Introduction:

Neuronavigation procedures demand high precision and accuracy. Despite this need there is still few literature analysing error in these procedures. There are many steps which could introduce significant position and orientation errors. One of these steps, is surface registration, when the image space is matched with the patient space.

Objective:

The objective of this study is to evaluate the influence of the point registration method in global procedure errors using a previously designed cranial phantom.

Methods:

The designed phantom consisted of a 3D plastic cranium with 10 spheres with different trajectories to their centres and 4 smaller spheres without trajectories. A cloud of points of the phantom was acquired, using the Polaris optical tracking system. Then, the centres and trajectories of the spheres were mathematically calculated and used as ground truth. Afterwards, Computed Tomography and Magnetic Resonance images of the phantom were acquired to be compared with the ground truth. A neuronavigation procedure (a biopsy using the Medtronic Vertek arm and biopsy needle) was planned using Medtronic Cranial 5.2.5 and framelink 5.4.1 software and subsequently simulated. In such procedure, a standard registration, in accordance to manufacturer specifications, as well as an improved registration, which included all sides of the phantom, were executed. The biopsy needle final position and trajectory were obtained for each of the spheres using the reference system and compared with the ground truth.

Results:

After execution of the procedure with a standard registration a mean position error of 5,41 mm (with standard deviation of 1,75 mm) and a mean orientation error of 3,99 degrees (with standard deviation of 1,72 degrees) was identified. After the procedure with the improved registration the mean position error decreased to 3,93mm (with standard deviation of 1,70 mm) and the mean orientation error to 3,65 degrees (with standard deviation of 1,29 degrees).

Conclusion:

Registration had an important influence on global position and orientation errors. A method which includes all the sides of the cranium was used and presented positive results in decreasing the magnitude of global errors, therefore suggesting the need for more accurate registration techniques. 

Introduction:
Classical DBS treatment of Parkinson disease (PD) involves constant stimulation by electrodes localized in deep brain structures such as the sub thalamic nucleus or the internal palladium. This constant stimulation may be suboptimal facing the fact that movement planning and execution involves sequential parietofrontal network activation that is expressed by modulation of EEG beta-band oscillation often referred as event-related desynchronization and synchronization (ERD/ERS). Indeed constant stimulation of the STN is not without secondary  sometimes dramatic secondary effects. In this context adaptive DBS, where stimulation is triggered by certain neurophysiological markers, has emerged as an improvement over conventional systems. It may reduce the amount of energy received by the patient and also theoretically the side effects.

Methods and Results:
We recorded brain activity from the subthalamic nucleus via the deep electrodes while stimulating simultaneously. This was made possible because of the design of a custom-made system that allowed filtering out stimulation artefacts from the underlying brain signals. Tests of the closed-loop adaptive DBS have been performed in 3 PD patients. The protocol consisted of three stimulation conditions over 3 days: continuous (normal), adaptive, and no stimulation, whereby each condition was 20 min long. A blinded neurologist evaluated the clinical assessments of the motor effect via the Unified Parkinson's disease rating scale (UPDRS) before, during, and after each condition. In the no-stimulation condition, we found a pathological synchronization of the beta band in both hemispheres and a strong coherence between both hemispheres in the high-beta and low-gamma band. Therefore, detection of increased (over the 50 percentile) beta band activity (22-28 Hz) was used to trigger the stimulation in the adaptive condition. When compared, subject-specific closed loop stimulation yielded similar efficiency to conventional continuous DBS

Discussion:
Our initial results demonstrate that we can successfully record local field potentials, detect the physiological biomarkers of motor symptoms in PD patients and adaptively trigger the DBS with the same efficacy as constant stimulation. Nevertheless, more subjects and tests over longer periods of time are necessary to confirm the preliminary observations on the efficacy of adaptive DBS.

Significance:
Closed-loop adaptive DBS is possible which opens up strategies to better tune the stimulation leading to increased battery life and better control over symptom variations

Backgorund: Targeting accuracy in DBS surgery is defined as the level of accordance between the selected target and the anatomically real target reflected by characteristic electrophysiological results of microelectrode recording (MER). We aimed to determine the correspondence between the preoperative predicted target based on modern 3 Tesla T2-weighted magnetic resonance imaging sequences and intraoperative MER results separately on the initial and consecutive second side of surgery.

Methods: We retrospectively analyzed 86 trajectories of DBS electrodes implanted into the subthalamic nucleus (STN) of patients with advanced Parkinson´s disease. The entrance point of the electrode into the STN and the length of the electrode trajectory crossing the STN was determined by intraoperative MER findings and 3T high-resolution T2-weighted magnetic resonance images with 1mm slice thickness.

Results: The average difference between MRI-based and MER-based trajectory length crossing the STN determined in each patient was 0.28 ± 1.02 mm (-0.51 to -0.05 mm 95% CI). There was a statistically significant difference between the MRI- and MER-based entry point on both the initial and second side of surgery (p = 0.04). 43 % of the patients had a difference of more than ±1 mm of the MRI-based predicted and the MER-based determined entry point into the STN with values ranging from -3.0 to + 4.5 mm.

Conclusions: MRI-based targeting of the STN is accurate in the majority of cases in both the first and second side of surgery. However, in 43% of implanted electrodes we found a relevant deviation of more than one millimetre supporting the concept of MER as an important tool to guide and optimise targeting and electrode placement.

Backgorund: Deep Brain Stimulation (DBS) is an approved treatment option for therapy of refractory essential tremor (ET). Although evidence exists about its therapeutic effect, the optimal location of stimulation to improve tremor is still a matter of debate. Apart from the Ventral Intermediate Thalamus (VIM), the posterior subthalamic area (PSA) including the Dentato-Rubro-Thalamic-Tract (DRTT) has more recently been proposed as an appropriate target. The objective of our study is to present our MRI-based targeting procedure and correlate it with the stimulation site, clinical outcome and DTI-based fiber tracking identification of the DRTT.

