Parkinsonism and Related Disorders 20S1 (2014) S192–S196

Contents lists available at SciVerse ScienceDirect

Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis

Deep brain stimulation: new techniques Marwan Hariz a,b, * a Unit

of Functional Neurosurgery, UCL Institute of Neurology, Queen Square, London, UK of Clinical Neuroscience, Ume˚ a University, Ume˚ a, Sweden

b Department

article info

summary

Keywords: Deep brain stimulation Techniques Innovations MRI Parkinson’s disease Stereotactic surgery

The technology of the hardware used in deep brain stimulation (DBS), and the mode of delivering the stimulation have not significantly evolved since the start of the modern era of DBS 25 years ago. However, new technology is now being developed along several avenues. New features of the implantable pulse generator (IPG) allow fractionation of the electric current into variable proportions between different contacts of the multi-polar lead. Another design consists in leads that allow selective current steering from directionally placed electrode contacts that would deliver the stimulation in a specific direction or even create a directional shaped electric field that would conform to the anatomy of the brain target aimed at, avoiding adjacent structures, and thus avoiding side effects. Closed loop adaptive stimulation technologies are being developed, allowing a tracking of the pathological local field potential of the brain target, and delivering automatically the stimulation to suppress the pathological activity as soon as it is detected and for as long as needed. This feature may contribute to a DBS therapy “on demand”, instead of continuously. Finally, advances in imaging technology are providing “new” brain targets, and increasingly allowing DBS to be performed accurately while avoiding the risks of microelectrode recording. © 2013 Elsevier Ltd. All rights reserved.

1. Introduction Since the start of the modern era of deep brain stimulation (DBS) more than 25 years ago, not much has happened in terms of new technology that can be readily used in day-to-day practice to deliver the electric current to the brain target. The DBS hardware and the mode and pattern of stimulation have remained essentially the same: a quadripolar lead with 1.5 mm electrodes with an inter-space of 1.5 mm or 0.5 mm, is implanted into a brain target and an implanted pulse generator (IPG) delivers continuous stimulation resulting in a spherical shape of the electric field around the electrodes [1]. Electrical parameters for chronic stimulation (frequency, voltage, pulse width and polarity) are decided based on screening of the four electrode contacts for effect and side effects, which is sometimes a laborious exercise for the programming clinician and the patient, and may need incremental repetitive adjustments after surgery, over periods of weeks and months in some patients. Even surgical targeting has remained essentially the same, relying on a stereotactic frame, conventional MR and/or CT imaging, and in most centres relying also on multiple microelectrode explorations of the brain target. In fact, rather than * Correspondence: Professor Marwan Hariz, Simon Sainsbury Chair of Functional Neurosurgery, Institute of Neurology, Box 146, Queen Square, London WC1N 3BG, UK. Tel.: +44 7985 642 852; fax: +44 20 7419 1860. E-mail address: [email protected] (M. Hariz). 1353-8020/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.

technological advances, what has taken priority starting at the turn of the century was the expansion of DBS indications beyond Parkinson’s disease (PD) and other movement disorders (dystonia, essential tremor), and beyond the classical brain targets (thalamus, pallidum, subthalamic nucleus), towards DBS of “novel” targets for surgical treatment of psychiatric, behavioural, cognitive and other brain disorders [2]. Real technical innovations of the DBS hardware have lagged behind. Most technical “innovations” in DBS that have received American FDA or European CE approval for routine use in the last decade have been mostly marginal and more or less cosmetic, such as a new burrhole anchoring device for the DBS lead, lower profile of connector cables, a double channel IPG, and IPGs that can be recharged and that can deliver a constant current rather than a constant voltage stimulation. The latter feature, that is, the delivery of constant current (milliamp) stimulation rather than the usual constant voltage stimulation, has been reported in one study detailing the clinical results of DBS using the St-Jude constant current DBS device (St Jude Medical, Plano, Texas, USA) in patients with PD [3]. It showed that constant current stimulation was efficacious, and that the clinical outcome was similar to previous studies that have used constant voltage-based stimulation. It is only in the last couple of years that the field of DBS has been witnessing a surge in true technological innovation. New electrode designs and new patterns of stimulation are being introduced. In parallel, new modalities for brain imaging and new targeting techniques are being developed. Most of these

