Deep Brain Stimulation

Damiaan Denys • Matthijs FeenstraRick SchuurmanEditors

Deep Brain Stimulation

A New Frontier in Psychiatry

123

EditorsDamiaan DenysDepartment of PsychiatryAcademic Medical Center, University of AmsterdamAmsterdamThe Netherlands

and

Neuromodulation and BehaviourNetherlands Institute for NeuroscienceAmsterdamThe Netherlands

Matthijs FeenstraNeuromodulation and BehaviourNetherlands Institute for NeuroscienceAmsterdamThe Netherlands

Rick SchuurmanDepartment of NeurosurgeryAcademic Medical Center, University of AmsterdamAmsterdamThe Netherlands

ISBN 978-3-642-30990-8 ISBN 978-3-642-30991-5 (eBook)DOI 10.1007/978-3-642-30991-5Springer Heidelberg New York Dordrecht London

Library of Congress Control Number: 2012946624

� Springer-Verlag Berlin Heidelberg 2012This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part ofthe material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission orinformation storage and retrieval, electronic adaptation, computer software, or by similar or dissimilarmethodology now known or hereafter developed. Exempted from this legal reservation are briefexcerpts in connection with reviews or scholarly analysis or material supplied specifically for thepurpose of being entered and executed on a computer system, for exclusive use by the purchaser of thework. Duplication of this publication or parts thereof is permitted only under the provisions ofthe Copyright Law of the Publisher’s location, in its current version, and permission for use must alwaysbe obtained from Springer. Permissions for use may be obtained through RightsLink at the CopyrightClearance Center. Violations are liable to prosecution under the respective Copyright Law.The use of general descriptive names, registered names, trademarks, service marks, etc. in thispublication does not imply, even in the absence of a specific statement, that such names are exemptfrom the relevant protective laws and regulations and therefore free for general use.While the advice and information in this book are believed to be true and accurate at the date ofpublication, neither the authors nor the editors nor the publisher can accept any legal responsibility forany errors or omissions that may be made. The publisher makes no warranty, express or implied, withrespect to the material contained herein.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Foreword

Deep brain stimulation (DBS) makes it possible for psychosurgery to get a secondchance. In particular, it does those proud who initiated the method and developedit for new indications. It permitted the rebirth of psychosurgery. But, on the otherhand and at the same time, the reappearance of psychosurgery because of themerits of DBS, also has threatening aspects since one should not forget the darkages of malpractice and overuse of psychosurgery, that led the discipline intodismay and even into oblivion. The advent of psychosurgery was a huge wave ofhope. One could get rid of the plague of mental disorders, affecting hundreds ofthousands of people who, because of lack of treatment as well as understanding,were confined by society to prisons or, camps, or later to more humane places suchas asylums and psychiatric hospitals. Unfortunately, because of the haste ofapplication, the lack of wisdom, the desire to exploit this new medical opportunity,and also the distorted manner in which it was applied to modify the brain and themind of patients, and even of healthy persons, psychosurgery was finally bannedfor the treatment of mental disorders.

For a long period, it was almost obscene to mention terms such as lobectomy,psychosurgery, and even electroconvulsive therapy. The new era, which isannounced by the current book, was possible only because surgeons and psychi-atrists in the USA and Europe kept pursuing the hard work to heal their patients inthe most cautious and meritorious way. The breakthrough came from neurosci-entists using the newly developed functional imaging methods to build up a newmodel to understand the anatomical structures involved in consciousness, themind, and behavior. Presenting a new scheme of the organization of the basalganglia to understand mood and anxiety disorders, is in my view of paramountimportance, similar to the work that has been done for movement disorders. It willbe a milestone on the future road for the treatment of mental disorders.

DBS was an appropriate tool to translate the newly formed anatomofunctionalconcepts into therapy-oriented surgical strategies. The experience acquired in alarge range of indications demonstrates the validity of the functional inhibitionconcept. The long-term follow-up of cohorts of patients will establish the merits ofDBS: precision and adaptability, leading to low morbidity and minimal mortality.

v

Above all, the reversibility of the effects of DBS is important since it minimizesthe risk of irreversible changes of behavior, mood, and personality, which hauntedthe outcome of early lesioning surgery. We now have the right tool, an adequatescalpel, to perform surgery for the treatment of mental disorders again. We musttake advantage of the unique opportunity to precisely locate the right targets, anddoing so to, increase our knowledge of the physiology of the mind and its dis-turbances. We have to do this only for the patients’ benefit. We have to resist thetemptation of achieving spectacular changes or to be driven by technology. Wehave to be extremely cautious in selecting patients, in analyzing indications, and inevaluating the results, and we have to report them carefully to the medical com-munity. We have to be aware of our huge societal responsibility, and being proudof this privilege, not to miss the second chance that DBS offers to treat, and whoknows, to cure, mental illnesses.

Clinatec Institute, CEA, Grenoble, France Alim Louis Benabid

vi Foreword

Preface

Deep brain stimulation (DBS) was introduced in the 1980s for the treatment oftherapy-resistant neurological disorders, and has been applied since 2000 on anexperimental basis for the treatment of therapy-resistant psychiatric disorders.Since its introduction, DBS has evolved into a well-accepted therapy to treatpatients with movement disorders, but the use of electrical stimulation to inten-tionally alter emotion, motivation, and cognition of psychiatric patients oftencauses amazement and even disbelief. Neurosurgery for the treatment of psychi-atric disorders has always been and still is surrounded by controversy.

The relation between DBS and psychiatry is fascinating because it is bothappealing and threatening. First, DBS for the treatment of psychiatric disorders isattractive because it offers an ultimate treatment option for a group of seriously ill,untreatable psychiatric patients. Second, the risk of the operation is relativelysmall and the technique renders the possibility of continual adjustment, which is animportant issue for psychiatric patients. Finally, DBS has the potential to increaseour understanding of the brain pathophysiology of psychiatric disorders; it offers aview into the pathological brain. DBS is also threatening because psychiatricdisorders are less discrete and objectifiable conditions than movement disorders.Second, psychiatric symptoms are more intimately connected with a person’sidentity and integrity than motor symptoms, therefore raising more challengingethical issues. Finally, the boundary between treatment and enhancement in psy-chiatry is vague. Altering cognition, emotion, and motivation is an intended goal inpsychiatry and not a side effect, and may result in changes beyond the natural self.

In the past decade, DBS has been applied in obsessive–compulsive disorder,major depressive disorder, Tourette syndrome, and addiction. The results haveconsistently shown a promising success rate. However, the number of patientstreated world-wide is still only limited and most reports deal with small-scalestudies or case reports. Moreover, little is still known about how DBS acts inpsychiatry, emphasizing the need for translational animal studies.

The purpose of this book is to conduct the first comprehensive overview of DBSin psychiatric disorders, with a particular emphasis on the relation between pre-clinical animal studies and clinical patient studies. The book starts with the basic

vii

principles of stimulation (Chap. 1), neuroanatomical circuits (Chap. 2), andhypotheses of the mechanism of action (Chap. 3). Separate chapters subsequentlyreview DBS in different psychiatric disorders and animal models: obsessive–compulsive disorder (Chaps. 4–7), major depressive disorder (Chaps. 8–11),Tourette syndrome (Chap. 12), addiction (Chaps. 13, 14), and psychiatric symp-toms in Parkinson’s disease (Chaps. 15, 16). We have also included a discussionon the role of intracranial recordings (Chap. 17), neurotransmitter changes(Chaps. 18), glial cells (Chap. 19), the significance of animal studies (Chap. 20),neuroimaging (Chap. 21), and optogenetics (Chap. 22). The future of next-gen-eration electrodes (Chap. 23) and nanotechniques (Chap. 24) is reviewed, and weend with a discussion of ethical issues of DBS in psychiatric disorders (Chap. 25)and a critical review of the history of DBS (Chap. 26).

We thank the authors, all experts in their field, for their excellent contributionsto this compendium of DBS in psychiatric disorders. We greatly appreciate theeditorial work of Renske van Dijk in the creation of this book. We sincerely hopethat this compilation of present-day knowledge will contribute to increasedunderstanding across the boundaries of separate specialties and research areas, andthat it may be of help in guiding future steps for all those involved to advance theknowledge and application of DBS in psychiatric disorders.

Damiaan DenysMatthijs FeenstraRick Schuurman

viii Preface

Contents

1 Basic Principles of Deep Brain Stimulation . . . . . . . . . . . . . . . . . 1F. L. H. Gielen and G. C. Molnar

2 Neural Circuits Affected by Deep Brain Stimulationfor the Treatment of Psychiatric Disorders . . . . . . . . . . . . . . . . . 11Suzanne N. Haber and Benjamin D. Greenberg

3 Mechanisms of Action of Deep Brain Stimulationfor the Treatment of Psychiatric Disorders . . . . . . . . . . . . . . . . . 21J. Luis Lujan and Cameron C. McIntyre

4 Deep Brain Stimulation in the Ventral Capsule/Ventral Striatumfor the Treatment of Obsessive–Compulsive Disorder:Role of the Bed Nucleus of the Stria Terminalis . . . . . . . . . . . . . 35Loes Gabriëls and Bart Nuttin

5 Deep Brain Stimulation in Obsessive–Compulsive DisorderTargeted at the Nucleus Accumbens . . . . . . . . . . . . . . . . . . . . . . 43Pelle P. de Koning, Pepijn van den Munckhof, Martijn Figee,Rick Schuurman and Damiaan Denys

6 What is the Role of the Subthalamic Nucleusin Obsessive–Compulsive Disorder? Elements and Insightsfrom Deep Brain Stimulation Studies . . . . . . . . . . . . . . . . . . . . . 53William I. A. Haynes and Luc Mallet

7 Obsessive–Compulsive Disorders in Animals . . . . . . . . . . . . . . . . 61Christine Winter

ix

8 Subcallosal Cingulate Cortex Deep Brain Stimulationfor the Treatment of Refractory Mood Disorders:Evidence and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71Peter Giacobbe, Nir Lipsman and Andres M. Lozano

9 Deep Brain Stimulation of the Human Reward Systemas a Putative Treatment for Refractory Major Depression . . . . . . 81T. E. Schlaepfer, V. A. Coenen and B. H. Bewernick

10 Depression in Humans: The VentralCapsule/Ventral Striatum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95Mayur Pandya, Andre Machado and Donald Malone

11 Deep Brain Stimulation in Animal Models of Depression . . . . . . . 103Brian W. Scott, José N. Nobrega and Clement Hamani

12 Deep Brain Stimulation in Tourette Syndrome . . . . . . . . . . . . . . 113L. Ackermans, I. Neuner, J. Kuhn and V. Visser-Vandewalle

13 Surgical Treatments for Drug Addictions in Humans. . . . . . . . . . 131Bomin Sun and Wei Liu

14 Manipulating Addictive Behaviour in Animal Models . . . . . . . . . 141Rolinka M. C. Schippers, Tommy Pattij and Taco J. De Vries

15 Neuropsychiatric Side Effects of Deep Brain Stimulationin Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159Christine Daniels and Jens Volkmann

16 Psychiatric Aspects of Parkinson’s Disease in Animal Modelsof Deep Brain Stimulation of the Subthalamic Nucleus . . . . . . . . 175S. K. H. Tan, H. Hartung, V. Visser-Vandewalle,T. Sharp and Y. Temel

17 Scientific Recordings in Deep Brain Stimulation . . . . . . . . . . . . . 183Michael X. Cohen

18 Neurotransmitter Release During Deep Brain Stimulation . . . . . . 193Osama A. Abulseoud, Emily J. Knight and Kendall H. Lee

19 The Potential Role of Nonneuronal Cells in the DeepBrain Stimulation Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . 205Vinata Vedam-Mai, Michael S. Okun and Elly M. Hol

x Contents

20 Animal Studies in Deep Brain Stimulation Research . . . . . . . . . . 217Matthijs G. P. Feenstra and Damiaan Denys

21 Neuroimaging Deep Brain Stimulationin Psychiatric Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225Martijn Figee, Pepijn van den Munckhof,Rick Schuurman and Damiaan Denys

22 Optogenetic Strategies for the Treatment of NeuropsychiatricDisorders: Circuit-Function Analysisand Clinical Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241Daniel L. Albaugh and Garret D. Stuber

23 Next-Generation Electrodes for Steering Brain Stimulation . . . . . 253H. C. F. Martens, M. M. J. Decré and E. Toader

24 Future Applications: Nanotechniques . . . . . . . . . . . . . . . . . . . . . 263Russell J. Andrews, Jessica E. Koehne and Meyya Meyyappan

25 Ethical Guidance for the Use of Deep Brain Stimulationin Psychiatric Trials and Emerging Uses:Review and Reflections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273Emily Bell and Eric Racine

26 History of ‘‘Psychiatric’’ Deep Brain Stimulation:A Critical Appraisal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289Marwan I. Hariz

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295

Contents xi

Chapter 1Basic Principles of Deep BrainStimulation

F. L. H. Gielen and G. C. Molnar

1.1 Introduction

All evidence-based understanding regarding the principles of neurostimulation includingdeep brain stimulation (DBS) is based upon basic biophysical, electrochemical, andneurophysiological concepts. Scientific evidence for these concepts, although relativelyold, provide a solid technical foundation regarding how to perform DBS (Ranck 1975;Merrill et al. 2005; Kuncel and Grill 2004; Agnew et al. 1990; Durand 2000; Rattay 1989;Tehovnik 1996; Rushton 1927; Holsheimer 2003). Undoubtedly the most strikingprogress in neuromodulation using DBS can be attributed to the enormous progress inanatomical, functional, and network visualization provided by MRI techniques. Only25 years ago, all DBS implants were performed using ventriculography, which visualizedonly two landmarks in the three-dimensional brain: the anterior commissure and posteriorcommissure. Via a short period of CT-based DBS targeting in the first half of the 1990s,state-of-the art DBS targeting is now based upon ever better and more revealing MRItechniques. Once the lead has been implanted in the patient’s brain, the device must beprogrammed to identify the optimal stimulation parameters that provide the most clinicalbenefit, the least amount of side effects, and ideally, utilize the lowest energy. This processis made easier with knowledge of the patient’s brain anatomy, stimulation-induced sideeffects of nearby structures, the lead trajectory, and basic concepts of extracellularstimulation. In this chapter the basic biophysical, electrochemical, and neurophysio-logical concepts pertinent to extracellular stimulation will be reviewed.

F. L. H. Gielen (&)Medtronic Bakken Research Center, Endepolsdomein 5,6229 GW, Maastricht, The Netherlandse-mail: frans.gielen@medtronic.com

G. C. MolnarMedtronic Inc, Rice Creek East 280, Minneapolis, MN 55432-3568, USAe-mail: gabi.c.molnar@medtronic.com

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_1, � Springer-Verlag Berlin Heidelberg 2012

1

1.2 Main Biophysical, Electrochemical,and Neurophysiological Concepts in DBS

Neuromodulation by DBS is a result of electrical currents that flow into and out ofneurological substrates, including cells, axons, dendrites, and glial cells, leading topolarization of these elements. The current is generated by a pulse generator and isdelivered to the tissue via electrodes implanted in the brain. The amount of currentdelivered to the tissue in a constant-voltage system is mainly determined by theelectrode impedance (Eimp), which includes the complex transition between themetal of the stimulation electrode and the immediately surrounding neurologicaltissue (Schwan 1992). As a result of engineering optimization of wires in a DBSsystem, one can disregard the resistance of these wires for the stimulation currentsin a first-order approximation. The main factors affecting Eimp are the electricalproperties of the electrode encapsulation and the bulk tissue medium (Butson et al.2006). Eimp plays a major role in the electronic design of a DBS system, especiallyfor determining device longevity. It is also an important factor in determining thesafety of chronic DBS; in fact, the stimulation pulse shape and the amount ofelectrical charge that passes through the surface of the stimulation electrode are themain parameters that determine the safety of chronic DBS.

1.3 Stimulation Pulse Shape

At the start of the era of long-term neuromodulation, electrical stimulators usedmonophasic waveforms. However, Lilly et al. (1955) determined that the long-term exposure of tissue to direct current will damage tissue. They proposed using astimulation pulse that had two phases of current flow, resulting in zero net flow ofcharge. Charge-balanced stimulation pulses have become the standard method ofdelivering stimulation pulses to excitable tissue in long-term clinical applications.The purpose of the reversal phase during biphasic stimulation is to reverse thedirection of the electrochemical processes that occurred during the stimulationphase to avoid tissue damage resulting from the accumulation of toxic products ormodifications of tissue pH (Merrill et al. 2005). Such pulses can have manydifferent pulse shapes, but in clinical applications some variation of biphasicrectangular pulse shapes is common.

1.4 Charge Density

Over the past 40 years research has identified the requirements for safe chronicelectrical stimulation in biological tissues such as neuronal tissue. A vital findingwas that charge-balanced stimulation pulses alone are not enough to ensure safechronic electrical stimulation. McCreery et al. (1990) found that charge density(CD) interacted synergistically with the charge per phase to determine the

2 F. L. H. Gielen and G. C. Molnar

threshold of stimulation-induced neural injury in cat cortex. Using the data fromMcCreery, Shannon (1992) derived an equation defining the boundary betweensafe and unsafe charge and CD levels:

log CD ¼ k � log Q

where CD is in microcoulombs per square centimeter per phase and Q is thecharge in microcoulombs per phase. The boundary occurs at approximatelyk = 1.85. From basic physics it is known that CD is not constant over the entiresurface area of a stimulation electrode, especially near the edges of the electrode.However, no clinically relevant quantitative information exists concerning thedistribution of damage over the electrode–tissue interface. In DBS, the CD level atwhich a warning is elicited by the clinician programmer is 30 lC/cm2 per phase,which corresponds to a value of k of approximately 1.7 that lies entirely within thesafe zone as describe by Shannon.

Postmortem studies indicate that there is minimal tissue damage associated withchronic DBS. Mechanical insertion of the DBS lead is associated with a classicforeign body response characterized by the presence of a fibrous tissue capsulesurrounded by a region of gliosis (Grill 2005). However, all these studiesemployed stimulation parameters well below the CD limit. So far no clinicallyrelevant irreversible and potentially negative effects of chronic high-CD DBS invarious neurological human substrates have been reported.

Published chronic high-CD DBS applications which offer therapeutic benefitsmay indicate that the appropriate functional brain target has not been found and thatin fact the implanted electrodes are too far from the optimal target since such highcharge must be delivered to achieve clinical benefit. Therefore, ongoing research infinding the ‘‘sweet spot’’ brain targets in currently high-CD DBS therapies may alsolead to lower-CD DBS without loss of therapy efficacy. Recently published diffusiontensor imaging (DTI) tractography applications have already shown that this tech-nique may be helpful in refining patient-specific DBS targets, potentially resulting inlower CD (Coenen et al. 2009, 2011a, b). This assumes that there is a sweet spot thatmay be modulated by the stimulation and that it is not necessary for a larger amountof neuronal tissue to be recruite to obtain the desired effect.

1.5 Current–Distance Relationship

Nerve cells near the electrode are more likely to be activated than neurons locatedfarther away. Research suggests that the current–distance relationship (CDR) canbe estimated by the following mathematical relation (Bagshaw and Evans 1976):

Ith ¼ aþ kD2

where Ith is the threshold current, D is the distance from the electrode, a is thethreshold when the electrode is in direct contact with the neural element, and k isthe strength–distance constant. The value of k for the activation of central nervous

1 Basic Principles of Deep Brain Stimulation 3

system neurons can range between 100 and 4,000 lA/mm2 using 0.2-ms pulses(Tehovnik 1996). This ‘‘constant’’ is actually a function of many parameters,including electrode size, pulse width, tissue impedance, nerve fiber size, andthe nerve membrane properties. This relationship indicates that for a small cathodethe activation threshold of a neuron increases as the square of the distance from theelectrode.

1.6 Strength–Duration Relationship

The strength–duration relationship (SDR) describes the relationship between thestimulation amplitude and pulse width and can be described by the Weiss equation(Weiss 1901):

Ith ¼ Irh 1þ Tch

PW

� �

where Ith is the threshold current, Irh is the rheobase current, PW is the pulse width,and Tch is the chronaxie. The threshold current to excite a neuron increases as thepulse width decreases. Less excitable neuronal elements will have a longerchronaxie compared with more excitable elements. Typical values of chronaxie forlarge myelinated fibers are 30–200 ls, whereas values for cell bodies are in the1–10 ms range (Ranck 1975). The SDR will depend on several factors, includingthe distance between the electrode and the target neurons, the polarity of thestimulus, the waveform of the stimulus, and the fiber diameter. Figure 1.1 showsthe most important implications of the CDR and SDR in DBS clinical practice.The key aspects are:

Fig. 1.1 The activation radius is shown for large (blue) and small (black) fiber diameters withshort (R1, r1, solid lines) and long (R2, r2, dashed lines) pulse widths. Smaller-diameter axonsrequire higher thresholds for activation relative to larger-diameter fibers. For a given fiberdiameter, axons are excited farther away with larger pulse widths

4 F. L. H. Gielen and G. C. Molnar

• Axons farther away from the electrode require higher amplitudes for activation.• Larger axons are stimulated at lower thresholds than smaller axons.• For a particular stimulation amplitude and pulse width, larger-diameter axons

are stimulated farther away from the stimulation electrode than smaller-diameteraxons.

• With increasing pulse widths the difference in activation radius around thestimulation electrode between larger-diameter and smaller-diameter axonsbecomes smaller.

This implies that SDR is an important relationship to potentially differentiatebetween different neurophysiological functions in a mixed structure such as thebrain.

1.7 Stimulation Current Distribution: Monopolarand Bipolar Stimulation

In an electrical circuit, stimulation current must be delivered by means of at least onepositive (anode) and one negative (cathode) stimulation electrode. A negative(cathodal) stimulus causes nearby neurons to depolarize, whereas a positive (anodal)stimulus causes nearby neurons to hyperpolarize. Stimulation current distributionsresulting from at least one cathode and at least one anode that do not interact from aphysiological point of view are by definition called ‘‘monopolar.’’ In clinical practicethis means that when the physical separation between the positive and negativeelectrodes is five to ten times larger than the largest dimension of the smalleststimulation electrode, the stimulation can be considered ‘‘monopolar.’’ For a typicalDBS electrode length of 1.5 mm, stimulation can be considered ‘‘monopolar’’ if thesmallest distance between positive and negative electrode contacts is 7.5–15 mm ormore. In DBS applications the metal housing of the stimulator serves as the anodeduring monopolar stimulation. Electrode configurations with one cathode and oneanode on a lead are considered ‘‘bipolar,’’ and all other distributions of anodes andcathodes may be considered ‘‘multipolar.’’ Monopolar stimulation results in a largercurrent spread than bipolar stimulation for a given stimulation amplitude; clinically,larger amplitudes with bipolar stimulation are required to obtain the same effects aswith monopolar stimulation (Deli et al. 2011).

1.8 Interaction of Stimulation Amplitude,Pulse Width, and Frequency

In addition to the electrode configuration (number and location of anodes andcathodes), there are various parameters that may be controlled that affect thevolume of tissue activated by DBS, including the stimulation amplitude, pulsewidth, and frequency.

1 Basic Principles of Deep Brain Stimulation 5

The main effect of increasing the stimulation amplitude is to increase the numberof neuronal elements that are activated or modulated by the stimulation. Neuronalelements located farther from the stimulation electrode will be activated, resulting ina larger volume of neuronal tissue being modulated as described by the CDR. On thebasis of findings using microelectrode as well as macroelectrode stimulation, e.g., aDBS electrode, a certain minimal volume of brain tissue needs to be stimulated toachieve a noticeable clinical effect. In DBS clinical practice this volume is typicallyin first-order approximation about 2.5 mm around the mathematical center of a DBSelectrode contact at a typical stimulation amplitude of 2.8 V or approximately2.8 mA with an Eimp of 1,000 X (Fig. 1.2). In a first-order approximation the radiusof the spherical activation volume increases proportionally to an inverse quadraticrelation (see Sect. 1.5). In current clinical practice this means that the activationradius is between 4.5 and 5 mm for a stimulation amplitude of 9 V (about 9 mA).These approximations for the spatial extent of stimulation have been confirmed intwo patients using patient-specific DTI to identify the distance between the elec-trode and the fiber pathways responsible for the generation of side effects at specificamplitudes of stimulation (Mädler and Coenen 2012).

The main effect of the stimulation pulse width was described in Sect. 1.6. For agiven applied electric field (defined by the amplitude), the pulse width determineswhich and how many neuronal elements are activated. In DBS, smaller pulsewidths are typically used as larger pulse widths decrease the therapy window(difference in amplitude needed to cause side effects and amplitude needed forclinical benefit) (Rizzone et al. 2001). The use of shorter pulse widths also min-imizes the charge delivered.

In current clinical practice, changes in stimulation frequency above a certainvalue have a relatively minor effect. Typically frequencies above 100 Hz may beused for therapy with no significant advantages of increasing the frequency(Rizzone et al. 2001). Research considering the brain as an assembly of a largenumber of functional networks, each with its own functional and temporal features,will most likely reveal more sophisticated uses of the stimulation frequency andduty cycle. For example, there is increasing evidence from DBS electroderecordings that there is an imbalance of beneficial and pathological oscillatoryactivity within basal ganglia circuits in movement disorders, and that theseoscillations occur at particular frequencies and are modulated by various factors,including movement, medication, and stimulation (Brown et al. 2001; Eusebioet al. 2011; Kuhn et al. 2004). Preliminary evidence indicates that selection ofappropriate stimulation parameters may decrease the relative power of the path-ological oscillations (Eusebio et al. 2011). These types of recordings may provideclues about where and how to provide stimulation for various disease states treatedby DBS, potentially including psychiatric conditions (McCracken and Grace 2009;Lega et al. 2011). In another example, DTI tractography has the appealingpotential to provide patient-specific DBS targeting which is tailored to reestablishor modulate balances in brain network structures. The imbalances of such brainnetworks may be specific for psychiatric diseases such as obsessive–compulsivedisorder, depression, and addiction (Coenen et al. 2012).

6 F. L. H. Gielen and G. C. Molnar

1.9 Different Activation Thresholds for Stimulationof Different Neurological Elements

The basic effects of extracellular stimulation of neuronal elements are summarizedbelow but have also been reviewed extensively elsewhere (Durand 2000; Ranck1975; Rattay 1989; Tehovnik 1996; Rushton 1927; Holsheimer 2003).

Fig. 1.2 Computer modeling of activation area during monopolar stimulation for the Medtronicmodel 3387 and 3389 deep brain stimulation leads

1 Basic Principles of Deep Brain Stimulation 7

• The threshold for stimulation is lower at the cathode (negative electrode) than atthe anode (positive electrode). Stimulation near an anode is still possible, butrequires higher currents than cathodal stimulation. Further, the site of actionpotential initiation is not directly under the anode, but is located a distanceaway. Another mechanism for excitation with anodal stimulation is anode breakexcitation, in which the axon is excited at the end of a long hyperpolarizingpulse.

• Nerve cells located farther away from the electrode require higher amplitudesfor stimulation than those located closer to the electrode. This follows from theCDR as previously described.

• Axons will be stimulated at lower amplitudes than cell bodies. Typically, thesite of action potential initiation is at the axon hillock or at a node of Ranvier(McIntyre and Grill 1999).

• Larger-diameter axons require lower amplitudes for stimulation than smaller-diameter axons.

• Axons with branching processes will be more easily activated than those withoutbranches

• Nerve fibers oriented parallel to the electric field (stimulation current flowingparallel to nerve fibers) have a lower threshold for stimulation than nerve fibersoriented perpendicular to the field (current flowing perpendicular to nervefibers).

• Uniform electric fields along a neuron are ineffective for stimulation. The firstspatial derivative of the electric field or equivalently the second spatial deriv-ative of the potentials in the direction along a nerve fiber drives neuronalpolarization; thus, if there is a uniform field along a nerve fiber, the secondspatial derivative is zero and therefore the transmembrane potential of theneuron will not change.

1.10 Conclusions

In summary, biophysical, electrochemical, and neurophysiological principles helpto explain some of the effects of DBS and thus to guide target and stimulationparameter selection. Future research utilizing neurophysiological and imagingtechniques will offer new insights regarding patient-specific networks in healthyand disease states. Such information of the brain structure, connectivity, function,and modulation with electrical stimulation will provide better information aboutwhere and how to optimally provide stimulation in a patient-specific manner totreat the desired disease. Such progress, however, requires a high-level multidis-ciplinary approach in which it is essential to combine neuroanatomy, neurophys-iology, neuroimaging, physics, and stochastic techniques.

8 F. L. H. Gielen and G. C. Molnar

References

Agnew WF, McCreeery DB, Yuen TG, Bullara LA (1990) Effects of prolonged electricalstimulation of the central nervous system. In: Agnew WF, McCreery DB (eds) Neuralprostheses: fundamental studies. Prentice Hall, Englewood Cliffs, pp 226–252

Bagshaw EV, Evans MH (1976) Measurement of current spread from microelectrodes whenstimulating within the nervous system. Exp Brain Res 25:391–400

Brown P, Oliviero A, Mazzone P, Insola A, Tonali P, Di Lazzaro V (2001) Dopaminedependency of oscillations between subthalamic nucleus and pallidum in Parkinson’s disease.J Neurosci 21(3):1033–1038

Butson CR, Maks CB, McIntyre CC (2006) Sources and effects of electrode impedance duringdeep brain stimulation. Clin Neurophysiol 117:447–454

Coenen VA, Honey CR, Hurwitz T, Rahman AA, McMaster J, Bürgel U, Mädler B (2009) Medialforebrain bundle stimulation as a pathophysiological mechanism for hypomania in subthalamicnucleus deep brain stimulation for Parkinson’s disease. Neurosurgery 64(6):1106–1114; discussion1114–1115

Coenen VA, Allert N, Mädler B (2011a) A role of diffusion tensor imaging fiber tracking in deepbrain stimulation surgery: DBS of the dentato-rubro-thalamic tract (drt) for the treatment oftherapy-refractory tremor. Acta Neurochir (Wien) 153(8):1579–1585

Coenen VA, Mädler B, Schiffbauer H, Urbach H, Allert N (2011b) Individual fiber anatomy ofthe subthalamic region revealed with diffusion tensor imaging: a concept to identify the deepbrain stimulation target for tremor suppression. Neurosurgery 68(4):1069–1075; discussion1075–1076. Erratum in: Neurosurgery 68(6):E1780–E1781

Coenen VA, Panksepp J, Hurwitz TA, Urbach H, Mädler B (2012) Human medial forebrainbundle (MFB) and anterior thalamic radiation (ATR): diffusion tensor imaging of two majorsubcortical pathways that may promote a dynamic balance of opposite affects relevant forunderstanding depression. J Neuropsychiatry Clin Neurosci 24:1–14

Deli G, Balas I, Nagy F, Balazs E, Janszky J, Komoly S, Kovacs N (2011) Comparison of theefficacy of unipolar and bipolar electrode configuration during subthalamic deep brainstimulation. Parkinsonism Relat Disord 17:50–54

Durand DM (2000) Electric stimulation of excitable tissue. In Bronzino JD (ed) The biomedicalengineering handbook, 2nd edn. CRC Press, Boca Raton

Eusebio A, Thevathasan W, Doyle Gaynor L, Pogosyan A, Bye E, Foltynie T, Zrinzo L, Ashkan K,Aziz T, Brown P (2011) Deep brain stimulation can suppress pathological synchronisation inparkinsonian patients. J Neurol Neurosurg Psychiatry 82(5):569–573

Grill WM (2005) Safety considerations for deep brain stimulation: review and analysis. ExpertRev Med Devices 2(4):409–420

Holsheimer J (2003) Principles of neurostimulation. In: Simpson BA (ed) Pain research andclinical management. Elsevier, Amsterdam

Kuhn AA, Williams D, Kupsch A, Limousin P, Hariz M, Schneider G, Yarrow K, Brown P(2004) Event-related beta desynchronization in human subthalamic nucleus correlates withmotor performance. Brain 127:735–746

Kuncel AM, Grill WM (2004) Selection of stimulus parameters for deep brain stimulation. ClinNeurophysiol 115:2431–2441

Lega BC, Kahana MJ, Jaggi J, Baltuch GH, Zaghloul K (2011) Neuronal and oscillatory activityduring reward processing in the human ventral striatum. NeuroReport 22:795–800

Lilly JC, Hughes JR, Alvord EC, Galkin TW (1955) Brief, noninjurious electric waveform forstimulation of the brain. Science 121:468–469

Mädler B, Coenen VA (2012) Explaining clinical effects of deep brain stimulation throughsimplified target-specific modeling of the volume of activated tissue. AJNR Am J Neuroradiol33(6):1072–1080

1 Basic Principles of Deep Brain Stimulation 9

McCracken CB, Grace AA (2009) Nucleus accumbens deep brain stimulation produces region-specific alterations in local field potential oscillations and evoked responses in vivo.J Neurosci 29(16):5354–5363

McCreery DB, Agnew WF, Yuen TG, Bullara L (1990) Charge density and charge per phase ascofactors in neural injury induced by electrical stimulation. IEEE Trans Biomed Eng37(10):996–1001

McIntyre CC, Grill WM (1999) Excitaiton of central nervous system neuron by nonuniformelectric fields. Biophysical J 76:878–888

Merrill DR, Bikson M, Jefferys JG (2005) Electrical stimulation of excitable tissue: design ofefficacious and safe protocols. J Neurosci Methods 141:171–198

Ranck JB Jr (1975) Which elements are excited in electrical stimulation of mammalian centralnervous system: a review. Brain Res 98(3):417–440

Rattay F (1989) Analysis of models for extracellular fiber stimulation. IEEE Trans Biomed Eng36(7):676–682

Rizzone M, Lanotte M, Bergamasco B, Tavella A, Torre E, Faccani G, Melcarne A, Lopiano L(2001) Deep brain stimulation of the subthalamic nucleus in Parkinson’s disease: effects ofvariation in stimulation parameters. J Neurol Neurosurg Psychiatry 71:215–219

Rushton WA (1927) The effect upon the threshold for nervous excitation of the length of nerveexposed and the angle between current and nerve. J Physiol 63:357–377

Shannon RV (1992) A model of safe levels for electrical stimulation. IEEE Trans Biomed Eng39(4):424–426

Schwan HP (1992) Linear and nonlinear electrode polarization and biological materials. AnnBiomed Eng 20(3):269–288

Tehovnik EJ (1996) Electrical stimulation of neural tissue to evoke behavioral responses.J Neurosci Methods 65(1):1–17

Weiss G (1901) Sur la possibilite de rendre comparables entre eux les appareils servant al’excitation electrique. Arch Ital Biol 35:413–446

10 F. L. H. Gielen and G. C. Molnar

Chapter 2Neural Circuits Affected by Deep BrainStimulation for the Treatmentof Psychiatric Disorders

Suzanne N. Haber and Benjamin D. Greenberg

AbbreviationsAC Anterior commissureACC Anterior cingulate cortexdACC Dorsal anterior cingulate cortexDBS Deep brain stimulationMD Major depressionOCD Obsessive–compulsive disorderOFC Orbitofrontal cortexPFC Prefrontal cortexSCGwm Subgenual cingulate gyrus white matterVC Ventral anterior internal capsulevmPFC Ventromedial prefrontal cortexvPFC Ventral prefrontal cortexVS Ventral striatum

2.1 Introduction

Although the pathophysiology of obsessive–compulsive disorder (OCD) and majordepression (MD) remains incompletely understood, converging lines of evidencepoint to abnormalities in the anterior cingulate cortex (ACC) and orbitofrontal

S. N. Haber (&)Department of Pharmacology and Physiology, School of Medicine and Dentistry,University of Rochester, 601 Elmwood Avenue, Rochester, NY 14642, USAe-mail: suzanne_haber@urmc.rochester.edu

B. D. GreenbergDepartment of Psychiatry and Human Behavior, Alpert Medical School,Butler Hospital, Brown University, Providence, RI, USA

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_2, � Springer-Verlag Berlin Heidelberg 2012

11

cortex (OFC)–basal ganglia circuit. Collectively, these brain regions are involvedin various aspects of incentive-based learning and good decision-making skills(Chase et al. 2008; Rudebeck et al. 2008; Haber and Knutson 2010). They are alsoassociated with sadness and pathological risk-taking (Mayberg 2007; Chamberlainet al. 2008). Changes in activity in the OFC and ACC associated with OCD andMD are accentuated during provocation of symptoms, but the activity often returnsto near normal following successful treatment, either pharmacological or cognitivebehavioral, or surgical therapies. Moreover, regional activity in OFC (for OCD) orACC (for MD) predicts the subsequent response to treatment with medication orbehavioral therapy (McGuire et al. 1994; Rauch et al. 1994; Mayberg 2003; Yucelet al. 2007; Greenberg et al. 2010a). Taken together, the data suggest thatabnormalities in OFC/ACC–basal ganglia–thalamus circuitry are central to thepathophysiology of OCD and MD and are consistent with the classic targets forablative neurosurgical therapies. Indeed, stereotactic neurosurgical lesions in theventral anterior internal capsule (VC), the ACC, or the subcaudate white matter,both of which interrupt these circuits, are effective in the treatment of refractoryOCD and depression.

Deep brain stimulation (DBS), a standard treatment for otherwise refractorymovement disorders, such as Parkinson’s disease (Vitek 2002), is currently beinginvestigated for the treatment of severe mental health disorders, in particular,medication-resistant MD and OCD (Nuttin et al. 2003; Mayberg et al. 2005;Greenberg et al. 2008). Patients appropriate for neurosurgical intervention forOCD and MD exhibit a high degree of severity and functional impairment despiteaggressive sustained efforts with conventional treatments, and thus represent verysmall subsets of OCD or MD patient populations. DBS targets for the treatment ofOCD and MD are centered in structures that interrupt subcomponents of the ACCor OFC networks, including their connections to the ventral striatum (VS), thethalamus, and closely connected brainstem regions, (McFarland and Haber 2002;Mayberg et al. 2005; Haber et al. 2006; Cecconi et al. 2008; Greenberg et al.2010a). Two promising targets are located within white matter tracts. One iscentered in the ventral part of the anterior limb of the internal capsule (VC/VS),extending caudally into the VS (nucleus accumbens), at the border of the anteriorcommissure (AC). The second site is located in the subgenual cingulate gyruswhite matter (SCGwm) in the ventromedial prefrontal cortex (vmPFC). Two othertargets are centered within the grey matter, one in the nucleus accumbens, over-lapping the VC/VS target, the other in the subthalamic nucleus (Mallet et al. 2008;Denys et al. 2010).

The mechanisms of action for DBS are not well understood and the specificpathways affected by DBS at these sites remain unknown. Moreover, regardless ofthe site or disorder treated, the effectiveness of DBS differs between patients.Small differences in specific electrode placement likely play a critical role inclinical outcomes, as is true for all clinical applications of DBS in movementdisorders. This emphasizes the importance of understanding more precisely whichpart(s) of the OFC/ACC–basal ganglia neural network plays a central role in the

12 S. N. Haber and B. D. Greenberg

effects of DBS for the treatment of OCD and MD (Greenberg et al. 2010b;Mayberg 2007; Lehman et al. 2011).

2.2 The Anterior Cingulate and Orbital Prefrontal Cortices

The ACC and OFC are complex and heterogeneous regions, each of which isfurther divided into specific cortical areas: the ACC includes areas 24, 25, and 32;the orbital cortex is divided into areas 11, 12, 13, and 14 (Brodmann 1909; Fuster2001). Several homologies have been developed based primarily on cytoarchi-tectonics between monkey and human prefrontal cortical areas (for a review, seeOngur and Price 2000). Although imaging studies cannot distinguish betweenthese relatively small cortical divisions, functional studies have defined three mainregions (Petrides et al. 2002; O’Doherty et al. 2003; Rushworth et al. 2007;Rudebeck et al. 2008): the vmPFC, OFC, and dorsal ACC (dACC). The vmPFCincludes areas 10, 11/14, 25, and 32. The OFC includes areas 13, 12, and parts of11, and the dACC is area 24. Since the main targets for DBS treatment ofpsychiatric disease focus on connections of the vmPFC and OFC, this chapter willaddress these specifically. Collectively the vmPFC and OFC are referred to as theventral prefrontal cortex (vPFC).

Overall vPFC fibers reach cortical targets primarily via the uncinate fasciculusand extreme capsule. The uncinate fasciculus occupies the ventral plate of thevPFC and connects the prefrontal cortex (PFC) with the temporal lobe(Schmahmann and Pandya 2006; Petrides and Pandya 2007). The extreme capsulelies between the insula and the claustrum, and carries association fibers betweenthe frontal, temporal, and parietal cortex. vmPFC and OFC projections tosubcortical regions travel primarily in the internal and external capsules.Subcortical fibers pass through the external capsule to the ventral anterior limb ofthe internal capsule (Beevor and Horsley 1890; Schmahmann and Pandya 2006;Petrides and Pandya 2007). Although these major PFC pathways are well defined,less is known about the organization of vPFC fibers within them. Of particularimportance is determining how fibers from the vmPFC and OFC are segmentedwithin these bundles, and the rules can be used to determine their trajectories. Thisinformation is fundamental for predicting where specific fibers should travel, and,thus for a more precise identification of the specific connections within a whitematter bundle that are affected by DBS.