Methods: We present a prospective series of 9 patients with unilateral or bilateral DBS implantation in the PSA. T2-weighted MRI was used to target the PSA on axial slices 2-3 mm below midcommissural point (MCP) within the white matter between red nucleus and subthalamic nucleus using iPlan Net 3.0 (BrainLab). Fiber tracking of the DRTT was performed in each patient. Stimulation site was obtained by calculation of the position of the active contact and its corresponding Volume of Tissue Activated (VTA). Active contact positions and VTA were correlated to clinical outcome.

Results: The mean position of the active contact was LAT 10.54; AP -3.80 and VERT -1.59 mm with reference to MCP. Projection of the mean active contact position and its corresponding VTA onto the Morel stereotactic atlas revealed a stimulation site within the PSA in the proximity of the DRTT. This was correlated with DTI fibertracking findings. DBS resulted in significant tremor reduction 3-6 months postoperatively on the Fahn–Tolosa–Marin Tremor Rating Scale (TRS)

Conclusions: DBS of the PSA is effective in the treatment of ET. Our MRI-based anatomical landmarks seem to be reliable to target the PSA in each individual case. Our DTI findings suggest that the DRTT is involved in the efficacy of PSA DBS, although DRTT is not clearly identifiable in each patient.

Key words

Real-time fMRI, Neurofeedback, Parkinson’s disease

Introduction:

Neurofeedback training using real-time functional magnetic resonance imaging (rtfMRI) enables learned control of brain activity of a target brain region. This emerging non-invasive intervention has been used to produce clinical improvements in disorders such as chronic pain and recently, Parkinson’s disease (PD). We report a unique intensive neurofeedback training paradigm, using rtfMRI, for PD patients undergoing deep brain stimulation surgery of the subthalamic nucleus (STN-DBS), and compare the efficacy of training against a control group of non-surgical patients also undergoing rtfMRI neurofeedback training.

Aims:

The study aims to establish the feasibility of using intensive rtfMRI neurofeedback in patients with PD, with the goals of demonstrating volitional control of higher order motor cortical activation (i.e. supplementary motor area, SMA), and linked clinical improvements, together with a potential interaction with the effects of STN-DBS.

Methods:

Over one week of intensive rtfMRI training in the scanner, patients were trained to use motor imagery linked to feedback provided by a visual interface in the form of a ‘thermometer bar’ coupled in realtime to BOLD activation in their SMA. Patients were trained specifically to increase activation of their SMA. UPDRS assessments both on and off medication, together with finger tapping paradigms, were performed before and after neurofeedback training to investigate any improvements linked to the intervention. Functional, connectivity and structural imaging correlates were also examined. Further, in surgical patients, a possible interaction of rtfMRI-led neurofeedback with DBS will also be investigated.

Results:

Behavioural, brain activation patterns, connectivity and structural correlates of rtfMRI neurofeedback training were examined in relation to observed improvements in UPDRS scores both on and off medication.

Conclusions:

We provide preliminary evidence that intensive rtfMRI neurofeedback training can be successfully implemented and tolerated in PD patients, with the potential for behavioural and clinical improvements.

Optimal stimulation site for subthalamic nucleus deep brain stimulation remains ill-defined. Previous studies have disagreed on whether contacts within the nucleus, or just superior to it, in the rostral zona incerta, give best improvement in motor symptoms. Furthermore, the effect that cortical connectivity has on predicting the success of deep brain stimulation is yet to be established. Here, we aimed firstly to identify stimulation clusters in the subthalamic region that predict maximum improvement in rigidity, bradykinesia and tremor, or emergence of side-effects; and secondly, to map-out the cortical fingerprint, mediated by the hyperdirect (subthalamic-cortical) pathway which predicts maximum efficacy. Twenty patients with Parkinson’s disease underwent preoperative high angular resolution diffusion imaging prior to bilateral subthalamic nucleus deep brain stimulation. One-year after MRI-guided and MRI-verified surgery all contacts were screened for efficacy and side-effects at different amplitudes. Voxel-based statistical analysis was used on models of corresponding volumes of tissue activated to identify significant treatment clusters. Probabilistic tractography was employed to identify patterns of cortical connectivity associated with treatment efficacy. All patients responded to treatment (off medication UPDRS-III mean improvement was 46% [p<0.0001] at one year) without adverse events. Clusters corresponding to maximum improvement in tremor were inside the nucleus in the posterior, superolateral portion. Clusters corresponding to improvement in bradykinesia and rigidity were closer to the superior border in a slightly more medial and posterior location. The rigidity cluster extended beyond the superior border to the area of the zona incerta and H₂ Forel’s field. Cortical connectivity to the primary motor area was predictive of higher improvement in tremor; whilst that to the supplementary motor area was predictive of improvement in bradykinesia and rigidity; and connectivity to the prefrontal cortex was predictive of improvement in rigidity.  Our findings support the presence of separate stimulation sites within the subthalamic nucleus and its superior border, with different patterns of cortical connectivity, associated with maximum improvement in tremor, rigidity and bradykinesia.

INTRODUCTION

The subthalamic nucleus became the gold standard target for deep brain stimulation in Parkinson’s disease in the last 25 years, but the exact mechanism of chronic stimulation is still poorly understood. Some still rely on the hypothesis that only proper placement of DBS leads within the subthalamic nucleus can alleviate most motor symptoms of the disease. As our focus gradually shifts to using segmented, directional leads, understanding anatomical and connectivity relations of the subthalamic nucleus is becoming even more essential.

METHODS

15 patients suffering from Parkinson’s disease who underwent deep brain stimulation were enrolled in this study. 3D T1 (1x1x1 mm, isotropic voxels), contrast enhanced T1, T2 (0,5x0,5x1,5 mm), and DTI (32 directions, 2x2x2 mm, isotropic voxels) MRI sequences were acquired preoperatively from each subject on a 3T Siemens Verio scanner. Contrast enhanced CT scans were acquired on the day of the surgery with a Leksell stereotactic frame placed on the patients’ head. CT and MRI images were fused to preserve accuracy by creating a semi distortion-free environment. Lead placement has been carried out with the guidance of intraoperative microelectrode recording and macrostimulation. Postoperative contrast enhanced CT scans were acquired at least 6 weeks after surgery to exclude pneumocephalus or brain shift during analysis. Postoperative CT scans and T2 sequences were registered to anatomical T1 reference images using Flirt (FSL 5.0.9, FMRIB, Oxford). Preprocessing of DTI images has been carried out using FSL. Cortical parcellation was done using Freesurfer (ver. 5.3., Martinos Center for Biomedical Imaging, Harvard University). Cortical regions were manually checked for artifacts or inconsistencies, the subthalamic nuclei has been delineated manually by too independent experts. Probabilistic segmentation of the subthalamic nucleus was carried out using FSL probtrackX, using default settings and modified Euler streaming. Postoperative outcome was measured using UPDRS III motor scale by an independent neurologist, at least one year after surgery.

