pro-inflammatory state driven by different circulating mediators. Conventional spinal cord stimulation (SCS) suppressed CRPS pain levels by 40-50% in the past. Most recently, an approach that appears to have a considerable promise for treating focal neuropathic pain has become available (dorsal root ganglion stimulation; DRGSTIM). Anatomically targeted DRGSTIM was found to be superior to conventional SCS in a Class I study as well as several controlled and uncontrolled observational clinical trials for a variety of pain conditions. Briefly, DRGSTIM may have the capability to restore the distorted filter function of the DRG, thus inhibiting hyperexcitability of DRG neurons and deeper layer compartments (laminae II/III) of the spinal cord. The precise mechanism of DRG-evoked effects on spino-nociceptive neural transmission as yet is not fully established

Dorsal root ganglion stimulation (DRGSTIM) suppressed pain levels and improved functional capacity in intractable CRPS in observational and randomized-controlled studies. However, in-human studies evaluating the impact of selective DRG stimulation on the neuro-immune axis in CRPS patients remains under-investigated. 

Methods/Materials: This observational pilot study enrolled 24 subjects (12 CRPS patients - 12 healthy controls) and performed immunoassays of inflammatory mediators in saliva/serum, gene expression assay (whole transcriptome) of neuroinflammatory genes (PantherTM pathway enrichment analysis) and evaluation of pain, mood and sleep at baseline and after 3 months of selective L4-DRGSTIM.  

Results: After L4-DRGSTIM CRPS pain significantly decreased with improved sleep and mood. Elevated levels were detected pre- and post L4-DRGSTIM for high-mobility group box 1, tumor-necrosis factor α, IL-6 and leptin indicating a pro-inflammatory state in CRPS patients. IL-1β was elevated pre-L4 DRGSTIM, but not post-treatment. IL-10 decreased after 3 months in serum, while saliva oxytocin increased after L4-DRGSTIM (Fig.1). Gene expression sub-group analysis of the CRPS subjects (PantherTM pathway enrichment analysis) revealed changes, which may be associated to the clinical effects observed after unilateral L4-DRGSTIM including up and down regulation of certain inflammatory markers (Figure 2).

Discussion: Although of preliminary character, L4-DRGSTIM evoked pain relief and improved functional impairment in CRPS patients revealing a pro-inflammatory molecular pattern. Serum IL-10 significantly declined, while saliva oxytocin increased after L4-DRGSTIM. Sub-group analysis demonstrated either upregulated or downregulated genes involved in immune host response and neural pain circuits.

Conclusions: Large biobank-based approaches are recommended to re-evaluate genetic phenotyping as a quantitative outcome measure for neurostimulation therapy in CRPS patients. The concept of a personalized and predictive neurostimulation therapy based on a comprehensive, preimplant mapping represents the next pivotal step in clinical neuromodulation research for pain.




Thomas KINFE (Erlangen, Germany), Michael BUCHFELDER, Thomas YEARWOOD
18:30 - 18:40 #24080 - Gasserian ganglion stimulation in neuropathic facial pain.
Gasserian ganglion stimulation in neuropathic facial pain.


Neuropathic facial pain is a debilitating disease in its medically intractable form. According to the 3rd Edition of International Classification of Headache Disorders it can be the result of multiple sclerosis, mass effect, or its origin can be associated with posttraumatic, postherpetic nerve injuries. Gasserian ganglion stimulation can provide a reliable tool to decrease neuropathic facial pain.


Materials and methods

13 patients suffering from medically intractable neuropathic facial pain were enrolled in this study. Indications were of postherpetic (n=3), posttraumatic (3), iatrogenic (5), and unknown (2) origin. Each patient was implanted with a custom-made Medtronic 3 contact anchored, curved lead, under light sedation and intraoperative trial stimulation. Leads were advanced through the oval foramen under fluoroscopy, lead extension has been tunneled and externalized on the neck. Each patient underwent an at most 3-week-long postoperative testing period with an external neurostimulator to evaluate the results of stimulation. VAS scores were obtained three times a day, stimulation parameters were adjusted accordingly.



Of the 13 patients 11 patients completed the trial period successfully. Mean age was 53.84±14.24 years, gender distribution was 8 female and 5 male patients. Years passed since onset of symptoms 8.69±10.15. Mean preoperative VAS 9.15±0.9 decreased to 2.64±1.5 in the first two weeks during testing on an external neurostimulator. At least three months after surgery VAS scores were kept in lower ranges - mean VAS 2.5±2,27. Stimulation amplitude and pulse width were inconsistent in the whole group.



Gasserian ganglion stimulation can be a reliable therapeutic option in medically intractable neuropathic facial pain, but the underlying mechanism is yet to be uncovered. Due to the need of highly variable, patient specific stimulation parameters sometimes an even 3 week-long trial period is advised before IPG implantation to achieve tailored therapeutic settings. The described method can reliably decrease neuropathic facial pain in the observed patient population. Enrollment of more patients is necessary to uncover specific parameters that can tuned to different indications to provide a disease specific guideline for programming. Implementing new fixation techniques can achieve more control during intraoperative lead positioning.

László HALÁSZ (Budapest, Hungary), Loránd ERŐSS
18:40 - 19:00 Discussion.

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
17:00 - 19:15

Parallel Session 7

Moderators: Patric BLOMSTEDT (Neurosurgeon) (Umeå, Sweden), Jean FARISSE (HOSPITAL PRACTITIONER) (MARSEILLE, France), Kuan Hua KHO (Neurosurgeon) (Enschede, The Netherlands), Mircea POLOSAN (MD, PhD) (Grenoble, France)
17:00 - 17:10 #23609 - Neuropsychological and neuropsychiatric outcome after anterior capsulotomy (including the repeated surgery) for obsessive-compulsive disorder.
Neuropsychological and neuropsychiatric outcome after anterior capsulotomy (including the repeated surgery) for obsessive-compulsive disorder.

Objective: Anterior capsulotomy (AC) is sometimes used as a last resort for treatment-refractory obsessive-compulsive disorder (OCD). Previous studies assessing neuropsychological outcomes in patients with OCD have identified several forms of cognitive dysfunction that are associated with the disease, but only a few have focused on changes in cognitive function in OCD patients who have undergone surgery.

Purpose: In the present study, the authors investigated the effects of AC on the cognitive performance and mood status of patients with treatment-refractory OCD.

Method: The authors presented 12 patients with treatment-refractory OCD who had undergone bilateral AC between 2012 and 2019. Patients (N=12 F:M 5:7 ; mean age 39.7 years; duration more than 5 years) were assessed 6 months after intervention at the Department of Clinical Psychology, Na Homolce Hospital, Prague.  Treatment-refractory OCD diagnosis was based on recommended criteria for surgical treatment. For purpose of this study, patients were examined using WAIS-III – short form, including subtests Block Design, Arithmetic, Similarities, Picture Completion; Digit Span; Digit Symbol; RAVLT; TMT A, B; ROCFT; VFT. Anxiety-depressive symptomatology was assessed through the Czech version of the BDI-II; BAI and MADRS. The Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) was used to measure OCD symptoms.

Results: We found significant decrease of OCD, anxiety and depressive symptomatology assessed by Y-BOCS scale, BDI-II and BAI (p< .05) 6 months after AC in 7 patients and  partial decrease in 5 patients in whom the satisfactory result was reached after the repeated AC. We found unchanged cognitive performance measured by battery of neuropsychological tests in all patients. We detected better immediate and delayed visual memory performance (p< .05) at 6 months follow-up.

Conclusion: The findings of this study suggest that AC not only reduces OCD and anxiety-depressive symptoms but doesn’t influence cognitive performance of patients, even if they underwent repeated surgery. Supported by MH CZ – DRO (NHH, 00023884), IG 161201

Dusan URGOSIK (Prague, Czech Republic), Lenka KRAMSKA, Jaroslava SKOPOVA, Lucia HRESKOVA, Roman LISCAK
17:10 - 17:20 #23937 - What do we learn from the prefrontal leucotomy undergone for psychiatric disorders in times past? – Study with the combination of tractographic and scalp EEG functional connectivity analysis –.
What do we learn from the prefrontal leucotomy undergone for psychiatric disorders in times past? – Study with the combination of tractographic and scalp EEG functional connectivity analysis –.


    Although prefrontal leucotomy was an obsolete remedy for treatment-refractory mental illness, some patients who have undergone prefrontal leucotomy continue to reside in psychiatric hospitals. The surgery achieved a certain therapeutic response, however, that concomitantly produced several complications, such as personality defect (Miller and Psych, 1967), because this extensive lesion was often associated with impairment of frontal lobe function.

    During the past few years, the research focus in brain imaging moved from localizing functional regions to understanding how different regions interact together. It is now widely accepted that some of the brain functions are not supported by isolated regions but rather by a dense network of nodes interacting in various ways. 

    In view of this, the purpose of this study was aimed to identify fibers and functional connectivities severed by the prefrontal leucotomy.



    Diffusion tensor imaging (DTI) scans were acquired from 5 schizophrenia patients and from 3 prefrontal leucotomized patients with schizophrenia, group-matched for age, on a 1.5T scanner. All leucotomized patients underwent the surgeries approximately more than 40 years previously. Voxelwise statistical analysis of the fractional anisotropy (FA) data in white matter tracts were carried out to compare between leucotomized schizophrenia group and non-leucotomized schizophrenia patient group.
The other method we adopted is oscillation-based functional connectivity analyzed with lagged non-linear coherence developed by Pascual-Marqui (2011). Routine scalp-EEG (19 electrodes) was recorded in 2 prefrontal lobotomized patients and 5 schizophrenia patients matched for age, while they were at rest with eyes closed. This measures coherence (connectivity) between all pairs of ROIs, for each frequency band, for each group.



    Compared with non-leucotomized schizophrenia group, leucotomized schizophrenia group had lower FA in the white matter of rostral and dorsal region of corpus callosum and the mediodorsal thalamic nucleus (MD). Analysis of EEG functional connectivity for the resting state network in leucotomized schizophrenia patients revealed decreased connectivity between prefrontal regions and the rest of the brain for generally almost frequency bands. In particular, the surgeries decreased connectivity of long distance paths  linking between the prefrontal and other cortical regions for low frequency bands, while that of short distance paths between the frontal areas was reduced for high frequency bands.



    The result indicates that the connectivity sacrificed by the prefrontal leucotomy involves fiber degeneration in MD. This was in accordance with the previous studies that demonstrated the prefrontal cortex has reciprocal relationships with a specific portion of MD. Prefrontal leucotomy covered the therapeutic targets the contemporary psychiatric surgeries use. Prefrontal cortical activities in leucotomized brain remained still active, while they were anatomically isolated from the rest of the brain by disconnection with surgical manipulations. Leucotomies decreased functional connectivity of the long distance paths for lower frequency bands, while that of the short distance paths reduced for high frequency bands.

    The specific relationships between activity patterns over the entire brain within individuals and aberrant emotional states remain to be elucidated for advancement of the surgery for psychiatric diseases.