2.3 Organization of Pathways from Different vPFC Regions

All vPFC axons enter the uncinate fasciculus immediately adjacent to their corticalregion. Within the uncinate fasciculus, fibers from each injection site split intoseparate bundles, each of which contains subsets of axons that travel in different

2 Neural Circuits Affected by Deep Brain Stimulation 13

white matter tracts, the specifics of which depend on the cortical location of origin(Fig. 2.1). Axons from all vPFC areas travel in the uncinate fasciculus, corpuscallosum, cingulum bundle, superior longitudinal fasciculus, internal capsule,external capsule, and extreme capsule. In addition, fibers from specific vPFCregions also travel in the middle longitudinal fasciculus, ventral amygdalofugalpathway, stria terminalis, and the medial forebrain bundle. Most axons passthrough the external capsule initially, before breaking into separate subcorticalbundles, including those that travel to the striatum, or enter the internal capsule.The uncinate fasciculus and the internal capsule are the main fiber bundles thatconnect the vPFC to cortical and subcortical regions, respectively.

Although the uncinate fasciculus is known for its vmPFC–temporal lobeconnection (Schmahmann and Pandya 2006; Petrides and Pandya 2007), thesefrontotemporal axons do not form a distinct bundle within the ventral plate of thevPFC. Rather, fibers from each cortical region travel through the uncinatefasciculus to connect distal regions of the vPFC. vPFC axons also use this bundleas a channel to enter other white matter tracts, including the corpus callosum,cingulum bundle, and superior longitudinal fasciculus. Thus, the uncinatefasciculus contains three components, connections between the vPFC and temporallobe, connections between distal parts of vPFC regions, and as a conduit to other

Fig. 2.1 Medial orbital fiber pathways. Illustration of how different bundles separate from theinjections site as they enter the white matter. Note fibers divide into medial, dorsal, and lateralpathways. a Three-dimensional rendering of a lateral view of a sagittal plane. b Inset to bettervisualize the separation of fiber bundles. External and extreme capsule pathways have beenremoved for clarity. AC indicates the location of the AC. Note axons traveling through theinternal capsule divide into dorsal thalamic fibers and ventral brainstem axons. AF ventralamygdalofugal pathway, Amyg amygdala, CB cingulum bundle, CC corpus callosum, EC externalcapsule, EmC extreme capsule, IC internal capsule, MFB medial forebrain bundle, MLF middlelongitudinal fasciculus, SLF superior longitudinal fasciculus, UF uncinate fasciculus

14 S. N. Haber and B. D. Greenberg

fiber bundles (Dejerine 1895; Nauta 1964; Lehman et al. 2011). These threecomponents are intertwined within the vPFC (Fig. 2.1).

Axons from the vPFC occupy the most ventral part of the rostral anterior limbof the internal capsule. The internal capsule has been classically defined as thedorsal nucleus accumbens and AC (Dejerine 1895; Schmahmann and Pandya2006). However, a significant proportion of the descending vPFC fibers travel inwhite matter fascicules embedded within the nucleus accumbens and AC(Fig. 2.2a). Thus, these fascicules, which can be seen in human histologicalpreparations (Dejerine 1895), constitute an integral part of the internal capsule andcarry descending vPFC internal capsule fibers (Fig. 2.3b). The vPFC fibers withinthe internal capsule are organized according to their destination. In particular,thalamic internal capsule fibers from each cortical region travel dorsal to theirbrainstem axons (Lehman et al. 2011) (Fig. 2.2b).

The medial/lateral position within the vPFC dictates both the route that fiberstake to enter major white matter tracts and the position they take within some ofthose tracts (Lehman et al. 2011). The medial vPFC fibers enter the internalcapsule (and striatum) ventrally, directly through the subcaudate white matter, andmove dorsally in the internal capsule as they travel caudally (Fig. 2.3a, b).In contrast, the lateral vPFC fibers enter the internal capsule from a lateral anddorsal position and move ventrally through the internal capsule as they travelposteriorly (Fig. 2.3c, d). This results in an organization in which fibers frommedial areas travel ventral to axons from lateral vPFC regions (Fig. 2.3a). Thus,fibers are stacked in the internal capsule with the vmPFC axons ventral orembedded within the AC and the lateral OFC regions positioned dorsal to the AC.This topography is maintained (albeit with a great deal of compression) as theyenter the inferior thalamic peduncle.

Superimposed on this topographic organization is the arrangement of thalamicaxons that travel dorsal to brainstem axons from the same cortical area. This createsa complex convergence between thalamic and brainstem fibers from different vPFC

Fig. 2.2 Photomicrograph and schematics of vmPFC and lateral orbitofrontal cortex (OFC)pathways through the internal capsule (parasagittal plane). a Fibers passing through the internalcapsule travel dorsal to, embedded within, and ventral to the anterior commissure. b The differentpositions of thalamic versus brainstem fibers of the vmPFC (red and purple) and the lateral OFC(dark blue and light blue) entering and traveling through the internal capsule. Brainstem fibers(purple and light blue) travel ventral to thalamic fibers (red and dark blue). AC anteriorcommissure, Cd caudate nucleus, Pu putamen, vPFC ventral prefrontal cortex

2 Neural Circuits Affected by Deep Brain Stimulation 15

regions. For example, medial PFC brainstem fibers are embedded within the VS; thefibers that pass through the AC travel to the thalamus. In contrast, axons thatterminate in the brainstem from more lateral vPFC regions travel within the AC,whereas those that lie dorsal to the AC terminate in the thalamus. Thus, within theAC region of the internal capsule, axons from thalamic fibers from the vmPFC travelwith brainstem fibers from more lateral parts of vPFC (see Fig. 2.2b). The mostlateral vPFC fibers all travel dorsal to the AC. Taken together, the medial/lateralorigin of vPFC fibers coupled with their thalamic versus brainstem organization,positions thalamic internal capsule fibers from medial vPFC regions with those thatterminate in the brainstem that arise from more lateral vPFC areas. These resultsimply that DBS at various internal capsule locations affects combinations ofthalamic and brainstem vPFC fibers from different vPFC regions.

2.4 Deep Brain Stimulation Sites: What Gets Stimulated?

To estimate the most likely set of fibers involved at the different DBS contacts, weused the fiber trajectories outlined from non-human-primate experiments withrepresentations of electrodes used in humans for the DBS targets adjusted for size

Fig. 2.3 Schematics and photomicrographs of ventral prefrontal cortex (vPFC) fibers through theinternal capsule. a, b Fibers from the vmPFC enter ventrally and move dorsally as they travelcaudally. c, d Fibers from the lateral OFC enter dorsally and move ventrally as they travel caudally.AC anterior commissure, Cd caudate nucleus, GP globus pallidus, IC internal capsule, Pu putamen

16 S. N. Haber and B. D. Greenberg

(40 % of human dimensions) (Lehman et al. 2011). The most effective SCGwmcontacts (1 and 2) are at the border between the subgenual cingulate gyrus and theinferior rostral gyrus (Hamani et al. 2009). Contact 1 is within the inferior rostralgyrus, contact 2 is within the subgenual cingulate gyrus, and contracts 0 and 3 areventral and dorsal, respectively. Thus, at this site, contacts 0–2 will involve (1) allconnections from vmPFC areas adjacent to the electrode contacts (both corticaland subcortical projections) (Fig. 2.4a); (2) uncinate fasciculus fibers fromnonadjacent vmPFC and medial OFC as they travel medially to other vPFC areasand/or enter the medial forebrain bundle; (3) a subset of central OFC fiberstraveling medially to innervate medial PFC areas (Fig. 2.4b); (4) a subset ofanterior vmPFC and medial OFC fibers on route to the corpus callosum through

Fig. 2.4 Modeled deep brain stimulation electrodes at the SCGwm and ventral anterior internalcapsule/ventral striatum (VC/VS) targets. a The SCGwm target involves all fibers from corticalareas adjacent to the electrode, including descending projections. b The SCGwm target alsoinvolves other vPFC fibers that pass through the site, including axons from lateral vPFC regionstraveling medially, and those from medial OFC areas traveling dorsally. c–e Sagittal view ofspecific vPFC bundles traveling in the internal capsule with an electrode representation embeddedat the VC/VS site. Each contact captures a different set of thalamic and/or brainstem fibers. AFventral amygdalofugal bundle, AC anterior commissure, C0 contact 0, C1 contact 1, C2 contact 2,C3 contact 3, CB cingulum bundle, CC corpus callosum, cOFC central orbital prefrontal cortex,EC external capsule, EmC extreme capsule, IC internal capsule, lOFC lateral orbital cortex,mOFC medial orbital cortex, SLF superior longitudinal fasciculus, UF uncinate fasciculus,vmPFC ventral medial prefrontal cortex

2 Neural Circuits Affected by Deep Brain Stimulation 17

the uncinate fasciculus; and (5) axons traveling from the contralateral vmPFC andmedial OFC (not illustrated). Contact 3 involves primarily fibers in the corpuscallosum. In addition, this site captures a subset of fibers traveling from the medialOFC and posterior central OFC to the cingulum bundle and superior longitudinalfasciculus.

The VC/VS electrode is implanted at an angle, positioning contact 0 mostposteriorly. Each contact activates a different subset of corticothalamic and brain-stem fibers (Fig. 2.4c–f). Axons from the vmPFC pass through contact 0, mosttraveling to the brainstem, with few traveling to the thalamus (Fig. 2.4c). In contrast,contact 1 captures fibers from the vmPFC traveling to the thalamus, but not thosetraveling to the brainstem. Contact 1 involves some central OFC brainstem axons,but few thalamic OFC fibers (Fig. 2.4d). Contact 2 captures central OFC brainstemfibers, whereas contact 3 captures both brainstem and thalamic central OFC fibersand brainstem fibers from the lateral OFC (Fig. 2.4d, e).

Finally, contact 0 at the nucleus accumbens site is placed in the shell of the VSand contact 1 is placed in the core. In contrast to the subgenual cingulate gyrus andVC/VS, stimulation of the two ventral contacts is located primarily in gray matter,but within corticostriatal fibers. Contacts 2 and 3 are within the VC and likelyinvolve cortical connections similar to those described above for the VC/VS target.

The effectiveness of DBS for the treatment of depression at the SCGwm andVC/VS (white matter) sites has not been directly compared in a randomized studyfor patient selection and other variables. Nonetheless, both sites have shownpromising initial efficacies in open-label trials in over 50 % of otherwise intrac-table patients (Malone et al. 2009; Kennedy et al. 2011; Greenberg et al. 2010a, b).Stimulation at the SCGwm site captures all cortical and subcortical projectionsfrom the area surrounding each contact site. However, it also capture fibers fromnonadjacent cortical areas passing through the target, both corticocortical andcorticosubcortical connections. In contrast, neither the VS nor the VC/VS sitedirectly involves corticocortical fibers. However, each contact in the VC/VSregion involves a different combination of thalamic and/or brainstem bundles.DBS at all three of the different stimulation targets will capture subsets of fibersthat include both thalamic and brainstem fiber projections from prefrontal corticalregions. Thus, an important part of the clinical effectiveness of DBS is likely torequire a combination of thalamic and brainstem fibers. Interestingly, there can benotable differences in fibers likely to be modulated by DBS within a given surgicaltarget. This is in accord with accumulating clinical experience suggesting, as inDBS in the subthalamic nucleus for the treatment of Parkinson’s disease, activa-tion of even adjacent stimulation contacts may be associated with very differentclinical effects.

Thus, the rules fibers use to reach their targets provides an important guide forunderstanding the relationship between specific contact stimulation and thefunctional connectivities likely to be involved at each contact, providing insightinto behavioral and therapeutic effects of DBS.

Acknowledgments This work was supported by NIH grants MH XXXXXX and MH73111.

18 S. N. Haber and B. D. Greenberg

References

Beevor CE, Horsley V (1890) An experimental investigation into the arrangement of theexcitable fibres of the internal capsule of the Bonnet Monkey (Macacus sinicus). Philos TransRoyal Soc Biol Sci 181:49–88

Brodmann K (1909) Vergleichende Lokalisationslehre der Grosshirnrinde. J.A. Barth, LeipzigCecconi JP, Lopes AC, Duran FL, Santos LC, Hoexter MQ, Gentil AF, Canteras MM, Castro CC,

Noren G, Greenberg BD, Rauch SL, Busatto GF, Miguel EC (2008) Gamma ventralcapsulotomy for treatment of resistant obsessive-compulsive disorder: a structural MRI pilotprospective study. Neurosci Lett 447:138–142

Chamberlain SR, Menzies L, Hampshire A, Suckling J, Fineberg NA, del Campo N, Aitken M,Craig K, Owen AM, Bullmore ET, Robbins TW, Sahakian BJ (2008) Orbitofrontaldysfunction in patients with obsessive-compulsive disorder and their unaffected relatives.Science 321:421–422

Chase HW, Clark L, Myers CE, Gluck MA, Sahakian BJ, Bullmore ET, Robbins TW (2008) The roleof the orbitofrontal cortex in human discrimination learning. Neuropsychologia 46:1326–1337

Dejerine J (1895) Anatomie des centres nerveux. Rueff, ParisDenys D, Mantione M, Figee M, van den Munckhof P, Koerselman F, Westenberg H, Bosch A,

Schuurman R (2010) Deep brain stimulation of the nucleus accumbens for treatment-refractory obsessive-compulsive disorder. Arch Gen Psychiatry 67:1061–1068

Fuster JM (2001) The prefrontal cortex—an update: time is of the essence. Neuron 30:319–333Greenberg BD, Askland KD, Carpenter LL (2008) The evolution of deep brain stimulation for

neuropsychiatric disorders. Front Biosci 13:4638–4648Greenberg BD, Rauch SL, Haber SN (2010a) Invasive circuitry-based neurotherapeutics:

stereotactic ablation and deep brain stimulation for OCD. Neuropsychopharmacology 35:317–336

Greenberg BD, Gabriels LA, Malone DA Jr, Rezai AR, Friehs GM, Okun MS, Shapira NA, FooteKD, Cosyns PR, Kubu CS, Malloy PF, Salloway SP, Giftakis JE, Rise MT, Machado AG,Baker KB, Stypulkowski PH, Goodman WK, Rasmussen SA, Nuttin BJ (2010b) Deep brainstimulation of the ventral internal capsule/ventral striatum for obsessive-compulsive disorder:worldwide experience. Mol Psychiatry 15:64–79

Haber SN, Knutson B (2010) The reward circuit: linking primate anatomy and human imaging.Neuropsychopharmacology 35:4–26

Haber SN, Kim KS, Mailly P, Calzavara R (2006) Reward-related cortical inputs define a largestriatal region in primates that interface with associative cortical inputs, providing a substratefor incentive-based learning. J Neurosci 26:8368–8376

Hamani C, Mayberg H, Snyder B, Giacobbe P, Kennedy S, Lozano AM (2009) Deep brain stimulationof the subcallosal cingulate gyrus for depression: anatomical location of active contacts in clinicalresponders and a suggested guideline for targeting. J Neurosurg 111:1209–1215

Kennedy SH, Giacobbe P, Rizvi SJ, Placenza FM, Nishikawa Y, Mayberg HS, Lozano AM(2011) Deep brain stimulation for treatment-resistant depression: follow-up after 3 to 6 years.Am J Psychiatry

Lehman JF, Greenberg BD, McIntyre CC, Rasmussen SA, Haber SN (2011) Rules ventralprefrontal cortical axons use to reach their targets: implications for diffusion tensor imagingtractography and deep brain stimulation for psychiatric illness. J Neurosci 31:10392–10402

Mallet L et al (2008) Subthalamic nucleus stimulation in severe obsessive-compulsive disorder.N Engl J Med 359:2121–2134

Malone DA Jr, Dougherty DD, Rezai AR, Carpenter LL, Friehs GM, Eskandar EN, Rauch SL,Rasmussen SA, Machado AG, Kubu CS, Tyrka AR, Price LH, Stypulkowski PH, Giftakis JE,Rise MT, Malloy PF, Salloway SP, Greenberg BD (2009) Deep brain stimulation of the ventralcapsule/ventral striatum for treatment-resistant depression. Biol Psychiatry 65:267–275

Mayberg HS (2003) Positron emission tomography imaging in depression: a neural systemsperspective. Neuroimaging Clin N Am 13:805–815

2 Neural Circuits Affected by Deep Brain Stimulation 19

Mayberg HS (2007) Defining the neural circuitry of depression: toward a new nosology withtherapeutic implications. Biol Psychiatry 61:729–730

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM,Kennedy SH (2005) Deep brain stimulation for treatment-resistant depression. Neuron45:651–660

McFarland NR, Haber SN (2002) Thalamic relay nuclei of the basal ganglia form both reciprocaland nonreciprocal cortical connections, linking multiple frontal cortical areas. J Neurosci22:8117–8132

McGuire PK, Bench CJ, Frith CD, Marks IM, Frackowiak RS, Dolan RJ (1994) Functionalanatomy of obsessive-compulsive phenomena. Br J Psychiatry 164:459–468

Nauta W (1964) Some efferent connections of the prefrontal cortex in the monkey. In: Waren J,Akert K (eds) The frontal granular cortex and behavior. McGraw-Hill, New York, pp 397–409

Nuttin BJ, Gabriels LA, Cosyns PR, Meyerson BA, Andreewitch S, Sunaert SG, Maes AF,Dupont PJ, Gybels JM, Gielen F, Demeulemeester HG (2003) Long-term electrical capsularstimulation in patients with obsessive-compulsive disorder. Neurosurgery 52:1263–1272;discussion 1272–1264

O’Doherty J, Critchley H, Deichmann R, Dolan RJ (2003) Dissociating valence of outcome frombehavioral control in human orbital and ventral prefrontal cortices. J Neurosci 23:7931–7939

Ongur D, Price JL (2000) The organization of networks within the orbital and medial prefrontalcortex of rats, monkeys and humans. Cereb Cortex 10:206–219

Petrides M, Pandya DN (2007) Efferent association pathways from the rostral prefrontal cortex inthe macaque monkey. J Neurosci 27:11573–11586

Petrides M, Alivisatos B, Frey S (2002) Differential activation of the human orbital, mid-ventrolateral, and mid-dorsolateral prefrontal cortex during the processing of visual stimuli.Proc Natl Acad Sci U S A 99:5649–5654

Rauch SL, Jenike MA, Alpert NM, Baer L, Breiter HC, Savage CR, Fischman AJ (1994)Regional cerebral blood flow measured during symptom provocation in obsessive-compulsivedisorder using oxygen 15-labeled carbon dioxide and positron emission tomography. ArchGen Psychiatry 51:62–70

Rudebeck PH, Bannerman DM, Rushworth MF (2008) The contribution of distinct subregions ofthe ventromedial frontal cortex to emotion, social behavior, and decision making. Cogn AffectBehav Neurosci 8:485–497

Rushworth MF, Behrens TE, Rudebeck PH, Walton ME (2007) Contrasting roles for cingulateand orbitofrontal cortex in decisions and social behaviour. Trends Cogn Sci 11:168–176

Schmahmann J, Pandya D (2006) Fiber pathways of the brain. Oxford University Press, New YorkVitek JL (2002) Mechanisms of deep brain stimulation: excitation or inhibition. Mov Disord

17:S69–S72Yucel M, Harrison BJ, Wood SJ, Fornito A, Wellard RM, Pujol J, Clarke K, Phillips ML, Kyrios

M, Velakoulis D, Pantelis C (2007) Functional and biochemical alterations of the medialfrontal cortex in obsessive-compulsive disorder. Arch Gen Psychiatry 64:946–955

20 S. N. Haber and B. D. Greenberg

Chapter 3Mechanisms of Action of Deep BrainStimulation for the Treatmentof Psychiatric Disorders

J. Luis Lujan and Cameron C. McIntyre

3.1 Introduction

Psychiatric disorders are typically characterized by a combination of affective,behavioral, cognitive, and perceptual traits that affect how individuals think, feel,and behave (Nemeroff 2007). Increasing evidence has accrued in recent yearsregarding the impact of psychiatric disease on the structural and functionalprocesses occurring in the brain. Major depressive disorder (MDD) and obsessive–compulsive disorder (OCD) are among the most devastating brain disorders, andare the result of genetic, chemical, electrical, structural, or traumatic problems inthe brain. Although most patients with MDD and OCD can be effectively treatedwith a combination of medications and psychotherapy, up to 20 % of patients failto respond to standard therapeutic interventions (Smith et al. 2011). For thesetreatment-resistant patients, more aggressive surgical strategies are needed. Deepbrain stimulation (DBS) represents a reversible alternative to conventional surgicallesions. DBS modulates brain activity by delivering high-frequency electricalpulses to subcortical structures (Benabid et al. 2005). Stimulating electrodesconnected to implanted pulse generators are permanently implanted into specificanatomical targets and used to stimulate the brain tissue. The extent of stimulationis adjusted to maximize the therapeutic efficacy of stimulation by varying thecontact configuration, stimulation frequency, stimulation amplitude, and pulse

J. L. Lujan � C. C. McIntyre (&)Department of Biomedical Engineering, Cleveland Clinic Foundation,9500 Euclid Avenue ND20, Cleveland, OH 44195, USAe-mail: mcintyc@ccf.org

J. L. Lujane-mail: lujanl@ccf.org

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_3, � Springer-Verlag Berlin Heidelberg 2012

21

duration. DBS has demonstrated encouraging results in clinical trials for thetreatment of psychiatric disorders (Lozano et al. 2008; Malone et al. 2009;Mayberg et al. 2005); however, its mechanisms are not yet fully understood, andits use remains an experimental procedure.

3.2 Neural Circuits of Psychiatric Disorders

Recent scientific efforts have focused on defining the organization and structuralconnectivity of neural circuits associated with psychiatric disease. Metabolicimaging studies have helped identify cortical and subcortical areas of the brainassociated with psychiatric diseases (Borairi and Dougherty 2011). Similarly,anatomical tracing studies in nonhuman primates have provided insight into theorganization of networks involved with these areas (Price 1999; Saleem et al.2008). Diffusion tensor imaging (DTI) studies measuring fractional anisotropy ofbrain tissue have shown abnormal white matter pathways connecting brain regionsassociated with MDD and OCD circuitry (Cannistraro et al. 2007; Szeszko et al.2005; Wakana et al. 2007). More recently, functional definition of these networkshas been augmented by the use of DTI tractography (Gutman et al. 2009;Johansen-Berg et al. 2008). These studies have shown corticostriatal–thalamo-cortical (CSTC) projections from the ventral anterior internal capsule/ventralstriatum (VC/VS) and subcallosal cingulate (SCC) white matter overlapping inmultiple regions of the brain associated with antidepressant responses.

Detailed tracing in nonhuman primates has shown that although the generaltrajectory of axonal pathways can overlap, anatomical segregation is typicallymaintained (Haber and Brucker 2009). Additionally, DTI tractography guided byexperimental data has been used to identify anatomical pathways in regions of theventral prefrontal cortex (vPFC) associated with psychiatric disease. Lehman et al.(2011) showed differences in the axonal trajectories projecting to and from distinctfunctional regions in the vPFC. Most importantly, they showed that these func-tionally distinct pathways are anatomically segmented within major fiber bundles.For example, axonal fibers from the vPFC course through the most ventral portionof the internal capsule and connect primarily to the medial dorsal thalamus but notto the motor or sensory thalamus.

A generalized model of MDD (Fig. 3.1) proposes that depression is not simplya dysfunction of any single region, but is a failure in coordination of interactionsbetween brain nuclei in CSTC networks (Kopell et al. 2004; Mayberg 1997). TheMDD model consists of three distinct anatomical and functional compartments.The dorsal compartment involves the premotor and prefrontal cortices, as well asthe dorsal segment of the anterior cingulate cortex. This compartment mediates thecognitive aspects of negative emotion. The ventral compartment, which involvesthe SCC, insula, and orbitofrontal cortex, is known to mediate circadian andvegetative aspects of depression. Finally, the rostral compartment involves thepregenual anterior cingulate cortex, amygdala, and hypothalamic–pituitary axis,

22 J. L. Lujan and C. C. McIntyre

and is thought to regulate the overall network activity by facilitating the interactionbetween the dorsal and ventral compartments.

A similar network model of OCD (Fig. 3.1) suggests that symptoms appearwhen striatopallidothalamic activity is abnormally decreased or when orbitofron-tothalamic activity is abnormally increased (Kopell et al. 2004; Haber and Brucker2009). The OCD model is characterized by three main building blocks. The firstblock involves an excitatory glutamatergic positive feedback loop between theorbital cortex, the prefrontal cortex, and the dorsomedial thalamic nucleus, passingthrough the anterior limb of the internal capsule. The second block is an inhibitory

Fig. 3.1 The corticosubcortical network involved in the models of depression and obsessive–compulsive disorder (OCD). In the model of depression, prefrontal, dorsal anterior cingulate, andpremotor cortices project to the dorsal striatum and continue on to the thalamus by means of thedorsomedial pallidum to form a corticothalamocortical loop. Similarly, the subgenual anteriorcingulate, orbitofrontal, and insular cortices project onto the ventral striatum, medial/rostralpallidum, and the anterior and dorsomedial thalamus. The amygdala sends excitatory projectionsand the pregenual anterior cingulate sends inhibitory projections to the nodes of these models.The model of OCD is formed by three principal pathways. The first is an excitatory and reciprocalpositive-feedback corticothalamic loop, in which the orbitofrontal and prefrontal cortices projectonto the dorsomedial thalamus via the anterior limb of the internal capsule. The second pathwayinvolves the orbitofrontal cortex and the ventral caudate, the dorsomedial pallidum, and theanterior, dorsomedial, and intralaminar nuclei of the thalamus. This loop is thought to modulatethe orbitofrontal–thalamic loop through inhibition from the dorsomedial pallidum to the thalamusvia GABAergic projections. This loop also involves inhibitory serotonergic projections from thedorsal raphe nuclei of the midbrain to the ventral striatum. The third pathway projects fromthe hippocampal formation to the mammillary body by means of the fornix and continues on tothe anterior thalamic nuclei and finally to the cingulate gyrus

3 Mechanisms of Action of Deep Brain Stimulation 23

c-aminobutyric acid (GABA)-ergic loop between the orbital and prefrontalcortices, the ventral caudate, the dorsomedial pallidum, and the anterior,dorsomedial, and intralaminar thalamic nuclei. This block is thought to mediatethe excitatory orbitofrontothalamic loop. The third block represents a loopbetween the limbic system and the circuit of Papez by means of the fornix and ontothe anterior thalamic nuclei.

The limited understanding of CSTC networks has prompted different stimula-tion targets to be tried in the pursuit of therapeutic benefits. For example, DBS ofSCC white matter has been used to generate long-term clinical improvement inMDD patients (Lozano et al. 2008; Mayberg et al. 2005). Similarly, DBS of theVC/VS has been successfully used to treat MDD and OCD patients (Malone et al.2009; Greenberg et al. 2010). Prevailing hypotheses suggest that DBS producestherapeutic benefits by regulating abnormal network activity within CSTCnetworks (McIntyre and Hahn 2010). Unfortunately, efforts to identify specificneural circuitry and targets associated with therapeutic clinical responses remainrestricted by the limited characterization of the effects of DBS on neuronalpopulations. Thus, given that current surgical targets for psychiatric DBS liewithin regions of white matter, it is crucial to characterize axonal response to DBSin the context of psychiatric disorders.

3.3 Neural Response to DBS

Studies have shown that DBS generates a complex three-dimensional electric fieldaccording to the anisotropic properties of brain tissue (Miocinovic et al. 2009;Chaturvedi et al. 2010). This electric field can in turn influence three types ofneural elements. The first type represents local neurons whose cell bodies lie closeto the electrode. The second represents afferent inputs from neurons whose axonterminals make synaptic connections with local cells near the electrode. The thirdcorresponds to axonal fibers passing close to the electrode, but originating fromneurons whose cell bodies and axon terminals are far away from the electrode. Thespecific response of individual neurons to an electric field will depend on thestimulation parameters and the extracellular potential distribution along its neuralprocesses (McNeal 1976; Rattay 1986).

Computational and experimental studies have shown that axons generatepropagating action potentials when stimulated with typical DBS settings (McIntyreet al. 2004a, b). The electric field generated by DBS (Fig. 3.2a) can be representedon the axon by a series of extracellular voltages at the nodes of Ranvier. If theelectrical stimulation is strong enough, the extracellular voltage will depolarize theaxon until its membrane potential, the voltage difference between the inside andthe outside of the axon, reaches its firing threshold and an action potential isgenerated (Fig. 3.2b). This basic response of an axon to a depolarizing extracel-lular electric field is related to the second spatial derivative of the voltage distri-bution along the axon (McNeal 1976). Action potentials start at the node of

24 J. L. Lujan and C. C. McIntyre

Ranvier, where the second spatial derivative of the extracellular potential islargest, and propagate in both directions along the axon. In turn, the sum of theseindividual neural responses will have a significant impact on the activity of theentire network.

3.4 Network Effects of DBS in Psychiatric Disorders

MDD and OCD are associated with abnormal activity in CSTC circuits (Mayberget al. 2005; Kopell et al. 2004). Studies have found OCD to be associated withabnormal metabolism and regional cerebral blood flow in striatal, anteriorcingulate, and orbital frontal regions (Figee et al. 2011; Greenberg et al. 2010).Similarly, abnormal neural activity has been reported in the amygdala, thalamus,and orbitofrontal and anterior cingulate cortices of depressed patients (Schulmanet al. 2011; Smith et al. 2011). Functional imaging studies have shown thatsuccessful treatment of MDD and OCD is associated with normalization ofabnormal basal metabolism in both local and remote regions of the brain. Inparticular, activity in the SCC and prefrontal cortex areas of the brain hasnormalized following chronic therapeutic SCC DBS (Mayberg et al. 2000).Similarly, normalization of activity in the medial orbitofrontal cortex, dorsalstriatum, SCC, ventral globus pallidus, and thalamus has been reported aftertherapeutic VC/VS DBS (Rauch et al. 2006). These results show that DBSgenerates complex changes throughout the network, and suggest that the antide-pressant benefits of DBS are correlated with the reversal of baseline abnormalities(Abelson et al. 2005; Mayberg et al. 2005). Furthermore, the overlap observed in

Fig. 3.2 Axonal activation model. a Electric field generated by deep brain stimulation representedby a group of isopotential contours. b Stimulation-induced extracellular voltages (Ve) interpolatedonto an axon model (red corresponds to the highest extracellular voltage magnitude and dark bluecorresponds to the lowest). The extracellular voltage depolarizes the cell and an action potential isgenerated. Action potentials start in the axon at the node of Ranvier, where the second spatialderivative of the extracellular voltage is largest (red trace). Once initiated, action potentialspropagate in both directions along the axon (blue traces)

3 Mechanisms of Action of Deep Brain Stimulation 25

neural circuitry and symptom improvement through chronic stimulation suggeststhat DBS may have similar therapeutic mechanisms in MDD and OCD.

It has been hypothesized that this CSTC dysregulation may be caused byreduced effectiveness of synaptic transmission in serotonergic, noradrenergic,and dopaminergic pathways (Scharinger et al. 2011). The amygdaloid complex,the mediodorsal thalamus, and the prefrontal cortex have all been implicated inthe pathophysiological processes of MDD and OCD (Saygin et al. 2011).Abnormal activity from the lateral amygdaloid nucleus can propagate to baso-lateral and central amygdaloid nuclei before converging in the nucleusaccumbens via the ventral amygdalofugal pathway and the extended amygdala(Tasan et al. 2010). In turn, the main efferents of the nucleus accumbensinnervate the pallidum, striatum, mediodorsal thalamus, prefrontal and cingulatecortices, and mesolimbic dopaminergic areas. Thus, disruption of pathologicalactivity or impulse gating via high-frequency stimulation of the nucleus ac-cumbens may explain the therapeutic benefits of VC/VS DBS in both MDD andOCD patients (Sturm et al. 2003).

However, brain networks associated with psychiatric disease are complexdynamical systems. As such, activity modulation in one node can result in acascade of nonintuitive changes throughout the brain on multiple timescales.Therefore, impulse gating at one node in the network may not be sufficient toobtain the maximum therapeutic benefits. It has been shown that clinicalimprovements and normalization of pathological metabolic activity requirechronic stimulation over several months (Greenberg et al. 2006; Mayberg et al.2005). For example, sleep disturbances in depressed patients tend to normalizewithin the first week of DBS. However, interest, energy, and other psychomotorimprovements occur after a few weeks of DBS. These changes are finally followedby increased interest and pleasure in social interactions, improved planning ability,and improved task initiation and completion. The slow progression of therapeuticeffect suggests that, unlike DBS for the treatment of movement disorders,immediate disruption of network activity is not the only, or even principal, ther-apeutic mechanism of DBS in psychiatric disorders.

A logical hypothesis that follows is that psychiatric DBS produces therapeuticeffects via both immediately acting and long-acting mechanisms. For example,immediate normalization of SCC activity, consistent with acute symptomimprovement in depression, could be the result of DBS-induced activation ofinhibitory GABAergic afferents and stimulation-induced synaptic failure (Mayberget al. 2005). This immediate disruption of pathological activity could also beachieved by inducing changes in membrane excitability and by creating imbalancesbetween excitatory and inhibitory inputs (Hallett 2000). Additionally, immediateactivation changes in remote regions of the brain could occur as an indirectconsequence of transsynaptic effects or direct activation of projection neurons

26 J. L. Lujan and C. C. McIntyre

(McCracken and Grace 2007). On the other hand, persisting metabolic changesassociated with delayed therapeutic benefits suggest that DBS may reverse large-scale reorganization of the brain. This type of large-scale reorganization of the brainhas been shown to occur in psychiatric disease (Chollet and Weiller 2000).Furthermore, it has been shown that it can be induced by manipulating the inputs toparallel, reciprocal, and overlapping brain networks through continuous excitation/inhibition of serotonergic and noradrenergic pathways (Vaidya and Duman 2001;Seitz et al. 1995). The therapeutic benefits observed after cessation of chronicstimulation support the hypothesis that synaptic potentiation and depression mayproduce long-term changes in brain circuitry (e.g., synapse formation andconnection growth).

A speculative and untested hypothetical example of how DBS-induced activitycould result in long-term network changes is given below. First, suprathresholdhigh-frequency DBS activates a large number of axons, resulting in an increase ofglutamatergic and serotonergic exposure in the striatum (McCracken and Grace2007). Second, activation of serotonin receptors in the striatum results in proteinkinase phosphorylation and activation of second messenger systems (Ward andDorsa 1999). The second messenger systems regulate gene transcription andinduce long-term potentiation and synaptic growth. Third, spine enlargementoccurs as a result of high-frequency stimulation, thereby increasing the number ofglutamate receptors (Bennett 2000). The net effect would be a significant alterationof striatum excitability and cortical integration, potentially changing the patho-logical CSTC network dynamics and alleviating disease symptoms. Variants ofthis hypothetical example of DBS-induced network changes could also beoccurring in the cortex, or other nodes of the CSTC network.

3.5 Identifying Target Pathways for Stimulation

Imaging and anatomical techniques alone can only partially describe the effects ofDBS. As such, computational models create unique opportunities to refine ourunderstanding of neural networks involved in psychiatric disease. In an attempt tocharacterize the neuronal response to patient-specific DBS, we analyzed axonalactivation in treatment-resistant MDD and OCD patients who were implantedbilaterally in the VC/VS with quadripolar DBS electrodes (1.27-mm diameter,3-mm contacts, and 4-mm spacing, Medtronic, Minneapolis, MN, USA) (Lujanet al. 2012). We characterized axonal activation using computational models ofDBS. These models included (1) a virtual DBS electrode, (2) an electric field finiteelement model including electrical properties of brain tissue and the electrode–tissue interface (Chaturvedi et al. 2010), and (3) multicompartment models ofmyelinated axons (McIntyre et al. 2002; McNeal 1976). Postoperative electrodelocations and associated electric field models were determined for each patient.The trajectories of white-matter axon fibers near the DBS electrodes were iden-tified using streamline tractography (Mori et al. 1999). We interpolated the

3 Mechanisms of Action of Deep Brain Stimulation 27

three-dimensional electric fields onto the multicompartment axon models andidentified active axonal pathways that were associated with therapeutic and non-therapeutic clinical outcomes across multiple patients.

Our analysis suggests that DBS would activate axonal pathways connecting theorbitofrontal and subgenual anterior cingulate cortices with the ventral striatum aswell as interhemispheric connections. Moreover, our results suggest there are fiveactive pathways (P1–P5) associated with therapeutic outcomes in MDD and fouractive pathways (P6–P9) associated with therapeutic responses in OCD (Fig. 3.3)(Lujan et al. 2012). Clinical response was defined using the criteria outlined inMalone et al. (2009) and Greenberg et al. (2010). Generally, active pathwayspassed through the ventral anterior internal capsule and coursed lateral and medial

Fig. 3.3 Outcome-specific active axonal pathways. Common active pathways in the responderand nonresponder clinical groups for major depressive disorder (MDD) (a and c, respectively)and OCD (b and c, respectively). The combinations of letters indicate the general location of theboundaries of each pathway with respect to the region of. The numbers indicate distinct pathwaysidentified using an algorithm described in Lujan et al. (2012). d Region of interest used to analyzeactive pathways identified by the diffusion tensor imaging tractography. The three-dimensionalsurfaces represent various nuclei of interest. D dorsal, V ventral, A anterior, P posterior,M medial, L lateral, Cau caudate nucleus, GP pallidum, Acc nucleus accumbens

28 J. L. Lujan and C. C. McIntyre

to the ventral striatum, or dorsal and lateral to the nucleus accumbens (Fig. 3.3).P1–P4 coursed along the ventromedial surface of the dorsal striatum, from thedorsolateral and posterior region of our region of interest (ROI). These continuedwith anterolateral (P1), ventrolateroposterior (P2), ventromedial–anterior (P3), andventromedial–posterior (P4) projections relative to the boundaries of the ROI. P5overlapped with the ventrolateroposterior segment of P2 in its course along theventromedial portion of the posterior nucleus accumbens. This pathway courseddorsally along the lateral head of the caudate nucleus, continuing in a lateral andanterior direction over the central caudate nucleus. P6–P8 coursed in an antero-posterior direction along the lateral head of the caudate nucleus, continuingventrally along the posterior nucleus accumbens. P6 and P7 overlapped at theirdorsal ROI boundaries and anterior segments before reaching the posterior nucleusaccumbens. P6 coursed medially and ventrally after passing by the posteriornucleus accumbens, and finally projecting in an anterior direction. However, P7continued medially along the posterior nucleus accumbens in a ventral directionwithin the ROI. P8 followed a more dorsal trajectory, continuing medially alongthe posterior nucleus accumbens in a ventral direction and overlapping with P7.Conversely, one active pathway (P10) was associated with nonresponders. Thispathway was adjacent to the ventromedial surface of the dorsal striatum andfollowed a trajectory similar to that of P1.

These trajectories are consistent with probabilistic tractography findings showingthat the effects of VC/VS DBS may be mediated via strong connections to orbito-frontal, anterior mid cingulate, hypothalamus, nucleus accumbens, and amygdala/hippocampus regions (Johansen-Berg et al. 2008; Gutman et al. 2009). The besttherapeutic outcomes were achieved when axonal pathways associated only withclinical responders were activated. Similarly, clinical outcomes deteriorated whentherapeutic pathways overlapping with pathways identified as nonresponders wereactivated. Our results suggest that pathways lateral and posterior to the middleportion (in a dorsoventral direction) of the ventral striatum should be the focus ofinvestigation in future psychiatric DBS studies. However, careful attention shouldbe paid to pathways coursing dorsal and lateral to the ventral striatum. Our findingssuggest that therapeutic improvements require unique and selective activation ofaxonal pathways associated with specific clinical benefits. Furthermore, theysuggest that simultaneous activation of optimal and nonoptimal pathways maydeteriorate, slow down the progression of, and even prevent clinical improvements.

3.6 Conclusions and Future Directions

From initial hypotheses stating that psychiatric disorders are caused solely bychemical imbalances in the brain, to a more complex theory that involves braincircuit interactions and plasticity (Krishnan and Nestler 2010), research aimed atunraveling the mysteries of the mind has allowed the development of networkmodels of disease (Kopell et al. 2004; Mayberg 1997). These models have helped

3 Mechanisms of Action of Deep Brain Stimulation 29

us obtain a better understanding of the pathophysiological processes of MDD andOCD. Nevertheless, we remain far from fully understanding the underlying net-work mechanisms of disease or the therapeutic mechanisms of DBS.

Quantitative outcome metrics and biophysical markers are needed that willallow a better definition and more accurate classification of psychiatric disorderson the basis of their etiological and pathological aspects. Perhaps as important isthe need to document clinical changes in stimulation settings and their corre-sponding neurobehavioral outcomes. Limited documentation of clinical titration ofstimulation settings makes it difficult to identify the direct clinical effects inducedby specific device settings. As a result, the optimal DBS target locations andstimulation settings remain elusive.