RESULTS

Results of probabilistic tractography showed statistically significant distinct functional subgroups within the subthalamic nucleus, connecting to limbic, dorsolateral prefrontal, pre-supplementer motor, supplementer motor, and motor regions respectively. Clinical outcome was measured at least one year after surgery by selecting the best stimulation contacts, chosen by a neurologist independently based on clinical signs only. Best clinical outcome could be achieved by using contacts closest to pathways traversing and terminating at portion of the subthalamic nucleus connecting to the supplementer motor area. Average outcome in the whole population was 8,92/108 on the UPDRS III scale.

CONCLUSION

Probabilistic tractography is able to reveal functional subgroups within the subthalamic nucleus. Estimating fiber distribution within and surrounding the nucleus can provide even more sophisticated targets, while also contributing to programming conventional and directional leads.

Francesco Sammartino MD ¹; Vibhor Krishna MD, 2¹ ; Nicolas Kon Kam King MD, PhD ¹ ; Andres M Lozano MD ¹ , PhD; Mojgan Hodaie MD, MSc ¹

1 – Division of Neurosurgery, University of Toronto. Toronto Western Hospital, 399 Bathurst Street, Toronto, M5T2S8, Canada.

2 – Center for Neuromodulation, Department of Neurosurgery and Neuroscience, The Ohio State University, Columbus, OH.

Introduction –

Diffusion tensor imaging can be useful to delineate the structural connectivity of surgical targets for deep brain stimulation in brain.  However the concerns surrounding imaging artifacts and preprocessing steps make it challenging to translate imaging findings into surgical targeting. We report the role of diffusion imaging algorithms for the purposes of targeting ventral intermediate (VIM) nucleus for tremor surgery, and corroboration of accuracy by comparison of location of tracts during intraoperative microelectrode recordings.

Methods –

Diffusion (60 gradient directions) and structural imaging in 3Tesla MR was used in a cohort of 6 patients (3 essential tremor, and 3 with tremor dominant Parkinson’s disease). Raw diffusion data was preprocessed offline with eddy-current and registration-based distortion corrections. After conversion to DICOM, the images were uploaded into the StealthViz workstation (Medtronic Inc., Minneapolis, MN). The diffusion data was rigidly aligned with the high resolution 3D T1 sequence, to allow for anatomical/diffusion mixed images. This also permitted determination of accuracy of coregistration by comparing the location of 10 different anatomical landmarks in the proximity to the thalamus (anterior commissure, posterior commissure, bilateral superior colliculi, bilateral body of fornix, bilateral superior cerebellar peduncle, and the posterior and inferior borders of the splenium of the corpus callosum in the midsagittal plane). The registration was considered adequate if the 7 out of these 10 landmarks were within ≤ 2mm between the T1 and diffusion-weighted images. After this stage deterministic FACT-based tractography was performed. Prior to ROI placement, the tensor calculation results were verified by looking at the direction of the principal eigenvector in selected location (color FA map). The pyramidal tract and medial lemniscus were tracked using standard tracking parameters (seed density = 1, FA stop value 0.2 and tracking angle 45-600. VIM tractography was only performed if all the previous preprocessing steps were satisfactorily completed. The location of the tracts were then retrospectively correlated to the microelectrode intraoperative recordings.

Results –

We were able to successfully determine the location of VIM by visualizing the dentatorubrothalamic tract (DRT) in all the patients.  The mean error in anatomical alignment between scans was less than 2 mm in 7 of the chosen landmarks for this cohort. We observed consistent discrepancy in the alignment of the anterior commissure, likely originating from the susceptibility effect from air-filled sphenoid sinuses despite being minimized by our pre-processing steps.  During MER, the areas with intraoperative stimulation-induced tremor suppression were consistently overlapping with the DRT projection in each patient. The entire preprocessing process take 80 minutes per subject.

Conclusions –

We present here a framework to perform tractography based targeting and include the necessary steps to maximize accuracy.  Use of this methodology also permits improved visualization of white matter tracts from small ROI for targeting purposes. Key steps in the use of tractography based targeting include the ability of mixing diffusion and anatomical images, assessment of coregistration accuracy and visualization of DRT.

Background: Deep brain stimulation (DBS) of the superolateral branch of the medial forebrain bundle (slMFB) emerges as an interesting alternative  - yet experimental - treatment for therapy refractory psychiatric diseases. First experiences have been reported from a pilot trial in major depression (1) and an uncontrolled case series for obsessive compulsive disorder (OCD) (2).

Objective: To describe the surgical technique for deep brain stimulation (DBS) of the supero-lateral branch of the medial forebrain bundle (slMFB). To report our experience with the successful bilateral implantation in a larger patient group.

Methods: Surgical experience from bilateral implantation procedures in n=27 patients is reported. The detailed procedure of diffusion tensor imaging magnetic resonance imaging fiber tracking (DTI FT) assisted targeting together with detailed descriptive electrophysiology in 144 trajectories of the target region (recording and stimulation) and intraoperative testing are addressed.

Results: In this early stage of experience, bilateral slMFB DBS requires DTI FT assisted targeting combined with in depth intraoperative electrophysiological investigation of the target region.

Conclusion: The slMFB is a promising target region for the treatment of some psychiatric disorders (1,2). DTI FT assisted DBS of the slMFB is based on an imaging technology that is readily addressed in other indications (3,4). To the authors’ knowledge the slMFB is the only target region for psychiatric disorders that allows for intra-operative testing with clear effects and side effects to guide implantation. In our eyes, this makes surgery of the slMFB is in many features comparable to typical movement disorder surgery.