Katsushige WATANABE (Tokyo, Japan), Shunsuke IKEDA, Yasushi OKAMURA, Makoto TANIGUCHI
17:20 - 17:30 #23955 - Structural MRI analyses in OCD: a review of the literature to set out possible predictive factors in patients candidate to anterior capsulotomy.
Structural MRI analyses in OCD: a review of the literature to set out possible predictive factors in patients candidate to anterior capsulotomy.

We analysed the literature to find out the status of art on structural imaging (MRI) in patients affected from OCD with a particular attention on comparison with health controls and with respect to symptoms.

MRI studies in OCD were initially based on region of interest, and later voxel‑based morphometry (VBM) was used (whole‑brain analysis approach) for studying volumetric differences.

Meta-analyses of the literature comparing OCD patients with health controls on cortical thickness and surface showed lower surface area in the temporal cortex and thinner inferior parietal cortex, in the right dorsolateral prefrontal cortex, left posterior cingulate cortex, and bilateral hippocampi other than those prevalent in the CSTC model.  Other meta‑analysis of VBM studies showed decreased grey matter (GM) in bilateral orbitofrontal cortex and anterior cingulate cortex and increased grey matter in the basal ganglia (caudate, putamen, and pallidum). Also Increase in cerebellar GM was found in one meta‑analysis.

Other meta-analyses showed differences with respect to symptom dimensions:

-        cortical thickness was increased in the left OFC in contamination/ cleaning,

-        right OFC, left cingulate cortex, right parietal cortex, and middle temporal cortex in sexual/religious;

-        right occipital and lingual gyrus in aggression/checking;

-        left insula, lingual, precentral, and postcentral gyrus with decrease in fusiform GM in symmetry/order symptom dimension

Even if many studies had recruited patients across heterogeneous symptom dimensions, contributory to the variability in imaging findings making comparisons difficult. Y‑BOCS severity scores have a positive correlation with increase in left dorsal ACC thickness, decrease in cortical thickness in bilateral occipital gyri, and negative correlation with GM volume of left ventral striatum, left frontal pole volume, right anterior cingulate gyrus volume and surface area, and right OFC gyrification.

Enlarging what literature demonstrated is possible to re-analyse the same regions in patients that underwent neurosurgical procedure to find out some predictive factors.

Giorgio SPATOLA (Brescia, Italy), Raphaelle RICHIERI, Viktor JIRSA, Costantin TULEASCA, Yassine BELTAIFA, Nadine GIRARD, Jean Marie REGIS
17:30 - 17:40 #26011 - Long-term results of deep brain stimulation for schizophrenia: a pilot study.
Long-term results of deep brain stimulation for schizophrenia: a pilot study.

Objective: Deep brain stimulation is spreading its indication within psychiatry. In this pilot study, authors present long-term results of DBS for treatment resistant schizophrenia (TRS) and target selection.


Methods: Eight patients with TRS underwent DBS, 4 of them in the accumbens nucleus (AcN) and 4 in the subgenual cingulate gyrus (SCG). Demographic and baseline characteristics were recorded. The Positive and Negative Symptoms Scales (PANSS) and its sub-scales (positive, negative and general symptoms) were used to measure the response to the therapy. Responder were defined as improvement of PANSS > 25% to baseline. The AC/PC coordinates of the active contacts and the total electrical energy delivered (TEED) were calculated and correlated with clinical outcomes. Patient-specific tractography-activation models (TAMs) were performed to identify potential pathways in responders.


Results: Five women (62.5%) and three men (37.5%) with a median age of 41 years [34 – 58] were included. Along 4 years follow-up, four patients (50%) improved their positive symptoms (two implanted in SCG and two in AcN), two patients (25%) improved their negative symptoms (two SCG) and three patients (37.5%) improved their general symptoms (two AcN and one SCG). TEED showed no correlation with clinical outcomes. In SCG patients the forceps minor (FM) and the frontopolar fascicle (FPF) were the most frequently stimulated pathway, while the superolateral middle forebrain bundle (slMFB) was frequently stimulated in AcN patients.


Conclusions: DBS for TRS is an investigational indication. These initial results show that for chronic schizophrenic patients, DBS in SCG might improve specific psychotic symptoms (positive and negative) while DBS in AcN might improve general symptoms. FM, FPFS and slMFB might be tracts involved in the complex circuitry of this disease. 

17:40 - 17:50 #26041 - Dual-target deep brain stimulation for depression informed by intracranial stereo-EEG.
Dual-target deep brain stimulation for depression informed by intracranial stereo-EEG.

Deep brain stimulation (DBS) for treatment-resistant depression (TRD) is investigational, with previous studies demonstrating heterogeneous results. We devised a novel personalized medicine platform for DBS therapy development focused on this essential aspect: the need to achieve an individualized understanding of the specific brain networks contributing to a patient’s particular depressive phenotype and their response to stimulation. To do so, we borrowed from the field of epilepsy surgery the well-established approach of using intracranial EEG to individualize the understanding of epileptic networks.

We report results from the first subject treated with DBS for TRD using this approach. The subject underwent implantation of DBS leads targeting both ventral capsule/ventral striatum (VC/VS) and sub-callosal cingulate (SCC) bilaterally, as well as 10 stereo-EEG (sEEG) electrodes targeting fronto-temporal regions putatively involved in depression-implicated networks. We then performed a series of studies using these externalized electrodes to gain an individual-specific understanding of mood-regulating brain networks and their response to stimulation. These recordings enable derivation of individually optimized stimulation parameters using a novel “inverse solution” approach: we first use this novel dual-target strategy to maximize accessibility to as much of the relevant TRD network as possible. We then use the intracranial sEEG recordings to narrow the possible stimulation parameter space and choose those parameters that engage network sub-regions most effectively for the individual subject. This “sEEG-informed DBS” platform enables therapeutic development with an emphasis on individualized understanding of network (patho-)physiology.We then implemented these customized DBS stimulation parameters in an outpatient trial. During the open-label phase, standard rating scales demonstrated progressive improvement in depression severity leading to remission of symptoms by week 22. The subject then entered the double-blind, randomized discontinuation phase of the trial to distinguish between true and sham response. Over this time, he reported steadily worsening mood and anxiety, and his symptom scores increased during this period until he met rescue criteria. At this point, stimulation was reinstated at pre-discontinuation levels, and his depression symptoms again quickly remitted. 

We envision this platform as one for DBS therapy development and optimization for challenging disorders during a critical period of knowledge acquisition. We do not use this platform primarily to find new stimulation targets, as doing so would potentially require an intracranial search in all future patients. Rather, we use the electrophysiological data to optimize identification of DBS stimulation parameters that engage pathological networks. Our results highlight the feasibility and potential value of doing so using an “inverse solution” approach.

The increased invasiveness of this platform over conventional DBS is meant to be a bridge to future less invasive approaches. This successful case is a critical first demonstration of this novel strategy using direct intracranial measurements. Future work can test the success of substituting non-invasive techniques as readouts of neural activity as they become increasingly reliable. More generally speaking, an important advantage of this platform is that stimulation configurations derived from future analyses can be readily implemented as new sets of stimulation parameters to employ and test. This approach enables repeated iteration between computational analysis and clinical testing, providing a long-term testbed for the neuroscientist and continued hope for symptomatic relief for the patient.

            In summary, our initial results demonstrate the feasibility of this novel platform. We propose that this approach, if consistently demonstrated safe and effective, can be used to develop and improve surgical neuromodulation for a vast array of neurological and psychiatric disorders.


Sameer SHETH (Houston, USA), Kelly BIJANKI, Brian METZGER, Anusha ALLAWALA, Victoria PIRTLE, Joshua ADKINSON, John MYERS, Raissa MATHURA, Denise OSWALT, Evangelia TSOLAKI, Jiayang XIAO, Angela NOECKER, Adriana STRUTT, Jeffrey COHN, Cameron MCINTYRE, Sanjay MATHEW, David BORTON, Wayne GOODMAN, Nader POURATIAN
17:50 - 18:00 #26057 - Targeting deep brain stimulation in obsessive-compulsive disorder: lessons learned form 84 consecutive cases.
Targeting deep brain stimulation in obsessive-compulsive disorder: lessons learned form 84 consecutive cases.

We present 84 consecutive patients with therapy-refractory obsessive-compulsive disorder (OCD) who underwent deep brain stimulation (DBS) from 2005-2019, and determined the relationship between the anatomical location of active electrode contacts and the clinical outcome. In the first 28 patients, electrodes were implanted in the nucleus accumbens (NAc), with a lateral angle approximately following the anterior limb of the internal capsule (ALIC): 11 (43%) responded to DBS. In patients 29-72, electrodes were implanted in the ventral part of the ALIC (vALIC), just dorsal of the NAc. Also, five previous non-responding NAc patients underwent repositioning of their electrodes to the vALIC. Twenty-nine patients (59%) responded to DBS. In patients 73-84, electrodes were implanted in the superolateral branch of the medial forebrain bundle (MFB), during its course through the ALIC: 10 (83%) responded to DBS. Five previous non-responding NAc & vALIC patients underwent repositioning of their electrodes to the MFB, and three of them became responders. We also retrospectively identified 8 patients among patients 29-72 in whom both centers of stimulation were in the MFB: 7 responded to DBS. Overall, 20/25 MFB patients (80%) responded to DBS. We therefore conclude that the MFB (during its course through the ALIC) is a better DBS target than vALIC and NAc in treatment-refractory OCD.

Pepijn VAN DEN MUNCKHOF (Amsterdam, The Netherlands), Maarten BOT, Luka LIEBRAND, Pelle DE KONING, Nienke VULINK, Martijn FIGEE, Guido VAN WINGEN, Damiaan DENYS, Rick SCHUURMAN
18:00 - 18:10 #26194 - Globus pallidus internus deep brain stimulation can provide durable benefits in severe Gilles de la Tourette syndrome.
Globus pallidus internus deep brain stimulation can provide durable benefits in severe Gilles de la Tourette syndrome.



To explore whether bilateral Globus pallidus internus (GPi) deep brain stimulation (DBS) provides sustained symptom and quality of life improvements in Gilles de la Tourette syndrome (GTS).




GTS is defined by the presence of multiple motor and vocal tics which can be associated with psychiatric comorbidities. In medical treatment refractory cases, DBS of the GPi has been shown to result in a broad range of symptom improvements though the durability of this effect remains uncertain. We previously demonstrated short term benefits in a cohort of GTS patients treated with GPi-DBS. In this study we explored long-term outcomes in these patients.



Patients previously implanted between 2010 and 2014 were contacted for reassessment. Those consenting were assessed using the following scales: Yale Global Tic Severity scale (YGTSS) sub scores (motor tics, vocal tics, and impairment); Yale- Brown Obsessive Compulsive Scale (YBOCS); Quality of life (QOL); Beck Depression Index (BDI) and State-Trait Anxiety Inventory (STAI). Total scores in the preoperative, and first available post-operative assessments were compared with the current delayed assessment. Clinical outcomes for continuous values are presented as mean ±standard deviations (SD). The Wilcoxon signed-rank test was used for comparison of non-parametric paired data. An exploratory univariate linear regression analysis was performed to identify if baseline factors contributed to the final YGTSS and QOL scores. A two-side p value of <0.05 was used to judge statistical significance.