By integrating multicenter patient-specific information derived from high-res-olution DTI, probabilistic tractography, and clinical outcome data, we should beable to establish a more complete and refined depiction of the optimal therapeuticpathways for stimulation. The resulting pathway–outcome mapping will allow usto statistically define optimal stimulation targets for different causes and disorders.Furthermore, definition of these outcome–pathway maps will enable computa-tional selection of the optimal stimulation settings to maximize therapeutic out-comes on a patient-specific basis without exhaustive trial-and-error searchesthrough the DBS parameter space. To achieve this goal, a critical step in theabove-mentioned process should be to correlate human pathways with their non-human-primate counterparts defined from detailed histological staining studies(Lehman et al. 2011). These correlation analyses will allow identification ofdirectly stimulated axons and their associated cortical and subcortical regionsdirectly affected by stimulation.

References

Abelson JL, Curtis GC, Sagher O, Albucher RC, Harrigan M, Taylor SF, Martis B, Giordani B(2005) Deep brain stimulation for refractory obsessive-compulsive disorder. Biol Psychiatry57(5):510–516

Benabid AL, Wallace B, Mitrofanis J, Xia R, Piallat B, Chabardes S, Berger F (2005) A putativegeneralized model of the effects and mechanism of action of high frequency electricalstimulation of the central nervous system. Acta Neurol Belg 105(3):149–157

Bennett MR (2000) The concept of long term potentiation of transmission at synapses. ProgNeurobiol 60(2):109–137

Borairi S, Dougherty DD (2011) The use of neuroimaging to predict treatment response forneurosurgical interventions for treatment-refractory major depression and obsessive-compulsive disorder. Harv Rev Psychiatry 19(3):155–161

Cannistraro PA, Makris N, Howard JD, Wedig MM, Hodge SM, Wilhelm S, Kennedy DN, RauchSL (2007) A diffusion tensor imaging study of white matter in obsessive-compulsive disorder.Depress Anxiety 24(6):440–446

Chaturvedi A, Butson CR, Lempka SF, Cooper SE, McIntyre CC (2010) Patient-specific modelsof deep brain stimulation: Influence of field model complexity on neural activationpredictions. Brain Stimulat 3(2):65–77

30 J. L. Lujan and C. C. McIntyre

Chollet F, Weiller C (2000) Recovery of neurological function. In: Toga AW, Frackowiak RS,Mazziotta JC (eds) Brain mapping: the disorders. Academic Press, New York

Figee M, Vink M, de Geus F, Vulink N, Veltman DJ, Westenberg H, Denys D (2011)Dysfunctional reward circuitry in obsessive-compulsive disorder. Biol Psychiatry69(9):867–874

Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, Salloway SP, Okun MS,Goodman WK, Rasmussen SA (2006) Three-year outcomes in deep brain stimulation for highlyresistant obsessive-compulsive disorder. Neuropsychopharmacology 31(11):2384–2393

Greenberg BD, Gabriels LA, Malone DA Jr, Rezai AR, Friehs GM, Okun MS, Shapira NA, FooteKD, Cosyns PR, Kubu CS, Malloy PF, Salloway SP, Giftakis JE, Rise MT, Machado AG,Baker KB, Stypulkowski PH, Goodman WK, Rasmussen SA, Nuttin BJ (2010) Deep brainstimulation of the ventral internal capsule/ventral striatum for obsessive-compulsive disorder:worldwide experience. Mol Psychiatry 15(1):64–79

Gutman DA, Holtzheimer PE, Behrens TE, Johansen-Berg H, Mayberg HS (2009) Atractography analysis of two deep brain stimulation white matter targets for depression.Biol Psychiatry 65(4):276–282

Haber SN, Brucker JL (2009) Cognitive and limbic circuits that are affected by deep brainstimulation. Front Biosci 14:1823–1834

Hallett M (2000) Plasticity. In: Mazziotta JC, Toga AW, Frackowiak RS (eds) Brain mapping:the disorders. Academic Press. New York

Johansen-Berg H, Gutman DA, Behrens TE, Matthews PM, Rushworth MF, Katz E, Lozano AM,Mayberg HS (2008) Anatomical connectivity of the subgenual cingulate region targeted withdeep brain stimulation for treatment-resistant depression. Cereb Cortex 18(6):1374–1383

Kopell BH, Greenberg B, Rezai AR (2004) Deep brain stimulation for psychiatric disorders.J Clin Neurophysiol 21(1):51–67

Krishnan V, Nestler EJ (2010) Linking molecules to mood: new insight into the biology ofdepression. Am J Psychiatry 167(11):1305–1320

Lehman JF, Greenberg BD, McIntyre CC, Rasmussen SA, Haber SN (2011) Rules ventralprefrontal cortical axons use to reach their targets: implications for diffusion tensor imagingtractography and deep brain stimulation for psychiatric illness. J Neurosci31(28):10392–10402

Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH (2008)Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. BiolPsychiatry 64(6):461–467

Lujan JL, Chaturvedi A, Malone DA, Rezai AR, Machado AG, McIntyre CC (2012) Axonalpathways linked to therapeutic and nontherapeutic outcomes during psychiatric deep brainstimulation. Hum Brain Mapp 33(4):958–968

Malone DA Jr, Dougherty DD, Rezai AR, Carpenter LL, Friehs GM, Eskandar EN, Rauch SL,Rasmussen SA, Machado AG, Kubu CS, Tyrka AR, Price LH, Stypulkowski PH, Giftakis JE,Rise MT, Malloy PF, Salloway SP, Greenberg BD (2009) Deep brain stimulation of theventral capsule/ventral striatum for treatment-resistant depression. Biol Psychiatry65(4):267–275

Mayberg HS (1997) Limbic-cortical dysregulation: a proposed model of depression. J Neuro-psychiatry Clin Neurosci 9(3):471–481

Mayberg HS, Brannan SK, Tekell JL, Silva JA, Mahurin RK, McGinnis S, Jerabek PA (2000)Regional metabolic effects of fluoxetine in major depression: serial changes and relationshipto clinical response. Biol Psychiatry 48(8):830–843

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM,Kennedy SH (2005) Deep brain stimulation for treatment-resistant depression. Neuron45(5):651–660

McCracken CB, Grace AA (2007) High-frequency deep brain stimulation of the nucleusaccumbens region suppresses neuronal activity and selectively modulates afferent drive in ratorbitofrontal cortex in vivo. J Neurosci 27(46):12601–12610

3 Mechanisms of Action of Deep Brain Stimulation 31

McIntyre CC, Grill WM, Sherman DL, Thakor NV (2004a) Cellular effects of deep brainstimulation: model-based analysis of activation and inhibition. J Neurophysiol91(4):1457–1469

McIntyre CC, Hahn PJ (2010) Network perspectives on the mechanisms of deep brainstimulation. Neurobiol Dis 38(3):329–337

McIntyre CC, Mori S, Sherman DL, Thakor NV, Vitek JL (2004b) Electric field and stimulatinginfluence generated by deep brain stimulation of the subthalamic nucleus. Clin Neurophysiol115(3):589–595

McIntyre CC, Richardson AG, Grill WM (2002) Modeling the excitability of mammalian nervefibers: influence of afterpotentials on the recovery cycle. J Neurophysiol 87(2):995–1006

McNeal DR (1976) Analysis of a model for excitation of myelinated nerve. IEEE Trans BiomedEng 23(4):329–337

Miocinovic S, Lempka SF, Russo GS, Maks CB, Butson CR, Sakaie KE, Vitek JL, McIntyre CC(2009) Experimental and theoretical characterization of the voltage distribution generated bydeep brain stimulation. Exp Neurol 216(1):166–176

Mori S, Crain BJ, Chacko VP, van Zijl PC (1999) Three-dimensional tracking of axonalprojections in the brain by magnetic resonance imaging. Ann Neurol 45(2):265–269

Nemeroff CB (2007) The burden of severe depression: a review of diagnostic challenges andtreatment alternatives. J Psychiatr Res 41:189–206

Price JL (1999) Prefrontal cortical networks related to visceral function and mood. Ann N YAcad Sci 877:383–396

Rattay F (1986) Analysis of models for external stimulation of axons. IEEE Trans Biomed Eng33(10):974–977

Rauch SL, Dougherty DD, Malone D, Rezai A, Friehs G, Fischman AJ, Alpert NM, Haber SN,Stypulkowski PH, Rise MT, Rasmussen SA, Greenberg BD (2006) A functional neuroim-aging investigation of deep brain stimulation in patients with obsessive-compulsive disorder.J Neurosurg 104(4):558–565

Saleem KS, Kondo H, Price JL (2008) Complementary circuits connecting the orbital and medialprefrontal networks with the temporal, insular, and opercular cortex in the macaque monkey.J Comp Neurol 506(4):659–693

Saygin ZM, Osher DE, Augustinack J, Fischl B, Gabrieli JD (2011) Connectivity-basedsegmentation of human amygdala nuclei using probabilistic tractography. Neuroimage56(3):1353–1361

Scharinger C, Rabl U, Pezawas L, Kasper S (2011) The genetic blueprint of major depressivedisorder: contributions of imaging genetics studies. World J Biol Psychiatry 12(7):474–488

Schulman JJ, Cancro R, Lowe S, Lu F, Walton KD, Llinás RR (2011) Imaging of thalamocorticaldysrhythmia in neuropsychiatry. Front Hum Neurosci 5:69

Seitz RJ, Huang Y, Knorr U, Tellmann L, Herzog H, Freund HJ (1995) Large-scale plasticity ofthe human motor cortex. NeuroReport 6(5):742–744

Smith R, Fadok RA, Purcell M, Liu S, Stonnington C, Spetzler RF, Baxter LC (2011) Localizingsadness activation within the subgenual cingulate in individuals: a novel functional MRIparadigm for detecting individual differences in the neural circuitry underlying depression.Brain Imaging Behavior 5:229–239

Sturm V, Lenartz D, Koulousakis A, Treuer H, Herholz K, Klein JC, Klosterkotter J (2003) Thenucleus accumbens: a target for deep brain stimulation in obsessive-compulsive- and anxiety-disorders. J Chem Neuroanat 26(4):293–299

Szeszko PR, Ardekani BA, Ashtari M, Malhotra AK, Robinson DG, Bilder RM, Lim KO (2005)White matter abnormalities in obsessive-compulsive disorder: a diffusion tensor imagingstudy. Arch Gen Psychiatry 62(7):782–790

Tasan RO, Nguyen NK, Weger S, Sartori SB, Singewald N, Heilbronn R, Herzog H, Sperk G(2010) The central and basolateral amygdala are critical sites of neuropeptide Y/Y2 receptor-mediated regulation of anxiety and depression. J Neurosci 30(18):6282–6290

Vaidya VA, Duman RS (2001) Depresssion—emerging insights from neurobiology. Br Med Bull57:61–79

32 J. L. Lujan and C. C. McIntyre

Wakana S, Caprihan A, Panzenboeck MM, Fallon JH, Perry M, Gollub RL, Hua K, Zhang J,Jiang H, Dubey P, Blitz A, van Zijl P, Mori S (2007) Reproducibility of quantitativetractography methods applied to cerebral white matter. Neuroimage 36(3):630–644

Ward RP, Dorsa DM (1999) Molecular and behavioral effects mediated by Gs-coupled adenosineA2a, but not serotonin 5-Ht4 or 5-Ht6 receptors following antipsychotic administration.Neuroscience 89(3):927–938

3 Mechanisms of Action of Deep Brain Stimulation 33

Chapter 4Deep Brain Stimulation in the VentralCapsule/Ventral Striatumfor the Treatment of Obsessive–CompulsiveDisorder: Role of the Bed Nucleusof the Stria Terminalis

Loes Gabriëls and Bart Nuttin

4.1 Introduction

Obsessive–compulsive disorder (OCD) is a clinically well defined anxiety disor-der. Hallmark symptoms are obsessions and compulsions, and for many patients,symptoms are accompanied by severe anxiety and far-reaching avoidance oftrigger situations. One of the characteristic symptoms of anxiety disorders ingeneral, and OCD specifically, is the imbalance toward negatively valencedconditions and anticipation of adverse outcomes (‘‘worst case scenarios’’).

The impact of OCD on multiple domains of quality of life (including socialfunctioning, education, employment, marriage and family relationships, socio-economic status) depends upon the severity of the disorder, with severer OCDresulting in poorer quality of life and social functioning.

The primary treatments for OCD include pharmacotherapy and cognitivebehavioral therapy. Alone or in combination, these treatment modalities areeffective in reducing OCD symptoms for most patients. But despite conscientiouscompliance and adherence to treatment according to internationally acceptedguidelines, some OCD patients remain refractory to conventional treatment.

In a small group of severe, treatment-resistant patients, neurosurgical proce-dures have been employed as a treatment of last resort. Until 1998, several ste-reotactic neurosurgical lesioning techniques were performed in strictly selectedOCD patients. They all aimed at selectively destroying part of a dysfunctionalcorticosubcortical brain circuit associated with pathological symptoms of OCD.One of these procedures is bilateral anterior capsulotomy, a lesioning technique

L. Gabriëls (&)Department of Psychiatry, University Hospitals Leuven, KU Leuven, Leuven, Belgiume-mail: loes.gabriels@uzleuven.be

B. NuttinDepartment of Neurosurgery, University Hospitals Leuven, KU Leuven, Leuven, Belgium

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_4, � Springer-Verlag Berlin Heidelberg 2012

35

shown to be successful in reducing symptom severity in roughly half of cases oftreatment-resistant OCD (Jenike 1998).

In anterior capsulotomy, an elongated lesion is made in the anterior limbs of theinternal capsule (ALIC) and part of the ventrally located nucleus accumbens, thusinterrupting ventral fibers originating from the orbitofrontal cortex (OFC) andsubgenual anterior cingulate cortex that project via the ventral striatum to medial,dorsomedial, and anterior thalamic nuclei (Kopell et al. 2004). The irreversibilityof side effects (e.g., apathy) in anterior capsulotomy was the main driving force toinvestigate the effect of deep brain stimulation (DBS) in the same brain region asan alternative treatment option for OCD.

In the first cases of DBS for the treatment of OCD (Nuttin et al. 1999), thechoice of the site for lead implantation was based on experience with anteriorcapsulotomy. The chosen target was the ALIC, with electrodes implanted in theALIC, with the most distal contact in the region of the ventral capsule–ventralstriatum, copying the trajectory of the lesion in anterior capsulotomy. The ratio-nale for this target was parallel to the choice of target for the treatment of tremorand Parkinson’s disease, where the identification of surgical lesions with thera-peutic benefits was followed by DBS applied with high frequencies to the samestructures.

The results were promising, but high stimulation amplitudes were required toinduce symptom relief (Nuttin et al. 2003). As the number of patients increased, atarget location versus outcome analysis revealed a better outcome with a moreposterior location of the electrodes (Greenberg et al. 2010). The patients com-monly had significant non-OCD anxiety symptoms at the baseline, and on average,these anxiety symptoms decreased by approximately 50 % between the baselineand the treatment phase as well. The response rates of DBS in severe and treat-ment-resistant OCD improved as the target was shifted to a more posterior andmore medial position, just posterior to the anterior commissure. According to Maiet al. (1998), this target region is the bed nucleus of the stria terminalis (BNST).

4.2 Brain Imaging and OCD

Brain-imaging studies provide accumulating insight into the neural circuitryunderlying OCD. Hyperactivity is frequently observed in corticostriatothalamo-cortical (CSTC) circuits, especially in the OFC and the caudate nucleus in OCDpatients, and this hyperactivity can be magnified by provocation of OCD symp-toms (Saxena and Rauch 2000). Some studies point to differences in the volumesof CSTC structures between patients with OCD and control volunteers (Jenikeet al. 1996). Additionally, the white matter tracts linking putative CSTC nodesmay be abnormal; a diffusion tensor imaging study found differences in the cin-gulum bundles and ALIC in patients with OCD compared with non-OCD controls(Cannistraro et al. 2007).

36 L. Gabriëls and B. Nuttin

The ALIC is a large and complex array of fiber tracts. It contains the anteriorthalamic radiation (or peduncle) as well as the prefrontal corticopontine tract andfibers connecting the caudate nucleus with the putamen (Axer and von Keyserlingk2000). The anterior thalamic peduncle forms a reciprocal connection between thedorsomedial thalamic nuclei and the dorsolateral and medial prefrontal cortex, andbetween the anterior thalamic nuclei and the cingulate gyrus. Other importantanatomical structures immediately adjacent to the internal capsule (Mai et al.1998) that could be influenced by selection of the optimal parameters are the striaterminalis and the BNST, which are part of the extended amygdala concept.

4.3 Amygdala and BNST and Fear

The neuroanatomy and neurochemistry of the BNST have been carefully reviewed(van Kuyck et al. 2009). Cyto- and chemo-architectonic studies demonstrate thatthe BNST is a highly complex structure. Currently there is no direct evidence thatthe neurotransmitters, neuropeptides, and receptors observed in the BNST areinvolved in the specific underlying pathologic mechanisms of OCD, but substantialevidence suggests that the activity of the BNST mediates many forms of anxietybehavior in humans and animals (Straube et al. 2007; Walker et al. 2003).

A widely adopted neural circuit model of fear places the amygdala in centerstage and assigns different functions to different amygdala subdivisions. Thebasolateral amygdala screens incoming sensory information for threat cues andpasses information on such cues to the central nucleus of the amygdala’s medialsubdivision, which mediates threat responses. This model should, however, beexpanded to include the BNST as an important component of anxiety circuitry.

In most mammals, the extended amygdala consists of a ring of neuronsencircling the internal capsule and basal ganglia. Remnants from an embryoniccontinuous structure connecting the BNST and the central nucleus of the amygdalaform interrupted cell columns within the stria terminalis as it takes a semicirculardetour above and behind the internal capsule and thalamus (Martin et al. 1991).The extended amygdala is directly continuous with the caudomedial shell of thenucleus accumbens and together they establish specific neuronal circuits with themedial prefrontal cortex–OFC (Heimer et al. 1997, 2003). They project signifi-cantly to many areas in the hypothalamus and the brainstem, including the ven-trolateral part of the periaqueductal gray matter, which has received considerableattention as a prominent staging area for the coordination of somatomotor andautonomic responses in affective defensive behavior. The amygdala and the BNSTare critically involved in the mediation of stimulus-specific fear and anxiety andboth receive processed sensory information from the basolateral nucleus of theamygdala and hence are in a position to respond to emotionally significant stimuli(Davis and Shi 1999).

4 Deep Brain Stimulation in the Ventral Capsule/Ventral Striatum 37

4.4 Differences in Responses to Short-Duration Versus Long-Duration Threats in the Amygdala and BNST

The central nucleus of the amygdala and the BNST are closely related and serve similarbut complementary functions in fear conditioning. The central nucleus of the amygdalamediates short-duration but not long-duration threat responses and the BNST mediateslong-duration but not short-duration threat responses (Walker et al. 2009). There isgood evidence in support of the hypothesis that the sustained fear system, and thus theBNST, exerts an inhibitory influence on the phasic (short-duration) fear system.

Lesion, stimulation, and pharmacological studies suggest that the centralnucleus of the amygdala is the main output station of the amygdala for the rapidgeneration of brief conditioned fear responses to discrete sensory cues. Althoughthe BNST does not seem to be involved in learning to fear an explicit stimulus(e.g., a tone or a light presented only in the presence of the aversive stimulus), it isinvolved in learning to fear more general, long-lasting cues (Walker et al. 2003;Davis et al. 1997). BNST lesions do not disrupt conditioned fear responses to cuesbut disrupt contextual fear responses and moreover, they mediate slowly devel-oping and long-lasting responses to diffuse threats (Sullivan et al. 2004). TheBNST contributes to coding the appetitive outcome of a given situation. It con-tributes to a general ‘‘awareness’’ loosely linked to a particular context, rather thanthe prediction of specific outcomes by discrete cues (Walker et al. 2009).

4.5 OCD and BNST

The BNST is a limbic forebrain structure and a subregion of the extendedamygdala. It receives heavy projections from the basolateral amygdala and pro-jects in turn to hypothalamic and brainstem target areas that mediate many of theautonomic and behavioral responses to aversive or threatening stimuli. As such,the BNST plays a complementary role in regulation of physiological changesassociated with chronic stress exposure.

Although the BNST may not be necessary for rapid-onset, short-durationbehaviors which occur in response to specific threats, the BNST may mediate slower-onset, longer-lasting responses that frequently accompany sustained threats, and thatmay persist even after threat termination. It has been implicated in longer-durationand sustained increases in anxiety-like behavior. Modulation of the BNST could thusmodulate these longer-lasting responses to a perceived sustained threat as in OCD.

The BNST is considered a relay and an integral regulator of the hypothalamic–pituitary–adrenal stress axis. It acts as a critical intermediary by receiving stressorinputs from the corticolimbic system and sending projections (primarilyGABAergic) to the paraventricular nucleus of the hypothalamus, where cortico-trophin-releasing hormone is released, inducing pituitary activation. It thus directlyinitiates and influences the peripheral stress response (Dunn 1987).

38 L. Gabriëls and B. Nuttin

Although learning to fear stimuli that predict danger promotes survival, theinability to inhibit fear responses to inappropriate cues leads to a pernicious cycleof avoidance behaviors. Clinical anxiety is often thought of as an inability toappropriately inhibit fear, and interventions that facilitate the ability of patients toinhibit fear offer an effective strategy for the treatment of anxiety and reduceavoidance behavior. Studies have revealed large interindividual variations inresponses to fear, with clinically anxious humans exhibiting a tendency to gen-eralize learned fear responses to safe stimuli or situations. There is evidence thatthese interindividual variations in fear generalization are determined by theinfluences of the BNST on the amygdala and/or its targets (Duvarci et al. 2009).

Aversive events are more debilitating when they occur unpredictably than whenthey occur predictably. Both a predictable and an unpredictable threat evoketransient activity in the dorsal amygdala, but only an unpredictable threat producessustained activity in a forebrain region corresponding to the BNST (Alvarez et al.2011). Exposure to triggers for anxiety provoking obsessive thoughts and com-pulsive rituals in OCD may be perceived as a predictable threat, but this is onlypart of the threat. What characterizes OCD patients is the concern with the risk thatthe threat is not brought under full control. For them, the uncertainty, the (smallbut unpredictable) chance that the (perceived) exposure, might lead to the ‘‘veryunlikely but possible’’ feared consequence is unbearable.

In the presence of threat stimuli, two classes of defensive behaviors are elicited;those that are associated with imminent danger and are characterized by flight orfight (fear), and those that are associated with temporally uncertain danger and arecharacterized by sustained apprehension and hypervigilance (anxiety). This dis-tinction between (phasic, short-lasting) fear of an imminent threat and (sustained,long-lasting) anxiety from temporally uncertain danger is suggested by evidencefrom ethological studies and can be traced back to distinct neuroanatomical sys-tems, the amygdala and the BNST (Grillon 2008).

Behavioral inhibition in response to unfamiliar individuals and/or novelty is amarker of an anxious temperament and an early predictor of subsequent devel-opment of anxiety disorders. OFC lesions in monkeys decrease behavioral inhi-bition. Metabolism in the BNST region and individual differences in BNSTactivity predict behavioral inhibition (Kalin et al. 2007). An important function ofthe OFC in response to a threat is to modulate the BNST, and thus more directlyinfluence the expression of behavioral inhibition (Fox et al. 2010).

Patients with OCD experience chronic apprehension and arousal related to thepotential occurrence of threats. The level of apprehension is inappropriate giventhe environment, leading to tension, behavioral impairments, and distress. OCDpatients are hypervigilant, as manifested by an enhanced state of arousal andreadiness to deal with potential threats and often accompanied by negative affectstates and activation of the autonomic nervous system. Typically, this hypervigi-lance is characterized by heightened monitoring of the environment for cues andtriggers related to one’s future level of threat or safety. Neurobiologically, cuedthreat processing is initiated by the amygdala, whereas sustained vigilance asso-ciated with ambiguous or distant threat cues is represented by tonic engagement of

4 Deep Brain Stimulation in the Ventral Capsule/Ventral Striatum 39

the BNST. Recently, elevated resting metabolism within the BNST has beenidentified as mediating the trait of anxious temperament in primates (Oler et al.2009) and BNST lesions disrupt individual variability in rodent anxiety-likebehavior (Duvarci et al. 2009). Animal models suggest that hypervigilant threatmonitoring is distinct from cued fear-like responses and is mediated by the BNST.The role of the human BNST in mediating environmental threat monitoring wasrecently investigated with functional MRI. Activity in the BNST and the insulacorrelates with continuous monitoring of changes in the environmental threat leveland subserved hypervigilant threat-monitoring processes in more highly traitanxious individuals (Somerville et al. 2010).

4.6 Conclusion

Research into DBS for the treatment of OCD was initiated in 1998 as a therapeuticinnovation, and DBS was investigated as an alternative to treatment by stereotacticbilateral anterior capsulotomy. Gradually, on the basis of observations of betteroutcome in patients with more posterior and medial lead placements, the targetshifted from the ALIC to the BNST.

The BNST is a structure at the crossroads of the CSTC circuit involved in OCDand the amygdala. Its activity is modulated by the OFC, and it is involved in manyaspects of anxiety, behavioral inhibition, and hypervigilance that are present in OCDpatients. The fact that DBS in this target area can suppress OCD symptoms mustprompt further research into the role of the (dys)function of the BNST in OCD.

References

Alvarez RP, Chen G, Bodurka J, Kaplan R, Grillon C (2011) Phasic and sustained fear in humanselicits distinct patterns of brain activity. Neuroimage 55(1):389–400

Axer H, von Keyserlingk D (2000) Mapping of fiber orientation in human internal capsule bymeans of polarized light and confocal scanning laser microscopy. J Neurosci Methods94:165–175

Cannistraro PA, Makris N, Howard JD et al (2007) A diffusion tensor imaging study of whitematter in obsessive–compulsive disorder. Depress Anxiety 24:440–446

Davis M, Shi S (1999) The extended amygdala: are the central nucleus of the amygdale and thebed nucleus of the stria terminalis differentially involved in fear versus anxiety? Ann N YAcad Sci 877:281–291

Davis M, Walker DL, Lee Y (1997) Amygdala and bed nucleus of the stria terminalis: differentialroles in fear and anxiety measured with the acoustic startle reflex. Philos Trans R Soc Lond BBiol Sci 352:1675–1687

Dunn JD (1987) Plasma corticosterone responses to electrical stimulation of the bed nucleus ofthe stria terminalis. Brain Res 407(2):327–331

Duvarci S, Bauer E, Paré D (2009) The bed nucleus of the stria terminalis mediates inter-individual variations in anxiety and fear. J Neurosci 29(33):10357–10361

40 L. Gabriëls and B. Nuttin

Fox AS, Shelton SE, Oakes TR et al (2010) Orbitofrontal cortex lesions alter anxiety-relatedactivity in the primate bed nucleus of stria terminalis. J Neurosci 30:7023–7027

Greenberg BD, Gabriels LA, Malone DA Jr et al (2010) Deep brain stimulation of the ventralinternal capsule/ventral striatum for obsessive–compulsive disorder: worldwide experience.Mol Psychiatry 15(1):64–79

Grillon C (2008) Models and mechanisms of anxiety: evidence from startle studies.Psychopharmacology 199(3):421–437

Heimer L (2003) A new anatomical framework for neuropsychiatric disorders and drug abuse.Am J Psychiatry 160:1726–1739

Heimer L, Harlan RE, Alheid GF et al (1997) Substantia innominata: a notion which impedesclinical-anatomical correlations in neuropsychiatric disorders. Neuroscience 76(4):957–1006

Jenike MA (1998) Neurosurgical treatment of obsessive–compulsive disorder. Br J PsychiatrySuppl 35:79–90

Jenike MA, Breiter HC, Baer L et al (1996) Cerebral structural abnormalities in obsessive–compulsive disorder: a quantitative morphometric magnetic resonance imaging study. ArchGen Psychiatry 53:625–632

Kalin NH, Shelton SE, Davidson RJ (2007) Role of the primate orbitofrontal cortex in mediatinganxious temperament. Biol Psychiatry 62:1134–1139

Kopell BH, Greenberg B, Rezai AR (2004) Deep brain stimulation for psychiatric disorders.J Clin Neurophysiol 21(1):51–67

Mai J, Assheuer J, Paxinos G (1998) Atlas of the human brain. Acadamic Press, Orlando,pp 162–172

Martin LJ, Powers RE, Dellovade TL, Price DL (1991) The bed nucleus-amygdala continuum inhuman and monkey. J Comp Neurol 309:445–485

Nuttin B, Cosyns P, Demeulemeester H, Gybels J, Meyerson B (1999) Electrical stimulation inanterior limbs of internal capsules in patients with obsessive–compulsive disorder. Lancet354:1526

Nuttin B, Gabriëls L, Cosyns P et al (2003) Long-term electrical capsular stimulation in patientswith obsessive–compulsive disorder. Neurosurgery 52(6):1263–1274

Oler JA, Fox AS, Shelton SE et al (2009) Serotonin transporter availability in the amygdala andbed nucleus of the stria terminalis predicts anxious temperament and brain glucose metabolicactivity. J Neurosci 29:9961–9966

Saxena S, Rauch SL (2000) Functional neuroimaging and the neuroanatomy of obsessive–compulsive disorder. Psychiatr Clin N Am 23:563–586

Somerville LH, Whalen PJ, Kelley WM (2010) Human bed nucleus of the stria terminalis indexeshypervigilant threat monitoring. Biol Psychiatry 68:416–424

Straube T, Mentzel HJ, Miltner WH (2007) Waiting for spiders: brain activation duringanticipatory anxiety in spider phobics. Neuroimage 37:1427–1436

Sullivan GM, Apergis J, Bush DE et al (2004) Lesions in the bed nucleus of the stria terminalisdisrupt corticosterone and freezing responses elicited by a contextual but not by a specific cueconditioned fear stimulus. Neuroscience 128:7–14

van Kuyck K, Gabriëls L, Nuttin B (2009) Electrical brain stimulation in treatment-resistantobsessive–compulsive disorder: parcellation, and cyto- and chemoarchitecture of the bednucleus of the stria terminalis, a review. In: Krames ES, Peckham PH, Rezai AR (eds)Neuromodulation. Academic Press, London

Walker DL, Toufexis DJ, Davis M (2003) Role of the bed nucleus of the stria terminalis versusthe amygdala in fear, stress, and anxiety. Eur J Pharmacol 463:199–216

Walker DL, Miles LA, Davis M (2009) Selective participation of the bed nucleus of the striaterminalis and CRF in sustained anxiety-like versus phasic fear-like responses. ProgNeuropsychopharmacol Biol Psychiatry 33:1291–1308

4 Deep Brain Stimulation in the Ventral Capsule/Ventral Striatum 41

Chapter 5Deep Brain Stimulation inObsessive–Compulsive DisorderTargeted at the Nucleus Accumbens

Pelle P. de Koning, Pepijn van den Munckhof, Martijn Figee,Rick Schuurman and Damiaan Denys

5.1 Introduction

Obsessive–compulsive disorder (OCD) is a chronic disabling anxiety disordercharacterized by recurrent intrusive thoughts and/or repetitive compulsorybehaviors. OCD has an estimated lifetime prevalence of 2 %, afflicting men andwomen equally (Ruscio et al. 2010). Although most patients may benefit from drugtherapies and/or cognitive behavioral therapy, about 10 % of patients are con-sidered therapy-resistant (Denys 2006). For a small proportion of these patients,deep brain stimulation (DBS) may be an appropriate intervention/treatmentapproach.

DBS is a neurosurgical treatment involving the implantation of electrodes thatsend electrical impulses to specific locations in the brain selected according to thetype of symptoms to be addressed. On the basis of published trials and casestudies, it is estimated that a total of approximately 100 patients with OCD havereceived experimental DBS using five different brain targets: (1) anterior limb ofthe internal capsule (ALIC); (2) ventral striatum/ventral capsule; (3) subthalamicnucleus (STN); (4) inferior thalamic peduncle; and (5) nucleus accumbens (NAc).

In this chapter we will shed light on the latter, hereby focusing on the anatomyand function of the accumbal region, followed by elucidating the rationale, effi-cacy, and side effects of NAc DBS.

P. P. de Koning (&) � P. van den Munckhof � M. Figee �R. Schuurman � D. DenysUniversity of Amsterdam, Amsterdam, The Netherlandse-mail: p.p.dekoning@amc.nl

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_5, � Springer-Verlag Berlin Heidelberg 2012

43

5.2 Functional Anatomy of the Nucleus Accumbens

The NAc is part of the ventral striatum. It is located where the head of the caudateand the anterior portion of the putamen meet just inferior to the ALIC. The NAccan be divided into two principal parts. The medial, ventral, and lateral parts of theNAc are considered to be the shell of the NAc (NAc shell), whereas the central anddorsal parts are commonly referred to as the core of the NAc (NAc core) (Groe-newegen et al. 1999). The NAc core preferentially projects to classic striataltargets such as the pallidal and nigral complex. In addition, the NAc shell connectswith output areas such as the lateral hypothalamic areas, dopaminergic cell groups,and caudal mesencephalic areas that have been associated with locomotor func-tions (Voorn et al. 2004) (Fig. 5.1).

The principal neuronal cell type found in the NAc is the medium spiny neuron(MSN). The neurotransmitter produced by these neurons is c-aminobutyric acid,one of the main inhibitory neurotransmitters of the central nervous system. MSNare responsible for integration of dopaminergic and glutamatergic signaling. MSN

1 Isolated fibre bundles of the corona radiata2 Tail of the nucleus caudatus3 Strands of grey matter (pontes grisei) con-

necting the caudate nucleus with the putamen4 Corpus of the caudate nucleus5 Putamen6 Outline of the thalamus7 Internal capsule, one isolated fibre

bundle indicated8 Globus pallidus, external segment9 Globus pallidus, internal segment

10 Anterior commissure11 Head of the caudate nucleus12 Nucleus accumbens13 Peduncle of the lentiform nucleus14 Junction of the tail of the caudate nucleus

with the peduncle of the lentiform nucleus15 Cerebral peduncle

Fig. 5.1 The basal ganglia in medial view (6/5x), with the nucleus accumbens (12) below theinternal capsule and ventromedial to the caudate nucleus (from Nieuwenhuys et al. 2007 withpermission)

44 P. P. de Koning et al.

receive glutamatergic inputs from prefrontal association cortices and the basolat-eral amygdala and dopaminergic axons from the ventral tegmental area, whichconnect via the mesolimbic pathway (Heimer et al. 1997). Its main efferentsinnervate the pallidum, striatum, mediodorsal thalamus, and, as mentioned above,the mesolimbic dopaminergic area. The NAc thus attains a central positionbetween limbic and mesolimbic dopaminergic structures, basal ganglia, medio-dorsal thalamus, and prefrontal cortex.

5.3 Function of the Nucleus Accumbens

The NAc has been described as a limbic–motor interface, where learned associ-ations of motivational significance are converted into goal-directed behavior(Mogenson et al. 1980). Neuroanatomical and neurophysiological studies haverevealed latent neural mechanisms by which the NAc and its dopaminergicinnervation may select and integrate inputs from limbic structures such as thehippocampus and amygdala, as well as the prefrontal cortex. This led to the ideathat the NAc mediates goal-directed behavior by integrating hippocampus-dependent contextual information and amygdala-dependent affective informationwith prefrontal cortex cognitive functions to select appropriate behavioralresponses (Goto and Grace 2005; Gruber et al. 2009). Through these mechanismsthe NAc plays a key role in stress-related, sexual, feeding, drug self-administra-tion, and reward-related behaviors, as well as motivation, learning, and adaptivebehavior (Haber and Knutson 2010).

5.4 Rationale of Nucleus Accumbens Deep Brain Stimulationfor the Treatment of Obsessive–Compulsive Disorder

OCD has been hypothesized to be associated with deregulation of the cortico-striatal circuitry. Initial neurosurgical interventions lesioned the internal capsuleand parts of this corticostriatal circuitry. Because of the central position of the NAcin the basal ganglia between the amygdaloid complex, basal ganglia, mediodorsalthalamic nucleus, and prefrontal cortex, which are all involved in the patho-physiology of anxiety disorders (Shumyatsky et al. 2002) and OCD (Saxena andRauch 2000), it was suggested that the beneficial effects of anterior capsulotomymight be caused by blocking of amygdaloid–basal ganglia–prefrontal circuitry atthe level of the NAc rather than by blocking of the fiber tracts in the internalcapsule (Sturm et al. 2003).

The NAc is considered a promising target for DBS because there is evidence ofdysfunction of the reward system in OCD. OCD is characterized by the presence ofrecurrent and anxiety-provoking thoughts, images, or impulses (obsessions),

5 Deep Brain Stimulation 45

typically followed by repetitive ritualistic behaviors (compulsions) to relieveanxiety. OCD has been conceptualized as a disorder of behavioral addiction, withobsessions and compulsions being related to loss of voluntary control and adependency on repetitious, self-defeating behavior (Denys et al. 2004). Compul-sions can be viewed as addictive because of their rewarding effects followingreduction of obsession-induced anxiety. Addictive behavior is associated withdefective processing of natural rewards. The NAc, as part of the ventral striatum,has been implicated as a brain region that is critically involved in reward pro-cessing. In healthy humans, the ventral striatum is activated particularly inanticipation of a reward and in proportion to its expected value (Knutson et al.2001).

In a study by Figee et al. (2011) using a monetary incentive delay task andfunctional MRI, OCD patients showed attenuated reward anticipation activity inthe NAc compared with healthy controls. Reduced brain activity of the NAc wasmore pronounced in OCD patients with contamination fear compared with patientswith high-risk assessment symptoms. Their findings suggest an important role forthe NAc in the pathophysiology of OCD. OCD patients may be less able to makebeneficial choices because of defective NAc activation when anticipating rewards.Furthermore, the NAc is important for focusing on potential alerting andrewarding environmental stimuli that can be used for modulation of behavior byreinforcement learning. Therefore, the NAc may be less responsive when recruitedduring conventional reward processing owing to its bias toward drugs of abuse inaddiction, as well as toward obsessions and compulsions in OCD, supporting theconceptualization of OCD as a disorder of behavioral addiction.

5.5 Efficacy of Nucleus Accumbens Deep Brain Stimulationfor the Treatment of Obsessive–Compulsive Disorder

In 2003 Sturm et al. (2003) published the first results of unilateral, right-sided NAcimplantation in four OCD patients. In this open study, after 24–30 months, three offour patients were considered responders with no Yale–Brown Obsessive–Com-pulsive Scale (Y-BOCS) scores mentioned (Sturm et al. 2003). The same grouppublished a double-blind study on unilateral right-sided NAc DBS in ten OCDpatients (Huff et al. 2010). However, the symptom improvement observed in thedouble-blind part (3 months of active stimulation, 3 months of sham stimulation)of the study was limited to an average of 10 %. The mean Y-BOCS score wentfrom 27.9 in the active stimulation to 31.1 during sham stimulation. At 1-yearfollow-up, only one patient showed 35 % or greater symptom improvement, andwas therefore considered a responder. Five patients were considered partialresponders (25 % or greater symptom improvement). The average Y-BOCSsymptom decrease for all subjects at 1-year follow-up was 21 % (6.8 points). Anopen stimulation case study (Aouizerate et al. 2004) on NAc/ventral caudate DBS

46 P. P. de Koning et al.

for the treatment of OCD and depression reported a delayed 52 % decrease ofsymptoms at 15-month follow-up. In 2010, Franzini et al. (2010) reported on NAcDBS for the treatment of OCD in two patients. They reported an average symptomimprovement of 38 % (12 points). Denys et al. (2010) published the most exten-sive study on NAc DBS in 2010. This study consisted of an open 8-month treat-ment phase, followed by a double-blind cross-over phase with randomly assigned2-week periods of active or sham stimulation. It ended with an open 12-monthmaintenance phase. Sixteen patients with treatment-resistant OCD were includedin this study. This resulted in an average 46 % symptom decrease after 8 months.Nine of 16 patients were responders during follow-up. These nine subjects had amean Y-BOCS score decrease of 72 % (23.7 points). The average symptomdecrease at follow-up at 21 months for all 16 subjects was 48 % (17.5 points). Inthe double-blind, sham-controlled phase (n = 14), the mean Y-BOCS differencebetween active and sham stimulation was 25 % (8.3 points).

5.6 Difference Between the Target Site and the StimulationSite

It is of importance to realize that the actual stimulation site of the electrodes withDBS may differ from the target site, which is always the lowest electrode.Although the target of the lowest electrode in the Amsterdam sample was locatedat the NAc core, in most of the patients the beneficial effect of DBS was achievedby active stimulation in the upper two contacts of the electrodes These activecontracts are actually located at the border between the NAc core and the ventralpart of the internal capsule rather than in the NAc itself. As was mentioned inanother article, strikingly, the patients did not benefit from activation of the lowestelectrodes, but only from activation of the upper electrodes (Denys et al. 2010). Itis currently unclear which brain areas are involved with NAc stimulation, butpresumably, the efficacy is due to activation of axonal fibers running through theventral part of the internal capsule eventually modulating the prefrontal cortexand/or amygdala (Cohen et al. 2012) (Table 5.1).