1)    Schlaepfer, T. E., Bewernick, B. H., Kayser, S., dler, B. M. X., & Coenen, V. A. (2013). Rapid Effects of Deep Brain Stimulation for Treatment-Resistant Major Depression. Biological Psychiatry, 1–9.

2)    Coenen, V. A., Schlaepfer, T. E., Goll, P., Reinacher, P. C., Voderholzer, U., Tebartz van Elst, L., et al. (2016). The medial forebrain bundle as a target for deep brain stimulation for obsessive-compulsive disorder. CNS Spectrums, 1–8.

3)    Coenen, V. A., Allert, N., Paus, S., Kronenbürger, M., Urbach, H., & Mädler, B. (2014). Modulation of the Cerebello-Thalamo-Cortical Network in Thalamic Deep Brain Stimulation for Tremor. Neurosurgery, 75(6), 657–670.

4)    Coenen, V. A., Rijntjes, M., Prokop, T., Piroth, T., Amtage, F., Urbach, H., & Reinacher, P. C. (2016). One-pass deep brain stimulation of dentato-rubro-thalamic tract and subthalamic nucleus for tremor-dominant or equivalent type Parkinson’s disease. Acta Neurochirurgica, 1–9.

Background: Several targets are currently investigated for their antidepressant efficacy in treatment resistant depression (TRD). Among them, DBS to the supero-lateral branch of the medial forebrain bundle (slMFB) lead to rapid and long-term antidepressant effects (1). We designed a larger study with a staggered onset sham-controlled design.

Methods: Sixteen patients suffering from TRD were treated with DBS bilaterally to the slMFB. Patients were either stimulated with DBS immediately (group A) or with a delay of two months (group B) (staggered onset) in a double blind protocol.

Outcome criterion was the difference in antidepressant response between groups during sham phase as well as long-term effect of unblended DB stimulation.

Results: A significant difference in antidepressant effect was found in month two after stimulation onset between groups. Surprisingly, this difference was not found in the first month, because both groups responded significantly after stimulation onset; in group B, this effect was time-limited (three weeks). A significant antidepressant effect at 12 months of DBS was observed on the group level. Main side effect was strabism at higher stimulation currents. No change in cognition was identified.

Conclusions: This study demonstrates a significant difference in efficacy of sham and real stimulation in a larger sample. This confirms previous efficacy data from our group (1) in an unblinded study in a smaller sample. For few weeks, some of the sham-stimulated patients showed a significant response. This can either be explained by a transient placebo effect but likely presents a lesioning effect such as described in DBS for movement disorders.

(1)  Schlaepfer, T. E., Bewernick, B., Kayser, S., Maedler, B., & Coenen, V. A. (2013). Rapid Effects of Deep Brain Stimulation for Treatment-Resistant Major Depression. Biological Psychiatry.

Introduction

Obsessive-compulsive disorder (OCD) is a psychiatric disorder characterized by recurrent persistent thoughts known as obsessions, and time-consuming ritualistic behaviors known as compulsions. Cognitive behavioral and pharmacological therapies may be required when symptoms interfere with the global functioning of the patient. However, up to 10% of OCD patients treated with these therapies show an insufficient improvement of symptoms. For those patients, surgery may be considered. Several targets have been implicated in the study of the effects of deep brain stimulation (DBS) on OCD symptomatology. The most common targeted areas are the nucleus accumbens (Nacc), the internal capsule (IC), the ventral capsule/ventral striatum, and the subthalamic nucleus. To these days, there is still an ongoing debate about the best target for DBS in OCD patients.  Furthermore, these structures cannot be considered as completely distinct targets as the NAcc is located immediately underneath the anterior limb of the IC and extends dorsolaterally into the ventral putamen and dorsomedially into the ventral caudate nucleus. Therefore, the differences between the position of contacts, within the same lead, may determine the clinical postoperative outcome. The objective of this study was to determine the association and differences between location of active contacts within the Nacc and the ventral IC and clinical outcome at long-term follow up.

Methods

Twenty-three patients who underwent implantation of unilateral (N=3) or bilateral (N=20) electrodes for Nacc DBS were included in this study. OCD symptoms were measured with the Yale-Brown Obsessive Compulsive Scale (Y-BOCS) scale before surgery and postoperatively for a long-term follow-up at different time points. For the lead location, preoperative MRI planning scans were fused with postoperative CT scans in order to assess the exact location of the active contacts in relation to the Nacc and the IC.

Results

The mean postoperative follow-up was 18±7 months (range from 9-51 months). Mean stimulation parameters were 3.9±1V (range from 2.5-6.5V), 96±17µs (range from 90-150µs), 135±7Hz (range from 130-145Hz). Patients with active contacts in the Nacc showed an average improvement on the Y-BOCS of 47%, whereas patients with active contacts located in the IC showed an average improve of 32%. However, if the active contacts were located within a maximumum distance of 1,5 mm of the transition between Nacc and the IC, the average improve on the Y-BOCS scores was of 58%.

Conclusions

OCD patients with active electrodes for DBS in the transitional zone between Nacc and IC showed the best clinical outcome on the Y-BOCS scores at long-term follow-up compared to active electrodes located in Nacc or IC. Further studies will be required to determine if different regions of the cortico-striato-pallido-thalamo-cortical network are activated during stimulation of region that could correlate with these clinical observations.

Despite the advent and popularity of high profile neurostimulation programmes, lesion surgery for psychiatric disorders is still performed in many centres around the world. For most centres, this is a low-volume activity and it remains challenging for any single centre to present compelling outcome data derived from large cohorts of treated patients. Similarly, it takes many years to collect data to test hypotheses relating to optimising patient selection, surgical technique and lesion topography. Globally, psychiatry, surgery, medicine, psychology and neuroscience communities, the media and the lay public tend to believe lesion surgery for psychiatric disorders is not only an outdated treatment approach, but also ineffective and harmful. Sporadic, small scale case series (n<20) from single centres are unlikely to influence significantly wider opinion and practice. 