Of the 21 patients with GTS previously implanted at our centre between 2014 and 2016, 11 consented to participate. A slight male predominance (7/11) was noted. The mean operative age was 37.3±13.6 years and age of symptom onset 8.0±2.6 years. Seven patients were implanted in the anteromedial GPi and 4 in the posteroventral GPi. The first post-operative assessment was performed a mean of 15.7±15.2 months after implantation and the delayed post-operative assessment after 8.5± 1.5 years. The YGTSS score improved post implantation (85.6±12.5 vs 48.2±23.9, p <0.01) and remained unchanged at the delayed assessment (48.2±23.9 vs 46.8±22.2, p=0.85). Similar patterns of improvement were also noted for the total motor and vocal tic burden sub-scores. Patients reported an improvement in mood post-surgery as measured by the BDI (30.6±6.0 vs 17.0±10.2, p=0.04). This effect was sustained at the delayed assessment (17.0±10.2 vs 23.6±10.8, p=0.45). No difference was noted in the YBOCS score at the early post-operative assessment (15.5±8.6 vs 11.0±10.1, p=0.21), nor the delayed assessment (11.0±10.1 vs 19.0±10.8, p=0.09). No changes in the STAI score were noted at either post-operative assessment. Patients reported an improvement in QOL at the first post-operative assessment (74.7±17.4 vs 45.1±28.5, p=0.02). This effect was sustained at the delayed assessment (45.1±28.5 vs 47.6± 20.4, p=0.48). Although no baseline patient characteristics predicted the YGTSS score at the final assessment, the age of onset of the disease predicted the final QOL score noted (β=5.90, 95% CI 2.02-9.79, p=0.01). 



The beneficial effects of GPi DBS in Gilles de la Tourette’s syndrome can be sustained. However, we cannot exclude the possibility that the patients who declined to participate may include individuals with lower response rates to DBS.

Olga PARRAS GRANERO (London, United Kingdom), Nirosen VIJIARATNAM, Marjan JAHANSHAHI, Joseph CANDELARIO, Catherine MILABO, Harith AKRAM, Jonathan HYAM, Patricia LIMOUSIN, Marwan HARIZ, Eileen JOYCE, Ludvic ZRINZO, Tom FOLTYNIE
18:10 - 18:20 #26290 - Preliminary Safety and Feasibility Outcomes of Nucleus Accumbens Deep Brain Stimulation (DBS) Clinical Trial for Opioid Use Disorder.
Preliminary Safety and Feasibility Outcomes of Nucleus Accumbens Deep Brain Stimulation (DBS) Clinical Trial for Opioid Use Disorder.

Introduction: Given high relapse rates and the prevalence of opioid overdose deaths, novel treatments for opioid use disorder (OUD) are desperately needed for those who are treatment refractory. We initiated a US National Institute on Drug Abuse (NIDA) sponsored clinical trial to evaluate the safety and feasibility of deep brain stimulation (DBS) of the Nucleus Accumbens (NAc) and ventral anterior internal capsule (VC) for treatment refractory OUD and its effects on substance use, craving, frontal, executive, and emotional functions.

Methods: Participants with at least a 5-year history of severe, treatment-refractory OUD with multiple overdoses were eligible for this study. Participants were implanted with bilateral Medtronic DBS electrodes (3387) within the NAc/VC. Clinical safety and the therapeutic response to DBS, including the effects on substance abstinence, craving, mood, and executive function, were assessed. 18fluoro-Deoxy-Glucose (FDG) PET was performed at 12-week post titration to evaluate metabolic changes following DBS.  Local field potentials (LFP) were assessed using the Medtronic Percept DBS recording and stimulation device.

Results: Two eligible participants underwent NAc/VC DBS. There were no surgical complications. The first participant was a 33-year-old male who has achieved over 600 days of continuous abstinence to date, compared to an average relapse time of approximately 1-2 weeks prior to DBS implantation. Cravings for opioid and other substances decreased significantly post-implantation with the most significant reduction in benzodiazepine craving (53.4±29.5 vs. 1.0±2.2; p<0.001). The DBS therapeutic response was associated with a concomitant increase in glucose metabolism in the dorsolateral prefrontal and medial premotor cortices through the FDG PET analysis. The participant also showed signs of improvement in depression, anxiety, impulsivity, and frontal/executive functioning. Diffusion Tensor Imaging (DTI) revealed that the location of the DBS contact with the most optimal therapeutic response corresponded to strong structural connectivity to the mesial frontal region. Stimulation-related adverse effects were associated with connectivity to the amygdala/temporal lobe.

The second participant was a 22-year-old male who underwent DBS implantation successfully. Due to non-compliance with the study protocol and follow-up treatment regiment, the DBS system was explanted at 15 weeks post-implantation.

Conclusion: In a participant with severe, treatment-refractory opioid and benzodiazepine use disorder, DBS of the NAc/VC is safe, well-tolerated, and can reduce substance use and craving, and improve frontal and executive functions. The use of DTI provides valuable insight to select the optimal target for stimulation.  We will present the latest results of this ongoing clinical trial including LFP data and additional subjects. DBS for addiction is promising but complex and challenging given the nature of the population and disease. 

(This study received funding from NIDA 1UG3DA047714-01; device and technical support from Medtronic)

Ali REZAI (Morgantown, USA), Manish RANJAN, Pierre-Francois D’HAESE, Mark HAUT, Wanhong ZHENG, Laura LANDER, Nicholas BRANDMEIR, Victor FINOMORE, Sally HODDER, Berry JAMES, James MAHONEY
18:20 - 18:30 #26306 - Why subthalamic DBS should not be ignored for refractory OCD: evidence from quality of life study.
Why subthalamic DBS should not be ignored for refractory OCD: evidence from quality of life study.

Background : While Deep Brain Stimulation (DBS) is an established therapy for several neurological disorders, effectiveness and safety of this strategy in refractory Obsessive Compulsive Disorder (OCD) is increasingly reported since first reports in 1999. Randomized sham-controlled short term (3 months) as well as open-label long term follow-up (2y) clinical improvement with subthalamic (STN) DBS have consistently been shown on 2 different cohorts respectively. As OCD may lead to significant disability and altered quality of life, the latter includes subjective self-assessment and represents optimal therapeutical objective beyond clinical outcome. We thus focused on quality of life evolution with long term STN DBS in a cohort of refractory OCD.


Methods: Fourteen refractory OCD patients were recruited from Grenoble University Hospital and treated by STN high-frequency stimulation. Patients had at least five years of treatment-resistant, severe, disabling OCD before DBS surgery. Clinical severity and quality of life was evaluated using the YBOCS and SF36 scale, respectively. Patients were evaluated in the pre-op phase (T0), at 2-3 years of stimulation (T1), and at minimum of 5 years of stimulation (T2). Clinical scores were compared between the different end-points using repeated measures ANOVA. 


Results : We highlight that STN DBS lead to a significant improvement of mental and physical quality of life as measured by SF36 with a mean improvement of 46,9% and 26% respectively, while the clinical severity improved of 53,1% at 5 years end point compared to pre-op. Among the different dimensions of the mental score of quality of life, social function is the one improving the most ( +29,9 points within the 5y period, p=.0022). Regarding the physical score of quality of life, physical activity and limitations due to physical state (“role physical”) are improving significantly of 20,3 points (p= .0041) and 26,4 points (p=.0016) respectively.

There was any statistically significant difference between quality of life measures at 2y and 5y, suggesting a long-term persistence of the acquired improvement obtained within the initial 2y stimulation.    

Conclusion : STN DBS in refractory OCD leads to persistent clinical and quality of life improvement, fostering its recognition within the therapeutical algorithms in this disabling medical condition. 

Mircea POLOSAN (Grenoble), Pauline MAZE, Brigitte PIALLAT, Julien BASTIN, Eric SEIGNEURET, Alexandre KRAINIK, Paul KRACK, Stephan CHABARDES
18:30 - 18:35 #26146 - Cellular anatomy of the Sano triangle and therapeutic hypothesis for psychosurgery.
Cellular anatomy of the Sano triangle and therapeutic hypothesis for psychosurgery.

Stereotactic lesions of the Sano triangle were successfully performed to treat severe aggressiveness in the 70’s (Sano et al. 1970). This emotional and vegetative brain area, located within the medial subthalamic area, is still a target for deep brain stimulation in rare patients with pathological aggressiveness (Torres et al. 2020) and for resistant cluster headaches (Nowacki et al. 2019). Here, we report our experience of bilateral Sano triangle lesions in a schizophrenic violent patient with a transient improvement, in whom MRI lesions were precisely located on a normal post-mortem human brain obtained from a body donation studied with 11.7T MRI and immunostained histological sections. Our aim was to identify neuronal populations, axon terminals and fiber bundles involved in the lesions in order to better understand the pathophysiology of pathological aggressiveness. A 40-year-old man had a paranoid schizophrenia with persecutory delusions and hallucinations for 25 years. Often, he experienced unpredictable and major episodes of auto- and hetero-aggressiveness. Since 2011, violent outbursts became uncontrollable requiring permanent hospitalization and physical restraints. Treatment failure was due to his allergic reaction to all neuroleptics with recurrent severe malignant syndromes. Electroconvulsive therapy was ineffective. In 2013, the patient’s family consented to psychosurgery, which was approved by a multidisciplinary team in accordance with French law. Bilateral lesions of the Sano triangle were performed in December 2013. The targets were calculated on the patient’s T1 MRI with our histological and deformable YeB atlas (Bardinet et al. 2009). The coordinates were X = 4, Y = 12 and Z = 3.5 mm with respect to the posterior commissure. Intra-operative electrical stimulation was applied at different depths and the lesion was performed (75°C for 60s) at the site where stimulation induced sympathomimetic signs. After surgery, the patient showed mild improvement with less frequent and less intense violent outbursts that lasted only for 6 weeks. He also gained weight moderately. MRI showed a one-millimeter right lesion and no definitive left lesion. We performed a second surgery using the same targets and technique in July 2014. Immediately after, the patient’s status improved and physical restraints were removed. He interacted better with his family and the medical staff. MRI showed bilateral lesions. After three months, nocturnal aggressiveness returned requiring neuroleptic medication. The patient died from a malignant syndrome in October 2014. To localize precisely the lesions, a multimodal map of the Sano triangle was adapted to the postoperative patient MRI. This map was built from a normal human post-mortem brain, whose region of interest was first scanned at 11.7T (anatomical and diffusion sequences), then serially sliced and immuno-stained. Fusion of the post-mortem (control) and post-operative (patient) data was performed manually using linear transformations. Adjacent series of sections were immunostained with orexin, histamine, dopamine, serotonin transporter and vesicular transporter of glutamate 1. We showed that lesions were well located in the Sano triangle within the medial reticular formation and posterior to the posterior hypothalamus (Fig 1 a. and b.). Within both lesion sites, we observed few dopaminergic neurons and scattered neurons without specific labeling. The lesions sites mostly contain a homogeneous distribution of serotoninergic terminals and numerous glutamatergic terminals with an increasing latero-medial gradient (Fig 1 c. to f.). We also identified some orexinergic terminals originating in the hypothalamus and ending on unlabeled cell bodies, and a small dopaminergic bundle passing through the lateral part of the lesions. Our preliminary results suggest that Sano triangle lesions involve scattered reticular neurons and many diverse axonal endings together with dopaminergic projections to the thalamus. The obtained clinical effect could be explained by the interruption / modulation of cortical, serotoninergic and hypothalamic projections in this complex area. The use of DBS instead of lesioning may have permitted a more sustained improvement by delivering continuous stimulation. A targeting based on deformable histological atlas and individualized tractography should be the next step to improve results in psychosurgery.