M. Hariz / Parkinsonism and Related Disorders 20S1 (2014) S192–S196

innovations, summarised in this paper, are still in the pipeline or at the early stages of proof of concept. 2. Techniques related to new DBS hardware and new patterns of stimulation DBS is performed in tiny deep subcortical structures that are anatomically, functionally and literally crowded with neurons and axons. These brain structures mediate symptoms that are to be alleviated by the DBS but the therapy entails often more or less unwanted side effects due to propagation of the electric current beyond the specific target aimed at, even if the DBS lead is quite accurately placed. Slurred speech for example is one of the main side effects of stimulation in the subthalamic nucleus almost regardless of how accurate is the location of the electrode in the target [4]. Also, the three-dimensional configuration of the subcortical brain target is not spherical, nor does it have a regular shape, that would match the spherical or ellipsoid shape of the electric field generated by the DBS electrode. Since the relevant brain anatomy is not shaped like a ball, and it is anatomically irregular with a mixture of neurons and axons than can react differently to electrical stimulation, there is a need to shape the electric current in such a way as to maximise the benefit and minimise the unwanted stimulation-induced side effects. Various technical strategies are being explored that would allow a stimulation using an electric field that can be made as conformal as possible to the shape of the structures aimed at, minimizing spill-over of current into adjacent structures, and thus theoretically avoiding side effects while maintaining effect. 2.1. “Interleaving” stimulation mode This is a novel and readily available feature in the currently used generation of FDA- and CE-approved IPGs, such as Medtronics “PC” or “RC” brands of IPGs (Medtronic, Minneapolis, Minnesota, USA). An interleaving mode allows the independent and alternated stimulation of 2 contacts of the quadripolar DBS lead with different values for voltage and pulse width, but with the same frequency. This mode allows stimulation of adjacent anatomical structures with different energies, and the shape of the resulting electric field will vary accordingly. Interleaved stimulation mode is used when anatomically adjacent targets need to be stimulated at different amplitudes, when classical monopolar, double monopolar of bipolar stimulations fail to provide the desired effect and/or when there are side effects from too high a stimulation amplitude in the vicinity of the area where stimulation at same amplitude is efficient on the symptoms. This technique has been reported to be successful in individual patients stimulated in the subthalamic nucleus (STN) for PD [5], in the globus pallidus internus (GPi) for dystonia [6] and in the STN and the ventrolateral thalamus in a patient with both PD and essential tremor [7]. Some drawbacks of this technique are that it is rather laborious for the clinician programming the stimulation, and it increases the battery drain. Also, the frequency of the stimulating current cannot be adjusted independently for each electrode contact. Additionally, apart for the anecdotic observations reported above, there are no conclusive studies showing its benefit. Finally, the fact that the electric field, even with interleaving stimulation, still cannot be steered in a predetermined direction to avoid side effects, showed that there is a need to develop new electrode designs that permit more focussing and more shaping of the current into the brain target. 2.2. DBS device with multiple source constant current A newly developed DBS device from Boston Scientific, called “Vercise” (Boston Scientific Corporation, Natick, Massachussets,