Objective: to establish under the auspices of, and on behalf of, the Psychiatric Neurosurgery Committee of the WSSFN, an international, anonymised, pooled, clinical and neuroimaging dataset from centres offering ablative neurosurgery for Mood Disorders and Obsessive-Compulsive Disorder, and, where available, for other indications.

Aims: to build the capacity to compare retrospectively the clinical outcomes for different lesional procedures, across different clinical populations and across different international centres. Also, to establish the infrastructure for prospective collaborative audits of clinical outcomes framed around a shared, minimum dataset.

Method: anonymised individual-level patient data will be hosted, on behalf of WSSFN, by the University of Dundee (Scotland, U.K) Health Informatics Centre (HIC), a component of the Farr Institute, a UK-wide, publicly funded research collaboration involving 21 academic institutions and health partners in England, Scotland and Wales. HIC is a secure, curated Safe Haven environment with robust data governance procedures for the provisioning of anonymous data to researchers for approved projects. Analyses are performed within the safe haven by approved investigators via secure remote access. The proposed initial ‘ideal’ dataset will include information acquired pre-surgery and at 12m follow up relating to patient demography, diagnosis, symptom severity and functional capacity alongside structural MR imaging – T1 and T2 weighted whole brain scans - stored in DICOM, ANALYSE or NIFTI format (where possible).

The inception date for the Registry is September 1st 2016 and the project will work to identify and collect suitable data for 12 months in the first instance. We seek expressions of interest from any centre wishing to collaborate.

Introduction: The modulation of consciousness processes with deep brain stimulation (DBS) in minimally consciousness state seems achievable. Between 1968 and 2016 nine teams have reported effects in 58 vegetative (VS) or minimally conscious (MCS) patients. We analyzed the literature focusing on methodological issues, willing to address clinically relevant key-points for the selection of targets and design of future studies.

Literature review: Half of the studies were case-reports. Most teams intended to place electrodes in the thalamus. All leads were implanted according to atlas-based coordinates. Five studies used low frequency stimulation, 25 to 50 pulses /sec, and three high frequency stimulation, at 100 and 250 pulses /sec. The most recent studies reported effects in continuing VS-MCS patients, followed up during several months or years. The clinical status and DBS effects were measured using simple clinical observations, up to JFK Coma Recovery Scale—Revised. Parallel to the clinical status, the most recent study analyzed the extent of brain lesions. No severe irreversible, stimulo-induced, adverse effects were reported, but one patient had post-operative intra cerebral hematoma. One clinical study had double-blinded on/off crossover phase, whereas the others were observational studies. From these studies it can be inferred that high or low frequency stimulation of deep gray structures, particularly of the central thalamus, can provoke overt conscious behaviors. Recent literature concerning models of consciousness related circuitry let us think that several deep brain regions and cortices are involved and could be future relevant spots of neuromodulation.   

Conclusion-Perspective: Future studies willing to modulate the deep brain circuitry should take into account the recent knowledge on altered dynamics of neural correlates of disorder of consciousness, the dynamics of spontaneous recovery, and the consequences of structural and functional lesions.

Objectives:

    Obsessive-compulsive disorder (OCD) is one of the most disabling of all psychiatric illness with a lifetime prevalence of 2% - 3% and an early onset in adolescence or young adulthood. Core symptoms of OCD are intrusive thoughts, feelings, or images (obsessions) and repetitive behaviors (compulsions) aimed at reducing anxiety associated with obsessions. Despite available pharmacological and psychotherapeutic treatments about 10% of OCD patients remain severely affected and are considered treatment-refractory. For some of these patients deep brain stimulation (DBS) offers an appropriate treatment method. In hopes of identifying better treatment options such as DBS, many attempts have been made to clarify pathological brain mechanisms. Alterations in brain functional cortical connectivity in resting-state networks have been detected with functional imaging techniques, but neurophysiological connectivity measures have not been systematically examined.  

    To address this question, we applied resting-state electroencephalography (EEG) to investigate the whole brain fundamental functional alterations in patients with OCD. In this study, we focused on oscillation-based functional connectivity.

Methods:

    We used the combined application of the independent component analysis (ICA) and the Granger causality (GC) analysis to examine resting state functional connectivity as measured by routine scalp-EEG in 10 patients with OCD and 10 healthy controls matched for age lying with eyes closed in a dark sound-attenuated room. After the power spectra was computed, GC spectra method, a part of spectral interdependency methods, was used to examine the strengths, directions, and frequencies of interactions between dynamic processes. The measures of spectral interdependency were derived from the time series recordings of dynamic systems by using autoregressive modeling. The direction of information flow between these EEG sources was then estimated using the directed transfer function (DTF).

Results:

    As compared to controls, the patients with OCD had decreased functional connectivity between medial prefrontal and occipital cortex, and between the left temporal and occipital cortex. The patients with OCD expressed lower information flow (DTF) in beta frequency range principally from the occipital to the medial prefrontal region, and from the occipital to the left temporal region.

Conclusions:

    We used resting-state EEG to study whole-brain functional connectivity patterns in patients with OCD and healthy controls. The primary finding was changed functional connectivity in large-scale networks including the medial prefrontal cortex, temporal cortex and occipital cortex in the patients with OCD. These results indicated EEG connectivity-based measurements could probably serve as reliable and valid biomarkers for diagnosis of OCD.

    It has been suggested that deficits in the ability to selectively attend to relevant information while concurrently suppressing competing irrelevant information is a central feature of OCD. The impairment of inhibitory control in OCD patients might be reflected in abnormal cognitive functions, which seem to be due to reduced attention.

We speculate that decrease in functional connectivity in large-scale networks among the medial prefrontal cortex, temporal cortex and occipital cortex may be related to functional deficits in cognitive domains in OCD.

    The impact of neurological and psychiatric diseases on the functional organization of the brain has been a topic of growing interest in the last decade. It is likely that most psychiatric disorders do not result from a deficit in a single brain area and that a network perspective is necessary to explain their complex etiology. OCD is associated with abnormal intrinsic functional connectivity in large-scale brain networks.

This network approach can shed some light on the pathophysiological mechanisms underlying neuropsychiatric disorders.