5.7 Side Effects of Nucleus Accumbens Deep BrainStimulation

5.7.1 Mood Effects

Acute mood changes during the first few days of stimulation of the NAc have beenreported, such as transient sadness, anxiety, and euphoria, sometimes to the extentof hypomanic and manic symptoms (Okun et al. 2007). Transient mania or

5 Deep Brain Stimulation 47

Tab

le5.

1N

ucle

usac

cum

bens

deep

brai

nst

imul

atio

nst

udie

sin

the

trea

tmen

tof

obse

ssiv

e–co

mpu

lsiv

edi

sord

er

Aut

hors

(n)

Bil

ater

alve

rsus

unil

ater

al

Stu

dyde

sign

Fol

low

-up

(mon

ths)

Fol

low

-up

DY

-B

OC

S(a

llsu

bjec

ts)

Res

pond

ers

atfo

llow

-up

(C35

%Y

-BO

CS;)

Y-B

OC

Sch

ange

(act

ive

vers

ussh

amph

ase

Stu

rmet

al.

(200

3)5

U(R

-NA

c)O

pen

24–3

0N

oY

-BO

CS

-sco

res

3(6

0%

)N

A

Aou

izer

ate

etal

.(2

004)

1B

Ope

n27

;13

poin

ts1

(100

%)

NA

Huf

fet

al.

(201

0)10

U(R

-NA

c)D

oubl

ebl

ind,

3m

inac

tive

,3

min

sham

12;

6.8

poin

ts1

(10

%),

(5pa

rtia

lre

spon

ders

:[25

%Y

-B

OC

S;)

27.9

vers

us31

.1

Den

yset

al.

(201

0)16

BD

oubl

ebl

ind,

2w

eeks

acti

ve,

2w

eeks

sham

21;

17.5

poin

ts9

(56

%)

8.9-

Poi

ntdi

ffer

ence

Fra

nzin

ieta

l.(2

010)

2B

Ope

n24

–27

;13

poin

ts1

(50

%)

NA

Bbi

late

ral,

Uun

ilat

eral

,R

-NA

cri

ght-

side

dnu

cleu

sac

cum

bens

,Y

-BO

CS

Yal

e–B

row

nO

bses

sive

–Com

puls

ive

Sca

le,

NA

not

avai

labl

e

48 P. P. de Koning et al.

hypomania after implantation of the DBs electrodes has been reported in severaltargets, including the globus pallidus, STN, and the ALIC–NAc region. Allhypomanic and manic episodes associated with DBS dissolved after the fielddensity had been readjusted by changing the voltage and/or the active contact.Transient hypomania is the most commonly observed side effect immediately afterstimulation. Transient hypomanic episodes seem to occur more often in the VC/VS–NAc region. The occurrence is estimated to be as high as 50–67 % in ALIC–NAc DBS patients, as contrasted with 4–8 % in STN DBS patients. Chronic moodimprovement is an unintended but favorable side effect of DBS because mosttreatment-resistant OCD patients have comorbid major depression. Denys et al.(2010), Abelson et al. (2005), and Greenberg et al. (2010) reported improvementof depressive symptoms after NAc, ALIC and ventral striatum/ventral capsulestimulation, respectively. Antidepressive effects seem thus to be especially relatedto DBS of the ventral striatum because no mood improvement was observedfollowing STN stimulation (Mallet et al. 2008).

5.7.2 Impulsivity

A case report by Luigjes et al. (2011) illustrated that DBS in the area of the NAcmay cause immediate changes in impulsivity related to the applied voltage inpatients with OCD. They suggested that increasing the voltage of DBS in the areaof the NAc may affect impulsivity in patients with OCD. In both patients, reducingthe voltage of the stimulation could redress increased impulsivity. The preciselocation and amplitude of stimulation might be critical in inducing these behaviors.In contrast with impulsivity in the context of a hypomanic episode, which iscommonly observed in the first 3–4 days after DBS stimulation of the effectivecontact points, the increased impulsivity was not associated with mood elevationor restlessness, thereby supporting the idea that impulsivity and hypomania afterDBS may be unrelated side effects. The precise location and amplitude of stim-ulation might be critical in inducing these behaviors. However, the exact mech-anisms by which the changes occurred remain to be investigated.

5.7.3 Cognitive Effects

Apart from transient diminished concentration and verbal perseverations, NAcDBS has not been associated with evident cognitive decline and/or cognitivefunction improvement. However, the literature on this topic is sparse. Analysis ofneuropsychological testing in the study by Aouizerate et al. (2004; one subject)showed no deterioration in memory, attentional, or executive function tests. Denyset al. (2010) reported mild forgetfulness in five of 16 patients and word-findingproblems in three of 16 patients following NAc DBS.

5 Deep Brain Stimulation 49

5.8 Conclusion

The NAc is an effective target in DBS for therapy-resistant OCD patients, althoughthe stimulation might actually be applied in the ventral internal capsule. Sideeffects are sparse, generally mild, and largely reversed by readjusting the stimu-lation. Interestingly, the beneficial effects on mood and anxiety, along withimprovement in obsessions and compulsions, are striking, as even nonrespondersoften experience substantial mood improvement.

References

Abelson JL, Curtis GC, Sagher O et al (2005) Deep brain stimulation for refractory obsessivecompulsive disorder. Biol Psychiatry 57(5):510–516

Aouizerate B, Cuny E, Martin-Guehl C et al (2004) Deep brain stimulation of the ventral caudatenucleus in the treatment of obsessive-compulsive disorder and major depression. Case report.Neurosurgery 101(4):682–686

Cohen MX, Bour L, Mantione M, Figee M, Vink M, Tijssen MA, van Rootselaar AF, van denMunckhof P, Schuurman PR, Denys D (2012) Top-down-directed synchrony from medial frontalcortex to nucleus accumbens during reward anticipation. Hum Brain Mapp 33(1):246–252

Denys D (2006) Pharmacotherapy of obsessive-compulsive disorder and obsessive-compulsivespectrum disorders. Psychiatr Clin N Am 29(2):553–584, xi

Denys D, Zohar J, Westenberg HG (2004) The role of dopamine in obsessive-compulsivedisorder: preclinical and clinical evidence. J Clin Psychiatry 65(Suppl 14):11–17

Denys D, Mantione M, Figee M et al (2010) Deep brain stimulation of the nucleus accumbens fortreatment-refractory obsessive-compulsive disorder. Arch Gen Psychiatry 67(10):1061–1068

Figee M, Vink M, de Geus F, Vulink N, Veltman DJ, Westenberg H, Denys D (2011) Dysfunctionalreward circuitry in obsessive-compulsive disorder. Biol Psychiatry 69(9):867–874

Franzini A, Messina G, Gambini O et al (2010) Deep-brain stimulation of the nucleus accumbensin obsessive compulsive disorder: clinical, surgical and electrophysiological considerations intwo consecutive patients. Neurol Sci 31(3):353–359

Goto Y, Grace AA (2005) Dopaminergic modulation of limbic and cortical drive of nucleusaccumbens in goal-directed behavior. Nat Neurosci 8:805–812

Greenberg BD, Gabriels LA, Malone DA et al (2010) Deep brain stimulation of the ventralinternal capsule/ventral striatum for obsessive-compulsive disorder: worldwide experience.Mol Psychiatry 15(1):64–79

Groenewegen HJ, Wright CI, Beijer AV et al (1999) Convergence and segregation of ventralstriatal inputs and outputs. Ann N Y Acad Sci 877:49–63

Gruber AJ, Hussain RJ, O’Donnell P (2009) The nucleus accumbens: a switchboard for goal-directed behaviors. PLoS One 4:e5062

Haber SN, Knutson B (2010) The reward circuit: linking primate anatomy and human imaging.Neuropsychopharmacology 35(1):4–26

Heimer L, Alheid GF, de Olmos JS et al (1997) The accumbens: beyond the core-shelldichotomy. J Neuropsychiatry Clin Neurosci 9(3):354–381

Huff W, Lenartz D, Schormann M, Lee SH et al (2010) Unilateral deep brain stimulation of thenucleus accumbens in patients with treatment-resistant obsessive-compulsive disorder:outcomes after one year. Clin Neurol Neurosurg 112(2):137–143

Knutson B, Adams CM, Fong GW (2001) Anticipation of increasing monetary reward selectivelyrecruits nucleus accumbens. J Neurosci 21:RC159

50 P. P. de Koning et al.

Luigjes J, Mantione M, van den Brink W et al (2011) Deep brain stimulation increasesimpulsivity in two patients with obsessive-compulsive disorder. Int Clin Psychopharmacol26(6):338–340

Mallet L, Polosan M, Jaafari N et al (2008) Subthalamic nucleus stimulation in severe obsessive-compulsive disorder. N Engl J Med 359(20):2121–2134

Mogenson GJ, Jones DL, Yim CY (1980) From motivation to action: functional interfacebetween the limbic system and the motor system. Prog Neurobiol 14:69–97

Nieuwenhuys R, Voogd J, van Huijzen C (2007) The human central nervous system: a synopsisand atlas, 4th edn. Springer, Berlin

Okun MS, Man G, Foote KD et al (2007) Deep brain stimulation in the internal capsule andnucleus accumbens region: responses observed during active and sham programming.J Neurol Neurosurg Psychiatry 78(3):310–314

Ruscio AM, Stein DJ, Chiu WT, Kessler RC (2010) The epidemiology of obsessive-compulsivedisorder in the National Comorbidity Survey Replication. Mol Psychiatry 15(1):53–63

Saxena S, Rauch SL (2000) Functional neuroimaging and the neuroanatomy of obsessive-compulsive disorder. Psychiatr Clin N Am 23(3):563–586

Shumyatsky GP, Tsvetkov E, Malleret G, Vronskaya S, Hatton M, Hampton L, Battey JF, DulacC, Kandel ER, Bolshakov VY (2002) Identification of a signaling network in lateral nucleusof amygdala important for inhibiting memory specifically related to learned fear. Cell111(6):905–918

Sturm V, Lenartz D, Koulousakis A et al (2003) The nucleus accumbens: a target for deep brainstimulation in obsessive-compulsive- and anxiety-disorders. J Chem Neuroanat 26(4):293–299

Voorn P, Vanderschuren LJ, Groenewegen HJ et al (2004) Putting a spin on the dorsal-ventraldivide of the striatum. Trends Neurosci 27(8):467–474

5 Deep Brain Stimulation 51

Chapter 6What is the Role of the SubthalamicNucleus in Obsessive–CompulsiveDisorder? Elements and Insightsfrom Deep Brain Stimulation Studies

William I. A. Haynes and Luc Mallet

6.1 Introduction

Obsessive–compulsive disorder (OCD) consists of a combination of intrusive,anxious thoughts (obsessions) and repetitive behaviours (compulsions), which arenot pathological in nature but are pathological because of their highly repetitiveand stereotyped expression. Given the dysfunction of a cortico-subcortical loop(Haynes and Mallet 2010), running from the orbitofrontal and anterior cingulatecortices to the medial thalamus through the limbic basal ganglia, it is now part ofthose psychiatric illnesses for which deep brain stimulation (DBS) is being tried.Of the three targets in use, we will focus on the subthalamic nucleus (STN), theother two being treated elsewhere (nucleus accumbens and ventral capsule/ventralstriatum; see de Koning et al. and Gabriëls and Nuttin, this volume).

6.2 STN Stimulation for the Treatment of OCD: Historyand Results

The use of high-frequency stimulation of the STN for the treatment of OCD was firstperformed fortuitously (Mallet et al. 2002). Two patients at the Pitié-SalpêtrièreHospital in Paris with a long history of OCD (33 and 40 years of evolution) receivedDBS of the STN as routine surgery in the context of their Parkinson’s disease (PD)and were subsequently included in a prospective study of the non-motor effects of

W. I. A. Haynes � L. Mallet (&)Team—Behavior, Emotion, and Basal Ganglia, Pitié-Salpêtrière Hospital,ICM—Brain and Spine Institute, UPMC—Inserm UMR_S 975, CNRS UMR 7225,Cedex 13, 75651 Paris, Francee-mail: luc.mallet@upmc.fr

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_6, � Springer-Verlag Berlin Heidelberg 2012

53

high-frequency stimulation of the STN (Houeto et al. 2006). The stimulationparameters were set to those normally used for the treatment of PD. After 2 weeks ofstimulation, both patients reported the disappearance of compulsions with a decreasein obsessions. The effect on OCD symptoms was deemed relatively independent ofthe neurological results of the stimulation because OCD and PD appeared unrelatedin both patients, and one patient presented experienced only a mild improvement ofmotor symptoms (specifically tremor), probably because of the very anterior locationof electrodes in this patient. Furthermore, because PD and OCD are viewed as theresult of opposite imbalances in the basal ganglia, it is theoretically unlikely that bothwould be affected in the same way by high-frequency stimulation of the STN (Voon2004). Fontaine et al. (2004) described a similar case, with marked improvement ofPD symptoms and the disappearance of OCD symptoms. As in the first case reports,the effects on OCD symptoms appeared after 1 week of stimulation. This effect oncompulsions and obsessions was attributed to a target more medial and anterior thanthe usual one. It was hypothesised that the therapeutic action on OCD symptoms wasachieved through inhibition of the anterior tip of the STN, which belongs to thelimbic basal ganglia–thalamocortical loop (Bevan et al. 1997). To test this hypoth-esis, we performed high-frequency stimulation of the STN at different plots of theelectrodes with the electrodes using the same intensities in PD patients. On the basisof current models of current diffusion (McIntyre et al. 2004; Chaturvedi et al. 2010)and of our clinical observations, we were able to confirm the functional existence ofthe limbic, associative and motor territories in the STN (Mallet et al. 2007).

These first clinical results prompted the design of a larger double-blind cross-over study of high-frequency stimulation of the STN as a therapeutic tool in casesof severe OCD resistant to conventional pharmacological and psychologicaltreatments (e.g. selective serotonin reuptake inhibitors and cognitive and behav-ioural therapy) (Mallet et al. 2008). Sixteen patients were randomised into twogroups. The first group received 3 months of active stimulation followed by3 months of sham stimulation interspaced by 1 month of washout. The secondgroup followed the opposite sequence of sham stimulation then active stimulation.The target used was 2 mm anterior and 1 mm medial to that used for PD surgery,in the limbic part of the STN as established by the 3D deformable atlas used by theteam (Yelnik et al. 2007). The stimulation parameters were similar to those usedfor PD (pulse width 60 ls, 130 Hz, mean voltage 2.0 ± 0.8 V). The effectivenessof subthalamic stimulation was shown in two ways. First, active stimulationinduced a fast and significant improvement in both OCD symptoms and globalfunctioning compared with sham stimulation within each group (mean Yale–Brown Obsessive–Compulsive Scale—Y-BOCS—score 19 ± 8 versus 28 ± 7,P = 0.01; mean Global Assessment of Functioning score 56 ± 14 versus 43 ± 8,P = 0.005). Second, after the first 3 months, 70 % of patients in the group withactive stimulation responded to treatment (reduction of 35 % or more of the Y-BOCS score) and 62 % achieved satisfactory global functioning compared to12 % (Y-BOCS and Global Assessment of Functioning) in the group with shamstimulation. There was no carryover effect of stimulation in the washout period,with Y-BOCS scores promptly returning to the baseline. Moreover, the stimulation

54 W. I. A. Haynes and L. Mallet

appeared to decrease metabolism of the left anterior cingulate gyrus as observedwith positron emission tomography, and reduction of the Y-BOCS score wascorrelated to a decrease of ventromedial prefrontal cortex metabolism (Le Jeuneet al. 2010).

Concerning side effects, we observed seven serious and seven minor psychi-atric/behavioural side effects linked to the stimulation (hypomania, anxiety andimpulsivity). These were all transient and corrected by a modification of thestimulation parameters, as were the eventual motor side effects (mainly dyskine-sia) (Mallet et al. 2008).

The preliminary results of the 3-year follow-up, together with open observa-tions of patients operated on after the STOC study (not published), are in favour ofa long-term efficacy of high-frequency stimulation of the STN for the treatment ofsevere and treatment-resistant OCD. High-frequency stimulation of the STNtherefore appears to be a valid therapeutic tool in the treatment of severe andresistant OCD.

6.3 How Can One Explain the Effect of STN Stimulationon OCD Symptoms?

As most of the literature on the STN’s role has focused on its motor functions, afirst mechanism for these clinical results could be that high-frequency stimulationof the STN has an effect on fibre pathways nearby. Modifications of activities inthe prefrontal areas could then be explained by the stimulation of fibres from/tothese areas travelling in the internal capsule and in fields H and H2 of Forell or bya loop effect through the reciprocal connections of the STN to the ventral pallidum(Bevan et al. 1997). Although modelling studies have focused on the PD target,one can surmise that, given the very low intensities used, fibres need to be in theimmediate vicinity of the electrode plot to be affected (Chaturvedi et al. 2010).Another option is for the stimulation to directly affect the STN and therefore itsafferences and efferences. Indeed, there is an increasing amount of literature on thecognitive (Frank et al. 2007; Sauleau et al. 2009; Eagle and Baunez 2010) andemotional (Huebl et al. 2011) functions of the STN as well as the well-docu-mented non-motor effects of DBS in PD (see Volkmann and Daniels, this volume),the said functions being supported by the STN’s reciprocal connections to theventral pallidum.

As DBS techniques allow one to record neuronal activities of the targetedstructures before stimulation is set up, we were able to show that a number ofparameters of the STN’s activity were modified in OCD in comparison with to PD(Welter et al. 2011) and animal data (PD models and controls) in the literature. Thefiring rate was lower in OCD, which would be more ‘normal’, but burst activity wasincreased in the anterior ventromedial area, in line with the findings of a previousstudy (Piallat et al. 2011) and with the associative and limbic functions associated

6 The Role of the Subthalamic Nucleus in Obsessive–Compulsive Disorder 55

with that area (Mallet et al. 2007). Furthermore, a number of burst parameters andoscillatory activities (delta and alpha bands) were correlated to the severity of thesymptoms, as assessed by the Y-BOCS, as well as to compulsion and obsessionsubscores; some of these characteristics were predictive of response to treatment by

56 W. I. A. Haynes and L. Mallet

DBS (Welter et al. 2011). It is therefore likely that at least part of the effect is due to aneffect directly on the STN’s activity. Also, increased bursting activities and oscil-latory activities are considered to reduce the transmission of information in thenetwork and therefore to amount to a form of functional inhibition. DBS in PD isthought to disrupt this pathological oscillatory activity to restore information flow(Kuhn et al. 2008). In OCD, alongside our observation of increased oscillation powerin the delta band, EEG studies have shown an increased power in the same band in themedial wall (Koprivova et al. 2011). Part of the pathophysiological mechanisms ofOCD could therefore be due to a synchronisation in the lower-frequency band of thecingulate and subthalamic activities, mediated by a hyperdirect prefronto-subtha-lamic pathway (Afsharpour 1985).

6.4 What Is the Role of the STN in the OCD Network?

A first, simple hypothesis is that, because of its loss of function in OCD, the STN isincapable of inhibiting unwanted motor programs through imbalance of the directand indirect pathways. OCD would then be some kind of impulse disorder. This isdifficult, if not impossible, to reconcile with what one knows of the proimpulsiveeffects of DBS in PD (Voon 2004; Frank et al. 2007), especially in the anterior tip(Hershey et al. 2010), or of subthalamic lesions in animal models (Eagle and Baunez2010) as well as clinical observations. OCD would, therefore, rather seem to be due toan excess of cognitive control (Bradbury et al. 2011; Meiran et al. 2011).

bFig. 6.1 Subthalamic nucleus (STN), decisions and obsessive–compulsive disorder. a Behav-ioural selection in the basal ganglia. 1a various contextual information transits via the hyperdirectpathway from the cortex to the STN. This information is integrated in the STN (2a) to set thedecisional threshold which is communicated to the globus pallidus pars internalis (3a). 1bdepending on the environment, different behavioural programs are activated cortically.Information in favour of each is accumulated in the striatum (2b) before proceeding to theglobus pallidus pars internalis. If the program has sufficient strength, it is able to go through thethreshold and be selected (3b). Once the behaviour has been expressed, outcome is integrated,leading to an update of contextual information available in the cortex (4). Program weights andthreshold are subsequently adapted to the new situation derived from (Bogacz and Larsen 2011;Frank et al. 2007). b Hypothesis 1. A primary hyperactivity of the STN results in an unusuallyhigh threshold (1). Most programs are, therefore, rejected (2). This is perceived as a signal togather more information, in order to choose/express a behaviour (3). The checking program thusreceives a ‘boost’ and is able to pass (4). The checking compulsion is expressed (5). The situationis assessed (6) and either found lacking again (3) or able to promote normal behaviour (7).c Hypothesis 2. A first ‘normal’ checking behaviour is performed. Owing to faulty action–outcome mechanisms (1), the result of this behaviour does not lead to an update of theenvironmental information available (2). Basal ganglia levels are not modified (3), and the samebehaviour (checking) is selected again (4). d Hypothesis 3. Pathological doubt (1) takespredominance over other contextual information (2a) and results in an unusually high threshold(3a). This doubt also promotes doubt-assuaging behaviours, including verification (2b), making itthe only program able to cross the doubt-driven threshold (3b) and be expressed. Because of thesame doubt, the outcome of the checking behaviour is not trusted (4) and cycle starts again

6 The Role of the Subthalamic Nucleus in Obsessive–Compulsive Disorder 57

Recent models of decision-making (action selection) in the basal ganglia positthat the STN’s role is to gate motor programs, to establish the amount of infor-mation needed for motor expression. This ‘decisional threshold’ would be theresult of the convergence of a variety of information from the motor, premotor andpre-supplementary motor area regions through the hyperdirect pathway (Bogaczand Larsen 2011) (Fig. 6.1a). Also, high-frequency stimulation of the STN andlesion studies in both humans and animals induce modifications in attentional,executive and limbic processes (Vicente et al. 2009; Eagle and Baunez 2010),while electrophysiological studies show that STN activity is influenced by thebehavioural relevance and emotional properties of environmental cues (Sauleauet al. 2009; Huebl et al. 2011). Combining that information, one can consider thatthe STN’s role in behavioural selection extends beyond simple motor selection.

The STN would therefore gather different types of information on the envi-ronment/context to set the amount of information any behavioural program needsin its favour to be expressed (Bogacz and Larsen 2011). In conflictual situations,more information would be required and behaviour would therefore be delayed(Frank 2006). One of the effects of high-frequency stimulation of the STN wouldbe to decrease this threshold and therefore induce impulsive behaviour and choices(Frank et al. 2007). With this model, the dysfunction of the STN in OCD can beexplained in one of three, non-exclusive, ways.

First, there could be a primary hyperactivity of the STN. This would result in apathologically high behavioural/decisional threshold and would thus triggerchecking compulsions in the attempt to gather more information to reach thethreshold. The obsessional content would then be created by the patient to ratio-nalise his or her need to check the compulsion (Fig. 6.1b).

Second, OCD patients would have difficulty integrating the outcome of theirbehaviour (i.e. processing reward) because of the nucleus accumbens dysfunction(Figee et al. 2011). They would therefore be unable to update the contextual informationavailable (cortical level). Thus, the subthalamic decisional threshold would not be resetand adapted for the following set of actions. The behaviour just expressed wouldcontinue to appear the better one (cross the threshold) and be repeated (Fig. 6.1c).

Finally, the pathological doubt central to OCD (Are my hands really clean? Ismy door really closed?), possibly caused by the cortical dysfunctions (Kepecs et al.2008; Chua et al. 2009), would modify the decisional threshold so that only thedoubt-assuaging behaviour could be expressed (Fig. 6.1d).

6.5 Conclusion

Subthalamic stimulation appears to be a promising option in severe, treatment-resistant OCD. However, one must remember that this is a delicate techniquewhich requires the presence of a highly expert multidisciplinary team. The lowvoltages used tip the scale slightly in favour of the subthalamic target because oflower energy consumption. Nevertheless, the effects of DBS at the three targets

58 W. I. A. Haynes and L. Mallet

(STN, nucleus accumbens, ventral capsule/ventral striatum) will need to becompared directly to conclude which is the best-suited approach.

Aside from clinical considerations, the electrophysiological recordings that arepossible provide one with a new model to study human STN function, and with apossibility to understand OCD pathophysiological mechanisms better. Preliminaryresults indicate that the nucleus is involved in behavioural gating, which would bedeficient in OCD.

References

Afsharpour S (1985) Topographical projections of the cerebral cortex to the subthalamic nucleus.J Comp Neurol 236:14–28

Bevan MD, Clarke NP, Bolam JP (1997) Synaptic integration of functionally diverse pallidalinformation in the entopeduncular nucleus and subthalamic nucleus in the rat. J Neurosci17:308–324

Bogacz R, Larsen T (2011) Integration of reinforcement learning and optimal decision-makingtheories of the basal ganglia. Neural Comput 23:817–851

Bradbury C, Cassin SE, Rector NA (2011) Obsessive beliefs and neurocognitive flexibility inobsessive-compulsive disorder. Psychiatry Res 187:160–165

Chaturvedi A, Butson CR, Lempka SF, Cooper SE, McIntyre CC (2010) Patient-specific modelsof deep brain stimulation: influence of field model complexity on neural activationpredictions. Brain Stimul 3:65–67

Chua EF, Schacter DL, Sperling RA (2009) Neural correlates of metamemory: a comparison offeeling-of-knowing and retrospective confidence judgments. J Cogn Neurosci 21:1751–1765

Eagle DM, Baunez C (2010) Is there an inhibitory-response-control system in the rat? Evidencefrom anatomical and pharmacological studies of behavioral inhibition. Neurosci BiobehavRev 34:50–72

Figee M, Vink M, de Geus F, Vulink N, Veltman DJ, Westenberg H, Denys D (2011)Dysfunctional reward circuitry in obsessive-compulsive disorder. Biol Psychiatry 69:867–874

Fontaine D, Mattei V, Borg M, von Langsdorff D, Magnie MN, Chanalet S, Robert P, Paquis P(2004) Effect of subthalamic nucleus stimulation on obsessive-compulsive disorder in apatient with Parkinson disease. Case report. J Neurosurg 100:1084–1086

Frank MJ (2006) Hold your horses: a dynamic computational role for the subthalamic nucleus indecision making. Neural Netw 19:1120–1136

Frank MJ, Samanta J, Moustafa AA, Sherman SJ (2007) Hold your horses: impulsivity, deepbrain stimulation, and medication in parkinsonism. Science 318:1309–1312

Haynes WIA, Mallet L (2010) High-frequency stimulation of deep brain structures in obsessive-compulsive disorder: the search for a valid circuit. Eur J Neurosci 32:1118–1127

Hershey T, Campbell MC, Videen TO, Lugar HM, Weaver PM, Hartlein J, Karimi M, TabbalSD, Perlmutter JS (2010) Mapping Go-No-Go performance within the subthalamic nucleusregion. Brain 133:3625–3634

Houeto JL, Mallet L, Mesnage V, Tezenas du Montcel S, Behar C, Gargiulo M, Torny F,Pelissolo A, Welter ML, Agid Y (2006) Subthalamic stimulation in Parkinson disease:behavior and social adaptation. Arch Neurol 63:1090–1095

Huebl J, Schoenecker T, Siegert S, Brucke C, Schneider GH, Kupsch A, Yarrow K, Kuhn AA(2011) Modulation of subthalamic alpha activity to emotional stimuli correlates withdepressive symptoms in Parkinson’s disease. Mov Disord 26:477–483

Kepecs A, Uchida N, Zariwala HA, Mainen ZF (2008) Neural correlates, computation andbehavioural impact of decision confidence. Nature 455:227–231

6 The Role of the Subthalamic Nucleus in Obsessive–Compulsive Disorder 59

Koprivova J, Congedo M, Horacek J, Prasko J, Raszka M, Brunovsky M, Kohutova B, Hoschl C(2011) EEG source analysis in obsessive-compulsive disorder. Clin Neurophysiol 122:1735–1743

Kuhn AA, Kempf F, Brucke C, Gaynor Doyle L, Martinez-Torres I, Pogosyan A, Trottenberg T,Kupsch A, Schneider GH, Hariz MI, Vandenberghe W, Nuttin B, Brown P (2008) High-frequency stimulation of the subthalamic nucleus suppresses oscillatory beta activity inpatients with Parkinson’s disease in parallel with improvement in motor performance.J Neurosci 28:6165–6173

Le Jeune F, Verin M, N’Diaye K, Drapier D, Leray E, Du Montcel ST, Baup N, Pelissolo A,Polosan M, Mallet L, Yelnik J, Devaux B, Fontaine D, Chereau I, Bourguignon A, Peron J,Sauleau P, Raoul S, Garin E, Krebs MO, Jaafari N, Millet B (2010) Decrease of prefrontalmetabolism after subthalamic stimulation in obsessive-compulsive disorder: a positronemission tomography study. Biol Psychiatry 68:1016–1022

Mallet L, Mesnage V, Houeto JL, Pelissolo A, Yelnik J, Behar C, Gargiulo M, Welter ML,Bonnet AM, Pillon B, Cornu P, Dormont D, Pidoux B, Allilaire JF, Agid Y (2002)Compulsions, Parkinson’s disease, and stimulation. Lancet 360:1302–1304

Mallet L, Schupbach M, N’Diaye K, Remy P, Bardinet E, Czernecki V, Welter ML, Pelissolo A,Ruberg M, Agid Y, Yelnik J (2007) Stimulation of subterritories of the subthalamic nucleusreveals its role in the integration of the emotional and motor aspects of behavior. Proc NatlAcad Sci U S A 104:10661–10666

Mallet L, Polosan M, Jaafari N, Baup N, Welter ML, Fontaine D, du Montcel ST, Yelnik J,Chereau I, Arbus C, Raoul S, Aouizerate B, Damier P, Chabardes S, Czernecki V, Ardouin C,Krebs MO, Bardinet E, Chaynes P, Burbaud P, Cornu P, Derost P, Bougerol T, Bataille B,Mattei V, Dormont D, Devaux B, Verin M, Houeto JL, Pollak P, Benabid AL, Agid Y, KrackP, Millet B, Pelissolo A (2008) Subthalamic nucleus stimulation in severe obsessive–compulsive disorder. N Engl J Med 359:2121–2134

McIntyre CC, Mori S, Sherman DL, Thakor NV, Vitek JL (2004) Electric field and stimulatinginfluence generated by deep brain stimulation of the subthalamic nucleus. Clin Neurophysiol115:589–595

Meiran N, Diamond GM, Toder D, Nemets B (2011) Cognitive rigidity in unipolar depressionand obsessive compulsive disorder: examination of task switching, Stroop, working memoryupdating and post-conflict adaptation. Psychiatry Res 185:149–156

Piallat B, Polosan M, Fraix V, Goetz L, David O, Fenoy A, Torres N, Quesada JL, Seigneuret E,Pollak P, Krack P, Bougerol T, Benabid AL, Chabardes S (2011) Subthalamic neuronal firingin obsessive-compulsive disorder and Parkinson disease. Ann Neurol 69:793–802

Sauleau P, Eusebio A, Thevathasan W, Yarrow K, Pogosyan A, Zrinzo L, Ashkan K, Aziz T,Vandenberghe W, Nuttin B, Brown P (2009) Involvement of the subthalamic nucleus inengagement with behaviourally relevant stimuli. Eur J Neurosci 29:931–942

Vicente S, Biseul I, Peron J, Philippot P, Drapier S, Drapier D, Sauleau P, Haegelen C, Verin M(2009) Subthalamic nucleus stimulation affects subjective emotional experience in Parkin-son’s disease patients. Neuropsychologia 47:1928–1937

Voon V (2004) Repetition, repetition, and repetition: compulsive and punding behaviors inParkinson’s disease. Mov Disord 19:367–370

Welter ML, Burbaud P, Fernandez-Vidal S, Bardinet E, Coste J, Piallat B, Borg M, Besnard S,Sauleau P, Devaux B, Pidoux B, Chaynes P, Tezenas du Montcel S, Bastian A, Langbour N,Teillant A, Haynes W, Yelnik J, Karachi C, Mallet L (2011) Basal ganglia dysfunction inOCD: subthalamic neuronal activity correlates with symptoms severity and predicts high-frequency stimulation efficacy. Transl Psychiatry 1:e5

Yelnik J, Bardinet E, Dormont D, Malandain G, Ourselin S, Tande D, Karachi C, Ayache N,Cornu P, Agid Y (2007) A three-dimensional, histological and deformable atlas of the humanbasal ganglia. I. Atlas construction based on immunohistochemical and MRI data.Neuroimage 34:618–638

60 W. I. A. Haynes and L. Mallet

Chapter 7Obsessive–Compulsive Disordersin Animals

Christine Winter

7.1 Introduction

Not all patients suffering from otherwise therapy-resistant obsessive–compulsivedisorder (OCD) benefit from deep brain stimulation (DBS), and among the patientsthat do, most of them only experience a delayed and partial relief of symptoms. Thisinconsistency in the demonstration of therapeutic effects suggests that the optimalDBS parameters and brain sites for the treatment of OCD have not been found yet.This, in turn, may result from insufficient understanding of the pathophysiologicalprocesses and resulting dysfunctional neuronal networks underlying OCD.

DBS delivered by means of intracerebrally implanted electrodes leads to atransient and specific modulation of neural function of selected brain nuclei andassociated networks via application of an electric current. As such, DBS serves notonly as a therapeutic alternative and clinical perspective in the treatment ofotherwise therapy-resistant neuropsychiatric disorders but also as a powerfulpreclinical tool for delineating functional circuitries in the healthy and diseasedbrain: Animal experimental and clinical data have converged to indicate that athigh frequencies DBS produces an overall net inhibition of the stimulated target,whereas overall net excitation may constitute the underlying mechanism of DBS atlow frequencies. This antipodal mode of action of DBS allows the detailedinvestigation of a specific brain region and associated networks and to drawconclusions on the (patho-)physiological activity of the areas investigated in thecourse of symptom manifestation and reduction.

C. Winter (&)Section Experimental Psychiatry, Department of Psychiatry and Psychotherapy,University Hospital Carl Gustav Carus, Technical University Dresden,Fetscherstrasse 74, 01307 Dresden, Germanye-mail: christine.winter@uniklinikum-dresden.de

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_7, � Springer-Verlag Berlin Heidelberg 2012

61

This chapter summarizes preclinical studies that have used DBS in animalmodels of OCD to systematically map brain regions at which DBS (1) has ther-apeutic effects, (2) does not have such effects, or (3) is even deleterious in order tofurther promote the establishment of DBS in the treatment of patients withotherwise therapy-resistant OCD. The translational quality of these studies will beevaluated in due consideration of the prevalent validity criteria of the respectiveanimal models.

7.2 Animal Models

OCDs manifest themselves with compulsions and/or obsessions as well as addi-tional neuropsychological cognitive and noncognitive deficits. As a consequenceof this heterogeneity, the identification of appropriate animal models that closelymimic the specific behavioral and neuronal manifestation of OCD is severelychallenged and even further aggravated by the fact that cognition-based deficitsand symptoms such as obsessions may not be directly pictured (Korff and Harvey2006). Despite these difficulties, various animal models of OCD have beendescribed during the last 30 years, more or less successfully and more or lesscomprehensively (Joel 2006a). In the following, only the three rat models of OCDwill be detailed and evaluated that have been used for investigating the effects ofDBS on OCD-related behavior in rats.

7.2.1 Quinpirole Model

In the quinpirole (QNP) model, the repeated application of the dopamine D2/D3receptor agonist QNP leads to the development of compulsive behavior that bestresembles compulsive checking behavior in men (Szechtman et al. 1998): on anopen field that is subdivided into 25 partially object equipped subareas, rats treatedintermittently over a long period with QNP (1) revisit one or two subareas/objects(home base) excessively often when compared with other subareas/objects as wellas with saline-treated controls, (2) perform ritualized movement patterns whenapproaching the preferred subareas/objects, (3) stop at only a few other subareas/objects before returning to their home base, and (4) direct this behavior to anotherlocation when the preferred object is moved there. From a phenomenological pointof view (face validity), this behavior fulfills the following criteria specific to OCD:(1) a excessive occupation with an object as well as the ambivalence to disengagefrom it, (2) ritualized motor behavior, and (3) context-dependent symptoms(Szechtman et al. 2001; Joel 2006a). The predictive validity is given by the factthat the tricyclic antidepressant clomipramine leads to a partial reduction ofsymptom presentation (Szechtman et al. 1998). With respect to its constructvalidity, the QNP model accommodates the dopaminergic system (Sesia et al.

62 C. Winter

2011). The importance of the dopaminergic system in the manifestation of OCDsymptoms has repeatedly been evidenced, for example, by the fact that antidop-aminergic drugs may yield beneficial effects when other pharmacological optionsare ineffective (Koo et al. 2010). Although the neural substrates of QNP-inducedcompulsive checking are not known, intermittent administration of QNP over along period has been shown to affect the functioning of basal ganglia–thalamo-cortical circuits involved in locomotor and compulsive behavior (Carpenter et al.2003; Richards et al. 2007), i.e., the balance between activity in the direct and theindirect pathways of this circuitry (Perreault et al. 2006).

7.2.2 Signal Attenuation Model

The signal attenuation (SA) model belongs to the group of animal models in whichobsessive symptoms develop after specific behavioral manipulation (Joel 2006a,b). It is based on the theoretical assumption that OCD-related behavior resultsfrom a disrupted feedback following the successful accomplishment of a naturalgoal-directed activity. In the model, the attenuation of an external feedback thatindicates reward as a result of a successfully performed task leads, in a subsequenttest session, to an excessive task performance that is not followed by a rewardrequest. The deficient demand for recompense indicates an inadequate andmeaningless character of the operation as well as a dysfunctional reward system,core features of OCD. Joel and coworkers demonstrated that this deficient behaviormay selectively be modulated by lesions to structures which have previously beenshown pathophysiologically relevant to OCD as well as anticompulsive drugs (forreview: Joel 2006b).

7.2.3 Schedule-Induced Polydipsia Model

The schedule-induced polydipsia (SIP) model belongs to the group of models inwhich a naturally occurring behavior becomes compulsive as a result of a specificbehavioral manipulation, a fact that provides strong face validity to it. In thismodel, rats are only intermittently fed under a fixed-time 60-s schedule, resultingin the induction of excessive drinking in the presence of water that does not serve aphysiological function as rats are not deprived of water (Platt et al. 2008), butrather presents itself as a nonsensical activity (see above). The predictive validityof the model is given by the fact that administration of selective serotonin reuptakeinhibitor decreases polydipsia without affecting water or food intake under controlconditions (Woods et al. 1993).

7 Obsessive–Compulsive Disorders in Animals 63

7.3 Deep Brain Stimulation

7.3.1 Brain-Site-Specific Effects

In these three models, the effects of DBS of several brain sites—all of which arepart of or interconnected with the basal ganglia–thalamocortical circuitry anddiscussed as relevant in the manifestation of OCD—have been tested: (1) thesubthalamic nucleus (STN), known as the ‘‘driving force’’ of the basal gangliasystem (Benazzouz and Hallet 2000), (2) the globus pallidus (GP), which is dif-ferentiated into the lateral GP (LGP; rodent equivalent of the human GP externus)and the nucleus entopeduncularis (EP; rodent equivalent of the human GP inter-nus), the latter being one of the output structures of the basal ganglia system, and(3) the nucleus accumbens (Nacc), a relay structure of the limbic system, dividedinto the functionally and anatomically distinct Nacc core and Nacc shell. Notably,in the respective studies, any observed anticompulsive effect of DBS of these brainsites was found to occur immediately with onset of DBS, whereas in OCD patients,therapeutic DBS effects only gradually develop. This discrepancy is well knownfrom various pharmacological studies and has been interpreted in the way thatsymptom-reductive effects observed after immediate intervention indicate themodel’s predictive validity (Bourin et al. 2001), whereas symptom-reductiveeffects observed after continuous or prolonged treatment adds to the model’s facevalidity (Willner 1991; Joel 2006a).

The effects of DBS of the STN were studied in both the QNP model (Winteret al. 2008a) and the SA model (Klavir et al. 2009) of OCD. DBS of the STNperformed at high frequencies (130 Hz) and a current of 150 lA specificallyattenuated compulsive checking in both models. The therapeutic effect was tran-sient, as evidenced by the fact that in QNP-treated rats, compulsive checkingreturned to its baseline when stimulation was discontinued. The overall effec-tiveness of high-frequency stimulation of the STN in ameliorating OCD symptomshas meanwhile been validated clinically (Mallet et al. 2008).

The effects of DBS of the GP were also studied in the QNP model (Djodari-Irani et al. 2011) as well as the SA model (Klavir et al. 2011) of OCD. In the QNPmodel, high-frequency stimulation (130 Hz) of the LGP did not have an effect,whereas high-frequency stimulation of the EP reduced one of four behavioralmeasures of OCD, suggesting that with the parameters tested (75–150 lA) high-frequency stimulation of the LGP had no effect, whereas high-frequency stimu-lation of the EP had a minor anticompulsive effect. Notably, in the SA model, DBSof both GP subareas and with equivalent parameters (high-frequency stimulation;75 lA for the LGP and 100 lA for the EP) led to a significant reduction incompulsive lever presses.