S193

USA) has recently received European CE approval. This rechargeable device allows delivery of a multiple source constant current with possibility to allocate completely different stimulation parameters independently to each of the eight contacts on the same lead. This would result in an electric field that can theoretically be tailored to the stimulated brain structure, by applying various amplitudes, frequencies and pulse widths to different electrode contacts in the target area. Additionally, this device allows stimulation with pulse widths below what is available with the other established brands of DBS in use today, i.e., below 60 ms. Two multicentre trials with the Vercise device are underway in patients with PD: one aims to evaluate if the therapeutic window of the stimulation can be increased by using shorter pulse widths than the classical 60 ms [8,9]; the other trial will investigate if the flexibilities in variation of stimulation parameters at different electrode contacts allowed by this device would improve outcome while decreasing side effects [10]. Additionally, this device has been used in a patient with phantom limb pain who was successfully stimulated using a trajectory that placed one electrode contact in the parafascicular thalamic nucleus delivering stimulation at 132 Hz and two contacts in the periventricular–periaqueductal grey matter delivering a 10 Hz stimulation [11]. An important drawback of this system is that the electrodes are not MRI-compatible. Also, as with interleaving, the fact remains that the current delivered will still encompass the tissue around the whole circumference of the electrode contact, without possibility for true field shaping and true selective directional steering. 2.3. Current steering and field shaping The need for true current steering and true electric field shaping arises from the need to conform the electric field to the variable anatomy in the brain target of interest and to circumvent side effects when the electric field affects structures adjacent to the target. For example, if a STN DBS lead is in the STN, but happens to lie close to its lateral border, the electric field around the electrode will be partly into the STN and partly into the internal capsule. Similarly, if the DBS electrode is into the posteroventral GPi but too close to its medial and posterior aspects, the internal capsule will receive the same amount of stimulation as the GPi proper. Motor and other side effects will occur especially at higher stimulation levels that would be eventually needed to control the symptoms that motivated the DBS surgery in the first place. Such side effects may not be readily detectable at surgery, and may mitigate in the long run the benefits of DBS. One way to circumvent this nuisance is to design a lead with electrode contacts that are split in two, three or even four parts along the circumference of the electrode, with each part being able to be stimulated specifically in a selective and predetermined cardinal direction. With this concept, an electrode in the STN or the GPi lying close to the internal capsule, as in the examples above, can be made to deliver the electric current only towards the STN or towards the GPi (like a spotlight) without affecting the internal capsule lying in the opposite direction. This is what is meant by current steering, also called directional electrode design. This concept has been contemplated for some time and has been computer simulated in three cases of ventral intermediate (Vim) thalamic DBS showing the possibility to avoid paresthesias by steering the current anteriorly away from the posteriorly adjacent sensory thalamus and towards the Vim proper [12]. Recently, two designs of directional stimulation electrode have been tried intraoperatively in few patients: – The “directSTIM” lead, manufactured by Aleva Neurotherapeutics (Lausanne, Switzerland), consists of a lead with four rings, where each ring consists of three independent electrodes with three different orientations allowing independent stimulation in any of the three directions (Fig. 1). Pollo et al. recently presented an

S194

M. Hariz / Parkinsonism and Related Disorders 20S1 (2014) S192–S196

Distal end of the directSTIM lead (Aleva Neurotherapeutics SA) Ø 1.3 mm

1.5 mm 0.5 mm

1.0 mm

Electrode 7 (tri-contact) Electrodes 4, 5, 6

0° 2 1 2 ode ctr e l E

Electrodes 1, 2, 3 Electrode 0 (tri-contact)

0° Electrode 1

Ele 2 4 0 ctr ° od e

3

Fig. 1. Design of a quadripolar DBS lead called “directSTIM” where each electrode pole is divided into three independent compartments that can deliver directionally the electric current in a given direction, so called current-steering. (Reproduced with permission from Aleva Neurotherapeutics SA, Lausanne, Switzerland.)

implementation of this device in four patients, three undergoing DBS in the STN and one in the Vim thalamus, who were assessed during intra-operative stimulation by a neurologist blinded to the mode of stimulation. It was shown that this device may allow increase in the therapeutic window with lower stimulation thresholds for effects and an avoidance of side effects, compared with stimulation with a regular electrode [13]. – The “SureStim” lead is another lead design, manufactured by Sapiens (Eindhoven, The Netherlands). It consists of 32 contacts distributed evenly around the circumference of the lead (Fig. 2). The contacts can be activated group-wise, resulting in both a directional current and a shaping of the electric field [14,15]. This device has been tested intra-operatively and was shown to also allow simultaneous recordings of local field potentials (LFP) across the brain target aimed at without need to move the electrode. The spatial information thus obtained from the recorded clusters of neurons may help predicting the steering of the electric current in the appropriate direction as well as the the shaping of the electric field to encompass the region of interest [16]. anterior

lateral

medial

CI

STN ZI posterior Fig. 2. Schematic view of the multipolar Sapiens lead called “SureStim”and a directional electric field allowing the electric field to be restricted into the target structure without spilling over laterally. STN, subthalamic nucleus; ZI, zona incerta; CI, capsula interna. (Reproduced with permission from Sapiens, Eindhoven, The Netherlands.)