Marie Des Neiges SANTIN (STRASBOURG), Marion PLAZE, Chantal FRANCOIS, Christophe DESTRIEUX, Eric BARDINET, Marwan HARIZ, Raphaël GAILLARD, Carine KARACHI
18:35 - 18:40 #26015 - Cocaine-induced changes in dopamine levels in nucleus accumbens as a potential biomarker for drug addiction neuromodulation.
Cocaine-induced changes in dopamine levels in nucleus accumbens as a potential biomarker for drug addiction neuromodulation.


To develop novel treatments, it is imperative to be able to measure the effect of drugs of abuse on neurotransmission. Here we present a novel electrochemical method, known as multiple cyclic square wave voltammetry (M-CSWV), which can accurately measure tonic dopamine levels, with temporo-spatial resolution superior to existing methods. Together with fast-scan cyclic voltammetry (FSCV), we utilized these techniques to elucidate changes in phasic and tonic dopamine at nucleus accumbens core (NAcc), after cocaine administration.



Carbon fiber microelectrode (CFM) and stimulating electrode were implanted into NAcc and medial forebrain bundle (MFB) of Sprague-Dawley rats, respectively. Locations were optimized via evoked phasic response by MFB stimulation. In the first group, the phasic response was measured after a dose of i.v. saline or cocaine hydrochloride (3mg/kg). In the second group, tonic levels were measured using M-CSWV after saline and cocaine.



Both the phasic (n=4) and tonic (n=5) dopamine responses were augmented by cocaine injection. The phasic and tonic levels changed by approximately x2.4 and x1.9, respectively. The minimal disruption/disturbance of neuronal tissue by CFM may explain why the measured baseline tonic values [134±32 nM] were 10-fold higher compared to conventional microdialysis values.



In this study, we elucidated the dopamine dynamics at NAcc with acute cocaine administration. This is the first time this has been explored with such a high temporo-spatial resolution. Our results demonstrated the exciting possibility of M-CSWV as a sensing component of a closed-loop neuromodulation system that could modulate dopamine levels similar to drugs of abuse.

Jason YUEN (Rochester, MN, USA), Abhinav GOYAL, Aaron RUSHEEN, Abbas KOUZANI, Michael BERK, Jee Hyun KIM, Susannah TYE, Charles BLAHA, Dong-Pyo JANG, Kevin BENNET, Hojin SHIN, Yoonbae OH, Kendall LEE
18:40 - 18:45 #23958 - Deep brain stimulation (DBS) of the median forebrain bundle in a rodent model of depression: Interaction between depressive-like phenotypes and estrogen in female rats.
Deep brain stimulation (DBS) of the median forebrain bundle in a rodent model of depression: Interaction between depressive-like phenotypes and estrogen in female rats.

Depression is the leading cause of disability worldwide and a major contributor to the global burden of disease. Women are at a higher risk than men to develop mood disorders and depression. The increased risk is associated with fluctuating estrogen levels that occur during reproductive cycle events. There is compelling scientific evidence indicating the neuromodulatory and neuroprotective effects of estrogen, which are directly relevant to mood symptomatology. Specifically, affective regulation has been linked to neural structures rich in estrogen receptors and estrogenic regulation of neurotransmitters. The limbic system, implicated in depression, is modulated by various neurotransmitters which are influenced by the circulating hormone estrogen. Various studies on deep-brain stimulation (DBS) of the median forebrain bundle (MFB) to treat refractory depression are currently ongoing. The current study is the first in a series addressing the issue of sexual dimorphisms in the Flinders Sensitive Line (FSL) rodent model of depression. The female estrous cycle on anti-depressant effects of DBS and the effect on specific neurotransmitters and hormones are being investigated.

Depressive-like FSLs and non-depressive Sprague Dawley (SD) received 8 days continuous MFB-DBS while monitoring the estrous cycle. Before and after DBS, behavioral comparison was assessed using the forced swim test, the sucrose consumption test and locomotion measurements. Post-mortem expression of neurotransmitters, estrogen and their respective receptors in implicated regions was assessed using qPCR, Western Blot and ELISA.

FSLs exhibited significant differential behaviors compared to controls. After MFB-DBS, a significant reduction in depressive-like phenotypes and a significant increase in serum estradiol concentrations was observed. Significant behavioral changes occurred alongside hormonal fluctuations in the estrous cycle throughout all groups. Behavioral results suggest an impact of estrogen fluctuations on the presence of depressive-like symptoms in both FSL and controls. Further analysis focuses on molecular changes to establish relevant interactions between hormonal cycles and neurotransmitter activity.

Anna TADROS (Freiburg, Germany), Wilf GARDNER, Yixin TONG, Tsvetan SERCHOV, Volker Arnd COENEN, Máté DÖBRÖSSY
18:45 - 18:50 #25945 - Deep Brain Stimulation for psychiatric disorders: long-term surgical management.
Deep Brain Stimulation for psychiatric disorders: long-term surgical management.


Deep Brain Stimulation (DBS) has implemented itself as a hallmark in movement disorder therapy and has been explored for psychiatric disorders in clinical trials as an adjunct treatment. Data on how to surgically manage these patients long after the clinical trial has ended is currently lacking.



A single center database analysis was performed to identify all cases of DBS for psychiatric indications. Epidemiologic data, number and type of follow-up surgeries after initial implantation, rate of complications, success in long-term therapy and documented stimulation parameters were analyzed.



Between 2003 and 2019 n=103 patients were implanted with a DBS system for a psychiatric indication (excluding dementias) with a mean follow-up of 106 months. Mean age was 43.1 years with two thirds being female. Indications were major depression (n=66), bipolar disorder (n=6), obsessive-compulsive disorder (n=6), anorexia nervosa (n=22) and Tourette’s syndrome (n=3). The predominant target structure was the subgenual cingulate gyrus (CG25, 91% for depression, bipolar disorder and anorexia). 48.5% of all patients still had an active DBS system with a mean follow-up of 94 months at the time of the study.  21.4% of patients had the system explanted with lack of efficacy being the most common one (77% of explants). IPG replacements were the most common scheduled surgery with an average of 2.3 replacements per patient. IPGs lasted for an average of 24.0 months with average stimulation parameters of 130Hz, 85µs and 5.3V. N=42 patients were switched to a rechargeable IPG with 24% being switched back to a non-rechargeable IPG later on. 37% of patients had unscheduled surgeries for wound-related complications (15.5% of patients), hardware related issues (10.6%) or suboptimal electrode placement (1.0%).



Patients with DBS for psychiatric disorders represent a separate entity compared to movement disorder patients. The rate of explants and unscheduled surgeries is higher. High stimulation parameters demand frequent IPG replacements generating a considerable rate of wound-related complications. Strategies to reduce the number of IPG replacements (optimization of stimulation parameters, use of rechargeable IPGs) could help to increase the rate of long-term responders in the future. When conceiving trials, strategies on how to enable long-term therapy for these patients should be considered.

Martin JAKOBS (Heidelberg, Germany), David Hernán AGUIRRE-PADILLA, Peter GIACOBBE, Andreas UNTERBERG, Andres LOZANO
18:50 - 18:55 #23953 - Investigation of sleep, behavioural and physiological deficits in the FSL model of depression and the related effects of medial forebrain bundle deep-brain stimulation.
Investigation of sleep, behavioural and physiological deficits in the FSL model of depression and the related effects of medial forebrain bundle deep-brain stimulation.

Depression is a common mental disorder, representing one of the largest health burdens worldwide. Despite its significance, the neurobiology underlying depressive symptoms is poorly understood. There remain no unique physiological biomarkers for the disease, and a significant proportion of patients fail to respond to conventional treatments. Sleep problems are an extremely common symptom in depression, the resolution of which may impact treatment outcomes. Deep Brain Stimulation of the medial forebrain bundle (MFB-DBS) represents a promising treatment for refractory depression. The Flinders Sensitive Line (FSL) rat is an established model for depressive symptoms, several aspects of its phenotype have not been fully examined. In the current study, sleep, behavioral and physiological abnormalities in the FSL are further investigated, to extend characterization of the model. In order to assess the anti-depressant action of MFB-DBS, and to provide insight into potential associated mechanisms of the treatment, the effect of MFB-DBS on various measures was examined in the FSL.

FSL rats and non-depressive Sprague Dawley controls were implanted via stereotactic surgery with electrodes for DBS in the MFB, and for electrophysiological recording in the prefrontal cortex, nucleus accumbens and CA1 hippocampus. Assessments were conducted pre- and post- 24-hour DBS. Behavioural phenotypes were measured by the forced swim and sucrose consumption tests. Post-mortem, tissue was collected and processed for analysis using in situ hybridization, HPLC, qPCR and western blotting.

FSLs exhibited previously unreported changes in slow-wave activity and oscillatory activity during sleep, and depressive-like behaviors including anhedonia. MFB-DBS improved behavioral phenotypes, alongside physiological changes including to pre-frontal extracellular monoamine levels and the expression of receptors of the dopaminergic system. Despite other observed anti-depressant effects, measures of sleep were unaffected.

The presence of previously unreported slow-wave sleep deficits and anhedonic-like phenotypes enhance the validity of the FSL as a model of depression. Improvement of behavioral phenotypes provides evidence of the anti-depressant effect of MFB-DBS, while insight into the biological substrate of these effects, potentially mediated via alterations to dopaminergic functioning, is provided by physiological changes. The lack of influence on sleep symptoms presents a challenge and potential clinical limitation of MFB-DBS, which must be addressed in future clinical and pre-clinical research.

Wilf GARDNER (Strasbourg), Laura DURIEUX, Anna TADROS, Fanny FUCHS, Tsvetan SERCHOV, Chantal MATHIS, Volker Arnd COENEN, Máté DÖBRÖSSY, Lucas LECOURTIER
18:55 - 19:00 #23912 - Longterm deep brain stimulation in treatment-resistant obsessive-compulsive disorder: outcome and quality of life at 4- 8 years follow-up.
Longterm deep brain stimulation in treatment-resistant obsessive-compulsive disorder: outcome and quality of life at 4- 8 years follow-up.