Finally, the effects of DBS of the Nacc were studied in the SIP model as well as inthe QNP model of OCD and basically revealed similar results, which have also beenvalidated clinically (Denys et al. 2010). In the SIP model, van Kuyck et al. (2008)demonstrated anticompulsive effects of DBS of the Nacc at 130 Hz and currents of

64 C. Winter

200 lA or higher (up to 500 lA). Equivalent therapeutic effects were also foundwhen DBS was performed in the mediodorsal thalamic nucleus, the filter station ofthe basal ganglia–thalamocortical circuit and the bed nucleus of the stria terminalis.Supporting the specificity of the observed effects of high-frequency stimulation,DBS of the Nacc at low frequencies (10 Hz) did not affect compulsive behavior in theSIP model of OCD (van Kuyck et al. 2008), but increased it in the 8-hydroxy-2-(di-n-propylamino)tetralin rat model of perseveration (van Kuyck et al. 2003). DBS at lowfrequencies has been demonstrated to be ineffective for most DBS indications in theclinic and in animal experimental models (Benabid et al. 1991; Limousin et al. 1995;Ushe et al. 2006; van Kuyck et al. 2003, 2008). In the QNP model, the effects of DBSof subregions of the Nacc were additionally studied and the data showed that high-frequency stimulation (130 Hz) of the Nacc shell was effective at 100 lA, but not atlower or higher currents, whereas high-frequency stimulation of the Nacc core wasmore effective at 150 lA than at 100 lA (Mundt et al. 2009). These differentialeffects were suggested to (1) be due to an unspecific mechanism such as currentspread to subregion-specific neighboring nerve fibers and brain areas that support orcounteract target-specific DBS effects or (2) reflect high-frequency-stimulation-dependent modulations of different subregion-specific efferents (Mundt et al. 2009).

7.3.2 Symptom-Specific Effects

The QNP, SA, and SIP models belong to the most widely applied and reviewedanimal models of OCD (Albelda and Joel 2012), which, as summarized earlier, allhave high face, construct, and predictive validity. Consequently, the fact thatanticompulsive effects of DBS of both the STN and the Nacc could be replicated ineach of the models strongly suggests that the therapeutic effects of DBS were areal phenomenon and not an artifact of the experimental setup or model. On theother hand, the differences in deficit induction and phenotype expression of theQNP, SA, and SIP models suggest that each of these models represents a specificaspect of OCD that may also differ in its responsiveness to a therapeutic inter-vention. Consequently, the lack of an anticompulsive effect of DBS of the GP inthe QNP model is most likely due to specific aspects of the model itself. In light ofthe facts that (1) high-frequency stimulation of the LGP/EP with the stimulationparameters tested is generally behaviorally effective, (2) the stimulation parame-ters tested were sufficient to produce anticompulsive effects when applied to brainsites other than the GP, and most crucially (3) DBS of the GP was shown to exertan anticompulsive effect in the SA model of OCD, it is most likely that the model-specific responsiveness depend on the manipulation tested, i.e. DBS of the STNversus DBS of the GP. To further elaborate such symptom- and region-specificeffects of DBS, additional systematic studies in other animal models of OCD aremandatory. Despite their specific characteristics, the QNP, SA, and SIP modelsshare a major deficiency, i.e., the nonconsideration and missing acknowledgementof OCD as a neurodevelopmental disorder. The recently described clomipramine

7 Obsessive–Compulsive Disorders in Animals 65

model, in which drug exposure during a sensitive period in the neonatal rat hasbeen shown to induce anxious and perseveration-like behavior, cognitive inflexi-bility, and hoarding only with a considerable postmanipulation delay in the adultrat, tried to meet these objections (Andersen et al. 2010). With the pendingdescription of its predictive validity and the absence of further behavioral andneurobiological characteristics phenotypic of OCD (Sesia et al. 2011), however, itremains to be discussed whether the clomipramine model may eventually evolveto represent a model for OCD powerful enough to enable also the elucidation ofprogressive mechanisms underlying behavioral manifestations as well as early andpreventive interventions.

7.3.3 Mechanism Effects

As a major drawback, none of the studies on DBS in animal models of OCDpublished so far investigated potential mechanisms underlying the effectiveness ofDBS in general and with respect to OCD pathology in particular.

However, pharmacological inactivation studies using intracerebral administra-tions of the GABA agonist muscimol that paralleled some of the DBS studiespresented here allow some speculations on the overall net effects of DBS.Equivalent anticompulsive effects following high-frequency stimulation andpharmacological inactivation of the STN in the QNP and SA models of OCDsupport the notion that similar effects at a system level, i.e., an overall net inhi-bition, may be engaged by both interventions. In this respect, DBS at high fre-quencies has been discussed to suppress neuronal activity partly as a result ofstimulating inhibitory GABAergic afferences and suppressing excitatory gluta-matergic afferences and/or neuronal cell bodies (Dostrovsky et al. 2000; Beurrieret al. 2001). Focal muscimol administration itself enhances the GABAergic inputto the targeted nucleus directly. High-frequency stimulation and pharmacologicalinactivation of the STN have both been reported to increase striatal dopaminetransmission (Bruet et al. 2003; Meissner et al. 2003; Lee et al. 2006; Winter et al.2008b), potentially providing a final common pathway by which STN-activitymodulation reduces compulsive behavior.

Contrasting these data, only pharmacological inactivation but not high-fre-quency stimulation of both GP subnuclei was effective in producing a clear an-ticompulsive effect in the QNP model of OCD. Obviously, the target-specificdistribution and arrangement of (1) individual cells differentially receptive toelectrical stimulation (Ranck 1975) and differing in their functional characteristicand (2) GABA-A receptors responsible for mediating the muscimol-dependenteffect affect the local effectivity of both DBS and pharmacological inactivation anddetermine whether the behavioral effects of the two interventions are similar ordifferent. Following this argumentative line, we have suggested that the specificcellular arrangement of the STN subserved both interventions to engage the samemechanism, whereas the cellular arrangement of the GP does not promote

66 C. Winter

equivalent behavioral effects of pharmacological inactivation and high-frequencystimulation (Djodari-Irani et al. 2011).

7.3.4 Network Effects

Interestingly, all of the afore-mentioned studies described basically equivalenteffects for activity modulation of the basal ganglia nuclei STN, LGP, and EP,complying with the idea that behavioral and neural deficits resulting from a dis-tinct malfunctioning station of the basal ganglia–thalamocortical circuit may beoutbalanced by a manipulation to any other station of the same circuitry.According to the current view of the basal ganglia circuitry, the respective nucleiare supposed to exert partly opposing effects on behavioral output (Albin et al.1989), suggesting that the therapeutic effects of DBS in the basal ganglia–thala-mocortical circuitry obtained may at least be partly due to less specific andselective activity modulation of passing or neighboring fibers that interact with thetarget-specific effects. In this respect, the evaluation of effective stimulationparameters and side effects may be carefully considered when the optimal stim-ulation site is to be defined for treatment of neuropsychiatric disorders.

7.4 Conclusion

We have described studies aiming at systematically mapping brain regions atwhich DBS affects neuropsychiatric symptoms in animal models of OCD. Theselected animal models were appropriate in that they fulfilled current criteria ofvalidity. They consequently allowed the translational and comparative analysis ofthe therapeutic and potentially even pathophysiological relevance of select brainsites in OCD as well as the investigation of DBS in comparison with other invasivemethods differentially modulating the activity of targeted and associated brain sitesin order to understand further the overall net effects underlying DBS effectivityand contributed to defining optimal stimulation parameters for DBS of a specificbrain site in the treatment of OCD.

However, the studies were all restricted in their methodological approach andprofile and consequently do not allow conclusions on neurobiological and func-tional networks underlying the manifestation of OCD. For this, understanding themechanisms by which DBS exerts effectiveness or fails to be effective in reducingthe specific symptoms is mandatory. Two major hypotheses are temporarily beingdiscussed to underlie DBS efficacy: (1) high-frequency stimulation suppressesactivity in the DBS target (Dostrovsky et al. 2000; Beurrier et al. 2001) and (2)high-frequency stimulation initiates/induces a new activity in neuronal networksthat are associated with the DBS target (Montgomery and Baker 2000; Vitek 2002;McIntyre et al. 2004). The neurobiological and functional effects in neuronal

7 Obsessive–Compulsive Disorders in Animals 67

networks that combine both hypotheses and may be studied in animal experimentalapproaches are altered activity (using electrophysiological recording and imaging),altered neurotransmission (using in vivo microdialysis), and altered plasticity andproliferation. As the detailed description of the pathophysiological neurobiologicalnetworks underlying OCD is crucial for efficient and scientifically based selectionof the most promising DBS targets, future studies will need to use combinedapproaches at different levels of neurobiological integrity. In addition, the effectsof acute DBS studied so far yield only little valuable information on the mecha-nisms engaged in clinically effective DBS considering that in the clinical conditionDBS is performed continuously. To best explore the mechanisms and effectivenessof DBS it is crucial to apply DBS chronically in appropriate animal models of therespective disorders. If these prerequisites are fulfilled, future studies may cru-cially contribute to a greater understanding of DBS technology so that its appli-cation is likely to improve the quality of life for a significant number of patientswith otherwise therapy-resistant psychiatric disorders, such as OCD.

References

Albelda N, Joel D (2012) Animal models of obsessive compulsive disorder: exploringpharmacology and neural substrates. Neurosci Biobehav Rev 36(1):47–63

Albin RL, Young AB, Penney JB (1989) The functional anatomy of basal ganglia disorders.Trends Neurosci 12(10):366–375

Andersen SL, Greene-Colozzi EA, Sonntag KC (2010) A novel, multiple symptom model ofobsessive-compulsive-like behaviors in animals. Biol Psychiatry 68:741–747

Benabid AL, Pollak P, Gervason C, Hoffmann D, Gao DM, Hommel M, Perret JE, de RougemontJ (1991) Long-term suppression of tremor by chronic stimulation of the ventral intermediatethalamic nucleus. Lancet 337:403–406

Benazzouz A, Hallett M (2000) Mechanism of action of deep brain stimulation. Neurology 55(12Suppl 6):S13–S16

Beurrier C, Bioulac B, Audin J, Hammond C (2001) High-frequency stimulation produces a transientblockade of voltage-gated currents in subthalamic neurons. J Neurophysiol 85:1351–1356

Bourin M, Fiocco AJ, Clenet F (2001) How valuable are animal models in defining antidepressantactivity? Hum Psychopharmacol 16(1):9–21

Bruet N, Windels F, Carcenac C, Feuerstein C, Bertrand A, Poupard A, Savasta M (2003)Neurochemical mechanisms induced by high frequency stimulation of the subthalamicnucleus: increase of extracellular striatal glutamate and GABA in normal and hemiparkin-sonian rats. J Neuropathol Exp Neurol 62:1228–1240

Carpenter TL, Pazdernik TL, Levant B (2003) Differences in quinpirole-induced local cerebralglucose utilization between naive and sensitized rats. Brain Res 964:295–301

Denys D, Mantione M, Figee M, van den Munckhof P, Koerselman F, Westenberg H, Bosch A,Schuurman R (2010) Deep brain stimulation of the nucleus accumbens for treatment-refractory obsessive-compulsive disorder. Arch Gen Psychiatry 67(10):1061–1068

Djodari-Irani A, Klein J, Banzhaf J, Joel D, Heinz A, Harnack D, Lagemann T, Juckel G, Kupsch A,Morgenstern R, Winter C (2011) Activity modulation of the globus pallidus and the nucleusentopeduncularis affects compulsive checking in rats. Behav Brain Res 219(1):149–158

Dostrovsky JO, Levy R, Wu JP, Hutchison WD, Tasker RR, Lozano AM (2000) Microstimu-lation-induced inhibition of neuronal firing in human globus pallidus. J Neurophysiol84(1):570–574

68 C. Winter

Joel D (2006a) Current animal models of obsessive compulsive disorder: a critical review. ProgNeuropsychopharmacol Biol Psychiatry 30:374–388

Joel D (2006b) The signal attenuation rat model of obsessive-compulsive disorder: a review.Psychopharmacology 186:487–503

Klavir O, Flash S, Winter C, Joel D (2009) High frequency stimulation and pharmacologicalinactivation of the subthalamic nucleus reduces ‘compulsive’ lever-pressing in rats. ExpNeurol 215:101–109

Klavir O, Winter C, Joel D (2011) High but not low frequency stimulation of both the globuspallidus and the entopeduncular nucleus reduces ‘compulsive’ lever-pressing in rats. BehavBrain Res 216:84–93

Koo MS, Kim EJ, Roh D, Kim CH (2010) Role of dopamine in the pathophysiology andtreatment of obsessive compulsive disorder. Expert Rev Neurother 10(2):275–290

Korff S, Harvey BH (2006) Animal models of obsessive compulsive disorder: rationale tounderstanding psychobiology and pharmacology. Psychiatr Clin N Am 29:371–390

Lee KH, Blaha CD, Harris BT, Cooper S, Hitti FL, Leiter JC, Roberts DW, Kim U (2006)Dopamine efflux in the rat striatum evoked by electrical stimulation of the subthalamicnucleus: potential mechanism of action in Parkinson’s disease. Eur J Neurosci 23:1005–1014

Limousin P, Pollak P, Benazzouz A, Hoffmann D, Le Bas JF, Broussolle E, Perret JE, BenabidAL (1995) Effect of parkinsonian signs and symptoms of bilateral subthalamic nucleusstimulation. Lancet 345:91–95

Mallet L, Polosan M, Jaafari N, Baup N, Welter ML, Fonatine D, du Montcel ST, Jelnik J,Chéreau I, Arbus C, Raoul S, Aouizerate B, Damier P, Charbardès S, Czernecki V, ArdouinC, Krebs MO, Bardinet E, Chaynes P, Burbaud P, Cornu P, Derost P, Bougerol T, Bataille B,Mattei V, Dormont D, Devaux B, Vérin M, Houeto JL, Pollak P, Benabid AL, Agid Y, KrackP, Millet B, Pelisollo A. (2008) N Engl J Med; 359(29):2121–2134

McIntyre CC, Savasta M, Kerkerian-Le GL, Vitek JL (2004) Uncovering the mechanism(s) ofaction of deep brain stimulation: activation, inhibition, or both. Clin Neurophysiol115:1239–1248

Meissner W, Harnack D, Reese R, Paul G, Reum T, Ansorge M, Kusserow H, Winter C,Morgenstern R, Kupsch A (2003) High-frequency stimulation of the subthalamic nucleusenhances striatal dopamine release and metabolism in rats. J Neurochem 85:601–609

Montgomery EB Jr, Baker KB (2000) Mechanisms of deep brain stimulation and future technicaldevelopments. Neurol Res 22:259–266

Mundt A, Klein J, Joel D, Heinz A, Djodari-Irani A, Harnack D, Kupsch A, Orawa H, Juckel G,Morgenstern R, Winter C (2009) High-frequency stimulation of the nucleus accumbens core andshell reduces quinpirole-induced compulsive checking in rats. Eur J Neurosci 29:2401–2412

Perreault ML, Graham D, Bisnaire L, Simms J, Hayton S, Szechtman H (2006) Kappa-opioidagonist U69593 potentiates locomotor sensitization to the D2/D3 agonist quinpirole: pre- andpostsynaptic mechanisms. Neuropsychopharmacology 31(9):1967–1981

Platt B, Beyer CE, Schechter LE, Rosenzweig-Lipson S (2008) Schedule-induced polydipsia: arat model of obsessive compulsive disorder. Curr Protoc Neurosci 9(9):27

Ranck JB Jr (1975) Which elements are excited in electrical stimulation of mammalian centralnervous system: a review. Brain Res 98:417–440

Richards TL, Pazdernik TL, Levant B (2007) Clorgyline-induced modification of behavioralsensitization to quinpirole: effects on local cerebral glucose utilization. Brain Res 1160:124–133

Sesia T, Bizup B, Schreiber S, Grace AA (2011) Quinpirole and clomipramine chronic injectionmodels for obsessive compulsive disorders: effect on ventral tegmentale activity and OCD-related behavioral paradigms. Society for Neuroscience Abstract Number 66.15

Szechtman H, Sulis W, Eilam D (1998) Quinpirole induces compulsive checking behavior in rats:a potential animal model of obsessive-compulsive disorder (OCD). Behav Neurosci112:1475–1485

Szechtman H, Eckert MJ, Tse WS, Boersma JT, Bonura CA, McClelland JZ, Culver KE, Eilam D(2001) Compulsive checking behavior of quinpirole-sensitized rats as an animal model ofobsessive-compulsive disorder (OCD): form and control. BMC Neurosci 2:4

7 Obsessive–Compulsive Disorders in Animals 69

Ushe M, Mink JW, Tabbal SD, Hong M, Schneider GP, Rich KM, Lyons KE, Pahwa R,Perlmutter JS (2006) Postural tremor suppression is dependent on thalamic stimulationfrequency. Mov Disord 21:1290–1292

van Kuyck K, Demeulemeester H, Feys H, De WW, Dewil M, Tousseyn T, De SP, Gybels J,Bogaerts K, Dom R, Nuttin B (2003) Effects of electrical stimulation or lesion in nucleusaccumbens on the behaviour of rats in a T-maze after administration of 8-OH-DPAT orvehicle. Behav Brain Res 140:165–173

van Kuyck K, Brak K, Das J, Rizopoulos D, Nuttin B (2008) Comparative study of the effects ofelectrical stimulation in the nucleus accumbens, the mediodorsal thalamic nucleus and the bednucleus of the stria terminalis in rats with schedule-induced polydipsia. Brain Res 1201:93–99

Vitek JL (2002) Mechanisms of deep brain stimulation: excitation or inhibition. Mov Disord17(Suppl 3):S69–S72

Willner P (1991) Behavioural models in psychopharmacology. In: Willner P (ed) Behaviouralmodels in psychopharmacology: theoretical, industrial and clinical perspectives. CambridgeUniversity Press, Cambridge, pp 3–18

Winter C, Mundt A, Jalali R, Joel D, Harnack D, Morgenstern R, Juckel G, Kupsch A (2008a)High frequency stimulation and temporary inactivation of the subthalamic nucleus reducequinpirole-induced compulsive checking behavior in rats. Exp Neurol 210:217–228

Winter C, Lemke C, Sohr R, Meissner W, Harnack D, Juckel G, Morgenstern R, Kupsch A(2008b) High frequency stimulation of the subthalamic nucleus modulates neurotransmissionin limbic brain regions of the rat. Exp Brain Res 185:497–507

Woods A, Smith C, Szewczak M, Dunn RW, Cornfeldt M, Corbett R (1993) Selective serotoninre-uptake inhibitors decrease schedule-induced polydipsia in rats: a potential model forobsessive compulsive disorder. Psychopharmacology 112:195–198

70 C. Winter

Chapter 8Subcallosal Cingulate Cortex Deep BrainStimulation for the Treatmentof Refractory Mood Disorders: Evidenceand Challenges

Peter Giacobbe, Nir Lipsman and Andres M. Lozano

8.1 Subcallosal Cingulate Cortex Deep Brain Stimulationfor Treatment-Resistant Depression: A Reviewof the Clinical Results from the Initial Cohort of Patients

Informed by the converging data suggesting that the subcallosal cingulate cortex(SCC) plays a role in the regulation of both normal and pathological mood states(Mayberg 2009; Giacobbe et al. 2009), a proof-of-principle trial of the efficacy ofSCC deep brain stimulation (DBS) in improving depressive symptoms in patientswith treatment-resistant depression (TRD) was initiated at the University of Tor-onto in 2003. Inclusion criteria required patients to meet the criteria for majordepressive disorder and to be in a current major depressive episode, characterizedby a minimum of 1-year duration and a score of at least 20 on the 17-itemHamilton Depression Rating Scale (HDRS-17). Patients were required to dem-onstrate a failure to respond to a minimum of four different treatments, includingantidepressant pharmacotherapy of sufficient dose and duration, evidence-basedpsychotherapy, and electroconvulsive therapy. Patients received open-label stim-ulation of the SCC, with the a priori primary outcome measure being the rates of

P. Giacobbe (&)Department of Psychiatry, University Health Network – Toronto General Hospital,University of Toronto, Toronto, Canadae-mail: peter.giacobbe@uhn.ca

N. Lipsman � A. M. LozanoDivision of Neurosurgery, Department of Surgery, University of Toronto, Toronto, Canadae-mail: nir.lipsman@utoronto.ca

A. M. Lozano (&)Division of Neurosurgery, University Health Network – Toronto Western Hospital,University of Toronto, Toronto, ON, Canadae-mail: lozano@uhnresearch.ca

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_8, � Springer-Verlag Berlin Heidelberg 2012

71

clinical response, defined as a 50 % or greater reduction in the HDRS-17 scorecompared with the presurgical baseline.

In the first report of the clinical effects of SCC DBS for the treatment of TRD, athird of the patients (two of six) met the response criterion at 1 month after DBSactivation (Mayberg et al. 2005). At the 6-month time point, four of the originalcohort of six patients (66 %) achieved an antidepressant response, with three ofthese subjects achieving remission, defined as an HDRS-17 score of 7 or less. In anextension of this original cohort, the 1-year outcomes of the first 20 patients whoreceived SCC DBS for the treatment of TRD at the University of Toronto werepublished (Lozano et al. 2008). Similar to the pilot SCC DBS study, an early andsustained antidepressant response to DBS was observed. After 1 month of activestimulation, the mean HDRS-17 score in the group was significantly decreasedfrom the pre-DBS baseline, and the antidepressant effect was observed at eachmonthly time point until the end of the 12-month study period. Seven of the 20patients (35 %) met the antidepressant response criterion after 1 month of stim-ulation of the SCC, and the number increased to 12 (60 %) by 6 months. At theend of the 12-month study period, 11 of the patients (55 %) were deemed to beresponders. Most of the patients (eight of 11) who achieved an antidepressantresponse at 6 months continued to meet this criterion at 12 months. It appearedthat not every aspect of the depressive symptoms that the patients were endorsingimproved at the same rates with SCC DBS, with the core mood symptoms mea-sured by HDRS-17 improving before the anxiety, sleep, and somatic symptoms.

The biological effects of chronic SCC DBS were assessed by the comparison ofthe regional cerebral blood flow changes before and after DBS, as measured bypositron emission tomography. After 3 months of stimulation of the SCC,decreased activity was observed in the SCC, hypothalamus, insula, and the medialand orbital frontal cortex (Lozano et al. 2008). Greater reductions in the metabolicactivity of the anterior insula and medial frontal and dorsolateral prefrontal cortexwere associated with a superior clinical response. Conversely, the prefrontal cortexand the dorsal cingulate increased in activity following SCC DBS.

In general, the surgery was well tolerated and no major perioperative complicationswere seen. Two patients developed infections on their scalp around site of the elec-trodes, requiring intravenous antibiotics and in one case explantation and reimplan-tation of the DBS electrodes (Lozano et al. 2008). Neuropsychological testing in the12 months following DBS did not reveal any deleterious effects of the implantation ofthe DBS electrodes or of continuous stimulation of the SCC, and most patientsexhibited improvements in their neuropsychological functioning from the ‘‘belowaverage’’ to ‘‘average’’ range after 12 months of DBS (McNeely et al. 2008).

The long-term outcomes of the cohort of 20 patients who received SCC DBSfor the treatment of TRD at our center have been examined (Kennedy et al. 2011).After the initial 12-month study of DBS (Lozano et al. 2008), patients wereassessed annually and at a last follow-up visit to measure depression severity,functional outcomes, and adverse events. The last follow-up visit ranged from 3 to6 years after SCC DBS, and represented over 70 patient-years of clinical follow-up. For patients who were lost to follow-up (seven of the 20 patients), the last

72 P. Giacobbe et al.

available clinical assessment was carried forward. Using an intent-to-treat analy-sis, we obtained response rates of 45 % (9/20) at year 2, 60 % (12/20) at year 3,and 55 % (11/20) at the last follow-up. Approximately a third of the patients (7/20)met the criterion for remission at their last follow-up visit. In addition to the long-term antidepressant effects of SCC DBS, patients reported functional gains in theirquality of life and work status after DBS. Improvements in all domains of self-reported quality of life were observed 6 months after DBS and these effectscontinued to accrue at the 1-year and last follow-up time points (Kennedy et al.2011). After 1 year of DBS, half of the patients were able to maintain employment(Lozano et al. 2008) and at the last follow-up 65 % engaged in work-relatedactivities (Kennedy et al. 2011).

Taken together, the initial open-label, proof-of-principle studies of SCC DBSfor the treatment of TRD provide evidence that SCC DBS is associated with arobust early antidepressant response that can be sustained for several years withchronic stimulation. Furthermore, the antidepressant effects of chronic stimulationresults in clinically meaningful benefits in health-related quality of life andemployment status for patients with TRD.

8.2 SCC DBS: Results from Other Centers

With the promising results from the initial proof-of-principle study of SCC DBSfor the treatment of TRD (Mayberg et al. 2005) and the data on the sustainabilityof the antidepressant effects with chronic stimulation (Lozano et al. 2008; Ken-nedy et al. 2011), the question of the ability to translate these results to othercenters is a crucial one. The inability to replicate the results of the initial Torontocohort in other centers with different psychiatrists and neurosurgeons would raisedoubts about the robustness of the antidepressant effects of SCC DBS and thepotential generalizability of this procedure. In the past year, positive clinical datahave emerged from others centers which are concordant with the initial findings.

The results of a multicenter Canadian trial of SCC DBS for the treatment ofTRD have been published (Lozano et al. 2012). This study represented a three-center (University of Toronto, University of British Columbia, and McGill Uni-versity) prospective, open-label trial of SCC DBS. Twenty-one patients with TRDunderwent bilateral implantation of electrodes in the SCC, with the outcomevariables being reduction in depressive symptom severity after 12 months ofcontinuous stimulation, and variations across the three sites in the localization ofthe DBS electrodes in the white matter target within the SCC. An early antide-pressant effect was seen at 4 weeks and remained consistent to the end of the 12-month period. At 2 months, the mean reduction in the HDRS-17 score across allthree centers was 40.3 % and the magnitude of the reduction of the pre-DBSdepressive symptoms at 6 months and 12 months was 43.3 and 41.4 %, respec-tively. Using a 50 % or greater reduction in the HDRS-17 score as the responsecriterion, 57 % of patients were deemed responders at 1 month, 48 % at 6 months,

8 Subcallosal Cingulate Cortex Deep Brain Stimulation 73

and 29 % at 12 months. Although the magnitude of the antidepressant effect ofSCC DBS as measured by the absolute scores on HDRS-17 was stable, theapparent drop in efficacy when measured in ‘‘response rates’’ from the sixth monthto the twelth month reflects the proportionately large number of patients (seven of21, or 33 %) with a reduction in their HDRS-17 scores in the 40–50 % range at the12-month time point. With use of previously published targeting guidelines(Hamani et al. 2009), no differences in the location of active contacts within theSCC were detected across the three centers. The results from this first multicenterevaluation of SCC DBS suggest that the antidepressant effects and surgical tar-geting of the electrodes were consistent and reproducible across the three centers.

More recently, the results of an open-label trial of SCC DBS with a sham lead-in phase have been reported (Holtzheimer et al. 2012). With use of similarinclusion and exclusion criteria to determine the severity of the depressivesymptoms and its resistance to conventional treatments as in the original Torontocohort and the Canadian multicenter trial (Mayberg et al. 2005; Lozano et al. 2008,2012), the effect of SCC DBS in a mixed group of mood disorder patients wasexamined. Ten patients with TRD and seven patients with bipolar depression (BD)received single-blind sham stimulation for 4 weeks followed by chronic activestimulation from DBS electrodes implanted bilaterally in the white matter target inthe SCC. A significant decrease in the severity of depressive symptoms andincrease in function were associated with chronic stimulation. At 1 year, 36 % ofpatients were antidepressant responders to active SCC DBS, and the proportionincreased to 92 % after 2 years of stimulation. In a similar fashion, the rates ofremission were reported to increase over time. The criterion for remission weremet by 18 % of patients at 6 months, 36 % of patients at 1 year, and 58 % ofpatients after 2 years of active stimulation. The effects of SCC DBS on improvingdepressive symptoms were reported to be similar for patients with TRD and forpatients with BD. The antidepressant effects of SCC DBS were enduring in thisstudy, with no relapses being reported in patients who achieved remission.

A European study of SCC DBS for the treatment of TRD has been published(Puigdemont et al. 2011). Eight patients with TRD received chronic, open-labelstimulation in the SCC. Approximately a third of patients (three of eight) met thecriterion for remission following 1 month of active stimulation, and the proportionincreased to 50 % at 1 year. The response rates at 1 month, 6 months, and 1 yearwere 37.5, 87.5, and 62.5 %, respectively. Although the rates of response to SCCDBS as well as the clinical and demographic characteristics of this group ofpatients share similarities with those of other SCC DBS cohorts (Mayberg et al.2005; Lozano et al. 2008, 2012; Holtzheimer et al. 2012), the authors noted somedesign differences in their study (Puigdemont et al. 2011). In contrast to the otherSCC DBS cohorts where monopolar stimulation was provided, the study of Pu-igdemont et al. (2011) used bipolar stimulation. Additionally, an analysis of therelationship between the antidepressant response and electrode placement revealedthat there was a correlation between placement of the electrode in the gray matterof Brodmann area (BA) 24, corpus callosum, and head of caudate, whereas non-responders had their electrodes localized predominantly near BA 25. Although it is

74 P. Giacobbe et al.

likely that stimulation from electrodes in the SCC can influence neuronal activityin both BA24 and BA25 (Hamani et al. 2011), previous studies have failed to finda correlation between electrode placement within the SCC and clinical outcomesin patients with TRD (Lozano et al. 2012; Hamani et al. 2009).

8.3 Psychiatric Adverse Effects of SCC DBS

To date, there have been no reported switches into hypomania or mania in thelong-term follow-up of patients who have received SCC DBS for the treatment ofTRD (Kennedy et al. 2011), nor have there been elevations in the ratings of manicbehavior for patients who have received this procedure for the treatment of BD(Holtzheimer et al. 2012). A case report on the use of SCC DBS for the treatmentof psychotic depression has been published (Puigdemont et al. 2009). Neuropsy-chological testing has consistently shown that SCC DBS does not have deleteriouseffects on neurocognition (McNeely et al. 2008; Holtzheimer et al. 2012;Puigdemont et al. 2011). There have been no published reports of SCC DBS forthe treatment of primary anxiety disorders. Existing data in patients with TRDsuggest a positive correlation between long-term improvement in depressive andanxiety symptoms (Lozano et al. 2008; Puigdemont et al. 2011) although longertimes were required to reach maximal improvements in anxiety symptoms ascompared with the core mood symptoms of depression (Lozano et al. 2008).However, short-term exacerbations in anxiety have been described in a minority ofpatients with SCC DBS (Lozano et al. 2012; Holtzheimer et al. 2012).

The most serious of the psychiatric adverse events reported in patients who havereceived SCC DBS are suicidal ideation and behavior. Although it is recognized thatincreased all-cause mortality, including completed suicides, may be an inherentfeature of TRD, and has been estimated in two studies to be 13 % over 4–8 years(Shergill et al. 1999) and 32 % over 7 years (O’Leary and Lee 1996) in this clinicalpopulation, the emergence of suicidal ideation in patients with SCC DBS is a psy-chiatric emergency (Giacobbe and Kennedy 2009). In the combined published caseseries of SCC DBS, less than 5 % of patients (three of 64) with TRD who havereceived this procedure have committed suicide (Kennedy et al. 2011; Lozano et al.2012; Holtzheimer et al. 2012; Puigdemont et al. 2011). In these reports the timing ofthe suicidal behavior, including both attempted suicide and suicide, have rangedfrom 1 week after DBS activation (Holtzheimer et al. 2012) to over 6 years aftersurgery (Kennedy et al. 2011). Given that it appears that suicidal behaviors can occurat any point in time following DBS, the clinician should be vigilant and inquiringabout the presence of suicidal ideation should be a routine element of ongoingpsychiatric follow-up of patients who have received this procedure.

The relationship between DBS and suicide in this population is likely to becomplex and multifactorial. As has been reported in TRD and other clinical popu-lations, the emergence of suicidal behaviors may occur even in those patients whohave had a marked improvement in their underlying condition (Holtzheimer et al.

8 Subcallosal Cingulate Cortex Deep Brain Stimulation 75

2012; Abelson et al. 2005; Albanese et al. 2005). The emergence of suicidal thoughtsin any patient with DBS should prompt an evaluation of the functionality of the DBSdevice (Giacobbe and Kennedy 2009). The depletion of the DBS battery or itsdeactivation may herald a rapid reemergence of depressive symptoms. Holtzheimeret al. (2012) reported that three patients experienced a relapse of their depressivesymptoms with suicidal ideation within 2 weeks of undergoing a single-blind dis-continuation of active DBS in their protocol. Interestingly, they described that forseveral months after the reactivation of stimulation, the depressive symptoms did notimprove to their previous level. In contrast, Howard et al. (2011) have reported a caseof a woman receiving SCC DBS for the treatment of TRD who experienced a rapidrelapse of her depressive symptoms with the reemergence of suicidality on twooccasions following the cessation of active stimulation, but they rapidly stabilizedwith the reintroduction of stimulation. The role of poor psychological readaptation inthe post-DBS period to one’s interpersonal and employment situation, which hasbeen suggested by some to be a potential contributor to suicide following DBS forParkinson’s disease (Houeto et al. 2006), merits further investigation in the TRDpopulation, especially given the younger age at which this group recieves DBScompared with those with Parkinson’s disease.

8.4 SCC DBS: How Can Patient Outcomes Be Improved?

To date there have been published results from 69 patients worldwide who havereceived SCC DBS for the treatment of a refractory mood disorder (Table 8.1).The fact that SCC DBS for the treatment of TRD is a nascent and emerging area ofinvestigation is underscored by the fact most of the studies have been publishedsince 2010. This rate of growth of our knowledge and experience with this pro-cedure for the treatment of TRD will to continue to progress in the years to comewith the exploration of the antidepressant effects of SCC DBS under methodo-logically rigorous, blinded, sham-controlled conditions.

By far the most evidence exists for the antidepressant effects of SCC DBS forpatients with TRD. The results of the use of SCC DBS for the treatment of BD havebeen mixed, with more recent publications reporting positive results (Mayberg et al.2005; Holtzheimer et al. 2012; McNab et al. 2009). The 1-year outcomes for SCC DBSfor the treatment of TRD appear to be comparable across centers and investigators(Lozano et al. 2008, 2012; Holtzheimer et al. 2012; Puigdemont et al. 2011)(Table 8.1), with progressively superior results being seen with long-term follow-upbeyond 1 year (Kennedy et al. 2011; Holtzheimer et al. 2012). The reason for theelevated rates of response observed over time is unclear. Although DBS exerts itselectrophysiological effects within milliseconds, positive clinical outcomes may onlybe evident weeks to months later. Understanding the short-term and long-termneurophysiological and psychological adaptations that occur with chronic SCC DBSmay help to elucidate the mechanisms of this putative treatment and improve patientoutcomes.

76 P. Giacobbe et al.

Tab

le8.

1P

ubli

shed

repo

rts

ofsu

bcal

losa

lci

ngul

ate

cort

exde

epbr

ain

stim

ulat

ion

for

trea

tmen

tof

refr

acto

rym

ood

diso

rder

s

Cit

atio

nD

iagn

osis

Num

ber

of pati

ents

Typ

eof

stim

ulat

ion

Sti

mul

atio

npa

ram

eter

sF

ollo

w-u

ppe

riod

(mon

ths)

Ant

idep

ress

ant

resp

onse

aC

omm

ents

Cas

ese

ries

May

berg

etal

.(2

005)

MD

D6

Con

stan

tvo

ltag

e4.

0V

,13

0H

z,60

ls6

4/6

Res

pond

ers

One

pati

ent

wit

hB

D

Loz

ano

etal

.(2

008)

MD

D20

Con

stan

tvo

ltag

e3.

5–5.

0V

,13

0H

z,90

ls

1211

/20

Res

pond

ers

Incl

udes

pati

ents

from

May

berg

etal

.(2

005)

Ken

nedy

etal

.(2

011)

MD

D20

Con

stan

tvo

ltag

e4.

3V

,12

4.7

Hz,

70.6

ls

36–7

211

/20

Res

pond

ers

atla

stfo

llow

-up

visi

tIn

clud

espa

tien

tsfr

omL

ozan

oet

al.

(200

8)L

ozan

oet

al.

(201

2)M

DD

21C

onst

ant

curr

ent

5.2

mA

,12

8.1

Hz,

93.9

ls

126/

21R

espo

nder

sM

ulti

cent

erC

anad

ian

tria

l

Pui

gdem

ont

etal

.(2

011)

MD

D8

Con

stan

tvo

ltag

e4.

2V

,135

Hz,

174.

4ls

125/

8R

espo

nder

sB

ipol

arst

imul

atio

nus

ed

Hol

tzhe

imer

etal

.(2

012)

MD

Dan

dB

D17

Con

stan

tcu

rren

t7.

25m

A,1

30H

z,91

ls24

5/14

Res

pond

ers

atm

onth

12,

11/1

2re

spon

ders

atm

onth

24

10M

DD

and

7B

Dpa

tien

ts

Cas

ere

port

sb

Nei

mat

etal

.(2

008)

MD

D1

Con

stan

tvo

ltag

e4.

0V

,13

0H

z,60

ls30

Yes

Pre

viou

sci

ngul

otom

y

McN

abet

al.

(200

9)B

D1

Con

stan

tvo

ltag

e5.

0V

,15

0H

z,21

0l

s16

No

chan

gein

depr

essi

vesy

mpt

oms

Rig

htth

alam

icst

roke

befo

reD

BS

Gui

njoa

net

al.

(201

0)M

DD

1C

onst

ant

volt

age

4.5

V,

120

Hz,

90ls

20Y

esR

espo

nse

wit

hbi

late

ral

stim

ulat

ion,

rem

issi

onw

ith

righ

tun

ilat

eral

stim

ulat

ion

BD

bipo

lar

depr

essi

on,

DB

Sde

epbr

ain

stim

ulat

ion,

MD

Dm

ajor

depr

essi

vedi

sord

era

Res

pons

eis

defi

ned

asa

50%

orgr

eate

rde

crea

sein

the

17-i

tem

Ham

ilto

nD

epre

ssio

nR

atin

gS

cale

from

the

pre-

DB

Sba

seli

neb

Cas

ere

port

sof

pati

ents

who

are

incl

uded

inth

epu

blis

hed

case

seri

esha

vebe

enex

clud

edfr

omth

eta

ble

8 Subcallosal Cingulate Cortex Deep Brain Stimulation 77

The optimal methods of combining the established biological and psychologicaltreatments for depression with SCC DBS are unknown. A synergistic or additiveeffect of concurrent changes in the treatments provided to these patients in long-term follow-up is a possible contributory factor to the improved response ratesobserved with chronic SCC DBS (Kennedy et al. 2011; Holtzheimer et al. 2012). Ithas been shown that the addition of another neuromodulation strategy to DBS canimprove clinical outcomes (Puigdemont et al. 2009), and the converse relationshiphas also been shown (Guinjoan et al. 2010). A future challenge is to identifywhether the addition of specific pharmacological and psychological treatmentsmay enhance or impair SCC DBS patients from achieving their maximal level ofsymptomatic relief and long-term functioning.

The optimal stimulation parameters have not yet been established. A comparisonof the results from studies utilizing constant current and constant voltage does notsuggest that either of these methods for stimulating the brain has clear superiorityover the other. Additionally, elucidating the presurgical neuroimaging predictors ofpositive clinical outcomes will help improve patient selection. There is preliminaryevidence that the integrity of the white matter projections from the SCC to theamygdala must be intact to produce an antidepressant response to SCC DBS. McNabet al. (2009) described a patient with TRD following a right thalamic stroke whofailed to respond to SCC DBS. Both in vivo diffusion tensor imaging and postmortemneuropathology revealed a reduced number of white matter fibers projecting from theSCC to the amygdala only in the right hemisphere, which was damaged by the stroke.Similarly, a superior effect of right unilateral compared with bilateral SCC stimu-lation has been associated with greater cross-hemispheric white matter projectionsseen with diffusion tensor imaging from the right SCC compared with the left SCC(Guinjoan et al. 2010). Given that using gross neuroanatomical landmarks of DBSelectrode placement may not be sufficient to predict long-term antidepressant out-comes with stimulation in the SCC, future studies should look at the potential role ofindividual differences in neuronal projections both to and from the SCC as a mediatorof response (Giacobbe et al. 2010).