2.4. Closed loop and adaptive stimulation Typical DBS is delivered continuously, that is during 24 hours (except in Vim DBS for tremor where the patient is asked to switch off stimulation during the night). In patients with advanced PD, who have DBS in the STN or the GPi to alleviate not only tremor but also and mainly akinesia, rigidity and motor fluctuations, the stimulation is kept on 24/7. However, these patients as a rule will still show fluctuations, albeit not as severe as preoperatively. This means that the patients will still have periods of the day with quite good mobility when medication is working – along with the stimulation –, and periods with less good mobility when medication is not working as good, but stimulation is still helping. These moment-to-moment fluctuations can also result from variable psychological and cognitive states of the patient in his/her everyday life. A question that has been debated is: do patients with advanced PD still need the stimulation activated even during the good periods during the day when their medication is working optimally and they feel “on”? An interesting experiment published by Chen et al. a few years ago [17] showed that when patients with STN DBS are in a good “on” period, stimulation during that period may not be necessary and in fact may even affect negatively, ever so slightly, bradykinesia. Also, it has been established through LFP recordings from the DBS leads, that during clinically off periods (with predominant akinesia and rigidity), there is a predominant beta activity in the STN, which disappears when the patient is clinically on, that is, when medications are working. These observations have opened the field for investigating whether it would be possible to deliver stimulation only when needed, that is “on demand”, with a device capable of recording continuously the LFP around the electrode and that will automatically activate itself to deliver the therapeutic stimulation as soon as beta activity is detected, and until beta activity subsides. This is the concept of the so-called “closed loop” or “adaptive” stimulation. The advantages would be that DBS therapy will be delivered only when needed, to avoid tolerance, and to save battery life. This is what has been done very recently by Little et al. [18], using a brain–computer interface system that tracks the beta activity recorded from the implanted conventional STN DBS electrode, and activates the delivery of stimulation that will then be automatically stopped when beta activity is no longer detected. The authors compared this so-called adaptive stimulation mode to continuous stimulation mode in 8 patients with externalised DBS leads and showed that adaptive stimulation was clearly more efficient on cardinal symptoms, and this was achieved at lower stimulation energies than in continuous stimulation mode. This exciting proof of principle has been verified “acutely” only in the very early postoperative period using custommade external non-implantable LFP detectors and stimulators. If this technique proves to be useful and efficacious on the symptoms of advanced PD also in the longer term and in the chronic situation, and if the hardware and the technology become fully implantable, it would be the first true revolution since the start of the modern era of DBS, and DBS would become a truly physiological, intelligent and “natural” way to treat malfunctions of brain circuitries, making functional neurosurgery functional in the full meaning of the word. 3. Techniques related to brain imaging in view of DBS and targeting In parallel with technological advances of the DBS hardware itself and in parallel with development of new patterns and modes of delivering the stimulation, technologies related to the surgical targeting for DBS are also witnessing new advances, related mainly to imaging of the brain target and to improving the marriage between tools used for targeting and tools used for imaging [19].

M. Hariz / Parkinsonism and Related Disorders 20S1 (2014) S192–S196

Functional imaging is an example of a dynamic physiological imaging technique that has already provided “targets” for DBS in cluster headache (the posteromedial hypothalamus) and in depression (the subgenual cingulum). Advances in visualisation of connectivities between various brain structures using diffusion tensor imaging (DTI) and tractography may prove to be a very useful tool to refine targeting within one structure [20]. DTI has also contributed to suggest “new” and potentially more appropriate targets for trials of DBS in other illnesses than movement disorders. An example is the targeting of the medial forebrain bundle for depression, based on its connectivity with the nucleus accumbens and the prefrontal cortex [21]. Tractography can also guide targeting of somatosensory fibres in DBS for chronic pain [22], and can visualize the lateral boundary of the thalamus, thus helping in defining the laterality of the individual target in the thalamic Vim nucleus and avoid placement of the DBS lead in the internal capsule [19]. The potential of these imaging modalities in unveiling brain dysfunctions as well as the circuitries underlying these dysfunctions, will undoubtedly contribute to increased use of DBS to treat a variety of neurological, psychiatric and behavioural disorders. Advances in structural imaging using high Tesla MRI will contribute to visually discriminate between various components of basal ganglia and thalamus that are or may be subjected to DBS [19,23]. The refinement of DBS targeting based on imaging is in fact already readily performed on conventional 1.5 Tesla MR machines using appropriate scanning sequences [24,25], and has allowed safe and efficient DBS surgery under general anaesthesia [26]. It is predicted that as the imaging becomes better, more sophisticated and adapted to the target aimed at, this will contribute to abandonment of the laborious, expensive and potentially dangerous techniques of multiple microelectrode penetrations during surgery [19,27]. Finally, the possibility to perform DBS surgery in a dedicated MRI suite, or even into the MRI bore proper [27], is certainly an advance in so far as it allows a real time, or near real time anatomical stereotactic verification of DBS targeting [19,27,28], and immediate correction of the lead location if needed. MRI-guided and MRI-verified DBS has already proven its exquisite safety and efficacy [19,24,28], and is gaining momentum worldwide. 4. Miscellaneous There are additional technical advances in DBS that merit mentioning, and that may gain popularity once their usefulness and versatility are proven. A couple of these are briefly mentioned here: – Chabardes ` et al. recently pioneered an intraventricular approach to DBS to stimulate targets lying very close to the wall and floor of the third ventricle [29]; in five patients with cluster headache, the electrode was inserted into the third ventricle and made to lie on its floor in close proximity to the posteromedial hypothalamus. Stimulation via this route resulted in significant pain relief in four patients. – Another French team has demonstrated that a DBS lead into the STN can be used in conjunction with a radiofrequency machine to generate a lesion around any of the 4 electrode contacts of that lead. This feature can be used for incremental lesioning through the DBS lead and in case of infection to perform a subthalamotomy before explanting the lead, thus allowing the patient to maintain some benefit on PD symptoms [30]. 5. Conclusions With the extension of DBS indications beyond movement disorders to several areas of neurology and psychiatry, the future of deep