BACKGROUND: Obsessive compulsive disorder (OCD) is a severe disabling disease, and around 10% of patients are considered to be treatment-resistant (tr) in spite of guideline based therapy. Deep brain stimulation (DBS) has been proposed as a promising treatment for patients with trOCD. However, the optimal site for stimulation is still a matter of debate, and clinical longterm follow-up observations including data on quality of life are sparse. We here present six trOCD patients who underwent DBS with electrodes placed in the bed nucleus of the stria terminalis/ anterior limb of the internal capsule (BNST/ALIC), followed for 4-8 years after lead implantation.

METHODS: In this prospective observational study, six patients (four men, two women) aged 32-51 years and suffering from severe to extreme trOCD underwent DBS of the BNST/ALIC. Symptom severity was assessed using the Yale-Brown Obsessive Compulsive Scale (Y-BOCS), and quality of life using the World Health Organization Quality of Life assessment scale (WHO-QoL BREF). Follow-up was obtained at least for 4 years in all patients.

RESULTS: With chronic DBS for 4 - 8 years 4/6 patients had sustained improvement. Two patients remitted, and two patients responded (defined as > 35% symptom reduction), while the other two patients were considered non-responders on longterm. Quality of life markedly improved in remitters and responders. We did not observe periinterventional side effects or adverse effects of chronic stimulation.

CONCLUSIONS: Chronic DBS of BNST/ALIC provides longterm benefit up to 4 - 8 years in trOCD, although not all patients take profit. Quality of life improves in DBS responders, documented by improved QoL scores and, even more important, by regaining of autonomy and improving psychosocial functioning.


Assel SARYYEVA, Lotta WINTER, Kerstin SCHWABE, Hans.e HEISSLER, Joachim RUNGE (Hannover, Germany), Mesbach ALAM, Ivo-Aleksander HEITLAND, Kai KAHL, Joachim K. KRAUSS
19:00 - 19:05 #23954 - Medial forebrain bundle DBS differentially modulates dopamine release in the nucleus accumbens in a rodent model of depression.
Medial forebrain bundle DBS differentially modulates dopamine release in the nucleus accumbens in a rodent model of depression.

Medial forebrain bundle (MFB) deep brain stimulation (DBS) has anti-depressant effects clinically and in depression models. Currently, therapeutic mechanisms of MFB DBS or how stimulation parameters acutely impact neurotransmitter release, particularly dopamine, are unknown. Experimentally, MFB DBS has been shown to evoke dopamine response in healthy controls, but not yet in a rodent model of depression.

The study investigated the impact of clinically used stimulation parameters on the dopamine induced response in a validated rodent depression model and in healthy controls. The stimulation-induced dopamine response in Flinders Sensitive Line (FSL, n = 6) rat model of depression was compared with Sprague Dawley (SD, n = 6) rats following MFB DSB, using Fast Scan Cyclic Voltammetry to assess the induced response in the nucleus accumbens. Stimulation parameters were 130 Hz (“clinically” relevant) with pulse widths between 100 and 350 μs. Linear mixed model analysis showed significant impact in both models following MFB DBS both at 130 and 60 Hz with 100 μs pulse width in inducing dopamine response. Furthermore, at 130 Hz the evoked dopamine responses were different across the groups at the different pulse widths.

The differential impact of MFB DBS on the induced dopamine response, including different response patterns at given pulse widths, is suggestive of physiological and anatomical divergence in the MFB in the pathological and healthy state. Studying how varying stimulation parameters affect the physiological outcome will promote a better understanding of the biological substrate of the disease and the possible anti-depressant mechanisms at play in clinical MFB DBS.

Danesh Vajari ASHOURI (Freiburg, Germany), Chockalingam RAMANATHAN, Yixin TONG, Stieglitz THOMAS, Volker Arnd COENEN, Máté DÖBRÖSSY

Introduction: A subgroup of patients with autism spectrum disorder (ASD) show self or heteroaggression, dyscontrol episodes, and others of obsessive-compulsive profile (OCD); some of them are resistant to medical and behavioural treatment. We describe the long-term outcome in a group of these patients, treated with radiofrequency brain lesions or combined stereotactic surgery and Gamma Knife (GK) radiosurgery.

Methods: We reviewed the medical records of ten ASD patients with pathological aggressiveness and OCD, who had undergone radiofrequency lesions and/or radiosurgery with GK in our institution.

Results: The ten patients had a significant reduction of their symptoms (PCQ 39.8 and 33.9, OAS 11.8 and 5.4, YBOCS 31.3 and 20.8, preoperatively and in the last follow-up, respectively; p<0.005 in all cases), although all but two needed more than one treatment to maintain this improvement.

Conclusions: We observed a marked improvement in behavior, quality of life and relationship with the environment in all our ten patients after the lesioning treatments, without long-lasting side effects.


Cristina TORRES (Madrid, Spain), Nuria MARTINEZ, Monica LARA, Marcos RÍOS-LAGO
Espace Vieux-Port

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
17:00 - 18:05

Parallel Session 8

Moderators: Sophie COLNAT-COULBOIS (PU-PH) (Nancy, France), Martin JAKOBS (Consultant) (Heidelberg, Germany), Krassimir MINKIN (Head of Center of Functional Neurosrgery) (Sofia, Bulgaria)
17:00 - 17:10 #23516 - sEEG mapping of basal temporal language area predicts postoperative language outcome.
sEEG mapping of basal temporal language area predicts postoperative language outcome.

sEEG mapping of basal temporal language area predicts postoperative language outcome 


To evaluate early and late post-operative naming outcome according to the resection status of the Basal Temporal Language Area (BTLA) identified by pre-operative cortical stimulations during Stereo-Electro-Encephalography (SEEG) in patients with intractable temporal lobe epilepsy.


Twenty patients who underwent SEEG for drug-resistant temporal lobe epilepsy met the inclusion criteria. During language mapping, a positive site was considered when stimulation of 2 contiguous contacts elicited at least one naming impairment during 2 remote sessions. After temporal lobe resection (TLR) ipsilateral to their BTLA, patients were classified as BTLA+ when at least one positive language-site was resected and as BTLA- when positive language sites were preserved. Outcomes in naming and verbal fluency tests were assessed using pre- postoperative (7 and 24 months after surgery) score changes and reliable change indices for clinically meaningful changes. Naming decline predictors were finally investigated using binary regression.


BTLA+ patients had significant greater naming score change compared to BTLA- patients. However, this difference was found only 6 months postoperatively. No difference in verbal fluency tests was observed. When RCI was used, 25% of patients had naming decline 6 months postoperatively (80% of them were BTLA+). A significant correlation was found between resection of the BTLA and this naming decline. In particular, a rate superior to 15% of resected positive language-sites predicted this decline.


Resection of the BTLA is associated with a specific and early naming decline. Even if this decline is transient, BTLA+ patients tend to keep lower naming scores compared to their baseline. SEEG mapping might help in the prediction of postoperative language outcome in dominant TLR. Removing visual naming sites identified by SEEG language mapping is possible but should be considered with caution.


Chifaou ABDALLAH, Helene BRISSART, Ludovic PIERSON, Olivier ARON, Natacha FORTHOFFER, Jean Pierre VIGNAL, Louise TYVAERT, Jacques JONAS, Louis MAILLARD, Sophie COLNAT-COULBOIS (Nancy)
17:10 - 17:20 #23666 - Distant Cortical Abnormalities in Epileptic Hypothalamic Hamartoma Patients.
Distant Cortical Abnormalities in Epileptic Hypothalamic Hamartoma Patients.

Background: We hypothesize that developmental or acquired cortical abnormalities could explain failure of surgery aiming at hypothalamic hamartomas (HH).

Objective: Our aim was to do quantitative surface-based MR study to detect distant cortical anomalies in HH epilepsy patients treated by Gamma Knife Surgery (GKS) and correlate it with the GKS response.

Methods: Forty-two HH epileptic patients were compared for sulcus-specific morphometry difference with closely matched controls. All patients weretreated by GKS and followed for at least 3 years with high quality MR studies before and after GKS. We used“surface-based”quantitative MR analysis using Brainvista/Morphologist pipeline (sulcal root/meridian parallel model).

Results (Key Findings):Non-responders comparison to control displayed bilateral multilobar cortical thinning in “pre-central” (marginal) and “frontal” (median and superior), and “ascending ramus” of Sylvian and unilateral thinning in left “insula”, and “pre-central” (superior).More specific thinning was observed in “collateral” (r>l) and “pre-central” (marginal) sulci (l) in non-responders than in responders. No thinning was found in the responders’ group. Pre-motor areas “pre-central” and limbic “collateral” sulcus (l) thinning was the most response-related significant and frequent cortical anomalies.

Conclusion: This is the first quantitative MR study in HH epilepsy patients, which is cross-sectional and controlled.We found distant cortical abnormalities related to the epilepsy response after GKS. Diffuse multilobar thinning was observed in non-responders HH in bilateral “pre-central” (marginal), “frontal” (median and superior), and “ascending ramus” of Sylvian and unilateral thinning in left “insula”. Pre-motor areas “pre-central” and limbic “collateral” sulcus (l) thinning was the most response-related significant and frequent cortical anomalies. There are different patterns of distant cortical abnormalities between GKS responders and non-responders suggesting different epilepsy networks involved in each group.

Hussein HAMDI ABOUELGHEIT (Marseille), Guillaume AUZIAS, Olivier COULON, Nadine GIRARD, Fabrice BARTOLOMEI, Jean REGIS
17:20 - 17:30 #23962 - Long term follow-up in VNS for resistant epilepsy.
Long term follow-up in VNS for resistant epilepsy.

Long term follow-up in VNS for resistant epilepsy


Freri E^, Marotta G ^, Porto E*, Tringali G*, Visani E°, Casazza M.°

Department of Neurosurgery*, Neuropediatrics^, Epileptology°

Istituto Neurologico C.Besta, IRCCS, Milan, Italy


          Speaker: Casazza M.

          Topic: Epilepsy

          Key words: VNS, pharmacoresistant epilepsy, long term follow-up


Vagal nerve stimulation (VNS) is a largely used palliative technique for treating pharmacoresistant focal epilepsy in adults and children. Reported results are encouraging, with a reduction  of seizure frequency higher than 50% in more than half of patients. Side effects are usually well tolerated. Unfortunately no data indicate in which seizure types and syndromes VNS is more effective.

Our results are worse than those published, but our follow-up is very long, more than 15 yrs for 12 patients. Mean follow up lasts 11.6 yrs (range 2-24): 8  died during follow-up, in 3 cases related to epilepsy.

We present a series of 61 patients (32 males and 29 females, affected by pharmacoresistant focal epilepsy, implanted with VNS at a mean age of 30.1years (range 3 to 62).

For four patients we miss many data, so we consider only 57 of the implanted patients of our Institute.

All patients had severe, long lasting epilepsy (mean disease duration 24.6 yrs, range 3-51), in 40 out of 57 cases associated with cognitive impairment, in 26 with neurological deficits.