References

Abelson JL, Curtis GC, Sagher O, Albucher RC, Harrigan M, Taylor SF, Martis B, Giordani B(2005) Deep brain stimulation for refractory obsessive-compulsive disorder. Biol Psychiatry57:510–516

Albanese A, Piacentini S, Romito LM, Leone M, Franzini A, Broggi G, Bussone G (2005) Suicideafter successful deep brain stimulation for movement disorders. Neurology 65:499–500

Giacobbe P, Kennedy SH (2009) Medical management and indications for surgery in depression.In: Lozano AM, Gildenberg PL, Tasker RR (eds) Textbook of stereotactic and functionalneurosurgery, 2nd edn. Springer, New York, pp 2925–2941

Giacobbe P, Mayberg HS, Lozano AM (2009) Treatment resistant depression as a failure of brainhomeostatic mechanisms: implications for deep brain stimulation. Exp Neurol 219:44–52

Giacobbe P, Lipsman N, Hamani C, Lozano AM, Kennedy SH (2010) Subgenual cingulate gyrusdeep brain stimulation: current status and future directions. Psychiatr Ann 40:485–491

78 P. Giacobbe et al.

Guinjoan SM, Mayberg HS, Costanzo EY, Fahrer RD, Tenca E, Antico J, Cerquetti D, Smyth E,Leiguarda RC, Nemeroff CB (2010) Asymmetrical contribution of brain structures totreatment-resistant depression as illustrated by effects of right subgenual cingulum stimula-tion. J Neuropsychiatry Clin Neurosci 22:265–277

Hamani C, Mayberg H, Snyder B, Giacobbe P, Kennedy S, Lozano AM (2009) Deep brain stimulationof the subcallosal cingulate gyrus for depression: anatomical location of active contacts in clinicalresponders and a suggested guideline for targeting. J Neurosurg 111:1209–1215

Hamani C, Mayberg H, Stone S, Laxton A, Haber S, Lozano AM (2011) The subcallosalcingulate gyrus in the context of major depression. Biol Psychiatry 69:301–308

Holtzheimer PE, Kelley ME, Gross RE, Filkowski MM, Garlow SJ, Barrocas A, Wint D,Craighead MC, Kozarsky J, Chismar R, Moreines JL, Mewes K, Posse PR, Gutman DA,Mayberg HS (2012) Subcallosal cingulate deep brain stimulation for treatment-resistantunipolar and bipolar depression. Arch Gen Psychiatry 69:150–158

Houeto JL, Mallet L, Mesnage V, Tezenas du Montcel S, Béhar C, Gargiulo M, Torny F,Pelissolo A, Welter ML, Agid Y (2006) Subthalamic stimulation in Parkinson disease:behavior and social adaptation. Arch Neurol 63:1090–1095

Howard A, Honey CR, Hurwitz TA, Ilcewicz-Klimek M, Woo C, Lam RW, Berman N (2011)Letter to the editor. J Neurosurg (in press)

Kennedy SH, Giacobbe P, Rizvi SJ, Placenza FM, Nishikawa Y, Mayberg HS, Lozano AM(2011) Deep brain stimulation for treatment-resistant depression: follow-up after 3 to 6 years.Am J Psychiatry 168:502–510

Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH (2008)Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. BiolPsychiatry 64:461–467

Lozano AM, Giacobbe P, Hamani C, Rizvi SJ, Kennedy SH, Kolivakis TT, Debonnel G, Sadikot AF,Lam RW, Howard AK, Ilcewicz-Klimek M, Honey CR, Mayberg HS (2012) A multicenter pilotstudy of subcallosal cingulate area deep brain stimulation for treatment-resistant depression.J Neurosurg 116:315–322

Mayberg HS (2009) Targeted electrode-based modulation of neural circuits for depression. J ClinInvestig 119:717–725

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, KennedySH (2005) Deep brain stimulation for treatment-resistant depression. Neuron 45:651–660

McNab JA, Voets NL, Jenkinson N, Squier W, Miller KL, Goodwin GM, Aziz TZ (2009)Reduced limbic connections may contraindicate subgenual cingulate deep brain stimulationfor intractable depression. J Neurosurg 111:780–784

McNeely HE, Mayberg HS, Lozano AM, Kennedy SH (2008) Neuropsychological impact ofCg25 deep brain stimulation for treatment-resistant depression: preliminary results over12 months. J Nerv Mental Dis 196:405–410

Neimat JS, Hamani C, Giacobbe P, Merskey H, Kennedy SH, Mayberg HS, Lozano AM (2008)Neural stimulation successfully treats depression in patients with prior ablative cingulotomy.Am J Psychiatry 165:687–693

O’Leary DA, Lee AS (1996) Seven year prognosis in depression: mortality and readmission riskin the Nottingham ECT cohort. Br J Psychiatry 169:423–429

Puigdemont D, Portella MJ, Pérez-Egea R, de Diego-Adeliño J, Gironell A, Molet J, Duran-Sindreu S, Alvarez E, Pérez V (2009) Depressive relapse after initial response to subcallosalcingulate gyrus-deep brain stimulation in a patient with a treatment-resistant depression:electroconvulsive therapy as a feasible strategy. Biol Psychiatry 66:e11–e12

Puigdemont D, Pérez-Egea R, Portella MJ, Molet J, de Diego-Adeliño J, Gironell A, Radua J,Gómez-Anson B, Rodríguez R, Serra M, de Quintana C, Artigas F, Alvarez E, Pérez V (2011)Deep brain stimulation of the subcallosal cingulate gyrus: further evidence in treatment-resistant major depression. Int J Neuropsychopharmacol 22:1–13

Shergill SS, Robertson MM, Stein G, Bernadt M, Katona CLE (1999) Outcome in refractorydepression. J Affect Disord 54:287–294

8 Subcallosal Cingulate Cortex Deep Brain Stimulation 79

Chapter 9Deep Brain Stimulation of the HumanReward System as a Putative Treatmentfor Refractory Major Depression

T. E. Schlaepfer, V. A. Coenen and B. H. Bewernick

9.1 Deciding on a Stimulation Target

Several brain structures presumably play a role in the development and mainte-nance of symptoms in depression. Current studies in therapy-resistant depression(TRD) are targeting the nucleus accumbens (NAc) (Schlaepfer et al. 2008;Bewernick et al. 2010), the medial forebrain bundle (MFB) (Coenen et al. 2011),the anterior cingulate cortex (Cg25) (Mayberg et al. 2005; Lozano et al. 2008;Puigdemont et al. 2011), and the anterior limb of the internal capsule (ALIC)(Malone et al. 2009). These targets are in close anatomical or functional rela-tionship (neural networks) and an overlap of effects is probable.

9.1.1 The Role of the NAc in Anhedonia and Reward Processing

Anhedonia—the inability to experience positive emotions from an activity thatwas previously associated with reward effects—is one of the core symptoms indepression (American Psychiatric Association 1994; Rush and Weissenburger1994; Argyropoulos and Nutt 1997).

T. E. Schlaepfer (&) � B. H. BewernickBrain Stimulation Group, Department of Psychiatry and Psychotherapy,University Hospital Bonn, Bonn, Germanye-mail: thomas@schlaepfer.org

T. E. SchlaepferDepartments of Psychiatry and Mental Health, The Johns Hopkins University,Baltimore, MD, USA

V. A. CoenenDepartment of Functional Neurosurgery, University Hospital Bonn, Bonn, Germany

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_9, � Springer-Verlag Berlin Heidelberg 2012

81

The NAc has been chosen as a target for deep brain stimulation (DBS) indepression because it is a key structure of the reward system. It is expected that thedysfunction of the reward system (anhedonia and loss of motivation) can be restoredby modulating the NAc (Schlaepfer et al. 2008) for three reasons: (1) the ventralstriatum is the most important brain region for normal and abnormal reward pro-cessing and pleasure information, (2) the NAc is known to be the ‘‘motivationgateway’’ between limbic systems involved in emotion and motor control, and (3) theventral striatum is uniquely located to modulate activity in other regions of the brain.

First, the NAc is a central region for processing reward and pleasure infor-mation. Increases in neuronal activity and dopamine release are observed in theNAc during expectations and experience of rewards (de la Fuente-Fernandez et al.2002; Adinoff 2004; Schultz 2004; Doyon et al. 2005). Neuroimaging studies showincreases in ventral striatal activity associated with euphoric responses to dex-troamphetamine (Drevets et al. 2001), cocaine-induced euphoria (Breiter et al.1997), monetary reward (Knutson et al. 2001; O’Doherty et al. 2001; Cohen et al.2005), and attractive faces (Aharon et al. 2001). In addition, the ventral striatumexhibits abnormal activity following administration of dextroamphetamine inpatients with major depression (Tremblay et al. 2005). Furthermore, animalresearch has demonstrated dysfunction of the reward system in mice subjected tosocial defeat stress (Berton and Nestler 2006). Together, converging evidenceexists that the NAc is a key region for the experience of reward and pleasure, andthat this region is dysfunctional in patients with depression.

Second, the NAc acts as a gateway to transmit, and therefore enhance or degrade,information from emotion centers to motor control regions of the brain. In humans,the ventral striatum is very active during reward-seeking behaviors (Knutson, et al.2003; Juckel et al. 2006), and this activation is reduced in certain clinical populations,e.g., patients with schizophrenia (Juckel et al. 2006). Depleting dopamine from theNAc in rats impairs reward-seeking behavior (Ito et al. 2004). Thus, the NAcmediates reward-related motivational behavior. As anhedonia can also be concep-tualized as lacking reward-motivated behavior, the functioning of the NAc seems inparticular relevant for the treatment of depression.

Third, DBS probably does not simply function as a lesion of one area of the brain.It is very plausible that DBS exerts its influence far beyond the target structure (Kleinet al. 2011). Thus, it is necessary to modulate a structure that has many connections toother key regions involved in depression. The ventral striatum is in a position tomodulate activity in other regions of the brain. The NAc receives dopaminergicprojections from the midbrain ventral tegmental area, from regions involved inemotion (e.g., amygdala, orbitofrontal cortex, medial prefrontal cortex), from motorregions (e.g., dorsal caudate and globus pallidus), and from regions involved inmemory (e.g., hippocampus) (Nauta and Domesick 1984). The NAc indirectlyprojects to cortical regions, including Cg25 and the medial prefrontal cortex, theventral pallidum, the thalamus, the amygdala, and the hypothalamus (Jones andMogenson 1980; Mogenson et al. 1983; Kelley and Stinus 1984). Many of theseregions are known to be implicated in normal and abnormal emotion processing(Mayberg 1997). These connections can be GABAergic (inhibitory) or

82 T. E. Schlaepfer et al.

glutamatergic (excitatory). Thus, DBS of the NAc can modulate neural activity inother emotion and motivation centers of the brain (Fig. 9.1).

9.1.2 The Role of the MFB in Depression

The MFB has been posited as a more effective target by Coenen et al. (2011) forseveral reasons: (1) DBS studies use higher amplitudes in depression than inneurological diseases, making it probable that the right target has not been found;(2) the MFB is neuroanatomically and functionally connected with the other DBStargets in depression; (3) fiber tracking and emulation of the electrical field andside effects of existing targets demonstrate an involvement of the MFB; and (4)new insights into the mode of action of DBS.

First, all three major DBS targets in depression research are effective as demon-strated in studies with small sample sizes. However, higher voltages and currents areneeded than in neurological disorders. Thus, large electric fields possibly stimulatestructures in the proximity of the intended target sites and not the target sites them-selves (Coenen et al. 2011). Especially for the ALIC and NAc targets, larger DBSelectrodes with more widely spaced contacts are needed, since they are more effective(Coenen et al. 2011). It is thus very probable that stimulation of a network mediatingresponses to emotional stimuli leads to the antidepressant effect (Coenen et al. 2011).

Second, the MFB is a structure that connects frontal areas (including Cg25) to theorigin of the mesolimbic dopaminergic ‘‘reward’’ system in the midbrain ventral teg-mental area (Schoene-Bake et al. 2010). Thus, the MFB connects all targets indepression: white matter surrounding the subcallosal cingulate gyrus (Johansen-Berget al. 2008; Lozano et al. 2008; Hamani et al. 2009), the NAc (Sturm et al. 2003;Schlaepfer and Lieb 2005; Schlaepfer et al. 2008), and the ALIC (Gutman et al. 2009).This leads to the hypothesis that these targets are most likely clinically effective becauseof a stimulation of the MFB (Coenen et al. 2011).

Fig. 9.1 Main trajectoriesfor current deep brainstimulation (DBS) targets indepression (yellow dots) asseen from the front left. scgsubgenual cingulate gyrus(Broadmann’s area 25), vc/vsventral capsule/ventralstriatum, NAcc nucleusaccumbens septi, 1 genu ofcorpus callosum, 2 caudatenucleus, 3 putamen

9 Deep Brain Stimulation of the Human Reward System 83

Third, electric field stimulation and fiber tracking (diffusion tensor imaging and fibertracking) has demonstrated the involvement of the superolateral branch of the MFB inDBS of the three current targets for the treatment of major depression (Coenen et al.2011). In addition, psychotropic side effects of subthalamic nucleus DBS in Parkinson’sdisease can be interpreted as an involvement of the MFB in positive affective stateappetitive motivation (Coenen et al. 2009). In theory, MFB could have widespreadeffects on the affective restructuring of the brain (Schoene-Bake et al. 2010).

Fourth, this goes in line with current insights into the mechanisms of DBS andits effects on fiber pathways in the vicinity of targeted brain regions. Activationand modulation of afferent fiber tracts, as opposed to other possible effects such asinhibition of nuclear structures (Gradinaru et al. 2009; Hamani and Nobrega 2010),are plausible mechanisms of action in DBS (Schoene-Bake et al. 2010). In thisrespect, excitatory modulation and not inactivation of the MFB is postulated as themechanism of action (Schoene-Bake et al. 2010). The first study exploring anti-depressant effects of stimulation of the MFB in TRD is under way (Fig. 9.2).

9.2 Efficacy of, Side Effects of, and Neurobiological ChangesCaused by NAc DBS

Eleven Patients have been followed for up to 4 years with DBS of the NAc. NAc DBSlead to immediate and long-term (days and months) antidepressant effects (Bewer-nick et al. 2010). Very preliminary data on three Patients stimulated at the MFB, areencouraging (current study of the Brain Stimulation Group Bonn, Germany).

9.2.1 Immediate Clinical Effects

The immediate clinical effects of DBS (60 s) are often more obvious to the psy-chologist than to the patient: more spontaneous engagement in conversation,positive change in mood, relaxation, more ideas and plans, and thus exploratorymotivation. Only a few patients experience immediate effects and most patientscannot tell if the parameters have been adjusted or even if stimulation has stopped.Interestingly, immediate effects have not been predictive for long-term effects(Bewernick et al. 2010).

9.2.2 Long-Term Clinical Effects

Robust antidepressant effects of DBS of the NAc have been demonstrated in onestudy group of 11 patients with extremely severe depression and extremelytreatment resistant depression (Bewernick et al. 2012). About 50 % of patients

84 T. E. Schlaepfer et al.

Fig. 9.2 a Human medial forebrain bundle (MFB; green). Transverse section through midbrainand prefrontal cortex (PFC). The MFB connects the ventral tegmental area (VTA) and theperiaqueductal gray (PAG) with the PFC, and is subdivided into two major branches, theinferomedial MFB (imMFB) and the superolateral MFB (slMFB). On its way to the PFC, ittraverses the anterior limb of the internal capsule (ALIC), from where it connects to the nucleusaccumbens septi (NAcc). It converges in its most distal projection to Brodmann’s area 25(subgenual cingulate gyrus, scg). The red dot indicates the origin of the slMFB and represents anexperimental target point for treatment of depression with DBS. b The human slMFB. Magneticresonance tractographic depiction. PT pyramidal tracts, RN red nucleus, SNr substantia nigra parsreticulata, STN subthalamic nucleus, mtt mammillothalamic tract, LH lateral hypothalamus, fxfornix (a Courtesy of Volker A. Coenen, Bonn University, Germany; b courtesy of Volker A.Coenen and Burkhard Mädler, Bonn University, Germany)

9 Deep Brain Stimulation of the Human Reward System 85

responded significantly during the first 6 months and remained stable during fol-low-up up to 4 years (Bewernick et al. 2012). This response rate was similar to thatin studies on other targets, such as 45 % when targeting Cg25 (Lozano et al. 2008;Kennedy et al. 2011) and 53.3 % when targeting the ventral striatum (Malone et al.2009) at the last follow-up. In addition to antidepressant effects, an anxiolyticeffect has been observed. A significant increase in positive activities and thus ahedonic effect was achieved as well as an amelioration of quality of life (Be-wernick et al. 2012).

9.2.3 Cognitive Effects

One study assessed cognitive effects of NAc DBS with a comprehensive neuro-psychological battery (attention, learning and memory, language, executivefunction, and visual perception) at the baseline and after 12 months (Grubert et al.2011). No detrimental cognitive effects were demonstrated. On the contrary, therewas improved cognitive performance on tests of attention, memory, executivefunction, and visual perception after 12 months (Grubert et al. 2011). Generally,there was a general trend towards cognitive enhancement from below average toaverage performance independent of the antidepressant effect. Thus, this study ofNAc DBS demonstrated safety regarding cognitive effects (Grubert et al. 2011).Larger samples will be used to investigate possible procognitive effects in severalcognitive domains.

9.2.4 Side Effects

Side effects related to DBS such as erythema, transient increase in anxiety, agitation,headache, and sweating could be counteracted by small adjustments of the stimu-lation settings (Bewernick et al. 2010). No worsening of psychiatric symptoms,reoccurrence of new symptoms, or cognitive impairments were reported. But onepatient committed suicide and one attempted suicide during the first year. The suicidewas caused by severe distress in a personal relationship and was judged as not relatedto the stimulation (Bewernick et al. 2010). This demonstrates the extreme importanceof following up patients included in DBS studies very closely (see later).

9.2.5 Changes in Brain Metabolism

The reported clinical improvements were associated with neurobiological changes.Depression is associated with pathological and abnormal functioning of brainregions, including striatum and prefrontal cortex. Thus, normalization of brain

86 T. E. Schlaepfer et al.

metabolism in frontostriatal networks was observed shortly after surgery (1 week)(Schlaepfer et al. 2008) and after 1 year (Bewernick et al. 2010) as a result ofstimulation: One week of DBS increased metabolism in the NAc, amygdala, anddorsal prefrontal cortex, and decreased metabolism in the ventral medial prefrontalcortex (Schlaepfer et al. 2008). Long-term effects showed a decreased metabolism inthe subgenual cingulate gyrus and in prefrontal regions, including the orbitofrontalcortex, which is consistent with the metabolic decreases observed in patientsundergoing DBS of the Cg25 (Lozano et al. 2008). In line with the theory of hy-perresponsiveness of the amygdala to fear signals in anxiety disorders (Nitschke et al.2009), a normalization of metabolism in the amygdala was correlated with thereduction in anxiety scores in the responder group (Bewernick et al. 2010).

It is thus very probable that DBS of the NAc, through its anatomical andfunctional connections with other limbic and prefrontal structures, restores activityin these connected regions.

9.3 Indications and Quality Standards

9.3.1 Patient Inclusion and Monitoring

Because of notable risks of surgery (e.g., intracerebral bleeding and woundinfection) and the lack of broad efficacy, DBS research needs to adhere to thehighest ethical standards. Obligatory rules for patient inclusion and target selectionare needed. Inclusion criteria based on severity, chronicity, disability, and treat-ment refractoriness (Nuttin et al. 2002) need to be internationally standardized anddefined for each psychiatric disease. The following criteria should be applied inmajor depression:

Inclusion criteria

• Major depression, severe, unipolar type• German mother tongue (in Germany)• Hamilton Depression Rating Scale (HDRS-24) score of more than 20• Global Assessment of Function score of less than 45• At least four episodes of major depression or an episode duration of more than 2 years• Five years after the first episode of major depression• Failure to respond to adequate trials (more than 5 weeks at the maximum recommended ortolerated dose) of primary antidepressants from at least three different classes;• Adequate trials (more than 3 weeks at the usually recommended or maximum tolerated dose) ofaugmentation/combination of a primary antidepressant using at least two different augmenting orcombination agents (lithium, triiodothyronine, stimulants, antipsychotics, anticonvulsants,buspirone, or a second primary antidepressant)

(continued)

9 Deep Brain Stimulation of the Human Reward System 87

(continued)

Inclusion criteria

• At least one adequate trial of electroconvulsive therapy (more than six bitemporal treatments)and at least one adequate trial of individual psychotherapy (more than 20 sessions with anexperienced psychotherapist)• Ability to give written informed consent• No medical comorbidity• Drug-free or on a stable drug regimen at least 6 weeks before entry to the study

Exclusion criteria

• Current or past nonaffective psychotic disorder• Any current clinically significant neurological disorder or medical illness affecting brainfunction, other than motor tics or Gilles de la Tourette syndrome• Any clinically significant abnormality on preoperative magnetic resonance imaging• Any surgical contraindications to undergoing DBS• Current or unstably remitted substance abuse (aside from nicotine)• Pregnancy and women of childbearing age not using effective contraception• History of severe personality disorder

Today, no symptom-specific prediction can guide the selection process, butNAc DBS specifically changed anhedonia and anxiety. In addition, patients thatwere less affected could profit more from DBS, possibly because of an earlierintervention.

Especially in depression with an elevated risk of suicide associated with thedisease, careful patient monitoring is necessary. Before surgery, patients need tobe seen in regular intervals over at least several months to control for changes inseverity and ensure the inclusion criteria are met. It is extremely important toclarify the patient’s expectations before surgery, and to closely follow the patientafter the operation to prevent stress, catastrophic thinking, hypomania, or suicidalideation, especially in the event of a suboptimal immediate therapy effect. Aftersurgery, visits should take place weekly and after amelioration of symptoms andafter parameter adjustment at monthly intervals for 1 year at least in order toevaluate the long-term effects. In the case of no response or immediate aggravationof symptoms, hospitalization or other treatment options (psychotherapy, change inmedication, and electroconvulsive therapy) should be offered.

9.3.2 Requirement for Study Center and Quality Standards

In psychiatric disorders the process of diagnosis is less verifiable and observable thanin neurology lacking neurobiological markers. Thus, it is essential to corroborate thepatient’s life history, course of illness, and psychopathology. Each case must bedocumented according to high scientific and administrative expectations (stan-dardized diagnostic with clinical scales, evaluation of cognitive parameters with

88 T. E. Schlaepfer et al.

psychological tests, quality of life, report of parameter changes, other therapies, etc.).In addition to the evaluation of the clinical effects, basic neuroscience (e.g., brainimaging, intracranial EEG, genetics, and anatomic estimation of the individualelectrical field) should be applied to learn the most about each patient.

This requires a team consisting of functional neurosurgeons, psychiatrists, andneuropsychologists who have developed long-standing and convincing experiencewith this patient group. These standards are most straightforwardly fulfilled intertiary-care academic centers where such resources are available.

In addition, the minimal requirements for using DBS in psychiatric conditions(Nuttin et al. 2002) should include an ethics committee to consider the studyprotocol and ongoing projects. Despite any review by committee, clinicalresponsibility remains with the patient’s clinicians and is not shared with reviewcommittees. Scientific quality standards for target selection need to be establishedwith clear anatomical and functional hypotheses.

9.3.3 NAc DBS for the Treatment of Bipolar Disorder

The symptoms of depression in the context of bipolar disorder are generally not thesame as in unipolar major depression (Belmaker 2004). Bipolar depression tends tobe atypical, with prominent fatigue, hypersomnia, and reverse diurnal mood var-iability (Berns and Nemeroff 2003). Nonetheless, current data point to the fact thatthe neurobiological aspects of bipolar depression are very similar to those ofunipolar major depression, especially regarding striatal dysfunction (Marchandand Yurgelun-Todd 2010; Kupferschmidt and Zakzanis 2011). In addition, anhe-donia and lack of motivation are also prominent in bipolar patients with TRD.

Similar to major depression, the pharmacological treatment of chronic bipolardisorder does not seem to be effective enough despite the availability of manypharmacological substances (Gijsman et al. 2004; Papadimitriou et al. 2007).Thus, bipolar patients with depression, especially anhedonia, could possibly profitfrom NAc DBS. It has to be kept in mind that DBS can possibly induce manic orhypomanic states (Bewernick et al. 2010; Haq et al. 2010); therefore, patients haveto be assessed carefully for hypomanic symptoms. A study is currently beingperformed in Bonn, Germany, exploring antidepressant effects and risk of maniainduction caused by NAc DBS in patients with bipolar disorder.

9.3.4 NAc as a Putative Target for the Treatmentof Nonaffective Disorders

Anhedonia is also often implicated with other chronic and debilitating psychiatricdisorders (Loas 1996), such as obsessive–compulsive disorders (OCD), substanceabuse disorders (Wise 1996), and schizophrenia (Wolf 2006). These initial

9 Deep Brain Stimulation of the Human Reward System 89

observations that DBS of striatal regions might indeed restore dysfunctionalprocessing of reward stimuli lays the ground for research into similar approachesfor the treatment of those other disorders. Evidence for efficacy of NAc DBS inobsessive–compulsive disorders has been demonstrated (Huff et al. 2010).

Single case reports of smoking cessation and remission of alcohol dependencythat occur as side effects of NAc DBS primarily used for the treatment of otherdisorders (Kuhn et al. 2007, 2009) do not alone justify the application of NAc DBSin substance addiction. Nonetheless, a DBS study targeting the NAc in opioidaddiction is currently under way.

9.4 Conclusions and Outlook

The NAc has been studied as a target in depression for DBS because of itsprominent role in the reward system. The first evidence has proven that NAc DBSinduces stable antidepressant (especially antianhedonic and anxiolytic) effects.Because of the small sample sizes in all DBS studies on depression, larger con-trolled studies (including a double-blind sham phase) have to be initiated beforeDBS can be seen as a treatment option also in less severe TRD.

Today it is not possible to decide on the optimal target for DBS in TRD as thestudies have been small and long-term data are only available for 5 years. Thereare also new targets under debate, such as the habenula (Sartorius and Henn 2007;Sartorius et al. 2010) and the MFB, which connects frontal DBS targets (CG25)with the NAc (Coenen et al. 2009, 2011).

The application of NAc DBS for the treatment of other psychiatric diseases(e.g., alcohol dependency, opioid addiction, schizophrenia) is currently underinvestigation.

DBS is a unique and promising method for the treatment of TRD. There havebeen no fundamental ethical objections to its use in psychiatric disorders (seeChap. 25), but until substantial clinical data are available, mandatory standards areneeded to prevent harming patients.

References

Adinoff B (2004) Neurobiologic processes in drug reward and addiction. Harv Rev Psychiatry12(6):305–320

Aharon I, Etcoff N et al (2001) Beautiful faces have variable reward value: fMRI and behavioralevidence. Neuron 32(3):537–551

American Psychiatric Association (1994) Diagnostic and statistical manual of mental disorders,4th edn. American Psychiatric Association, Arlington

Argyropoulos SV, Nutt DJ (1997) Anhedonia and chronic mild stress model in depression.Psychopharmacology (Berl) 134(4):333–336; discussion 371–337

Belmaker RH (2004) Bipolar disorder. N Engl J Med 351(5):476–486

90 T. E. Schlaepfer et al.

Berns GS, Nemeroff CB (2003) The neurobiology of bipolar disorder. Am J Med Genet C SeminMed Genet 123(1):76–84

Berton O, Nestler EJ (2006) New approaches to antidepressant drug discovery: beyondmonoamines. Nat Rev Neurosci 7(2):137–151

Bewernick BH, Hurlemann R et al (2010) Nucleus accumbens deep brain stimulationdeep brainstimulation decreases ratings of depression and anxiety in treatment-resistant depression. BiolPsychiatry 67(2):110–116

Bewernick B, Kayser S et al. (2012) Long-term effects of nucleus accumbens deep brainstimulation in treatment resistant depression—evidence for sustained efficacy. Neuropsycho-pharmacology (in press)

Breiter HC, Gollub RL et al (1997) Acute effects of cocaine on human brain activity and emotion.Neuron 19(3):591–611

Coenen VA, Honey CR et al (2009) Medial forebrain bundle stimulation as a pathophysiologicalmechanism for hypomania in subthalamic nucleussubthalamic nucleus deep brain stimula-tiondeep brain stimulation for Parkinson’s disease. Neurosurgery 64(6):1106–1114; discus-sion 1114–1105

Coenen VA, Schlaepfer TE et al (2011) Cross-species affective functions of the medial forebrainbundle–implications for the treatment of affective pain and depression in humans. NeurosciBiobehav Rev 35(9):1971–1981

Cohen MX, Young J et al (2005) Individual differences in extraversion and dopamine geneticspredict neural reward responses. Brain Res Cogn Brain Res 25(3):851–861

de la Fuente-Fernandez R, Phillips AG et al (2002) Dopamine release in human ventral striatumand expectation of reward. Behav Brain Res 136(2):359–363

Doyon WM, Anders SK et al (2005) Effect of operant self-administration of 10 % ethanol plus10 % sucrose on dopamine and ethanol concentrations in the nucleus accumbens.J Neurochem 93(6):1469–1481

Drevets WC, Gautier C et al (2001) Amphetamine-induced dopamine release in human ventralstriatum correlates with euphoria. Biol Psychiatry 49(2):81–96

Gijsman HJ, Geddes JR et al (2004) Antidepressants for bipolar depression: a systematic reviewof randomized, controlled trials. Am J Psychiatry 161(9):1537–1547

Gradinaru V, Mogri M et al (2009) Optical deconstruction of parkinsonian neural circuitry.Science 324(5925):354–359

Grubert C, Hurlemann R et al (2011) Neuropsychological safety of nucleus accumbens deep brainstimulationdeep brain stimulation for major depression: effects of 12-month stimulation.World J Biol Psychiatry 12(7):516–527

Gutman DA, Holtzheimer PE et al (2009) A tractography analysis of two deep brain stimulationdeepbrain stimulation white matter targets for depression. Biol Psychiatry 65(4):276–282

Hamani C, Mayberg H et al (2009) Deep brain stimulation of the subcallosal cingulate gyrus fordepression: anatomical location of active contacts in clinical responders and a suggestedguideline for targeting. J Neurosurg 111(6):1209–1215

Hamani C, Nobrega JN (2010) Deep brain stimulation in clinical trials and animal models ofdepression. Eur J Neurosci 32(7):1109–1117

Haq IU, Foote KD et al (2010) A case of mania following deep brain stimulationdeep brainstimulation for obsessive compulsive disorderobsessive compulsive disorder. Stereotact FunctNeurosurg 88(5):322–328

Huff W, Lenartz D et al (2010) Unilateral deep brain stimulationdeep brain stimulation of thenucleus accumbens in patients with treatment-resistant obsessive-compulsive disorder:outcomes after one year. Clin Neurol Neurosurg 112(2):137–143

Ito R, Robbins TW et al (2004) Differential control over cocaine-seeking behavior by nucleusaccumbens core and shell. Nat Neurosci 7(4):389–397

Johansen-Berg H, Gutman DA et al (2008) Anatomical connectivity of the subgenual cingulateregion targeted with deep brain stimulationdeep brain stimulation for treatment-resistantdepression. Cereb Cortex 18(6):1374–1383

9 Deep Brain Stimulation of the Human Reward System 91

Jones DL, Mogenson GJ (1980) Nucleus accumbens to globus pallidus GABA projection:electrophysiological and iontophoretic investigations. Brain Res 188(1):93–105

Juckel G, Schlagenhauf F et al (2006) Dysfunction of ventral striatal reward prediction inschizophrenia. Neuroimage 29(2):409–416

Kelley AE, Stinus L (1984) The distribution of the projection from the parataenial nucleus of thethalamus to the nucleus accumbens in the rat: an autoradiographic study. Exp Brain Res54(3):499–512

Kennedy SH, Giacobbe P et al (2011) Deep brain stimulation for treatment-resistant depression:follow-up after 3 to 6 years. Am J Psychiatry 168(5):502–510

Klein J, Soto-Montenegro ML et al (2011) A novel approach to investigate neuronal networkactivity patterns affected by deep brain stimulationdeep brain stimulation in rats. J PsychiatrRes 45(7):927–930

Knutson B, Adams CM et al (2001) Anticipation of increasing monetary reward selectivelyrecruits nucleus accumbens. J Neurosci 21(16):RC159

Knutson B, Fong GW et al (2003) A region of mesial prefrontal cortexprefrontal cortex tracksmonetarily rewarding outcomes: characterization with rapid event-related fMRI. Neuroimage18(2):263–272

Kuhn J, Bauer R et al (2009) Observations on unaided smoking cessation after deep brainstimulationdeep brain stimulation of the nucleus accumbens. Eur Addict Res 15(4):196–201

Kuhn J, Lenartz D et al (2007) Remission of alcohol dependency following deep brainstimulationdeep brain stimulation of the nucleus accumbens: valuable therapeutic implica-tions? J Neurol Neurosurg Psychiatry 78(10):1152–1153.

Kupferschmidt DA, Zakzanis KK (2011) Toward a functional neuroanatomical signature ofbipolar disorder: quantitative evidence from the neuroimaging literature. Psychiatry Res193(2):71–79

Loas G (1996) Vulnerability to depression: a model centered on anhedonia. J Affect Disord41(1):39–53

Lozano AM, Mayberg HS et al (2008) Subcallosal cingulate gyrus deep brain stimulationdeepbrain stimulation for treatment-resistant depression. Biol Psychiatry 64(6):461–467

Malone DA Jr, Dougherty DD et al (2009) Deep brain stimulation of the ventral capsule/ventralstriatum for treatment-resistant depression. Biol Psychiatry 65(4):267–275

Marchand WR, Yurgelun-Todd D (2010) Striatal structure and function in mood disorders: acomprehensive review. Bipolar Disord 12(8):764–785

Mayberg H, Lozano A et al (2005) Deep brain stimulation for treatment-resistant depression.Neuron 45(5):651–660

Mayberg HS (1997) Limbic-cortical dysregulation: a proposed model of depression. J Neuro-psychiatry Clin Neurosci 9(3):471–481

Mogenson GJ, Swanson LW et al (1983) Neural projections from nucleus accumbens to globuspallidus, substantia innominata, and lateral preoptic-lateral hypothalamic area: an anatomicaland electrophysiological investigation in the rat. J Neurosci 3(1):189–202

Nauta WJ, Domesick VB (1984) Afferent and efferent relationships of the basal gangliabasalganglia. Ciba Found Symp 107:3–29

Nitschke JB, Sarinopoulos I et al (2009) Anticipatory activation in the amygdala and anteriorcingulate in generalized anxiety disorder and prediction of treatment response. Am JPsychiatry 166(3):302–310

Nuttin B, Gybels J et al (2002) Deep brain stimulation for psychiatric disorders. Neurosurgery51(2):519

O’Doherty J, Kringelbach ML et al (2001) Abstract reward and punishment representations in thehuman orbitofrontal cortex. Nat Neurosci 4(1):95–102

Papadimitriou GN, Dikeos DG et al (2007) Non-pharmacological treatments in the managementof rapid cycling bipolar disorder. J Affect Disord 98(1–2):1–10

Puigdemont D, Perez-Egea R et al (2011) Deep brain stimulation of the subcallosal cingulategyrus: further evidence in treatment-resistant major depression. Int J Neuropsychopharma-col:1–13

92 T. E. Schlaepfer et al.

Rush AJ, Weissenburger JE (1994) Melancholic symptom features and DSM-IV. Am JPsychiatry 151(4):489–498

Sartorius A, Henn FA (2007) Deep brain stimulation of the lateral habenula in treatment resistantmajor depression. Med Hypotheses 69(6):1305–1308

Sartorius A, Kiening KL et al (2010) Remission of major depression under deep brainstimulationdeep brain stimulation of the lateral habenula in a therapy-refractory patient. BiolPsychiatry 67(2):e9–e11

Schlaepfer TE, Lieb K (2005) Deep brain stimulation for treatment of refractory depression.Lancet 366(9495):1420–1422

Schlaepfer TE, Cohen MX et al (2008) Deep brain stimulation to reward circuitry alleviatesanhedonia in refractory major depression. Neuropsychopharmacology 33(2):368–377

Schoene-Bake JC, Parpaley Y et al (2010) Tractographic analysis of historical lesion surgery fordepression. Neuropsychopharmacology 35(13):2553–2563

Schultz W (2004) Neural coding of basic reward terms of animal learning theory, game theory,microeconomics and behavioural ecology. Curr Opin Neurobiol 14(2):139–147

Sturm V, Lenartz D et al (2003) The nucleus accumbens: a target for deep brain stimulationdeepbrain stimulation in obsessive-compulsive- and anxiety-disorders. J Chem Neuroanat26(4):293–299

Tremblay LK, Naranjo CA et al (2005) Functional neuroanatomical substrates of altered rewardprocessing in major depressive disorder revealed by a dopaminergic probe. Arch GenPsychiatry 62(11):1228–1236

Wise RA (1996) Addictive drugs and brain stimulation reward. Annu Rev Neurosci 19:319–340Wolf DH (2006) Anhedonia in schizophrenia. Curr Psychiatry Rep 8(4):322–328

9 Deep Brain Stimulation of the Human Reward System 93

Chapter 10Depression in Humans: The VentralCapsule/Ventral Striatum

Mayur Pandya, Andre Machado and Donald Malone

10.1 Introduction

For many years, ablative procedures were the only available neurosurgical optionfor the management of severe neuropsychiatric conditions. Interventions such ascapsulotomy, cingulotomy, and limbic leukotomy improved the quality of life formany patients and provided significant relief to distressed patients and families.However, the irreversible nature of these lesions put patients at potential risk of adecline in executive or other cognitive functions as well as possible behavioraladverse effects (Ruck et al. 2003). With the advent and success of deep brainstimulation (DBS) for the treatment of movement disorders in the late 1990s, thepossibility of a parallel opportunity in the field of psychiatry was later explored.

The use of DBS of the ventral capsule/ventral striatum (VC/VS) for the treatmentof major depressive disorder (MDD) stems from pioneering work in obsessive–compulsive disorder (OCD). The anatomical underpinnings of mood and behavioralneural networks in humans, namely, thalamocortical connections and the prefrontal–ventral striatal–ventral pallidal–thalamocortical circuit, overlap with those of OCD(Modell et al. 1989; Remijnse et al. 2006; Carballedo et al. 2011).

M. PandyaDepartment of Psychiatry, Cleveland Clinic, 9500 Euclid Avenue – P57,Cleveland, OH 44195, USA

A. MachadoCleveland Clinic, Center for Neurological Restoration, 9500Euclid Avenue – S31, Cleveland, OH 44195, USA

D. Malone (&)Department of Psychiatry, Cleveland Clinic, 9500 Euclid Avenue – P57,Cleveland, OH 44195, USAe-mail: maloned@ccf.org

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_10, � Springer-Verlag Berlin Heidelberg 2012

95

The potential use of DBS for targeting the VC/VS, replacing capsulotomies forthe treatment of OCD, was pioneered by Nuttin et al. (1999), and was subsequentlypropagated by Greenberg et al. (2006). Capsulotomy, or ablation of the anteriorlimb of the internal capsule (ALIC), had been extensively studied by Leksell,Meyerson, and the Karolinska group for several decades (Mindus and Meyerson1995; Leksell and Backlund 1978). The work with gamma capsulotomies wasadvanced in the USA and further investigated at Brown University (Rasmussenet al. 2000). Anterior radiofrequency thermocapsulotomy, a stereotactic procedurein which the fibers of the ALIC are lesioned, has been shown to alleviate obsessivesymptoms and anxiety. This corroborates the importance of the fibers projectingthrough the ALIC, including those projecting to the cingulate and to mesialthalamus, in the control of emotions and behavior. Demonstration of functionalactivation of this circuitry with acute DBS of the VC/VS in six patients with OCD(Rauch et al. 2006) supported abnormalities in this region of interest anddemonstrated that DBS could modulate its function. Greenberg et al. (2006)demonstrated improved functionality in six of ten patients with OCD after chronicstimulation of the ventral ALIC and noted improvements in mood, opening thedoor for investigation of VC/VS DBS in mood disorders. DBS has been shown tobe a safe, reversible, and adjustable alternative to lesioning procedures in move-ment disorders (Tasker et al. 1997). DBS allows adverse effects to be managed byadjusting stimulation, and the effects can be activated or deactivated as needed.Additionally, unlike prior lesioning techniques used for this brain region,irreversible adverse effects have not been seen (Greenberg et al. 2010).

10.2 Anatomical Overview

Understanding the rationale for the VC/VS target chosen for investigation of MDDrequires an appreciation of the neuroanatomical structures and their positioningrelative to each other. Stimulation intended for a single region can result in a cascadeof changes in neighboring structures. Additionally, the expression of receptor typesand density in the region may augment or alter an affective response.

The orbitofrontal cortex projections and nucleus accumbens are believed toplay a significant role in emotional processing. The orbitofrontal cortex sharesmassive reciprocal excitatory projections with the thalamic mediodorsal nucleus,through the ventral ALIC. This excitatory intercommunication is modulated by alonger loop, initiated by fibers that project from the anterior prefrontal cortex to theventral striatum and then processed via the ventral pallidum and thalamus. Thiscircuitry—long loop and short loop originating at the orbitofrontal cortex—ishighly involved in the individual’s control of emotion and affective behavior, aswell as in the postulated pathogenesis of psychiatric disorders (Modell et al. 1989).