S195

brain stimulation as a therapeutic concept – and as a research tool – is bright. Current DBS technology is hampered, among other issues, by limited flexibility and by the difficulty to steer and shape the electric field to maximize effect and decrease side effects. Therefore, there is a need for DBS companies to develop new hardware that will permit flexibility in shaping the electric current according to the individual brain target. Most of the technical innovations mentioned above, especially those related to new DBS hardware and modes of delivering the electric current, are still in the pipeline or at the early stage of proof of concept, but nevertheless, they represent long-awaited qualitatively innovative technical advances that may contribute to improve the outcome of deep brain stimulation at many levels. Tremendous efforts remain to be done by clinicians in testing the new DBS hardware to demonstrate its usefulness and its advantages compared to what is available today. Since it is evident that MR imaging is becoming increasingly used both to guide DBS lead placement and to verify targeting accuracy postoperatively, manufacturers of new DBS systems would do well to ensure compatibility of the hardware with MRI, at least with the conventional 1.5 T MRI machines. This is today possible and quite safe with the currently used DBS brands (Medtronic, St Jude), on the condition of using a 1.5 T MR machine, a low SAR value and a transmit–receive head coil [28]. If the new brands and new designs of DBS leads that permit current steering and shaping of electric field prohibit the use of MRI on implanted patients, much of the very idea of providing a DBS that is adapted to, and conformal with, the individual visualized brain target will be lost. Acknowledgements The UCL Unit of Functional Neurosurgery is supported by the UK Parkinson Appeal and the Monument Trust. Conflict of interests The author has occasionally received from Medtronic and St Jude Medical travel expenses and fees for speaking at meetings. References ˚ om [1] Astr ¨ M, Zrinzo LU, Tisch S, Tripoliti E, Hariz MI, Wardell ˚ K. Method for patient-specific finite element modeling and simulation of deep brain stimulation. Med Biol Eng Comput 2009;47:21–8. [2] Krack P, Hariz MI, Baunez C, Guridi J, Obeso JA. Deep brain stimulation: from neurology to psychiatry? Trends Neurosci 2010;33:474–84. [3] Okun MS, Gallo BV, Mandybur G, Jagid J, Foote KD, Revilla FJ, et al. Subthalamic deep brain stimulation with a constant-current device in Parkinson’s disease: an open-label randomised controlled trial. Lancet Neurol 2012;11:140–9. ˚ om [4] Astr ¨ M, Tripoliti E, Hariz MI, Zrinzo LU, Martinez-Torres I, Limousin P, et al. Patient-specific model-based investigation of speech intelligibility and movement during deep brain stimulation. Stereotact Funct Neurosurg 2010;88:224–33. [5] Wojtecki L, Vesper J, Schnitzler A. Interleaving programming of subthalamic deep brain stimulation to reduce side effects with good motor outcome in a patient with Parkinson’s disease. Parkinsonism Relat Disord 2011;17:293–4. [6] Kovacs ´ N, Janszky J, Nagy F, Balas ´ I. Changing to interleaving stimulation might improve dystonia in cases not responding to pallidal stimulation. Mov Disord 2012;27:163–5. [7] Baumann CR, Imbach LL, Baumann-Vogel H, Uhl M, Sarnthein J, Sur ¨ uc ¨ u¨ O. Interleaving deep brain stimulation for a patient with both Parkinson’s disease and essential tremor. Mov Disord 2012;27:1700–1. [8] Steigerwald F, Reese R, Matthies C, Volkmann J. Increased therapeutic window with shorter pulse widths (