Epilepsy had unknown etiology in 17 cases, was associated with cortical malformations (10), progressive genetically determined encephalopathy (8), perinatal hypoxia (6), vascular malformations (4), inflammatory diseases (7), tuberous sclerosis (2), other causes (3). 

Seizures had focal onset, mostly with impaired awareness, in 40 patients they determined falls, usually tonic, and often with trauma.

Seizure frequency was daily or more in 68 % of cases. Patients were all treated polypharmacologically: 10 with 2, the other with 3 or more AEDs.

Seizures were recorded in most patients (50); in 8 cases a focal onset was not recognizable. In 28 patients recurrent focal stati or seizure clusters were present.

Interictal EEG was focal or multifocal in all patients, with rapid synchronism in 25.

Seventeen patients were excluded from resective epilepsy surgery , 3 refused it, 11 were operated on before implantation of VNS without results on seizure frequency.

After VNS, one patient underwent hemispherotomy, one anterior callosotomy, one deep brain stimulation and one coagulation of nodular periventricular heterotopia.

Seizure frequency was unchanged in 16 patients, in 21 cases seizures were reduced more than 50% and in 21 of them we observed a marked reduction of falls.

AEDs were added or modified during the years of VNS. Therefore we waited for some months after VNS depletion to verify seizure frequency. If there was an increase in number, VNS was replaced. This happened in 24 patients. In 15 patients no seizure frequency increase was observed after second VNS depletion, so that they were not reimplanted. In 9 patients a third implant was made for marked increase of seizures.

Side effects include stimulation adverse effects, as hoarseness, voice change and cough, differently present in most patients. Surgical side effects are rare, including infections (2). In 3 patients we observed electrode rupture, probably due to fall at least in one case: they requested a reimplantation.

We project to better analyze and discuss the reasons for the bad outcome of our patients at a long distance focusing on seizure type and epileptic syndrome, in order to try the identification of better candidates to VNS.

What do you want to do ?
New mailCopy

Elena FRERI, Marina CASAZZA, Guia MAROTTA, Edoardo PORTO (Milan, Italy), Giovanni Umberto TRINGALI, Elisa VISANI
17:30 - 17:35 #23917 - Awake epilepsy surgery.
Awake epilepsy surgery.

Awake epilepsy surgery

Minkin Krasimir, Gabrovski Kaloyan, Karazapryanov Petar, Dimova Petya, Karakostov Vasil



Awake craniotomy (AC) and direct electrical stimulation (DES) emerged together with epilepsy surgery more than 80 years ago in the seminal work of Wilder Penfield. In recent years AC gained important role in glioma surgery, but it is rarely used and reported in modern epilepsy surgery. Contemporary epilepsy surgeons rely more on functional MRI, transcranial magnetic stimulations and extraoperative brain mapping through invasive EEG than on awake surgery. The goal of our study was to investigate a series of awake epilepsy surgeries in patients investigated and operated on in an epilepsy center.



Material and Methods:

Our clinical material included 23 patients operated on for drug-resistant epilepsy using asleep-awake-asleep technique of AC with DES during a 10-year period. We investigate the following variables: age distribution, indications, brain mapping success rate, epilepsy surgery success rate, intraoperative complications, postoperative complications and intraoperative change of the preoperative resection plan according to the information obtained during the brain mapping using DES.


The mean age at surgery was 29 years (16- 47 years) with 4 patients under 18 years of age. The epileptogenic zone was localized in the frontal lobe in 12 patients, in the temporal lobe in 7 patients and in the frontotemporoinsullar region in 2 patients. The remaining 2 patients has had epileptogenic zones in the supramarginal gyrus and in the temporoparietoocipital region. Surgery was in the left hemisphere in all but 2 cases. The most frequent indication for AC was language mapping – 19 patients ( 83%). The most frequent histopathological findings were focal cortical dysplasia (14 patients) and ischemic or posttraumatic  gliosis (6 patients). We achieved satisfactory functional mapping in all patients. Intraoperative seizures were observed in 6 patients (23%) without further intraoperative complications and permanent loss of cooperativeness. Transient mild neurological deficit was observed in 5 patients. There were no persistent neurological deficits, hematomas or infections. Our preoperative plan based on preoperative fMRI was changed because of functional constraints in 8 patients (35%). The most striking finding was the localization of eloquent speech areas in the most anterior part of the superior temporal gyrus.


Awake craniotomy provides additional functional information which may change the preoperative plan based on preoperative fMRI, lower the incidence of persistent functional impairment, making it a useful but forgotten tool in epilepsy surgery.

Krasimir MINKIN (Sofia, Bulgaria), Kaloyan GABROVSKI, Petar KARAZAPRYANOV, Petya DIMOVA, Vasil KARAKOSTOV
17:40 - 17:45 #23972 - Single-center experience with frameless personalized stereotaxy for SEEG.
Single-center experience with frameless personalized stereotaxy for SEEG.

Drug resistant epileptic patients are a complex challenge for modern functional neurosurgeons. The optimal resection limits are planed on more criteria, including Stereo EEG recordings and analysis. StarFix microTargeting Platform (FHC Inc., Bowdoin, ME, USA),  is a frameless patient customized device. It is a lightweight fixture, incorporates guides aligned with all electrode trajectories, simplifying the entire surgical workflow.


Patients and method

A total of 23 patients were implanted. Age was ranging 3 to 46 yrs. All patients were on general anesthesia during procedure. Targets were placed in both hemispheres in different structures of the brain.  A vascular safety index, characterizing the proximity of the planned trajectories to the blood vessels was calculated for each trajectory. Number of implanted electrodes (DIXI Medical, Chaudefontaine France), varied between 8 to 20 and the number of contacts 92 to 258. The procedure will be presented in detail during presentation. A post-implantation CT was performed to check for the accuracy of electrode positioning and for possible complications.


All  electrodes reached their intended targets. There were no intracranial hemorrhages or other implantation-related complications. The implantation errors were less then2 mm. Time in the OR was represented by the average time per implanted electrode which was about 7 minutes; this value is shorter than 21 minutes reported for past frame-based implantation procedures. The patients underwent a  3 to 14-days monitoring sessions, during which we managed to record multiple habitual seizures.  No periprocedural adverse events were recorded.  All patients tolerated the implanted electrodes well. The patients who underwent surgery, with the resection of the epileptogenic zone as discussed in the multidisciplinary meeting, presented a good outcome, without any discernable neurological postoperative deficits. The histopathological analysis of the resected tissue showed different types of pathology which will be detailed. All patients presented Engel IA except one who was III.


StarFix microTargeting Platform have several advantages over classical, frame-based systems for epilepsy surgery patients. It is easy to use, light weight and does need frame reconfigure  for each electrode trajectory, excluding any possibility of human error. The total OR time is shorter then with metallic frame. StarFix greatly simplifies the implantation workflow. The accuracy of electrode positioning is comparable with the other methods. The flexibility of intraoperative trajectory adjustments available in classical frame is solved by planning additional backup trajectories in case an unexpected event prevents an electrode from being implanted. The anchors are not removed until the end of the SEEG implantation, to allow repositioning of the electrodes following the postoperative CT scan, in case they have not reached their intended targets due to incorrect depth setting or excessive curvature. Re-attaching the platform can be performed in a matter of minutes without any additional CT scan. Another advantage of StarFix is for young patients presenting a thinner cranium, there is an added risk of skull fracture, especially when using a general-purpose, rather heavy, metal frame.



Patient-customized stereotactic fixtures that scale to the patient’s anatomy is a safe and accurate option for SEEG exploration in subjects diagnosed with drug-resistant epilepsy.






Andrei BARBORICA, Jean CIUREA, Ioana MANDRUTZA, Rasina ALIN, Tatiana CIUREA, Costi PISTOL, Felix BREHAR (Bucharest, Romania)
17:45 - 17:50 #26142 - Reduction of penicillin-induced seizure in a non-human primate mesio- temporal lobe epilepsy model by deep brain cooling.
Reduction of penicillin-induced seizure in a non-human primate mesio- temporal lobe epilepsy model by deep brain cooling.


Introduction: It is estimated that 1% of the world population suffers from epilepsy and 30% of epileptic patients are resistant to all pharmacological therapies. For medication resistant patients that are not candidates for resective or ablative procedures, there is a great need of far better therapies that could lead to high rates of patients becoming free of seizures without side effects. In order to help this group of patients, we propose to develop an innovative implantable medical device that allows the cooling of deep epileptogenic area. Cooling effect on epilepsy attenuation is known since decades. Rothman et al. [1]  have shown suppression of epileptic discharges when rapid cooling is applied in brain neocortical surface.  Nevertheless, no antiepileptic medical device based on cooling for deep-seated areas is currently available. There is real technical difficulty in transporting the cooling effect to deep-seated areas, so, it has been difficult to study the effect of cooling at the hippocampal level.  Here we present the preliminary results of a prototype of implantable device capable to deliver the cooling effect to hippocampal areas in mesial temporal lobe seizures epilepsy (MTLE) in non-human primates model².Methods: Study was performed in a female Macaca fascicularis. The animal was implanted unilaterally with a deep cooling lead   and sEEG electrodes in the hippocampus. After post-operative recovery, penicillin was injected (10 min at 2µl/min, 1000UI/ml) into the hippocampus and animal was recorded during 5-7 hours periods. After a 45 min-period of seizure stabilization, focal cooling was applied and temperature and seizures frequency and duration were monitored. Injections were repeated once a week over a period of 203 days. To evaluate cooling safety and   characterize the changes occurring within the hippocampus, we performed a histological analysis, including neuronal nuclei and glial fibrillary acid protein immunostaining. Results: After each penicillin injection, we observed that seizure characteristics were reproducible between experiments as already reported by our group².Seizures duration was stable (58.5± 11 s) and the frequency of seizures reached a plateau with at least two seizures each 20 min per trial. MLTE seizures were reproduced similar to the ones observed in similar previous experiments and patients[2]. Sixteen trials were analyzed and seizures were detected by visual analysis of SEEG allowing electro-clinical correlation. We divided the trials in cooling sessions at 20°C, 24°C and no cooling (control). A reduction in the number of hippocampal seizures was seen at temperatures of 20°C when compared with control (32-34°C). There was no change in seizure duration during focal cooling.  Hippocampal sclerosis similar to that encountered in epileptic patients was found at the end of the study after 16 penicillin injections without additional damage.  Conclusions: Our results suggest for the first time, that focal cooling in deep areas of the brain can reduce the number of focal hippocampal seizures induced by penicillin. Deep brain cooling could be an alternative for control of focal seizures, lowering the risk of irreversible functional loss of the resective surgery and with potentially better efficacy than neuromodulation. The study suggest that this approach could be a safe and effective alternative for drug resistance epilepsy.

[1] Rothman SM., Smyth MD., Yang X-F., Peterson GP. Focal cooling for epilepsy: An alternative therapy that might actually work. Epilepsy Behav. 2005; 7: 214–221.