The densely arranged fiber bundles through the ALIC—connecting the orbito-frontal cortex to areas including the mesial thalamus and to the ventral striatum—makes it a favorable stereotactic target to affect the pathways related to the control ofemotional behavior. The ventral striatum area also has neighboring structures, such

96 M. Pandya et al.

as the bed nucleus of the stria terminalis, the anterior commissure, and the nucleusaccumbens, which are believed to be involved with stress-related and reward-motivation components of depression (Forray and Gysling 2004), presumablythrough afferent inputs from the limbic lobe. Given the likely spread of current tothese areas, these structures and related pathways may be involved in the mecha-nisms underlying the response to DBS.

The current leads for DBS are quadripolar. The height of each electrode contact andthe spacing between contacts differ from one lead model to the next. Some lead modelshave a long span of contacts, which will extend from the ventral part of the ventralstriatum to the dorsal ALIC. Stimulation can be adjusted so that each electrode contactcan be activated independently. Hence, the topographical organization of the region ofthe VC/VS plays a vital role in the response that is generated at various stimulationsettings. In addition, the programmer can adjust the amplitude of stimulation as well asthe frequency and width of each electrical pulse. In our experience, the most distalcontacts, presumably those at the site of the ventral striatum and ventral ALIC, gen-erate the most robust mood changes—including mood elevation, laughter, andincreased speech—as well as increased energy and alertness, and other common, butadverse, responses may include dizziness and anxiety (Machado et al. 2009). Differ-ences may also be noted on the basis of laterality. In addition to direct stimulation ofthese target sites, it is postulated that acute mood changes are a result of spread toneighboring white matter bundles such as the fornix, anterior commissure, internalcapsule, or extended amygdala. Isolated responses may also include sensations ofwarmth and flushing, which may be explained by projections to the hypothalamus. Themore dorsal electrode contacts, positioned along the main axis of the ALIC, resulted inlimited mood changes with inconsistent and relatively minimal responses of any kindwhen activated as cathodes.

In addition to its rich anatomical connections, the VC/VS has a high expression ofserotonin receptors which may facilitate a mood response. The ALIC has fibers ofpassage connecting the subgenual cingulate and orbitofrontal cortices, and it is pos-sible that stimulation of those fibers results in modulation of serotonin (Lujan et al.2008). Other investigations have supported the regulation of mood by the ventralstriatum by confirming abnormally reduced function of ventral striatal serotoninreceptors in affective disorders, such as MDD (Murrough et al. 2011), and demon-strating poorly sustained ventral striatal response to reward stimulus in mothers withpostpartum depression (Moses-Kolko et al. 2011). As discussed in the next section, theextent and variety of response at the VC/VS target may be further dependent on thestimulation mode used for DBS, the active contacts selected, and the settings.

10.3 Programming of the VC/VS for the Treatmentof Depression

Presently there are no consensus guidelines or algorithms for DBS programming indepression. With our experience over the past decade in 16 patients who underwentDBS electrode implantation at the VC/VS for the treatment of depression at our

10 Depression in Humans 97

clinic, we have been able to delineate trends and common responses that may assist inthe programming of this population.

The initial step of programming begins prior to implantation with appropriatepatient selection. Patients with MDD affected by another major comorbidpsychiatric illness (such as bipolar disorder, schizophrenia, schizoaffective disorder,active substance dependence) have not been evaluated at this target. Furthermore,although the presence of a personality disorder was not an exclusionary criterion inour cohort, those with severe personality disorders should be evaluated critically,especially if the risk of impulsivity is high.

Prior to any therapeutic programming, it is advisable to perform an initial cathodesurvey (Malone et al. 2009a, b), both intraoperatively and then more formally in theoffice. The intraoperative survey may be brief but is intended to ensure there are nosignificant adverse effects within the stimulation target. Following postoperativerecovery, an intensive survey of various combinations of contact settings, withdifferent pulse widths, should be undertaken with the goal of establishing anddocumenting the most therapeutic responses as well as thresholds to side effects.The survey should begin with a unilateral monopolar investigation, gradually pro-gressing to a bilateral survey which would include bipolar configurations. Thisideally should include a combination of distal cathodes paired with more proximalanode contacts, including the internal pulse generator. A systematic and compre-hensive evaluation of stimulation responses, both patient-reported and clinician-observed, is essential to guide future programming decisions. In our experience, thesurvey is ideally performed over a 3–5-day period with a maximum of 2 h per day ofprogramming to limit patient fatigue and minimize loss of accurate feedback toprogramming effects. It is helpful to keep a log so that each contact and stimulationsetting can be later correlated with the behavioral effects in order to allow selection ofthe best settings for long-term programming.

Once the stimulation parameters have been assessed and the responses carefullydocumented, the most effective settings should be programmed. Although the mostrobust responses appear to occur with bilateral stimulation, one must keep in mind thatin some cases stimulation may involve unilateral stimulation and/or monopolar con-figurations, which are typically utilized to minimize adverse effects. In addition,aggressive upward titration of stimulation during a single visit (i.e., typically anincrease of more than 3 V) should be done with caution as responses may occasionallybe delayed and not immediately evident upon programming. A more conservativetitration with a subsequent ‘‘wait time’’ (i.e., 30–45 min) following programmingsessions is recommended to minimize negative responses and to capture any delayedeffects. Although the intent is to improve mood to treat depression, it is possible totrigger hypomania. Increments in mood improvement should be made cautiously andaimed at restoring normality and not at reaching an abnormal plateau. Patients whoexhibit significantly improved mood during a visit with the programming physician areat risk of progressing into hypomania later if the same settings are maintained. This isone of the reasons why the authors prefer a more cautious approach.

In some cases long-term mood improvement may be inadequate or transient innature. Various strategies, such as activating multiple cathode contacts, manipulating

98 M. Pandya et al.

the pulse width, and/or employing intermittent (‘‘cycling’’) stimulation, may beworthwhile to achieve a more sustained response. At other times, stimulation at highervoltages may produce a desired response, along with overlapping somatosensoryeffects or autonomic hyperactivity. In these cases, adjustment of the stimulation fre-quency (i.e., rate), reversal of contact polarity, and/or utilization of stimulation titrationin small increments (i.e., 0.1 V), may attenuate these responses with relative preser-vation of the affective improvement. Lastly, although there appears to be a linearrelationship between voltage titration and mood response in many individuals, somepatients may exhibit a plateau effect. Further increases beyond a threshold voltage may,in some cases, result in a paradoxical effect. Establishing these thresholds during theinitial survey testing may minimize any unexpected responses.

The frequency of programming visits is variable and dependent upon manyfactors, including severity of symptoms or illness, rate of response to programmingchanges, and patient availability. Close monitoring of changes (either by telephoneor by face-to-face visits) following any programming change is helpful indetecting side effects. Furthermore, parallel medication adjustments should beperformed to optimize the response and minimize further mood deterioration. Solereliance on DBS to achieve a mood response for an extended period is notadvisable at this time. For most patients, it serves to augment the treatment theyare receiving with medications and/or psychotherapy.

Regular hardware maintenance involves frequent evaluation of therapyimpedances, monitoring for battery replacements, and ensuring there are safetymeasures to limit disruption of therapy. Impedances should be evaluated to ensureadequate and safe therapy is being delivered. A check of the impedances at allcontacts should be documented at each programming visit. Battery replacementsare dependent on the amount of current being delivered and the impedanceinvolved. Stimulation at higher amplitudes employing multiple contacts maydeplete the battery more quickly than stimulation at lower voltages. The averagebattery life for most patients in our cohort was 10–18 months, making batteryreplacement a relatively frequent occurrence. Advances in battery technology,such as rechargeable capabilities, may allow extended battery use betweenreplacements. Finally, every patient should be educated on the safety measures forthe implanted DBS device, ranging from damage and bodily injury from magneticresonance imaging scans to other environmental electromagnetic sources that maycause interruption in therapy or impose risk of injury to the patient.

10.4 Latest Results

Malone et al. (2009a, b), in a multicenter, open-label study, demonstrated the long-term clinical outcomes of VC/VS stimulation for the treatment of MDD. Fifteenlong-term, refractory patients (four males and 11 females, ranging in age from 18to 55 years) with MDD at three sites underwent DBS electrode implantation in theVC/VS region. The leads were implanted following the dorsoventral trajectory of

10 Depression in Humans 99

the ALIC, with the most distal contact (0) at the ventral striatum below the level ofthe anterior commissure. The patients were blind to the stimulation settings, withlong-term parameter selection based on a positive mood benefit and the absence ofadverse effects. Most of the patients in the study had the most distal electrodes(0,1, or both) programmed as a cathode and the neurostimulator or electrode 3configured as the anode. The stimulation frequency was either 100 or 130 Hz, andthe pulse width was typically 90 or 210 ls. The mean stimulation parameters at thelast follow-up were as follows: amplitude 6.7 (±1.8) V; pulse width 113.0(±45.0) ls, and frequency 127.0 (±11.1) Hz.

At the time of publication, the longest follow-up period was 51 months, with amean last follow-up of 23.5 (±14.9) months. There was a notable reduction indepressive symptoms in this highly treatment resistant cohort over the course ofthe study. In parallel, mean Global Assessment of Functioning scores increasedsignificantly for the group as a whole. A total of 25 serious adverse events werereported in six patients over a period equal to 353 patient-months of experience.Four (of six) were identified as related to DBS, including three incidents of moodchanges (hypomania and/or depression) which responded to adjustments in stim-ulation parameters. All patients had a minimum of 6 months of active stimulationand over two-thirds had a 1-year follow-up. On the basis of primary outcomemeasures at the last observation, five patients met the criterion for remission andeight met the criterion for categorical response.

Supplemental data—with inclusion of two additional patients (n = 17) and anaverage follow-up of 37.4 months—demonstrated the average reduction in theMontgomery–Asberg Depression Rating Scalescore to be 52.7 % at 3 months,48.8 % at 6 months, 54.8 % at 12 months, and 59.2 % at the last follow-up(Malone et al. 2009a, b). In addition, there was a significant reduction in suici-dality (73.3 %) over the course of the first year after implantation, with the mostsignificant drop occurring over the first month of stimulation (Malone et al. 2008).These are results from open-label studies. They indicate that DBS of the VC/VS issafe in patients with treatment-resistant depression. Additional controlled studies,including an ongoing study, will further assess the efficacy of the therapy.

10.5 Conclusion

The development of this emerging therapy depends on advances in neurosurgicaltechnique and patient selection, as well as furthering our understanding of thecircuitry involved in the targeted region. The safety profile thus far with DBS atthe VC/VS for the treatment of psychiatric disorders compares favorably to that oflesioning procedures; however, the continued prospective evaluation of cognitivefunctioning in patients undergoing long-term DBS is essential. Further experiencemay allow investigators to improve delivery of the therapy and improve outcomesthrough targeting and/or manipulation of stimulation parameters.

100 M. Pandya et al.

References

Carballedo A, Scheuerecker J, Meisenzahl E, Schoepf V, Bokde A, Möller HJ, Doyle M,Wiesmann M, Frodl T (2011) Functional connectivity of emotional processing in depression.J Affect Disord 134(1–3):272–9

Forray MI, Gysling K (2004) Role of noradrenergic projections to the bed nucleus of the striaterminalis in the regulation of the hypothalamic-pituitary-adrenal axis. Brain Res Rev47(1–3):145–160

Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, Salloway SP, OkunMS, Goodman WK, Rasmussen SA (2006) Three-year outcomes in deep brain stimulation forhighly resistant obsessive-compulsive disorder. Neuropsychopharmacology31(11):2384–2393

Greenberg BD, Gabriels LA, Malone DA Jr, Rezai AR, Friehs GM, Okun MS, Shapira NA, FooteKD, Cosyns PR, Kubu CS et al (2010) Deep brain stimulation of the ventral internal capsule/ventral striatum for obsessive-compulsive disorder: worldwide experience. Mol Psychiatry15(1):64–79

Leksell L, Backlund EO (1978) [Radiosurgical capsulotomy—a closed surgical method forpsychiatric surgery]. Lakartidningen 75(7):546–547

Lujan JL, Chaturvedi A, McIntyre C (2008). Tracking the mechanisms of deep brain stimulationfor neuropsychiatric disorders. Front Biosci 13:5892–5904

Machado A, Haber S, Sears N, Greenberg B, Malone D, Rezai A (2009) Functional topography ofthe ventral striatum and anterior limb of the internal capsule determined by electricalstimulation of awake patients. Clin Neurophysiol 120(11):1941–1948

Malone DA, Dougherty DD, Rezai AR, Carpenter LL, Friehs GM, Eskandar EN, Rauch SL,Rasmussen SA, Machado AG, Kubu CS et al (2009a) Deep brain stimulation of the ventralcapsule/ventral striatum for treatment-resistant depression. Biol Psychiatry 65(4):267–275

Malone DA, Dougherty DD, Rezai AR, Carpenter LL, Friehs GM, Eskandar EN, Rauch SL,Rasmussen SA, Machado AG, Kubu CS et al (2009b) Deep brain stimulation of the ventralcapsule/ventral striatum for treatment-resistant depression. American Psychiatric AssociationAnnual Meeting, San Francisco, CA, May 2009

Mathews M, Greenberg B, Dougherty D, Rezai A, Carpenter L, Kubu C, Malone D (2008)Change in suicidal ideation in patients undergoing DBS for depression. American Society forStereotactic and Functional Neurosurgery Biannual Meeting, Vancouver, Canada, June 2008

Mindus P, Meyerson BA (1995) Anterior capsulotomy for intractable anxiety disorders. In:Schmidek H, Sweet W (eds) Operative neurosurgical techniques, 3rd edn. W.B. SaundersCompany, Philadelphia, pp 1443–1455

Modell JG, Mountz JM, Curtis GC, Greden JF (1989) Neurophysiologic dysfunction in basalganglia/limbic striatal and thalamocortical circuits as a pathogenetic mechanism of obsessive-compulsive disorder. J Neuropsychiatry Clin Neurosci 1(1):27–36

Moses-Kolko EL, Fraser D, Wisner KL, James JA, Saul AT, Fiez JA, Phillips ML (2011) Rapidhabituation of ventral striatal response to reward receipt in postpartum depression. BiolPsychiatry 70(4):395–399

Murrough JW, Henry S, Hu J, Gallezot JD, Planeta-Wilson B, Neumaier JF, Neumeister A (2011)Reduced ventral striatal/ventral pallidal serotonin1B receptor binding potential in majordepressive disorder. Psychopharmacology 213(2–3):547–553

Nuttin B, Cosyns P, Demeulemeester H, Gybels J, Meyerson B (1999) Electrical stimulation inanterior limbs of internal capsules in patients with obsessive-compulsive disorder. Lancet354(9189):1526

Rasmussen S, Greenberg B, Mindus P, Friehs G, Noren G (2000) Neurosurgical approaches tointractable obsessive-compulsive disorder. CNS Spectr 5(11):23–34

Rauch SL, Dougherty DD, Malone D, Rezai A, Friehs G, Fischman AJ, Alpert NM, Haber SN,Stypulkowski PH, Rise MT et al (2006) A functional neuroimaging investigation of deepbrain stimulation in patients with obsessive-compulsive disorder. J Neurosurg 104(4):558–565

10 Depression in Humans 101

Remijnse PL, Nielen MM, van Balkom AJ, Cath DC, van Oppen P, Uylings HB, Veltman DJ(2006) Reduced orbitofrontal-striatal activity on a reversal learning task in obsessive-compulsive disorder. Arch Gen Psychiatry 63(11):1225–1236

Ruck C, Andreewitch S, Flyckt K, Edman G, Nyman H, Meyerson BA, Lippitz BE, Hindmarsh T,Syanborg P, Mindus P et al (2003) Capsulotomy for refractory anxiety disorders: long-termfollow-up of 26 patients. Am J Psychiatry 160(3):513–521

Tasker RR, Munz M, Junn FS, Kiss ZH, Davis K, Dostrovsky JO, Lozano AM (1997) Deep brainstimulation and thalamotomy for tremor compared. Acta Neurochir Suppl 68:49–53

102 M. Pandya et al.

Chapter 11Deep Brain Stimulation in Animal Modelsof Depression

Brian W. Scott, José N. Nobrega and Clement Hamani

11.1 Introduction

Depression is a major and disabling disorder that has a 6-month prevalence ofabout 5 % (Depression Guideline Panel 1993a) and entails enormous economicand social costs to society. First-line treatments include medication as well aspsychotherapy, with 60–70 % of patients responding well to these modalities. Thismeans, however, that 30–40 % of patients are refractory to treatment, necessitatingan alternative strategy, such as other classes of medications, electroconvulsivetherapy, and/or augmentative regimens designed to enhance the therapeutic effectof conventional antidepressant medications (Depression Guideline Panel 1993b;Guze and Robins 1970). In patients who continue to be unresponsive to treatment,deep brain stimulation (DBS) has shown promising results in investigationalstudies (Bewernick et al. 2010; Lozano et al. 2008; Malone et al. 2009). Theprocedure involves the delivery of electrical current to a specific brain site throughchronically implanted electrodes. A major advantage of DBS over stereotaxiclesions is that potential side effects of stimulation can usually be managed byaltering the stimulation parameters (e.g., current amplitude) or by discontinuingtreatment. One advantage of DBS over noninvasive stimulation procedures is theability to target small, anatomically defined brain areas. The stimulation targetsinvestigated to date include the subcallosal cingulate gyrus (Lozano et al. 2008;Mayberg et al. 2005), the inferior thalamic peduncle (Jimenez F et al. 2005), the

B. W. Scott � J. N. Nobrega � C. Hamani (&)Neuroimaging Research Section, Centre for Addiction and Mental Health,250 College Street, Toronto, ON M5T 1R8, Canadae-mail: clement.hamani@uhn.on.ca

C. HamaniDivision of Neurosurgery, Toronto Western Hospital, 399 Bathurst Street,Toronto, ON M5T 2S8, Canada

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_11, � Springer-Verlag Berlin Heidelberg 2012

103

nucleus accumbens (NAc) (Bewernick et al. 2010; Schlaepfer et al. 2008), theanterior limb of the internal capsule (Malone et al. 2009), the and lateral habenula(Sartorius et al. 2010).

Most of the research examining DBS has been conducted in clinical studieswith depressed patients. Few studies have attempted to examine the antidepres-sant-like mechanisms of DBS in animal models. In this chapter, we will reviewanimal data examining the use of DBS in experimental models of depression.

11.2 DBS in Animal Models

Important considerations when designing DBS studies using animal modelsinclude the choice of the target, stimulation parameters, and animal model.Rodents (rats and mice) are often used because of the wealth of accumulatedknowledge on their brain anatomy, physiological processes, and behavior.

11.2.1 DBS Targets

Structures within the orbital and medial prefrontal networks or the fiber pathwaysconnecting these structures are typical targets for DBS treatment in depression(Ongur and Price 2000; Price and Drevets 2010). The rationale for targeting thesestructures is based on knowledge gained from brain imaging and clinical reports ofbrain lesions, as well as on current understanding of the involvement of specificneurotransmitter systems in depression (Malone et al. 2009; Mayberg et al. 2005;Jimenez F et al. 2005; Schlaepfer et al. 2008; Sartorius et al. 2010; Hamani andNobrega 2010).

Although the selection of DBS targets is partially based on current theories ofdepression mechanisms in humans, transferring these theories to animal models inorder to select target structures might not be straightforward. This is mainly due todifferences in brain anatomy between species. For example, the anatomy of theprefrontal cortex differs considerably across species, and the structural correspon-dence between humans and rodents is somewhat controversial (Heidbreder andGroenewegen 2003; Uylings et al. 2003). However, on the basis of anatomicalconnections and cytoarchitectural features, the ventral aspect of the medial prefrontalcortex (vmPFC; including the infralimbic cortex—ILC—and the ventral prelimbiccortex—vPLC) has been commonly suggested as the anatomical correlate of thehuman subgenual cingulum (Gabbott et al. 2003; Takagishi and Chiba 1991).

The NAc is composed of two main regions, the core and the shell. These aresignificantly distinct from each other on the basis of morphological, neurochem-ical, and anatomical projections. Whereas the NAc core resembles the dorsalstriatum, the shell is more closely associated with the extended amygdala. Ascompared with the core, the shell shows weaker calbindin staining but is richer in

104 B. W. Scott et al.

mu opioid receptors, as well as in dopamine receptors (Basar et al. 2010; Heimeret al. 1997). Human DBS studies have mainly targeted the NAc core.

In rodents, the main fiber pathways of frontal regions include the forceps minorand the anterior commissure. In rodents the anterior limb of the internal capsuleand the inferior thalamic peduncle are not developed.

The lateral habenula is an important structure in the circuitry of reward andreinforcement. It sends extensive projections to the amygdala and the ventraltegmental area (VTA).

11.2.2 Stimulation Settings

Preclinical studies using DBS in animal models of depression have been used tostudy the mechanisms involved in the antidepressant effects of this therapy as wellas to characterize the optimal settings to induce a behavioral response. As in theclinical scenario, the effects of changing three main parameters—frequency, pulsewidth, and current amplitude—have been studied in greater detail.

By altering the stimulation frequency, one may influence the neural elementsrecruited by stimulation as well as the outcome. Overall, all neural elements canfollow stimulation patterns in a time-locked manner during short pulses of cath-odal low-frequency stimulation. However, at high frequencies (e.g., above100 Hz) DBS exerts a complex pool of effects ranging from functionally inhibitingneuronal populations to exciting axonal pathways in the vicinity of the electrodes(Vitek 2002). DBS has been shown to modulate activity in brain regions distantfrom the stimulation site either through this latter mechanism or through long-termcompensatory changes.

To determine the appropriate current amplitude, a few approaches have beenproposed. One strategy involves determining the threshold for the appearance ofside effects in each individual rat and then using a slightly lower intensity duringtreatment. When no side effects are noticed, one might consider delivering currentat levels that would generate a charge density approximating that used in humans.Charge density reflects the product of current amplitude and pulse duration(amperes 9 seconds = coulombs) delivered in each pulse through the exposedsurface of the electrodes. The pulse widths commonly used in animal research aresimilar to those used in clinical practice (i.e., 60–210 ls). Although calculating thecharge density in order to select a current is not ideal, it is useful for avoidingamplitudes that would generate high charge densities and the delivery of doses thatcould be significantly higher than the ones used in humans.

Also of importance in translational studies between species is the cytoarchi-tecture of target structures, the proportion of gray and white matter at the stimu-lation site, and the distance of the neural elements from the electrode. Theorientation of the neural elements at the site with respect to the electrode can alsoaffect how those elements are activated (Ranck 1975).

11 Deep Brain Stimulation 105

11.2.3 Behavioral Testing

With respect to the choice of behavioral tests to asses the antidepressant effects ofDBS, no animal model adequately mimics all aspects of depressive states in humans.The tests commonly used to investigate the effects of DBS in rodents are consideredsuitable to measure antidepressant-like and/or antianhedonic-like behavior.

The forced swim test (FST) has been pharmacologically validated to a con-siderable degree and has been widely used to screen for antidepressant activity of avariety of interventions (Cryan et al. 2005; Detke et al. 1995; Krahl et al. 2004; Liet al. 2007; Porsolt et al. 1977; Temel et al. 2007). The application of the FST inrats typically involves two sessions conducted on consecutive days or a few daysapart. In the first session, the rat is placed for 15 min in a transparent cylinder filledwith water (the hind limbs and tail cannot touch the bottom of the container). Onthe second day, the rat is again placed in the water for a 5-min swimming session,during which the following behaviors are scored: (1) immobility—minimal fore-paw movement with occasional movements of the hind limbs and tail to keep theanimal afloat; (2) swimming—movements mainly in the horizontal plane; (3)climbing—movements mainly along the walls of the cylinder as the rat attempts toclimb up them (Detke et al. 1995).

Treatments are normally given to the rats between the two sessions and an anti-depressant-like effect is inferred when a decrease in immobility scores is observed.Unlike the clinical effect of antidepressant medications in patients, which can takeweeks to emerge, an antidepressant-like response in the FST is seen within 1 day.This makes the test unsuitable for the study of long-term mechanisms of antide-pressant treatments. However, the FST has been shown to have strong predictivevalidity, in that most clinically effective treatments for depression are detected bythis test. Because of this, and because of its relative simplicity, the FST has becomethe most commonly used screener for antidepressant activity.

Another commonly used model is learned helplessness (LH). As typically used,animals undergoing an LH experiment are first exposed to a single inescapablestress (e.g., footshocks). This is followed by sessions in which the animal canescape from or avoid the stress. Rodents exposed to inescapable stress have sig-nificant deficits in learning an escape response as compared with nonstressedcontrols.

The effects of DBS have also been investigated using paradigms to studyhedonic states. One of these is the so-called chronic mild stress, also known aschronic unpredictable stress (CUS). During CUS, rodents are subjected to a seriesof unpredictable stressors over a period of weeks. A decline in their naturalpreference for sucrose or other sweet solutions over time is thought to reflect ananhedonic-like behavior (Banasr et al. 2007; Willner 2005; Willner et al. 1987). Asin the clinical scenario, antidepressant medications are effective in animalsundergoing CUS only when they are administered over a long period. As CUSextends over a period of weeks, chronic mechanisms involved in a behavioralresponse may be explored.

106 B. W. Scott et al.

11.2.4 Deep Brain Stimulation: Outcome in ExperimentalPreparations

In our laboratory, we have mainly focused on studying the effects of vmPFCstimulation. Overall, we have found that DBS in this region resulted in a 45 %decrease in immobility scores in the FST when compared with control subjects(Fig. 11.1) (Hamani et al. 2010a). This response was similar in magnitude to thatseen with the antidepressant imipramine. As mentioned already, possible mecha-nisms for the effects of high-frequency stimulation may be related to the functionalinactivation of local neurons and the modulation of structures distal to the targetsite via the stimulation of fiber pathways near the electrode (Vitek 2002). Theeffects of vmPFC inactivation in the FST have been investigated by chemical or

(b)

(c)

(a)

A

5-H

T (

% b

asel

ine)

DBS

time (min)

0

10

20

30

40

50

60

Immobility Swimming ClimbingImmobility Swimming Climbing

Beh

avio

ural

cou

nts

Control + Veh (8)

DBS + Veh (6)

Control + 5,7-DHT (12)

DBS + 5,7-DHT (12)

*

*

Fig. 11.1 Outcome of ventral medial prefrontal cortex (vmPFC) deep brain stimulation (DBS) inthe forced swim test (FST). a During behavioral experiments, rats were treated with vmPFC DBSat 130 Hz, 90 ls, and 100 lA. For scoring, the predominant behavior (immobility, swimming, orclimbing) during the 5 min of the FST was recorded every 5 s (maximal score of 60). Animalsreceiving vmPFC DBS had a significant reduction in immobility, the hallmark of anantidepressant-like response, as compared with controls (p = 0.006). b Antidepressant-likeeffects of vmPFC DBS were not observed in animals given raphe microinjections of 5,7-dihydroxytryptamine (5,7-DHT), a toxin for serotonergic neurons. Decrease in immobility scoresonly occurred in animals treated with DBS and raphe vehicle injections (p = 0.02 as comparedwith controls). c DBS of the vmPFC for 1 h (horizontal bar) was associated with a fourfoldincrease in the levels of hippocampal serotonin (5-HT) as assessed with microdialysis. A return tobaseline levels was only observed 150 min after stimulation had been discontinued. In a and b thenumbers in parentheses represent the number of animals per group. Veh ascorbic acid vehicleinjections. (Reproduced from Hamani et al. 2010 with permission of Elsevier)

11 Deep Brain Stimulation 107

radiofrequency lesions. Focal injection of the GABAA agonist muscimol tran-siently inhibits local cells, whereas radiofrequency lesions completely destroycells and disrupt the local cytoarchitecture of the region. Although both treatmentsinduce antidepressant-like effects to some extent (Hamani et al. 2010a; Slatteryet al. 2010), the magnitude of these effects is less pronounced than that observedafter vmPFC DBS. Thus, it appears that mechanisms other than, or in addition to, alocal target inactivation may be important for the antidepressant-like effects ofDBS in the vmPFC.

Ibotenic acid (IBO) is a well-known toxin that primarily injures neuronal cellbodies while leaving passing fibers somewhat intact. Animals with IBO vmPFClesions do not seem not have an antidepressant-like effect in the FST (Hamaniet al. 2010a; Banasr et al. 2010). By contrast, rats with IBO vmPFC lesions andreceiving DBS in the same target showed an antidepressant-like response com-parable to that caused by DBS alone (Hamani et al. 2010a). These data suggest thatmodulation of fiber pathways near the stimulation site may have a primary role inthe effect of vmPFC DBS in the FST. It is interesting to note, however, that IBOlesions also spare local glial cells, which have also been implicated in themechanisms of depression and antidepressant treatments (Banasr et al. 2010;Banasr and Duman 2008).

Following our initial findings, we decided to explore stimulation targets withinthe medial prefrontal cortex for the antidepressant-like response caused by DBS inthe FST. Electrodes were implanted in different groups of animals either in theprelimbic cortex or the in ILC. Whereas stimulation of the former region wasassociated with a positive outcome, animals treated with ILC DBS only had a trendtowards reduction in immobility scores (Hamani et al. 2010b). In addition to thestimulation site, we found that the effectiveness of DBS may be a function of thestimulation current and frequency. In the FST, the current range for an antide-pressant-like effect in rats approximated that used in humans (e.g., equivalentcharge density). In addition, increasing the current amplitude beyond a certainthreshold was associated with a decreased antidepressant-like response. Increasingthe current amplitude is a common strategy in the attempt to improve patientoutcome in clinical settings. Our results suggest that this strategy may not nec-essarily be the best approach. When we varied the frequency of stimulationdelivered to rats, we found that 130-Hz stimulation was more effective than 20 Hzin improving immobility scores in the FST (Hamani et al. 2010b). This result isconsistent with the results of reports on the use DBS in applications such asmovement disorders, pain, and epilepsy, where high-frequency stimulation hasalso been found to be more effective than low-frequency stimulation (Hamani andNobrega 2010). In clinical practice, DBS is typically applied bilaterally. We havefound, however, that unilateral stimulation of the left vmPFC was as effective asbilateral stimulation, whereas unilateral stimulation of the right vmPFC wasineffective (Hamani et al. 2010b). If unilateral stimulation in patients is effective,then the implantation of a single electrode may suffice, possibly reducing thelikelihood of complications associated with surgery.

108 B. W. Scott et al.

Antidepressant-like and antianhedonic-like effects of DBS have also beenexplored in other targets. In an initial study, Friedman et al. (2009) stimulated theVTA in naïve and Flinders rats (prone to depressive-like behavior). At settingsfashioned to mimic the neuronal firing pattern of the nucleus (Friedman et al.2009), DBS administered 20 min prior to behavioral testing induced an antide-pressant-like response, particularly in Flinders rats, in several different paradigms,including the FST, novelty exploration, the social interaction test, and sucroseconsumption (Friedman et al. 2009). More recently, electrodes were implanted inthe vPLC or the dorsal prelimbic cortex as well as the NAc of rats undergoing CUS(Gersner et al. 2010). Stimulation for 10 min/day, 5-s pulses at 20 Hz with 20-spauses between trains for 10 days in either the vPLC or the NAc, but not the dorsalprelimbic cortex, induced an antidepressant-like/antianhedonic-like effect (Gersneret al. 2010). Recently, Friedman et al. (2010, 2011) have studied the effects of LHstimulation in models of drug addiction and sucrose self-administration. In suchstudies, LH stimulation did lead to an antidepressant-like effect in the FST andinduced a decrease in sucrose self-administration (Friedman et al. 2010, 2011).

11.2.5 Neurochemical Substrates of DBS Effects

Selective serotonin (5-HT) and norepinephrine reuptake inhibitors are effectiveantidepressants in clinical practice and in animal models of depression (Cryanet al. 2005; Porsolt et al. 1978). Projections from the vmPFC to raphe nuclei andthe locus coeruleus modulate activity in these structures (Takagishi and Chiba1991; Gabbott et al. 2005; Vertes 1991; Jodo et al. 1998) and influence 5-HTrelease in multiple brain regions (Segal et al. 2007; Juckel et al. 1999).

Studies from our laboratory have found that the antidepressant-like effects ofvmPFC DBS in the FST could be completely abolished by 5-HT- but not by nor-epinephrine-depleting lesions (Fig. 11.1) (Hamani et al. 2010a). Further, we andothers found that electrical stimulation of the vmPFC induced a significant increasein 5-HT release in different brain regions (Fig. 11.1) (Hamani et al. 2010a; Juckelet al. 1999). Exactly how vmPFC stimulation can influence 5-HT release is unclear.However, it may involve the modulation of projections from the prefrontal cortex toraphe nuclei, which are involved in 5-HT synthesis and release. This hypothesis issupported by the fact that a strong antidepressant-like response in the FST is obtainedby stimulation of prefrontal regions with a high density of neurons projecting to theraphe nuclei (Hamani et al. 2010b, 2011; Gabbott et al. 2005). Direct evidence ofvmPFC modulation of raphe 5-HT release has yet to be obtained.

Brain-derived neurotrophic factor (BDNF) has also been suggested to play animportant role in depression and antidepressant treatments (Friedman et al. 2009;Gersner et al. 2010). Patients with depression show a reduction in the levels ofBDNF, as do rodents undergoing chronic stress (Shimizu et al. 2003; Smith et al.1995; Nibuya et al. 1995). Antidepressant treatments have been shown to upreg-ulate BDNF in several brain regions (Nibuya et al. 1995; Altar et al. 2003). DBS in

11 Deep Brain Stimulation 109

the vPLC, NAc, and VTA can increase BDNF levels in rodents undergoingchronic mild stress as well as in Flinders rats (Friedman et al. 2009; Gersner et al.2010). In addition, DBS in the vPLC has been shown to result in a positivecorrelation between BDNF levels and sucrose consumption in rodents (Friedmanet al. 2009; Gersner et al. 2010).

11.3 Summary

Clinical trials using DBS in several brain targets have shown promising results inpatients with treatment-resistant depression. The mechanisms for these effectsremain unknown, as do the optimal structures and stimulation parameters for themost effective treatment outcome. Animal models of DBS allow a somewhat rapidand detailed investigation of the mechanism and clinically relevant treatmentparadigms, which may then be used to help design treatment strategies.

Acknowledgments and Conflicts of Interest Experimental work conducted by the authors hasbeen supported in part by funds from the Brain & Behavior Research Foundation (NARSAD), theOntario Mental Health Foundation, and the Canadian Institutes for Health Research. C.H. is aconsultant to St. Jude Medical.

References

Altar CA, Whitehead RE, Chen R, Wortwein G, Madsen TM (2003) Effects of electroconvulsiveseizures and antidepressant drugs on brain-derived neurotrophic factor protein in rat brain.Biol Psychiatry 54:703–709

Banasr M, Duman RS (2008) Glial loss in the prefrontal cortex is sufficient to induce depressive-like behaviors. Biol Psychiatry 64:863–870

Banasr M, Valentine GW, Li XY, Gourley SL, Taylor JR, Duman RS (2007) Chronicunpredictable stress decreases cell proliferation in the cerebral cortex of the adult rat. BiolPsychiatry 62:496–504

Banasr M, Chowdhury GM, Terwilliger R, Newton SS, Duman RS, Behar KL et al (2010) Glialpathology in an animal model of depression: reversal of stress-induced cellular, metabolic andbehavioral deficits by the glutamate-modulating drug riluzole. Mol Psychiatry 15:501–511

Basar K, Sesia T, Groenewegen H, Steinbusch HW, Visser-Vandewalle V, Temel Y (2010)Nucleus accumbens and impulsivity. Prog Neurobiol 92:533–557

Bewernick BH, Hurlemann R, Matusch A, Kayser S, Grubert C, Hadrysiewicz B et al (2010)Nucleus accumbens deep brain stimulation decreases ratings of depression and anxiety intreatment-resistant depression. Biol Psychiatry 67:110–116

Cryan JF, Valentino RJ, Lucki I (2005) Assessing substrates underlying the behavioral effects ofantidepressants using the modified rat forced swimming test. Neurosci Biobehav Rev 29:547–569

Depression Guideline Panel (1993a) Depression in primary care: vol 1 detection and diagnosis(Clinical Guideline No 5, AHCPR Publication No 93-0550). US: Department of Health andHuman Services, Public Health Service, Agency for Health Care Policy and Research, Rockville

Depression Guideline Panel (1993b) Depression in primary care: vol 2 treatment of majordepression (Clinical Guideline No 5, AHCPR Publication No 93-0551). US: Department ofHealth and Human Services, Public Health Service, Agency for Health Care Policy andResearch, Rockville

110 B. W. Scott et al.

Detke MJ, Rickels M, Lucki I (1995) Active behaviors in the rat forced swimming testdifferentially produced by serotonergic and noradrenergic antidepressants. Psychopharma-cology (Berl) 121:66–72

Friedman A, Frankel M, Flaumenhaft Y, Merenlender A, Pinhasov A, Feder Y et al (2009)Programmed acute electrical stimulation of ventral tegmental area alleviates depressive-likebehavior. Neuropsychopharmacology 34:1057–1066

Friedman A, Lax E, Dikshtein Y, Abraham L, Flaumenhaft Y, Sudai E et al (2010) Electricalstimulation of the lateral habenula produces enduring inhibitory effect on cocaine seekingbehavior. Neuropharmacology 59:452–459

Friedman A, Lax E, Dikshtein Y, Abraham L, Flaumenhaft Y, Sudai E et al (2011) Electricalstimulation of the lateral habenula produces an inhibitory effect on sucrose self-administra-tion. Neuropharmacology 60:381–387

Gabbott PL, Warner TA, Jays PR, Bacon SJ (2003) Areal and synaptic interconnectivity ofprelimbic (area 32), infralimbic (area 25) and insular cortices in the rat. Brain Res 993:59–71

Gabbott PL, Warner TA, Jays PR, Salway P, Busby SJ (2005) Prefrontal cortex in the rat:projections to subcortical autonomic, motor, and limbic centers. J Comp Neurol 492:145–177

Gersner R, Toth E, Isserles M, Zangen A (2010) Site-specific antidepressant effects of repeatedsubconvulsive electrical stimulation: potential role of brain-derived neurotrophic factor. BiolPsychiatry 67:125–132

Guze SB, Robins E (1970) Suicide and primary affective disorders. Br J Psychiatry 117:437–438Hamani C, Nobrega JN (2010) Deep brain stimulation in clinical trials and animal models of

depression. Eur J Neurosci 32:1109–1117Hamani C, Diwan M, Macedo CE, Brandao ML, Shumake J, Gonzalez-Lima F et al (2010a)

Antidepressant-like effects of medial prefrontal cortex deep brain stimulation in rats. BiolPsychiatry 67:117–124

Hamani C, Diwan M, Isabella S, Lozano AM, Nobrega JN (2010b) Effects of differentstimulation parameters on the antidepressant-like response of medial prefrontal cortex deepbrain stimulation in rats. J Psychiatr Res 44:683–687

Hamani C, Diwan M, Raymond R, Nobrega JN, Macedo CE, Brandao ML et al (2011) Electricalbrain stimulation in depression: which target(s)? Biol Psychiatry 69:e7–e8

Heidbreder CA, Groenewegen HJ (2003) The medial prefrontal cortex in the rat: evidence for adorso-ventral distinction based upon functional and anatomical characteristics. NeurosciBiobehav Rev 27:555–579

Heimer L, Alheid GF, de Olmos JS, Groenewegen HJ, Haber SN, Harlan RE et al (1997) Theaccumbens: beyond the core-shell dichotomy. J Neuropsychiatry Clin Neurosci 9:354–381

Jimenez F, Velasco F, Salin-Pascual R, Hernandez JA, Velasco M, Criales JL et al (2005) Apatient with a resistant major depression disorder treated with deep brain stimulation in theinferior thalamic peduncle. Neurosurgery 57:585–593; discussion 585–593

Jodo E, Chiang C, Aston-Jones G (1998) Potent excitatory influence of prefrontal cortex activityon noradrenergic locus coeruleus neurons. Neuroscience 83:63–79

Juckel G, Mendlin A, Jacobs BL (1999) Electrical stimulation of rat medial prefrontal cortexenhances forebrain serotonin output: implications for electroconvulsive therapy andtranscranial magnetic stimulation in depression. Neuropsychopharmacology 21:391–398

Krahl SE, Senanayake SS, Pekary AE, Sattin A (2004) Vagus nerve stimulation (VNS) iseffective in a rat model of antidepressant action. J Psychiatr Res 38:237–240

Li B, Suemaru K, Cui R, Araki H (2007) Repeated electroconvulsive stimuli have long-lastingeffects on hippocampal BDNF and decrease immobility time in the rat forced swim test. LifeSci 80:1539–1543

Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH (2008)Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. BiolPsychiatry 64:461–467

Malone DA Jr, Dougherty DD, Rezai AR, Carpenter LL, Friehs GM, Eskandar EN et al (2009)Deep brain stimulation of the ventral capsule/ventral striatum for treatment-resistantdepression. Biol Psychiatry 65:267–275

11 Deep Brain Stimulation 111

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C et al (2005) Deepbrain stimulation for treatment-resistant depression. Neuron 45:651–660

Nibuya M, Morinobu S, Duman RS (1995) Regulation of BDNF and trkB mRNA in rat brain bychronic electroconvulsive seizure and antidepressant drug treatments. J Neurosci 15:7539–7547

Ongur D, Price JL (2000) The organization of networks within the orbital and medial prefrontalcortex of rats, monkeys and humans. Cereb Cortex 10:206–219

Porsolt RD, Le Pichon M, Jalfre M (1977) Depression: a new animal model sensitive toantidepressant treatments. Nature 266:730–732

Porsolt RD, Anton G, Blavet N, Jalfre M (1978) Behavioural despair in rats: a new modelsensitive to antidepressant treatments. Eur J Pharmacol 47:379–391

Price JL, Drevets WC (2010) Neurocircuitry of mood disorders. Neuropsychopharmacology35:192–216

Ranck JB Jr (1975) Which elements are excited in electrical stimulation of mammalian centralnervous system: a review. Brain Res 98:417–440

Sartorius A, Kiening KL, Kirsch P, von Gall CC, Haberkorn U, Unterberg AW et al (2010)Remission of major depression under deep brain stimulation of the lateral habenula in atherapy-refractory patient. Biol Psychiatry 67:e9–e11

Schlaepfer TE, Cohen MX, Frick C, Kosel M, Brodesser D, Axmacher N et al (2008) Deep brainstimulation to reward circuitry alleviates anhedonia in refractory major depression.Neuropsychopharmacology 33:368–377

Segal S, Tetens J, Kegeles L, Castrillon J, Steinfeld S, Krueger K et al (2007) The effects of localhigh frequency electrical stimulation on monoamine efflux in the subgenual cingulate cortex(Brodmann Area 25) and its striatal and thalamic projection regions. Program No 26722/W1,Society for Neuroscience, San Diego

Shimizu E, Hashimoto K, Okamura N, Koike K, Komatsu N, Kumakiri C et al (2003) Alterationsof serum levels of brain-derived neurotrophic factor (BDNF) in depressed patients with orwithout antidepressants. Biol Psychiatry 54:70–75

Slattery DA, Neumann I, Cryan JF (2010) Transient inactivation of the infralimbic cortex inducesantidepressant-like effects in the rat. J Psychopharmacol 25:1295–1303

Smith MA, Makino S, Kvetnansky R, Post RM (1995) Stress and glucocorticoids affect theexpression of brain-derived neurotrophic factor and neurotrophin-3 mRNAs in thehippocampus. J Neurosci 15:1768–1777

Takagishi M, Chiba T (1991) Efferent projections of the infralimbic (area 25) region of themedial prefrontal cortex in the rat: an anterograde tracer PHA-L study. Brain Res 566:26–39

Temel Y, Boothman LJ, Blokland A, Magill PJ, Steinbusch HW, Visser-Vandewalle V et al (2007)Inhibition of 5-HT neuron activity and induction of depressive-like behavior by high-frequencystimulation of the subthalamic nucleus. Proc Natl Acad Sci U S A 104:17087–17092

Uylings HB, Groenewegen HJ, Kolb B (2003) Do rats have a prefrontal cortex? Behav Brain Res146:3–17

Vertes RP (1991) A PHA-L analysis of ascending projections of the dorsal raphe nucleus in therat. J Comp Neurol 313:643–668

Vitek JL (2002) Mechanisms of deep brain stimulation: excitation or inhibition. Mov Disord17(Suppl 3):S69–S72

Willner P (2005) Chronic mild stress (CMS) revisited: consistency and behavioural-neurobio-logical concordance in the effects of CMS. Neuropsychobiology 52:90–110

Willner P, Towell A, Sampson D, Sophokleous S, Muscat R (1987) Reduction of sucrosepreference by chronic unpredictable mild stress, and its restoration by a tricyclicantidepressant. Psychopharmacology (Berl) 93:358–364

112 B. W. Scott et al.

Chapter 12Deep Brain Stimulation in TouretteSyndrome

L. Ackermans, I. Neuner, J. Kuhn and V. Visser-Vandewalle

12.1 Introduction

Tourette syndrome (TS) is a complex chronic neuropsychiatric disorder character-ized by motor and vocal tics. Motor tics are sudden, repetitive, stereotyped move-ments such as eye blinking, facial twitching, and head or shoulder movements,whereas vocal or phonic tics are sounds produced by air moving through the nose,mouth, or throat (e.g. coughing and throat clearing) as well as repeating syllables,words, or phrases (Mink 2001). TS typically has an onset in early childhood, andboys are more commonly affected than girls. Symptoms usually start with transientbouts of simple motor tics. Tics can become more ‘‘complex’’ in nature and appear tobe purposeful. A fleeting feeling of relief often follows the performance of a tic or aseries of tics (Leckman et al. 1993; Woods et al. 2005) Tics typically follow awaxing and waning pattern of severity, intensity, and frequency (Leckman 2002).Tic severity usually peaks between 8 and 12 years of age, with many patientsshowing a marked reduction in severity by the end of adolescence (Leckman et al.1998; Coffey et al. 2004; Bloch et al. 2006) Approximately 20 % of children with TScontinue to experience a moderate level of impairment of global functioning by the

L. Ackermans � V. Visser-Vandewalle (&)Department of Stereotactic and Functional Neurosurgery,University of Cologne, Cologne, Germanye-mail: veerle.visser-vandewalle@uk-koeln.de

L. AckermansMaastricht Institute for Neuromodulative Development (MIND), Maastricht,The Netherlands

I. NeunerDepartment of Psychiatry and Psychotherapy, RWTH Aachen University, Aachen, Germany

J. KuhnDepartment of Psychiatry and Psychotherapy, University of Cologne, Cologne, Germany

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_12, � Springer-Verlag Berlin Heidelberg 2012

113

age of 20 years (Bloch et al. 2006). TS alone is the exception rather than the rule.Attention deficit–hyperactivity disorder and obsessive–compulsive behaviour arethe commonest comorbidities. The presence of these comorbidities can add anotherlayer of complexity, which may make it more difficult to develop a treatment planthat addresses not only the tics, but also the co-occurring disorders.