Background: Obsessive compulsive disorder (OCD) has a lifetime prevalence of 1-2 %. Standard treatments are ineffective in up to 40% of cases. Even with the best treatments, there remains a subgroup with severe symptoms and significant disability. Studies of deep brain stimulation (DBS) for OCD have shown improvement in both symptoms and quality of life in severe OCD.  Two targets in particular have shown promise: the anteromedial subthalamic nucleus (STN) and the ventral capsule/ventral striatum (VC/VS) [1-3]. It is not clear however if one site has advantages over the other and, with regard to the VC/VS site, whether stimulation of the anterior capsule white matter or ventral striatum/nucleus accumbens grey matter is critical for improvement [2, 3].   

Aims: We report a within subject comparison of the effect of DBS on OCD symptoms at STN and VC/VS sites both individually and together (ClinicalTrials.gov #NCT02655926). The aims of the study were to determine: a) the efficacy of DBS at each site; b) whether stimulation of both sites improves the response compared to either site alone; and c) the critical stimulation contacts at the VC/VS site.

Methods: Six participants, with severe treatment refractory OCD, were recruited via the UK specialist OCD service and underwent implantation of bilateral electrodes at both the VC/VS and anteromedial STN sites. A Leksell frame-based MRI-guided and MRI-verified approach under general anaesthesia was used. The subthalamic nucleus was localised on axial T2-weighted stereotactic images and the VC/VS localised on coronal and axial proton density images (Siemens, 1.5T). Using a double blind cross-over design, 12-weeks stimulation at STN and VC/VS sites was compared, followed by stimulation at both sites for 12 weeks. The primary outcome measure was YBOCS: an improvement of greater than or equal to 35% was the predefined response.

Results: Accurate stereotactic and anatomical lead location was confirmed on immediate postoperative stereotactic MR images in all patients. For the VC/VS target, the deepest DBS lead contact was within the nucleus accumbens, the one superior to that in the “shell” of the nucleus accumbens while the superior two contacts were within the inferior aspect of the anterior limb of the internal capsule. The response rates (defined as the number of patients with >35% reduction in YBOCS) were: STN 3/6; VC/VS: 5/6; STN + VC/VS: 5/6. In the one non-responder YBOCS reduction was 32% after the combined STN+VC/VS stimulation phase. For the whole group, the mean reduction in YBOCS scores were: STN 16.3; VC/VS: 19.2; STN + VC/VS: 22.0 which represents a mean reduction of 42%, 53%, 62% from their own baseline scores and a reduction to predefined mild/subclinical symptoms of 0%, 50% 50% respectively. The top two DBS contacts of the quadripolar lead were found to be the most effective at the VC/VS target in all 6 patients.  

Conclusion: These results suggest that: a) The VC/VS site may be superior to the STN site for the amelioration of severe OCD symptoms; b) There is only a modest advantage of stimulating both sites together; and c) The effective stimulation site for the VC/VS target is the inferior aspect of the anterior limb of the internal capsule and not the ventral striatum/nucleus accumbens grey matter.

The study was funded by the MRC (MR/J012009/1)

1. Mallet et al, 2008 N Eng J Med 359: 2121
2. Denys et al, 2010 Arch Gen Psychiatry 6: 1061                                                    
3. Greenberg et al,  2010 Molecular Psychiatry 15:64                    

Background: Several targets for deep brain stimulation (DBS) have proven antidepressant efficacy for the treatment of otherwise treatment-resistant depression (TRD): a reduction of symptom severity of 50% in about 50% of patients could be demonstrated. The supero-lateral branch of the medial forebrain bundle (slMFB) was hypothesized to be a more efficacious target and rapid antidepressant effects were observed in seven patients. Long-term clinical data including quality of life, side effects and cognition covering four years are presented in order to evaluate clinical efficacy.

Methods: Eight patients suffering from TRD were treated with DBS bilaterally to the slMFB. Primary outcome criterion was a 50% reduction in depression severity at 12 months compared to baseline. Secondary measures were general functioning, quality of life, safety and cognition as assessed for up to for 4 years.

Results: Six of eight patients were responders at 12 months (75%), among them, four patients were remitters (50%). Long-term results revealed a stable antidepressant effect for up to four years. Clinical efficacy was also reflected in a substantial improvement of the global assessment of functioning. Main side effect was strabism at higher stimulation currents. No change in cognition was identified. Timeline analysis revealed a significant reduction in depression for 7/8 patients in all months (87.5%).

Conclusions: Long-term results of DBS to the slMFB for TRD suggest a rapid and sustained antidepressant effect; timeline analysis may be a more meaningful approach in assessing long-term outcome in TRD clinical studies. In order to validate the antidepressant effect of slMFB-DBS, the inclusion of a sham phase in further studies is needed.

Introduction: An effective treatment of patients in a minimally conscious state (MCS) or vegetative state (VS), caused by hypoxic encephalopathy (HE) or traumatic brain injury (TBI), is not yet available. Deep brain stimulation (DBS) of the thalamic reticular nuclei, as a therapeutic procedure, has been attempted mainly in patients with TBI.

Methods: Fourteen out of 49 patients were included in this study (4 patients had TBI and 10 patients had HE, 4 being in MCS and 10 patients in VS). The selection criteria for DBS, evaluating status of cerebral cortex and thalamocortical reticular formation, included: neurological evaluation, electrophysiological evaluation and the use of imaging techniques such as positron emission tomography (PET) and magnetic resonance imaging (MRI). The target for DBS was the centromedian-parafascicular nucleus (CM-pf) complex. Patient follow-up was between 38 and 60 months. 

Results: Two MCS patients regained consciousness and regained their ability to walk, to speak fluently and live independently. One MCS patient reached the level of consciousness, but currently is still in a wheelchair. One VS patient (after cerebral ischemic lesion) improved to the level of consciousness and currently responds to simple commands. Three VS patients died from respiratory infection, sepsis or cerebrovascular insult, respectively. Other patients remained without substantial improvement of consciousness.

Conclusion: The spontaneous recovery of MCS/ VS to the level of consciousness with no or minimal need for assistance in everyday life is very rare, therefore if a patient is a candidate according to the above mentioned criteria, DBS could be a treatment option.