[2] Sherdil A, Chabardès S, Guillemain I, Michallat S, Prabhu S, Pernet-Gallay K, et al. An on demand macaque model of mesial temporal lobe seizures induced by unilateral intra hippocampal injection of penicillin. Epilepsy Res. 2018; 142: 20–28.


Napoleon TORRES (GRENOBLE), Quentin BORNTRAGER, Nicolas AUBERT, Fabien SAUTER, Claude CHABROL, David RATEL, Jenny MOLET, Brigitte PIALLAT, Stephan CHABARDES
Salle 120

Thursday 09 September

Added to your list of favorites
Deleted from your list of favorites
18:05 - 19:05

Parallel Session 10
Flash - Movement disorders

Moderators: Harith AKRAM (Associate Professor) (London, United Kingdom), Carine KARACHI (MEDECIN) (PARIS, France), Oystein TVEITEN (Neurosurgeon) (Bergen, Norway)
18:05 - 18:15 #23359 - Long-term motor function and quality of life outcomes from a prospective, international DBS registry.
Long-term motor function and quality of life outcomes from a prospective, international DBS registry.

Objective: Here we report the collected outcomes from a large-scale registry of a Deep Brain Stimulation (DBS) system capable of Multiple Independent Current Source Control (MICC) in the management of symptoms of levodopa-responsive Parkinson's disease (PD).

Background: The effectiveness of Deep Brain Stimulation (DBS) for reducing motor complications of Parkinson's disease (PD) has been substantiated by randomized controlled trials (Schuepbach et al., 2013). Additionally, motor improvement is sustained for up to 10 years (Deuschl et al. 2013). Large patient data registries may facilitate insights regarding real world, clinical use of DBS. Furthermore, no registry database currently exists for a multiple-source, constant current DBS system.

Methods: The Vercise DBS Registry is a prospective, on-label, multi-center, international registry sponsored by Boston Scientific Corporation. The Vercise DBS system (Boston Scientific) is a multiple-source, constant-current system. Subjects were followed up to 3 years post-implantation where their overall improvement in quality of life and PD motor symptoms was evaluated. Clinical endpoints evaluated at baseline and during study follow included Unified Parkinson's disease Rating Scale (UPDRS), MDS-UPDRS, Parkinson's disease Questionnaire (PDQ-39), and Global Impression of Change. Subjects underwent either sleep or awake DBS implantation procedures. 

Results: To date, 822 patients have been enrolled (752 implanted). Improvement in quality of life (QoL), as assessed by PDQ-39,demonstrated improvement following implant at 6-months (-6.1-point change, p<0.0001) and up to 2-years (-2.4-point change) as compared with Baseline. Higher improvement in QoL (-15.7-point change) was noted in patients with worse QoLat Baseline (PDQ-39 SI >45). This trend was also noted in subjects with worse disease state at Baseline (Hoehn & Yahr ≥3) who reported a greater improvement in PDQ-39 summary index. At 1-year post-implant (n=272), a 32% improvement in MDS-UPDRS III scores (stim on/meds off) compared with baseline was reported and sustained up to 2-years (n=51). Stable neuropsychometric status (BDI-II, MoCA) was also reported. The safety profile was comparable to other published reports. Additional data collection and analysis is ongoing and will be presented. Data from those undergoing sleep versus awake DBS-implantation procedures will be presented. 

Conclusions: This DBS registry represents the first comprehensive, large scale collection of real-world outcomes and evaluation of safety and effectiveness of a multiple-source, constant-current DBS system.

Jan VESPER (DUSSELDORF, Germany), Roshini JAIN, Heleen SCHOLTES, Alex WANG, Michael T. BARBE, Steffen PASCHEN, Jens VOLKMANN, Chong-Sik LEE, Andrea KÜHN, Monika PÖTTER-NERGER, Günther DEUSCHL
18:15 - 18:25 #23412 - Directional versus Omnidirectional Deep Brain Stimulation: Results of a Multicenter Prospective Blinded Crossover Study.
Directional versus Omnidirectional Deep Brain Stimulation: Results of a Multicenter Prospective Blinded Crossover Study.

Introduction: Published reports on directional DBS have been limited to small single-center investigations. Therapeutic window (TW) has been introduced in DBS to describe the range of stimulation amplitudes achieving symptom relief without side effects. The PROGRESS study evaluated whether directional DBS provides a wider TW in a large prospective trial.

Methods: Participants receiving STN DBS for Parkinson’s disease were programmed with omnidirectional stimulation for 3 months, followed by directional stimulation for 3 months. The subject was blinded to stimulation type and a blinded evaluator assessed TW and motor symptoms. The primary endpoint was based on blinded off-medication evaluation of TW for directional vs. conventional stimulation at 3 months. Additional endpoints at 3, 6 and 12 months included adverse events, subject and clinician stimulation preference, therapeutic current strength (TCS), medication reduction, quality of life and UPDRS part III motor score.

Results: A directional DBS system was implanted in 234 subjects (62±8 years, 33% female). At 3 months, TW was wider using directional stimulation in 90.6% of subjects, satisfying the primary endpoint for superiority (p<0.001). The mean increase in TW with directional stimulation was 41% (2.98±1.38mA, compared to 2.11±1.33mA for omnidirectional, p<0.001). UPDRS part III motor score on medication was improved with either stimulation at each time point (p<0.001). After 6 months, 53% of subjects blinded to stimulation type (102/193) preferred the period with directional stimulation, 26% (50/193) preferred the omnidirectional period and 21% (41/193) had no preference. The directional period was preferred by 59% of clinicians (113/193) vs. 21% (41/193) who preferred the omnidirectional period. Additional results including 12-month data will be available.

Conclusion: A double-blind randomized comparison of directional and omnidirectional stimulation found that 90.6% of subjects had a wider TW using directional stimulation at 3 months and 89.3% at 12 months. There were improvements in the minimum amplitude required to achieve therapeutic benefit and the side effect threshold, leading to a 40% wider TW at 3 months and 32% at 12 months. For the first time, we demonstrated superiority of TW for directional stimulation over omnidirectional stimulation with sustained UPDRS III motor scores and quality of life improvements.

18:25 - 18:30 #23360 - Real-world clinical outcomes using a novel directional lead from a multicenter registry of deep brain stimulation for Parkinson's disease.
Real-world clinical outcomes using a novel directional lead from a multicenter registry of deep brain stimulation for Parkinson's disease.

Objective: In this report, initial real-world outcomes using a directional lead with a Deep Brain Stimulation (DBS) system capable of multiple independent current source control (MICC) for use in managing symptoms of levodopa-responsive Parkinson's disease (PD) are reported.

Background: Early Deep Brain Stimulation (DBS) systems used ring-shaped electrodes to achieve axial selectivity in stimulation of targettissue. However, directional current steering allows for rotational selectivity (in addition to axial) and has the potential tofurther improve patient outcomes by avoiding off-target stimulation due to the ability to create a well-defined field around the intended target. Several pilot studies have corroborated the use of directionality and its impact on therapeutic window and adverse effects.

Methods: The Vercise DBS Registry ( Identifier: NCT02071134) is a prospective, on-label, multi-center, international registry sponsored by Boston Scientific. Subjects were implanted with a directional lead included as part of a multiplesource, constant-current directional DBS system (Vercise Cartesia, Boston Scientific). Subjects were followed up to 3-years post-implantation where their overall improvement in quality of life and PD motor symptoms was evaluated. Clinical endpoints evaluated at baseline and during study follow-up included Unified Parkinson's disease Rating Scale (UPDRS), MDS-UPDRS, Parkinson's disease Questionnaire (PDQ-39), and Global Impression of Change.

Results: As of February 2021, 627 patients (mean age: 60.9 years, 68.4% male) included were implanted with a DBS directionallead. Improved Quality of Life, as assessed by PDQ-39 (p<0.001) following implant was noted up to 1-year post-implant(n=368). Improvements in motor function (change in MDS-UPDRS III scores-meds off condition) versus baseline were alsonoted (31% at 1-year (n=200), p<0.001). Over 80% of subjects, physicians reported an improvement in PD symptoms during long-term follow-up. Additional data collection and analysis is ongoing and will be presented.

Conclusions: This on-going registry represents the first comprehensive, large scale collection of real-world, long-term outcomes using a directional lead and an MICC-based DBS system. Using directional stimulation, it may be possible to achieve a bigger therapeutic window, thereby facilitating enhanced programming flexibility when optimizing for efficacy, while decreasing the likelihood of surpassing the adverse effect threshold.

Jan VESPER (DUSSELDORF, Germany), Roshini JAIN, Heleen SCHOLTES, Alex WANG, Michael T. BARBE, Jens VOLKMANN, Steffen PASCHEN, Chong-Sik LEE, Andrea KÜHN, Monika PÖTTER-NERGER, Günther DEUSCHL
18:30 - 18:35 #23392 - Directional DBS leads implanted under general anesthesia vs. sedation: Findings from an international prospective study.
Directional DBS leads implanted under general anesthesia vs. sedation: Findings from an international prospective study.

Introduction: Traditionally, deep brain stimulation (DBS) leads are implanted under local anesthesia or conscious sedation to allow intraoperative testing for effects (awake procedure). There are also two distinct approaches to perform lead implantation under general anesthesia with intubation (asleep procedure): 1) using intraoperative image-guided targeting only and 2) reducing anesthesia to allow for microelectrode recording.

Methods: A total of 234 subjects were enrolled in PROGRESS between January 2017 and January 2019 in 37 different sites located in Europe, North America and Australia. PROGRESS compared therapeutic window (TW), the difference in amplitude between side effect threshold and minimum therapeutic current, for directional vs. omnidirectional stimulation in STN DBS for Parkinson’s disease. This post-hoc analysis compares therapeutic window and UPDRS motor scores at 3 and 6 months in subjects with awake vs. asleep procedures. Both approaches of intraoperative image-guided targeting and microelectrode recording were integrated in the asleep group for this analysis.

Results: The PROGRESS met its primary endpoint of superiority with 90.6% of subjects (183/202) having a wider TW using directional stimulation. Leads were implanted in an awake procedure for 163 subjects and asleep procedure for 69 subjects. In Europe, 70% of subjects (90/129) were implanted with an awake procedure; in the United States, 73% (64/88) and in Australia, 60% (9/15). Three months after initial programming, directional stimulation increased TW by 0.84 mA for the awake implants (2.18±1.36 mA for omnidirectional stimulation; 3.02±1.30 mA for directional; p<0.001)) and 0.88 mA for the asleep implants (1.92±1.22 mA for omnidirectional stimulation; 2.83±1.53 mA for directional; p<0.001). The increase of TW between awake and asleep procedure was not statistically significant (p=0.57). At baseline, off medication UPDRS scores were 34.6±12.6 for awake subjects and 31.9±12.5 for asleep subjects. At the 6-month follow-up visit, on-medication UPDRS III motor scores improved by 44% in awake subjects (from 31.6±14.0 with DBS off to 17.7±10.4 with DBS on) and 42% in asleep subjects (29.2±13.5 to 16.8±8.5).

Conclusion: In a large international study, directional stimulation was associated with a similar increase in therapeutic window, regardless if the DBS system was implanted using an asleep versus awake procedure. UPDRS motor score improved similarly in both groups.

18:35 - 18:40 #23544 - Primary Cell IPG survival analysis and modelling in Directional Deep Brain Stimulation.
Primary Cell IPG survival analysis and modelling in Directional Deep Brain Stimulation.

Objective: To use explant data from a single center to generate survival plot and model survival curves for Abbott SJM Infinity 7 primary cell IPG.

Background: Primary cell IPGs have distinct advantages in patients with Movement or Neuropsychiatric disorders. However, shortened survival times, less than manufacturer guidance, have been reported for at least one primary cell IPG model. Modelling based on real world explant rates can support manufacturer guidance on IPG lifespan for new primary cell DBS systems.

Methods:  Case records for all Abbott SJM Infinity 7 IPGs implanted prior to 06/2019, at a single expert center, were reviewed. Implanted targets include STN, VIM and GPi, for Dystonia, Tremor and Parkinson’s Disease. Explants due to IPGs reaching end of service were included. Explants due to other causes, infection, or medical complications requiring explant were excluded. A Kaplan-Meier plot of time to explant (IPG survival) was done. Additionally, modelling fitting was done using JMP™ Life Distribution platform. The most conservative distribution model for survival rate is reported.

Results: 105 Infinity 7 IPGs were implanted between 04/2016 and 05/2019. 3 IPGs were replaced due to IPG depletion and included in the analysis. IPGs replaced due to failure other than IPG depletion (known CAPA) (n=2), due to pocket site infection (n=2) and 1 system explant for medical reasons were excluded. Maximum duration of implant at time of analysis was 3 years 10 months. Due to the low number of explants within the duration of follow up, KM survival plot is unable to define mean survival times. The SEV distribution model (BIC=38) estimates mean survival time of 5.201 years (SE±1.0), 95% CI [3.1, 7.2](most conservative estimate at time of analysis). 

Conclusions: With advancements in technology, programming tools and programming technique, real world evidence is necessary to monitor lifespan of primary cell DBS systems. For newer DBS systems, modelling based on real world explant data can provide additional confidence to manufacturer guidance on estimated IPG lifespan. Data and analysis will be updated prior to poster presentation to include validation of best model fit.

Philipp SLOTTY, Binith CHEERAN, Phyllis Sarah MCPHILLIPS, Jan VESPER (DUSSELDORF, Germany)
18:40 - 18:45 #23634 - Adaptive DBS Algorithm for Personalized Therapy in Parkinson’s Disease: ADAPT-PD Trial: a prospective single-blind, randomized crossover, multi-center trial of deep brain stimulation adaptive algorithms in subjects with Parkinson’s disease.
Adaptive DBS Algorithm for Personalized Therapy in Parkinson’s Disease: ADAPT-PD Trial: a prospective single-blind, randomized crossover, multi-center trial of deep brain stimulation adaptive algorithms in subjects with Parkinson’s disease.

Objective: To demonstrate safety and effectiveness of adaptive deep brain stimulation (aDBS) algorithms in subjects with Parkinson’s disease (PD).

Background: DBS is an effective therapy for PD symptoms, though opportunities exist to improve the efficiency and efficacy. Commercially approved DBS is programmed to run continuously (cDBS) at specified programming parameters. In contrast, adaptive DBS (aDBS) algorithms may individualize and optimize PD therapy by adjusting stimulation based on objective signals. The algorithm technology used in this study is uniquely embedded in the device, which allows for out-of-clinic assessments. Local field potentials (LFPs) represent population-level neuronal oscillations surrounding the DBS electrode and can be used as aDBS control signals. This study will evaluate the safety and effectiveness of aDBS in PD subjects with stable cDBS therapy.

Methods: Subjects will have been implanted with DBS leads either in the GPi or STN connected to a commercial DBS system capable of sensing LFPs. An investigational feature will be unlocked to allow programming of two different aDBS modes using low frequency (8-30 Hz) LFP control signals. Subjects will enter a 30-day Baseline Phase in their current cDBS programming configuration, followed by an aDBS Set-up and Adjustment Phase. Subjects tolerating both aDBS modes will then enter a 2-period randomized crossover Evaluation Phase and receive each aDBS mode over 30-day periods, followed by a Long-Term Follow-up Phase over 10 months. The aDBS evaluations will involve measures of On time, quality of life, speech, movement, sleep, patient preference and satisfaction, and total electrical energy delivered (TEED).

Results: The primary effectiveness endpoint will measure On time without troublesome dyskinesias from the motor diary. Other endpoints will include TEED, output from a wearable device, Voice Handicap Index, UPDRS, EQ-5D-5L, PDSS-2, PDQ-39, and patient preference and satisfaction. Safety will include evaluation of stimulation-related adverse events (AEs), AEs, and device deficiencies.

Conclusions: This international, multi-center, chronic aDBS study is expected to generate data to support safety and effectiveness for both aDBS modes in PD subjects.

Andrea KUHN (Berlin, Germany), Lisa TONDER, Robert RAIKE, Scott STANSLASKI, Kassa LYNCH, Helen BRONTE-STEWART
18:45 - 18:50 #23641 - Directional DBS is associated with fewer surgical revisions than traditional DBS: a nationwide real-world study.
Directional DBS is associated with fewer surgical revisions than traditional DBS: a nationwide real-world study.

Objective: To investigate the relative occurrence of complications necessitating surgical revision after de novo implantation of a directional deep brain stimulation (DBS) system compared to traditional omni-directional DBS systems in US patients with movement disorders.

Background: Real-world complication rates from DBS implantation have been poorly characterized and have yet to consider the impact of modern systems. These newer systems include significant advancements in the design of both the intracranial electrode and the lead extension. Specifically, modern systems offer directional shaping of the stimulation field that increases flexibility in delivering therapy while avoiding side effects. We therefore aimed to assess if these advancements resulted in fewer complications necessitating revisions.

Methods: Medicare fee for service claims were used to identify patients undergoing DBS implantation for Parkinson’s Disease or Essential Tremor between Jan 1, 2016 - Dec 31, 2018. Claims records were linked to manufacturer device registration data to identify which patients been implanted with Abbott Infinity, at the time the only commercially available system in the US with directional stimulation capability; linked patients were assigned to the Treatment group. Records that did not uniquely link were classified as omni-directional, non-Abbott and assigned to the Control group. ICD-9/10 diagnosis and procedure codes were used to identify reason for implant and to assess comorbidities. Patients were excluded if they had less than 1 year of enrollment prior to or less than 3 months enrollment after the index implant, and if they had evidence of prior DBS implant. Patients younger than 18 years old were excluded. Patients enrolled in Medicare Health Maintenance Organization (HMO) were also excluded because of the lack of necessary data for those patients in the Medicare database. Over a pre-specified maximum 2-year follow up and with a pre-specified 3-month acute period, device-related complication rates resulting in lead or implantable pulse generator (IPG) revision/removal were compared using the Andersen-Gill modification to Cox-proportional hazard model with correction for age, gender, year of implant, indication, number of leads, number of generators, and staged vs. non-staged implant procedure.

Results: A total of 7,788 patients were identified in the Medicare database, of which 5,188 fulfilled the aforementioned inclusion criteria. 603 patients (71 ± 7 years old, 40% female) implanted with Abbott Infinity systems were assigned to the Treatment group, and 4,585 (71 ± 7 years, 38% female) with traditional omni-directional implants were assigned to the Control group. Patients were followed to a maximum of 2 years, with a mean follow-up of 611 ± 160 days.  The cumulative rate of lead revision/removal over the follow up was 7% in the Treatment group versus 12% in the Control group (hazard ratio [HR] 0.5 [95% CI 0.4 - 0.7], p < 0.001).  The cumulative rate of IPG revision/removal over the follow up was 5% in the Treatment group versus 7% in the Control group (HR 0.6 [95% CI 0.4 - 1.0], p = 0.05). The overall revision/removal rate for leads or IPGs was 9% in the Treatment group versus 15% in the Control group (HR 0.5 [95% CI 0.4 - 0.7], p < 0.001). [Figure]

Conclusions: These findings represent the first real-world report of device-related complications in directional DBS. Using US insurance claims linked with manufacturer device registration data, we found that complication rates in patients implanted with Infinity directional DBS system were nearly half of those compared to those with traditional omni-directional DBS systems. Although both significantly lower, the magnitude of the difference in lead revisions was greater than that for IPG revisions, which may be attributable to the benefits of segmented electrodes and electrical field shaping capabilities. Further investigation and analysis are required to better understand the relationships between DBS system implanted, complication rates, and patient comorbidities.

Chengyuan WU (Philadelphia, PA, USA, USA), Monika PÖTTER-NERGER, Rahul AGARWAL, Stuart ROSENBERG, Allison CONNOLLY, Binith CHEERAN, Wolfgang HAMEL, Ashwini SHARAN
18:50 - 18:55 #23945 - Peak characteristics in the beta band of the human subthalamic nucleus: a case for low and high beta activity.
Peak characteristics in the beta band of the human subthalamic nucleus: a case for low and high beta activity.

Objective: Deep brain stimulation (DBS) of the subthalamic nucleus (STN) in patients suffering Parkinson’s disease allows for the recording of local field potentials (LFPs) in this part of the basal ganglia. Such recordings are of great interest since they may serve as possible biomarkers for the current physiological ”status” of the patient and thus might be used for “brain-sensing” and consecutive adaptive stimulation (“closed-loop DBS”).

Methods: 38 Parkinson-patients (mean age 60.1 years (range 47-71), 11 female/27 male, mean disease duration 11.6 years (range 7-20) having undergone STN-DBS were included. LFP-recordings were performed via externalized leads in all patients and additionally with a Medtronic ACTIVA PC+S™ INS in 10 of these 38 patients. Recordings (both with and without stimulation) were done with the patient in a recumbent position during rest (awake patient, eyes open), during right/ left hand opening and closing (frequency 2/s; 5 min.), during standing for 3-5 min, slow walking 30 m and fast walking 30 m. LFPs were amplified, recorded and digitally stored (resolution 0.1 µV, sampling rate 2000 Hz, filter 0.1 Hz – 500 Hz). Raw data were high-pass filtered (3rd order Butterworth, 1 Hz) and resampled to a sampling frequency of 422 Hz. The fast Fourier algorithm of Malta and averaging the resulting spectra were used to compute the frequency spectrum of all data

Results: Fifty-one of 76 (67.1%) recordings had one peak, eight (10.5%) recordings showed two peaks, and 17 (22.4%) recordings showed no peak. Movement of either hand did not reliably suppress beta peaks. Walking reduced the peaks in the high beta band (above 20.2 Hz) but not the peaks in the low beta band. Stimulation caused a stimulation-strength dependent suppression of most, but not all peaks. 

Conclusion: Beta-peaks can be detected in a high percentage of LFP-recordings using DBS-electrodes. Beta suppression caused by movement is dependent on the type of movement and the frequency of the peak. Further studies should consider the fact that the beta frequency band may host different physiological processes - that mig