There is currently no pharmacological treatment for cocaine addiction, therefore it’s important

to look for alternative treatment strategies. One possibility could be a surgical approach.

Indeed, it has been shown in the rat that the inactivation of the subthalamic nucleus (STN), by

either lesions or high frequency Deep Brain Stimulation (DBS), reduces motivation for cocaine

while increasing motivation for food. It has thus been suggested that STN high frequency DBS

could be a good strategy to treat cocaine addiction. Before testing in human addicts, the aim

of the present study was to validate this hypothesis in non-human primates. We have trained

two monkeys to work under various schedules of reinforcement (Fixed Ratio 15 (FR15) and

Progressive Ratio (PR)) for either apple sauce or cocaine (intraveinous 0.1 mg/kg/injection).

After stabilisation of performances, electrodes have been implanted bilaterally in the STN, and

chronic stimulation has been further applied (130 Hz, 2V). All conditions (apple saucestimulation

ON, apple sauce-stimulation OFF, cocaine-stimulation ON, cocaine-stimulation

OFF) have been tested in alternance. Results have first shown that the level of motivation was

higher for cocaine than for apple sauce before stimulation. Then, after STN DBS, the

motivation for apple sauce was significantly increased while that for cocaine was significantly

decreased. These results confirm the opposite effect of STN DBS on motivation that has been

previously demonstrated in rats. Since decreasing the motivation for the drug, without

diminishing other forms of motivation is the goal for a possible treatment of cocaine addiction,

STN DBS may thus be the appropriate strategy.

Objective: Unwanted side effects, induced by current spread into adjacent structures, limit the efficacy of conventional deep brain stimulation (DBS). In Parkinson’s disease (PD), the corticospinal tract and the medial lemniscus are correlated with motor and sensory side effects from subthalamic nucleus (STN) stimulation, while the internal capsule is also most often affected in globus pallidus internus (GPi) stimulation. In essential tremor (ET), DBS targeting the Vim/PSA can lead to inadvertent stimulation of the internal capsule and cerebellothalamic fibres, among others, resulting in paresthesia, dysarthria and disequilibrium.

Intraoperative studies in DBS of the STN provide evidence that directional steering of the electrical field can modify thresholds for both beneficial and side effects. In September 2015, a DBS system for chronic directional stimulation (Boston Scientific) has become available for clinical use.

After one year of experience with directional DBS, we aim to preliminary assess the therapeutic potential of this new technology and explore the surgical and radiologic specifics.

Methods: After classic target planning, DBS leads with a radiopaque marker were implanted employing standard procedures, aiming to implant the three segments of the directional electrodes facing anteriorly, posterolaterally and posteromedially. Intra- and postoperative x-ray and postoperative CCT were conducted to evaluate lead positioning and orientation.

In a pilot study for DBS of the STN, randomized, double-blinded monopolar reviews were conducted and therapeutic vs side effect thresholds were compared between uni- and omnidirectional stimulation. With the respective optimal settings established, patients entered a two-day double-blinded crossover phase and were then asked to opt for their preferred mode of stimulation. As of now, 10 consecutive patients were included into the ongoing treatment study.

Outside the study, three patients were implanted in Vim/PSA for ET, and one patient with PD had revision in GPi to alleviate early side effects under the use of conventional leads.

Results: Lead Implantation required extra effort only to aim for desired orientation of directional contacts per visual judgment. However, postoperative verification of lead orientation proved difficult, as the orientation of the marker cannot be objectively evaluated on X-ray and conventional CT scans. An algorithm for verification of axial orientation through analysis of CT artefacts is being developed.

During double-blinded clinical testing, unidirectional steering led to increased side effect thresholds in at least one direction compared to omnidirectional stimulation. The blindedly selected favorable direction of stimulation in STN DBS was posteromedial or anterior in most patients. After the blinded two-day crossover phase, the majority of patients opted for the directional stimulation, and all patients so far remain on directional settings for at least one hemisphere.

Conclusions: This study confirms that chronic directional DBS in movement disorders is safe, well tolerated, and may increase the threshold for side effects, thus widening the therapeutic window in comparison to conventional DBS. In an ongoing pilot study for DBS of the STN, most patients opted for directional stimulation after a blinded two-day trial. Larger studies and long-term follow-up are needed to confirm and quantitate the results. As conventional circular stimulation is still feasible with segmented electrodes, it appears safe to explore their use in diverse targets.

Background and Objective     

Adverse events (AEs) related to surgery or implanted devices are feared the most by patients to be treated with deep brain stimulation (DBS) and such AEs have a detrimental impact on clinical outcome and quality of life.

Methods         

The 'PubMed' database was searched using the following search term: ((deep brain stimulation[Title]) AND ((complication*[Title]) OR (adverse event*[Title]) OR (hemorrh*[Title]) OR (infect*[Title]) OR (explant*[Title]))). In addition, relevant studies assessing AEs, in particular prospective and monitored trials for Parkinson's disease, tremor, or dystonia, were included. The following three categories were used to assess AEs: (1) hemorrhage and other intracranial complications, such as brain infarction or brain abscess; (2) infections or erosions resulting in complete or partial hardware removal as this resulted in interruption of DBS therapy; (3) lead revisions due to fracture, misplacement, migration, or loss of effect since this involves an additional intracranial procedure. All three categories can be accurately assessed even in a retrospective manner provided that postoperative imaging is performed and that a complete set of surgical reports is available for all patients. The number of AEs was related to the number of patients but not the number of procedures or the number of electrodes implanted as this generates the most informative rates and does not result in the dilution of AEs rates. Additional analysis based on patient-years (mean follow-up x patients) served to account for long-term complications and to put studies with different follow-ups on a common denominator. AE rates were calculated (i) among studies ('non-pooled' analysis) and (ii) also determined by summing up respective AEs of all applicable studies and by dividing this number by the total number of patients included in these studies ('pooled' analysis).

Results            

Although, all three categories represented complications that had to be rated as serious AE and at least severe AE in many instances exact numbers could not be derived, even in high-ranked thoroughly monitored trials. The average rate of intracranial complications was 3.8% and 3.4% for non-pooled and pooled analysis, respectively. It did not differ between 9 studies involving >300 patients and 31 studies involving <100 patients (3.8% each; non-pooled). However, 'pooled' analysis of complications rates revealed that the incidence of intrcranial complications was lower in larger studies (3.2 vs 3.7% for studies having included >300 vs <100 patients, respectively). Interestingly, in monitored trials the reported rate of intracranial complications was the lowest (3.0%). This may be due to the fact that the inclusion criteria of those studies exclude patients exhibiting a higher risk for hemorrhages and centers may tend to suggest study participation rather to healthier patients. The rate of infections that required hardware removal (4.3% and 3.8%, pooled and non-pooled, respectively) or lead revisions (5.8% and 4.4%) was markedly lower in larger studies with >1000 patient-years of cumulative follow-up (3.9 and 3.4% for infections and 5.8 and 4.4% for lead revision) compared to smaller studies with <100 patients and/or <250 patient-years of follow-up (4.7 and 5.0% for infections and 7.7 and 8.3% for lead revision). These data were in agreement with a lower incidence of infections and lead revisions per patient-years. The rates for infections and lead revisions were always lower with 'pooled' analysis as opposed to 'non-pooled' analysis. This also indicates that infection and lead revision rates differed depending on study size with higher rates in series with fewer patients and shorter follow-up. Notably, the risk of infection with complete or partial hardware removal was comparably high in monitored trials (>5% for all assessments).

Conclusions

We propose three categories for the assessment of DBS surgery and hardware-related AEs for the following advantages: unequivocal definition of AEs; coverage of the most relevant complications from a patient's perspective; possibility of accurate retrospective assessment; suited for the assessment of different studies as performed in this meta-analysis. All hardware-related complications should also be expressed in patient-years serving as the best common denominator for in-between study comparison. With the proposed approach valid data can be gathered at any time and such data are suited for proper patient counseling by treating centers. 

Background: The best surgical procedure for subthalamic nucleus (STN) deep brain stimulation (DBS) implantation remains a subject of debate.

Objectives: To study the impact of the absence of awake clinical evaluation during STN-DBS image guided and verified on parameters of stimulation and clinical outcome for Parkinson’s disease (PD) patients.

Methods: 13 PD patients who underwent bilateral STN-DBS image guided and verified under general anaesthesia (GA) with a minimal intraoperative evaluation (side effects only), were compared to 10 patients operated under local anaesthesia (LA) with a complete testing. The primary endpoint was the therapeutic window between the mean threshold of intensity for motor improvement and the mean threshold of intensity for stimulation side effects, on the active contacts at 1 year post-operative, expressed in volts (V). Motor scores were also measured.

Results: The mean therapeutic window on the right STN was 2.06 V in the LA group and 2.4 V in the GA group (p= 0.316). For the left STN, it was 2.06 V in the LA group vs 2.16V in the GA group (p=0.811).The Unified PD rating scale III score in the “off drug-on stim” condition improved at short term in both groups (40.3 % in the LA group and 49% in the GA group), with no significant difference between the 2 groups (p= 0.336). UPDRS III  " on stim - on drug" improved from 70.7% in the GA group and 57% in the G group (p = 0.36).

Conclusion: Asleep, image-guided and verified STN-DBS provide the same motor results as awake surgery and at the same time maintain the therapeutic window, if pre and intraoperative imaging and surgical techniques enable an accurate STN identification and positioning of the DBS lead.

Introduction

Deep brain stimulation (DBS) of the subthalamic nucleus (STN) is advocated in patients with advanced Parkinson’s disease (PD). Microelectrode recording (MER) is one of the intraoperative targeting modalities performed in conjunction with stimulation. In addition, the occurrence of β oscillations of the local field potentials (LFP) has been suggested as another targeting modality. The goal of this observational study was to evaluate which intraoperative neurophysiological markers are most predictive of the STN-DBS efficacy.

Methods
Thirty-nine consecutive patients with PD and with a follow up of to at least one year post surgery were included. The efficacy of STN-DBS was evaluated using the MDS-UPDRS part III (score OFF antiparkinsonian medication and ON DBS at one year, versus preoperative OFF antiparkisonian medication).

MER were recorded and intraoperative stimulation thresholds were determined for the therapeutic effects on rigidity and for the internal capsule stimulation ( Neurostar, GE; 130Hz, 60µs). LFP were recorded from the macrocontact of the microelectrodes (FHC, USA) at the time it was positioned for stimulation (MR plus, GE). β power, computed from LFP [11-31Hz], recorded at the site chosen for stimulation were compared to that recorded above the STN. Coefficients of determination (R²) were computed to analyze which proportion of the variance of the STN-DBS efficacy could be explained by these neurophysiological markers.

Results

None significant linear regression was found between the STN-DBS efficacy and the stimulation thresholds of either the effects on rigidity, the magnitude of rigidity improvement, the threshold of internal capsule stimulation or the stimulation range (R²=0.03, p=0.34; R²=0.09; R²=0.09, p=0.07; R²=0.07; p=0.16; respectively). Again none significant linear correlation was found between the STN-DBS efficacy and the lenght over which STN cells were recorded (R²=0.03, p=0.33).

On the contrary, linear regression was significant between the change in β power and the STN-DBS efficacy (R²=0.35, p<0.05). To better analyze the role of b oscillations, additional sigmoidal regression was performed between the change in β power and the STN-DBS efficacy (R²=0.75).

Discussion and Conclusion

Here intraoperative stimulation thresholds had no determinant predictive value of the STN-DBS success. New directional electrodes, allowing adaptation of the electrical field dedicated to avoid stimulation of the internal capsule, should reduce again the role of intraoperative stimulation. Changes in β power accounted for 75% of the variance of the STN-DBS efficacy, confirming the major interest of this intraoperative neurophysiological marker, which could contribute to simplify intraoperative procedure and improve patients comfort.

Acknowledgements

This work was supported by the Swiss National Science Foundation [SNF grants no. 320030-149804].
None of the authors has any conflict of interest to disclose.

Impossible to past the abstract here for some reason. The abstract is attached below as a file.