TS might be interpreted as an overactive abnormal neural activity of both thesensorimotor and the limbic circuits, involving multiple outputs, of the basalganglia (Babel et al. 2001).

12.2 Treatment of TS

Frequently, TS is found to be a self-limiting disorder, whereas in a small proportionof patients the tics continue into adult life and require long-term behavioural or drugtreatment. Behavioural and drug treatment may provide temporary control ofsymptoms but certain patients are medically untreatable or experience unbearableside effects from the medication. For these patients, surgery may be an option.

Various attempts have been made to treat patients with TS through neurosur-gical ablative procedures (Temel and Visser-Vandewalle 2004). The target siteshave been diverse, including the frontal lobe (prefrontal lobotomy and bimedialfrontal leucotomy), the limbic system (limbic leucotomy and anterior cingulot-omy), the thalamus, and the cerebellum. The results have often been unsatisfactoryor major side effects have occurred, such as hemiplegia and dystonia. Hassler andDieckmann reported on the beneficial effects of lesioning the intralaminar andmidline thalamic nuclei in patients with TS, and, in patients with facial tics, alsothe nucleus ventralis oralis internus (VOI) (Hassler 1970).

DBS has been introduced in the field of neuropsychiatry to modulate neuronalactivity in the same areas as those targeted for lesioning in the past, but in areversible way.

In 1999, deep brain stimulation (DBS) was introduced as a new surgicaltechnique in the treatment of intractable TS (Vandewalle et al. 1999). Vandewalleet al. (1999) performed chronic bilateral stimulation of the medial part of thethalamus, at the cross point of the centromedian nucleus (CM), substantia peri-ventricularis (SPV), and the VOI. This target was chosen on the basis of the goodresults of thalamotomies described by Hassler and Dieckmann in 1970 (Hassler1970). Since this first report, about 70 patients who have had DBS for treatment ofTS have been reported. Nine different targets have been described, and can bedivided in four brain areas: (1) the medial part of the thalamus, (2) the globuspallidus internus (GPi), (3) the globus pallidus externus (GPe), (4) the internalcapsule (IC)/nucleus accumbens (NAc), and (5) the subthalamic nucleus. Theanatomical locations and corresponding studies between 1999 and 2011 (Van-dewalle et al. 1999; Visser-Vandewalle et al. 2003; Ackermans et al. 2010, 2011;Maciunas et al. 2007; Bajwa et al. 2007; Shields et al. 2008; Idris et al. 2010;Servello et al. 2008; Vernaleken et al. 2009; Van der Linden et al. 2002; Diederich

114 L. Ackermans et al.

et al. 2004; Gallagher et al. 2006; Shahed et al. 2007; Dehning et al. 2008; Duecket al. 2009; Foltynie et al. 2009; Houeto et al. 2005; Welter et al. 2008; Martínez-Fernández et al. 2011; Vilela Filho et al.2010; Flaherty et al. 2005; Kuhn et al.2007; Zabek et al. 2008; Neuner et al. 2009, 2010; Servello et al. 2009; Burdicket al. 2010; Martinez-Torres et al. 2009; Porta et al. 2009) are illustrated inTable 12.1.

12.3 Targets

12.3.1 Medial Part of the Thalamus

After the promising results of DBS in the first TS patient described by Vandewalleet al. (1999), the same group reported on the beneficial effects of DBS of the sametarget in three patients in 2003 (Visser-Vandewalle et al. 2003). They stated thatstimulation of the VOI leads to diminished motor and vocal tics by inhibitingprojections to the facial parts of the premotor (and motor) cortex. Stimulation ofthe intralaminar nuclei reduces the activity of the dorsal, sensorimotor parts of thestriatum, whereas stimulation of the midline thalamic nuclei reduces activity in theventral, limbic striatum. In total, 32 patients have received thalamic DBS fortreatment of intractable TS, although within the thalamic target there has beensome variety:

1. The Cm/Voi/Spv cross point has been targeted most frequently (Vandewalleet al. 1999; Visser-Vandewalle et al. 2003; Ackermans et al. 2010, 2011;Maciunas et al. 2007; Bajwa et al. 2007; Shields et al. 2008; Idris et al. 2010),with a range of tic reduction between 24 and 90 %. In 1999 (Vandewalle et al.1999) and 2003 (Visser-Vandewalle et al. 2003), Visser-Vandewalle et al.reported results for the first three TS patients. There was a good effect not onlyon tics but also on associated behavioural disorders, such as obsessive–com-pulsive behaviour (OCB) and self-injurious behaviour (SIB). In 2008 the samegroup reported a decrease in tic frequency of 78 and 92.6 % in two patients atlong-term follow up (Ackermans et al. 2010). An average tic reduction of 50 %in three of five patients has been described by Maciunas et al. (Maciunas et al.2007). The secondary outcome measures anxiety, depression, and OCB showeda trend towards improvement. Also a good effect of 66 % on tics and 76 % onOCB in a single case after this thalamic stimulation was reported by Bajwaet al. (2007). Idris et al. (2010) reported one patient with bilateral corticalhaematomas after thalamic DBS, with a short note that complex motor andvocal tics improved. Most recently, Ackermans et al. (2011) reported a double-blind randomized clinical trial of six TS patients with 49 % improvement fortics and no significant difference in associated behaviour.

2. The results for 18 patients ranged between 24 and 79 % as reported by Servelloet el. (2008) with the target being located 2 mm more anterior than the cross

12 Deep Brain Stimulation in Tourette Syndrome 115

Tab

le12

.1A

nato

mic

allo

cati

ons

and

corr

espo

ndin

gst

udie

sbe

twee

n19

99an

d20

11

Tar

get

Aut

hor

Num

ber

Eff

ect

onti

csE

ffec

ton

asso

ciat

edbe

havi

our

Sid

eef

fect

sF

ollo

w-

up (mon

ths)

Tha

lC

m/V

oi/

Spv

Van

dew

alle

(Van

dew

alle

etal

.19

99,

Vis

ser-

Van

dew

alle

etal

.20

03,

2008

)

372

–90

%G

ood

Dro

wsi

ness

,an

dch

ange

sin

sexu

alfu

ncti

onin

g8–

72

Mac

iuna

s(M

aciu

nas

etal

.20

07)

5D

ecre

ase

40%

mot

orti

csT

rend

tow

ards

impr

oved

moo

d,re

duce

dan

xiet

y,P

sych

otic

even

t3

Incr

ease

21%

voca

lti

csA

ndfe

wer

obse

ssio

nsan

dco

mpu

lsio

nsB

ajw

a(B

ajw

aet

al.

2007

)1

66%

76%

inO

CD

Not

men

tion

ed24

Shi

elds

(Shi

elds

etal

.20

08)

141

%N

otm

enti

oned

Mor

efr

eque

nt(1

–2ye

ars)

puls

ege

nera

tor

repl

acem

ents

3

Idri

s(I

dris

etal

.20

10)

1C

ompl

exm

otor

and

voca

lti

csim

prov

ed

Not

men

tion

edB

ilat

eral

intr

acer

ebra

lhe

mat

omas

2

Ack

erm

ans

(Ack

erm

ans

etal

.20

11)

637

%N

osi

gnifi

cnan

tch

ange

Red

uced

ener

gyan

dsu

bjec

tive

ocul

omot

ordi

stur

banc

es12

Tha

lC

m/V

oi/

Spv

+2

Ser

vell

o(S

erve

llo

etal

.20

09)a

1824

–79

%G

ood

effe

ctT

empo

rary

dist

urba

nces

ofoc

ulom

otio

nw

ere

men

tion

ed3–

18

Tha

l dors

omed

nuc

Ver

nale

ken

(Ver

nale

ken

etal

.20

09)

136

%Im

prov

emen

tin

obse

ssiv

eth

ough

tsan

daf

fect

ive

sym

ptom

s

No

clin

ical

impr

ovem

ent

afte

rG

pist

imul

atio

nin

the

past

6

(con

tinu

ed)

116 L. Ackermans et al.

Tab

le12

.1(c

onti

nued

)

Tar

get

Aut

hor

Num

ber

Eff

ect

onti

csE

ffec

ton

asso

ciat

edbe

havi

our

Sid

eef

fect

sF

ollo

w-

up (mon

ths)

GP

ivp

lV

ande

rL

inde

n(V

ande

rL

inde

net

al.

2002

)b

195

%N

otm

enti

oned

Non

e12

Die

deri

ch(D

iede

rich

etal

.20

04)

173

%N

oef

fect

Pro

nati

on/

supi

nati

onbr

adyk

ines

iaof

the

left

extr

emit

ies

14

Gal

lagh

er(G

alla

gher

etal

.20

06)

1D

isap

pear

ance

ofvo

cal

tics

and

mar

ked

impr

ovem

ent

inne

ckm

ovem

ents

Not

men

tion

edR

emov

alle

ftpu

lsge

nera

tor

due

toin

fect

ion,

resu

ltin

gin

ongo

ing

mot

orti

csof

righ

tfa

cean

dar

m

Unk

own

Sha

hed

(Sha

hed

etal

.20

07)

184

%69

%N

otm

enti

oned

6

Deh

ning

(Deh

ning

etal

.20

08)

188

%N

otm

enti

oned

Dep

ress

ive

moo

ds,

bein

gat

trib

uted

todi

fficu

ltie

sin

adju

stin

gto

the

new

situ

atio

n

12

Due

ck(D

ueck

etal

.20

09)

1N

otef

fect

ive

Not

men

tion

edU

nsuc

cesf

ulou

tcom

ein

pati

ent

wit

hm

enta

lre

tard

atio

nM

arti

nez-

Fer

nand

ez(M

artí

nez-

Fer

nánd

ezet

al.

2011

)

337

%N

otm

enti

oned

inab

stra

ctN

otm

enti

oned

inab

stra

ct18

(con

tinu

ed)

12 Deep Brain Stimulation in Tourette Syndrome 117

Tab

le12

.1(c

onti

nued

)

Tar

get

Aut

hor

Num

ber

Eff

ect

onti

csE

ffec

ton

asso

ciat

edbe

havi

our

Sid

eef

fect

sF

ollo

w-

up (mon

ths)

GP

iam

Hou

eto

(Hou

eto

etal

.20

05)

170

%A

boli

shed

SIB

Bot

hth

alam

ican

dpa

llid

alst

imul

atio

nha

da

sim

ilar

effe

cton

tics

,bu

tth

alam

icst

imul

atio

nha

da

mor

ebe

nefi

cial

effe

cton

asso

ciat

edbe

havi

our

dist

urba

nces

24

Wel

ter

(Wel

ter

etal

.20

08)

365

–96

%D

epre

ssiv

em

ood,

anxi

ety

and

impu

lsiv

enes

sim

prov

edw

ith

thal

amic

but

not

wit

hpa

llid

alst

imul

atio

n

Anx

iety

20–6

0

Fol

tyni

e(F

olty

nie

etal

.20

09)

188

%m

otor

and

90%

voca

lN

otm

enti

oned

Let

harg

y,a

sens

eof

inne

rte

nsio

nan

dsu

bjec

tive

lyfe

elin

gof

noim

prov

emen

tin

qual

ity

ofli

fe

6

Mar

tine

z-F

erna

ndez

(Mar

tíne

z-F

erná

ndez

etal

.20

11)

354

%N

otm

enti

oned

inab

stra

ctN

otm

enti

oned

inab

stra

ct18

GP

eV

ilel

aF

ilho

(Vil

ela

Fil

hoet

al.

2010

)7

61–9

6%

Not

men

tion

edA

sym

ptom

atic

peri

elec

trod

eoe

dem

ain

one

pati

ent

and

tran

sien

tde

pres

sion

inan

othe

r

3–71

(con

tinu

ed)

118 L. Ackermans et al.

Tab

le12

.1(c

onti

nued

)

Tar

get

Aut

hor

Num

ber

Eff

ect

onti

csE

ffec

ton

asso

ciat

edbe

havi

our

Sid

eef

fect

sF

ollo

w-

up (mon

ths)

IC/N

AC

Fla

hert

y(F

lahe

rty

etal

.20

05)

125

%S

elf-

inju

ryst

oppe

dIn

func

tion

ofth

eac

tive

pole

chos

enfo

rst

imul

atio

n,hy

pom

ania

orde

pres

sion

coul

dbe

evok

ed

18

Kuh

n(K

uhn

etal

.20

07)

141

%64

%A

utoa

gres

sive

beha

viou

rce

ased

30

Zab

ek(Z

abek

etal

.20

08)

1A

llev

iate

tics

All

evia

teO

CD

Not

men

tion

ed28

Neu

ner

(Neu

ner

etal

.200

9,20

10)

143

%50

%D

isap

pera

nce

ofS

IB18

–38

Ser

vell

o(S

erve

llo

etal

.20

09)

449

–82

%8–

60%

Thr

eeou

toff

our

pati

ents

rece

ived

thal

amic

stim

ulat

ion

inad

vanc

e10

–26

Bur

dick

(Bur

dick

etal

.20

10)

1W

orse

ning

of17

%N

och

ange

Not

men

tion

ed30

ST

NM

arti

nez-

Tor

res

(Mar

tine

z-T

orre

set

al.

2009

)

197

%D

ecre

ased

park

inso

nsy

mpt

oms

Not

men

tion

ed12

Tha

lT

hala

mus

,C

m/V

oi/S

pvce

ntro

med

ian

nucl

eus/

nucl

esve

ntro

-ora

lis

inte

rnus

/sub

stan

tia

peri

vent

ricu

lari

s,G

piG

lobu

spa

llid

usin

tern

us,

vpl

vent

ropo

s-te

rola

tera

l,G

Pe

Glo

bus

pall

idus

exte

rnus

,am

ante

rom

edia

l,IC

/NA

CIn

tern

alca

psul

e/nu

cleu

sac

cum

bens

,ST

NS

ubth

alam

icnu

cleu

sa

15ou

tof

18pa

tien

tsar

eal

sode

scri

bed

byP

orta

etal

.(2

009a

)w

ith

afo

llow

upof

24m

onth

sb

Sam

epa

tien

tis

desc

ribe

dby

Ack

erm

ans

etal

.(2

006)

wit

ha

long

term

foll

owup

12 Deep Brain Stimulation in Tourette Syndrome 119

point described above. Porta et al. (2009) described a 52 % tic reduction after2 years’ follow-up of 15 of these 18 patients.

3. The dorsomedial nucleus of the thalamus as a target for DBS in TS has beendescribed by Vernaleken et al. (2009). In this single case report, there was a36 % improvement of tics.

12.3.2 Globus Pallidus Internus

12.3.2.1 Globus Pallidus Internus (Posteroventrolateral)

Van der Linden et al. (2002) were the first to describe the effects of DBS of theventroposterolateral (vpl) (motor) part of the GPi, in 2002. At 6-month follow-up,a tic reduction of 95 % was noticed. The choice of the pallidal target was based onthe beneficial effects of DBS of the same brain region on hyperkinetic movementsinduced by medication in patients with Parkinson disease. Nine patients havereceived vpl GPi stimulation, with a range of 34–88 % tic reduction (Van derLinden et al. 2002; Diederich et al. 2004; Gallagher et al. 2006; Shahed et al. 2007;Dehning et al. 2008; Dueck et al. 2009; Foltynie et al. 2009). In 2005, Diederichet al. (2004) described the beneficial effects of chronic stimulation of the sametarget, with a follow-up of 14 months. However, there was no change in the ‘‘verymild compulsive tendencies’’. As a complication, a small haematoma at the tip ofthe right electrode was described, resulting in a deficit of alternating pronation/supination movements of the left hand. In 2006 Gallagher et al. (2006) reported ona right-handed man with ongoing motor tics of the right side of the face and theright arm after removal of the left pulse generator because of an infection. In a 16-year-old adolescent there was a reduction of 84 % for tics and 69 % for OCB asdescribed by Shahed et al. (2007). However, the patient had to wear a shield toprotect himself from compulsive harming with the pulse generator. Dehning et al.(2008) also reported on the beneficial effects of vpl GPi DBS, with a decrease of88 % on the Yale Global Tic Severity Scale (YGTSS) (Leckman et al. 1989). Forthe first few months postoperatively, the patient had depressive moods, which wereattributed to difficulties in adjusting to the new situation.

An unsuccessful outcome of vpl GPi DBS was described by Dueck et al. (2009)in a 16-year-old boy with TS and mental retardation.

12.3.2.2 Globus Pallidus Internus (Anteromedial)

Because of the proposed dysfunction of the associative limbic component of thebasal ganglia circuitry in TS, the anteromedial part of the GPi has been consideredto be another potential target. The results of DBS of this target have been describedin several reports (Foltynie et al. 2009; Houeto et al. 2005; Welter et al. 2008;

120 L. Ackermans et al.

Martínez-Fernández et al. 2011). The effect on tic reduction ranged from 54 to90 %.

Both Houeto et al. (2005) and Welter et al. (2008) described the effects ofbilateral anteromedial GPi and thalamic (CM) stimulation in one patient and threepatients, respectively. In all four patients, there was a better effect on associatedbehaviour disturbances after thalamic stimulation in comparison with pallidalstimulation. Houeto et al. reported that both thalamic and pallidal stimulation had asimilar effect on tics. Welter et al. stated that GPi DBS alone led to a tic reductionbetween 65 and 96 % and thalamic DBS alone yielded an improvement of30–64 %.

12.3.3 Globus Pallidus Externus

On the basis of the hypothesis that the GPe is hyperactive in TS, Vilela Filho et al.performed high-frequency stimulation of the GPe in seven patients with TS andevaluated the results in a double-blind prospective controlled study, and found amean tic reduction of 74 % (Vilela Filho et al. 2010).

12.3.4 Internal Capsule/Nucleus Accumbens

The ventral part of the IC, and the ventral striatum nucleus accumbens (NAc), wasthe fourth area targeted for DBS in TS. The rationale lies in the fact that TS andobsessive–compulsive disorder (OCD) share many clinical similarities and show astrong comorbidity. A study with event-related brain potentials indicated thatfrontal inhibitory mechanisms are altered like in TS and OCD (Muller-Vahl et al.2003). DBS of the NAc has been successfully performed in patients with OCD(Sturm et al. 2003).

IC/NAC DBS in TS has been described in several single cases (Flaherty et al.2005; Kuhn et al. 2007; Zabek et al. 2008; Neuner et al. 2009, 2010; Servello et al.2009; Burdick et al. 2010).

In 2005, Flaherty et al. (2005) described a tic reduction of 25 % on the YGTSSafter bilateral stimulation of the IC in one patient. Depending on of the active polechosen for stimulation, hypomania or depression could be evoked in this patient. In2008, after hardware failure, the same patient received thalamic DBS, with a 46 %decrease on the YGTSS, as reported by Shields et al. (2008).

Kuhn et al. (2007) reported on a 26-year-old patient who received DBS of theventral IC/NAC for treatment TS, with SIB and OCD as comorbidities. After30 months’ follow-up, there was a 41 % decrease on the YGTSS and a 64 %decrease on the Yale–Brown Obsessive–Compulsive Scale (Y-BOCS). The effectsof unilateral (right) NAc stimulation were described by Zabek et al. (2008).

12 Deep Brain Stimulation in Tourette Syndrome 121

Neuner et al. (2009) confirmed these results in a 38-year-old male TS patientwith comorbid OCD and SIB over a follow-up period of up to 36 months. TheYGTTS score was reduced by 43 %, and the Y-BOCS score was reduced by 50 %.SIB (self mutilation of the lips, forehead, and fingers coupled with the urge tobreak glass) completely ceased. It is also noteworthy that during active stimulationin this patient a depressive episode resulting in a suicide attempt was not preventedby NAc DBS (Neuner et al. 2010).

Finally, Servello et al. (2009) reported on four TS patients who had had tha-lamic DBS but who additionally underwent IC/NAc DBS because of persistentOCD comorbidity. A fourth patient received only IC/NAc DBS. The results wererather unsatisfactory.

Burdick et al. (2010) described a negative outcome after IC/NAc DBS. Thepatient had TS and OCD and showed a 17 % deterioration on the YGTSS and nochange on the Y-BOCS.

12.3.5 Subthalamic Nucleus

Martinez-Torres et al. (2009) reported on a patient with Parkinson disease and acomorbid tic disorder diagnosed in childhood who received subthalamic nucleusstimulation for his Parkinson disease. Stimulation led to a 97 % reduction of tics(video counting) after 1 year.

12.4 Clinical and Surgical Evaluation

12.4.1 Patient Selection

Careful patient selection is absolutely mandatory for DBS in TS (Rabins et al.2009). The TS patients considered for DBS should comprise only patient with verysevere cases who have already fruitlessly received standard therapies. Publishedguidelines include the following selection criteria.

12.4.1.1 Inclusion of Patients

1. The patient has a definite TS, established by two independent clinicians. Thediagnosis is established according to Diagnostic and Statistical Manual ofMental Disorders, fourth edition, text revision criteria (American PsychiatricAssociation 2000) and with the aid of the diagnostic confidence index(Robertson et al. 1999).

2. The patient has severe and incapacitating tics as the primary problem.

122 L. Ackermans et al.

3. The patient is treatment-resistant. This means that the patient either has not orhas only partially responded to three different medication regimes in adequatedoses over a period of at least 12 weeks, or has not tolerated the medicationbecause of side effects. Four different groups of medications that should havebeen tried are

i. ‘‘Classic’’ dopamine2 antagonists (haloperidol or pimozide)ii. Second generation neuroleptics with more or less proven efficacy (e.g.

risperidone)iii. Second generation neuroleptics without proven efficacy and experimental

character (e.g. quetiapine, aripiprazole)

Finally, a trial of at least ten sessions of behavioural therapy for tics, suchas habit reversal or exposure in vivo, have been attempted and have beenunsuccessful.

iv. Centrally acting a2 adrenergic agonist (clonidine, guanfacine)

Age is very much a subject of debate. There is agreement among all experts thatDBS should be performed only in adult patients. A minimum age between 18 and25 years has been suggested (Visser-Vandewalle et al. 2006; Mink et al. 2006;Müller-Vahl et al. 2011). However, beneficial results in two TS patients youngerthan 18 years have been described by Servello et al. (2008) and Shahed et al.(2007), and a unsuccessful outcome in one patient was described by Dueck et al.(2009). It remains unclear whether severely affected patients at the age of 18 yearsmay for a significant improvement until the age of 25 years (Müller-Vahl et al.2011). Consideration of broader economical and environmental issues, includingsocial support that is available to the patient, is recommended (Kuhn et al. 2009;Cavanna et al. 2011).

12.4.1.2 Exclusion of Patients

Patients are excluded from neurosurgical treatment if they have a tic disorder otherthan TS, severe psychiatric comorbid conditions (other than associated behaviouraldisorders), or a mental deficiency that could impede operative and postoperativerecovery, care, and assessment. Other contraindications for surgical treatment byDBS in TS are severe cardiovascular, pulmonary, or haematological disorders andstructural MRI abnormalities.

12.4.1.3 Surgical Procedure

Surgeons should have substantial experience in DBS treatment to enhance efficacyand minimize complications. The technique of DBS applied to TS is broadlysimilar to the one used for more classic indications. The target for TS, such as thenuclei of the medial part of the thalamus, is mostly invisible with current imaging

12 Deep Brain Stimulation in Tourette Syndrome 123

techniques, so only indirect targeting can be used. Another point of attention is thatTS patients might pull themselves out of the stereotactic frame because of the highratio of motor tics occurring in the head region. One solution is to operate with thepatient under general anaesthesia. Because of the uncertainty of the ideal targetand the importance of intraoperative findings, the patient should be cooperativeduring surgery. Sedating the patient to obtain tic suppression with maintenance ofthe possibility to communicate with the patient is preferable. The patient can besedated with a combination of lormetazepam and clonidine (Visser-Vandewalleet al. 2003), or with a propofol target-controlled infusion (Ackermans et al. 2006),sufficiently reducing the tics and their implications for the stereotactic procedure.At the same time, the patient can be interrogated so that immediate adversestimulation-induced side effects can be detected and the position of the electrodeadapted.

12.4.2 Perioperative Evaluation

It is of paramount importance that for all TS patients treated with DBS the exactlocation of the electrode is precisely determined and all effects are meticulouslydescribed. A more comprehensive survey of guidelines for the perioperativeassessment of the effects of DBS in TS can be found elsewhere (Mink et al. 2006).

12.4.3 Postoperative Evaluation

First, the execution of DBS should be restricted to neurosurgical units experiencedin DBS treatment with established collaborations with neurology and psychiatrydepartments specializing in the diagnosis and treatment of TS. For the assessmentof clinical effects, a description of the effect on tics, on associated behaviouraldisorders, the stimulation-induced side effects, and complications is mandatory.The most commonly used scale for tic rating is the YGTSS (Leckman et al. 1989).For a more objective evaluation, a video recording of the patient with and withoutstimulation should be made. From these recordings, the tics should be rated by twoindependent investigators. Ideally the patient and investigator are blinded to thestatus of the stimulation. A careful psychiatric and neuropsychological evaluationshould be performed at regular intervals. The clinical effects should be correlatedto the exact position of the electrode. The most prudent approach is to perform aCT scan postoperatively, and fuse the images with preoperative MRI images.

Only if these prerequisites are fulfilled and the maximum amount of data areexchanged between centres can the optimal target be established.

124 L. Ackermans et al.

12.4.4 Complications

Three major complications have been described, consisting of two haematomas atthe tip of the electrode, both leading to transient neurological deficits, with achange in rapidly alternating hand movements (Diederich et al. 2004), and avertical gaze palsy (Ackermans et al. 2007). One patient had intracerebral hae-matomas located around both electrodes (Idris et al. 2010).

12.5 Conclusion

In the last 10 years, about 70 patients having received DBS for TS have beenreported, with ten different brain targets, the Cm/Voi/Spv cross point of thethalamus being the first one described (Vandewalle et al. 1999; Visser-Vandewalleet al. 2003; Ackermans et al. 2010, 2011; Maciunas et al. 2007; Bajwa et al. 2007;Shields et al. 2008; Idris et al. 2010). Servello et al. (2008) and Porta et al. (2009)targeted the same area but at a point 2 mm more anteriorly. Houeto et al. (2005)targeted the centre of the CM, and one case was described with DBS of thedorsomedial thalamus (Vernaleken et al. 2009). Besides the thalamus, the GPe(Vilela Filho et al. 2010) and both the ventroposterolateral motor (Van der Lindenet al. 2002; Diederich et al. 2004; Gallagher et al. 2006; Shahed et al. 2007;Dehning et al. 2008; Dueck et al. 2009; Foltynie et al. 2009) and the anteromediallimbic part of the GPi have been targeted by DBS in refractory TS (Houeto et al.2005; Welter et al. 2008; Martínez-Fernández et al. 2011) Also the nucleus ac-cumbens—(Kuhn et al. 2007; Burdick et al. 2010) and the IC (Flaherty et al. 2005;Servello et al. 2009). Burdick et al. (2010) have been targeted, mostly in TSpatients with OCD. Finally, in a patient with both Parkinson disease and tics, therewas an improvement of tics after DBS of the subthalamic nucleus (Martinez-Torres et al. 2009).

Given the many different targets used for DBS in TS, and the small number ofpatients with the intractable syndrome, continuous exchange of clinical experienceand an ongoing evaluation are important. A uniform approach with standardinclusion criteria and outcome measures is warranted to find out which is the optimaltarget, or whether ‘‘tailored’’ targeting is needed, with a specific target for a specificsubtype of patients, as also suggested by Porta et al. (2009).

Given the consequences of TS for social, familial, and professional life, patientshave to deal with many challenges after surgery. Anticipating these postoperativechanges prior to surgery will be helpful in assisting patients and their families benefitfrom tic reduction and in maximizing the overall outcome and success of surgery.

Determination of the optimal surgical target and stimulation parameters willrequire close multicentre collaboration and standardized methods for evaluation.Another question still to be addressed is whether tolerance should play a role.

12 Deep Brain Stimulation in Tourette Syndrome 125

Therefore, a prospective, multicentre double-blind study to evaluate the effects ofDBS in selected TS patients would be the ideal approach.

References

Ackermans L, Temel Y, Cath D, van der Linden C, Bruggeman R, Kleijer M, Nederveen P,Schruers K, Colle H, Tijssen MA, Visser-Vandewalle V (2006) Dutch-Flemish TouretteSurgery Study Group. Deep brain stimulation in Tourette’s syndrome: two targets? MovDisord 21(5):709–13

Ackermans L, Temel Y, Bauer NJC, Visser-Vandewalle V (2007) Vertical gaze palsy afterthalamic stimulation for Tourette syndrome: case report. Neurosurgery 61(5):E1100

Ackermans L, Duits A, Temel Y (2010) Long-term outcome of thalamic deep brain stimulation intwo patients with Tourette syndrome. J Neurol Neurosur Psychiatry 81:1068–1072

Ackermans L, Duits A, van der Linden C, Tijssen M, Schruers K, Temel Y, Kleijer M, NederveenP, Bruggeman R, Tromp S, van Kranen-Mastenbroek V, Kingma H, Cath D, Visser-Vandewalle V (2011) Double-blind clinical trial of thalamic stimulation in patients withTourette syndrome. Brain 134:832–844

American Psychiatric Association (2000) Diagnostic and statistical manual of mental disorders,4th ed (Text revision). American Psychiatric Association, Washington

Babel TB, Warnke PC, Ostertag CB (2001) Immediate and long term outcome after infrathalamicand thalamic lesioning for intractable Tourette’s syndrome. J Neurol Neurosurg Psychiatry70:666–671

Bajwa RJ, de Lotbiniere AJ, King RA, Jabbari B, Quatrano S, Kunze K, Scahill L, Leckman JF(2007) Deep brain stimulation in Tourette’s syndrome. Mov Disord 22:1346–1350

Bloch MH, Peterson BS, Scahill L, Otka J, Katsovich L, Zhang H, Leckman JF (2006) Adulthoodoutcome of tic and obsessive-compulsive symptom severity in children with Tourettesyndrome. Arch Pediatr Adolesc Med 160:65–69

Burdick A, Foote KD, Goodman W, Ward HE, Ricciuti N, Murphy T, Haq I, Okun MS (2010)Lack of benefit of accumbens/capsular deep brain stimulation in a patient with both tics andobsessive-compulsive disorder. Neurocase 2010(16):321–330

Cavanna AE, Eddy CM, Mitchell R, Pall H, Mitchell I, Zrinzo L, Foltynie T, Jahanshahi M,Limousin P, Hariz MI, Rickards H (2011) An approach to deep brain stimulation for severetreatment-refractory Tourette syndrome: the UK perspective. Br J Neurosurg 25:38–44

Coffey BJ, Biederman J, Geller D, Frazier J, Spencer T, Doyle R, Gianini L, Small A, Frisone DF,Magovcevic M, Stein N, Faraone SV (2004) Reexamining tic persistence and tic-associatedimpairment in Tourette’s disorder: findings from a naturalistic follow-up study. J NervousMental Dis 192:776–780

Dehning S, Mehrkens JH, Mueller N, Botzel K (2008) Therapy-refractory Tourette syndrome:beneficial outcome with globus pallidus internus deep brain stimulation. Mov Disord23:1300–1302

Diederich NJ, Bumb A, Mertens E, Kalteis K, Stamenkovic M, Alesch F (2004) Efficient internalsegment pallidal stimulation in Gilles de la Tourette syndrome: a case report. Mov Disord19:S440

Dueck A, Wolters A, Wunsch K, Bohne-Suraj S, Mueller J, Haessler F, Benecke R, Buchmann J(2009) Deep brain stimulation of globus pallidus internus in a 16-year-old boy with severeTourette syndrome and mental retardation. Neuropediatrics 40:239–242

Flaherty AW, Williams ZM, Amimovin R, Kasper E, Rauch SL, Cosgrove SL, Eskander EN(2005) Deep brain stimulation of the internal capsule for the treatment of Tourette syndrome:technical case report. Neurosurgery 57:E40

126 L. Ackermans et al.

Foltynie T, Martinez-Torres I, Zrinzo L, Joyce E, Cavanna A, Jahanshahi M, Limousin P, HarizM (2009) Improvement in vocal & motor tics following DBS of motor GPi for Tourettesyndrome, not accompanied by subjective improvement in quality of life—a case report. MovDisord 24:S497–S498

Gallagher CL, Garell PC, Montgomery EB (2006) Hemitics and deep brain stimulation.Neurology 66:E12

Hassler R, Dieckmann G (1970) Traitement stéréotaxique des tics et cris inarticulés oucoprolaliques considérés comme phénomène d’obsession motrice au cours de la maladie deGilles de la Tourette. Rev Neurol Paris 123:89–100

Houeto JL, Karachi C, Mallet L, Pillon B, Yelnik J, Mesnage V, Welter ML, Navarro S, PelissoloA, Damier P, Pidoux B, Dormont D, Cornu P, Agid Y (2005) Tourette’s syndrome and deepbrain stimulation. J Neurol Neurosurg Psychiatry 76:904

Idris Z, Ghani AR, Mar W, Bhaskar S, Wan Hassan WN, Tharakan J, Abdullah JM, Omar J,Abass S, Hussin S, Abdullah WZ (2010) Intracerebral haematomas after deep brainstimulation surgery in a patient with Tourette syndrome and low factor XIIIA activity. J ClinNeurosci 17:1343–1344

Kuhn J, Lenartz D, Mai JK, Huff W, Lee SH, Koulousakis A, Klosterkoetter J, Sturm V (2007)Deep brain stimulation of the nucleus accumbens and the internal capsule in therapeuticallyrefractory Tourette-syndrome. J Neurol 254:963–965

Kuhn J, Gaebel W, Klosterkoetter J, Woopen C (2009) Deep brain stimulation as a newtherapeutic approach in therapy-resistant mental disorders: ethical aspects of investigationaltreatment. Eur Arch Psychiatry Clin Neurosci 259(Suppl 2):S135–S141

Leckman JF (2002) Tourette’s syndrome. Lancet 360:1577–1586Leckman JF, Riddle MA, Hardin MT, Ort SI, Swartz KL, Stevenson J, Cohen DJ (1989) The Yale

Global Tic Severity Scale: initial testing of a clinician-rated scale of tic severity. J Am AcadChild Adolesc Psychiatry 28:566–573

Leckman JF, Walker DE, Cohen DJ (1993) Premonitory urges in Tourette’s syndrome. Am JPsychiatry 150:98–102

Leckman JF, Zhang H, Vitale A, Lahnin F, Lynch K, Bondi C, Kim Y-S, Peterson BS (1998)Course of tic severity in Tourette syndrome: the first two decades. Pediatrics 102:14–19

Maciunas RJ, Maddux BN, Riley DE, Whitney CM, Schoenberg MR, Ogrocki PJ, Albert JM,Gould DJ (2007) Prospective randomized double-blind trial of bilateral thalamic deep brainstimulation in adults with Tourette syndrome. J Neurosurg 107:1004–1014

Martínez-Fernández R, Zrinzo L, Aviles-Olmos I, Hariz M, Martinez-Torres I, Joyce E,Jahanshahi M, Limousin P, Foltynie T (2011) Deep brain stimulation for Gilles de la Tourettesyndrome: a case series targeting subregions of the globus pallidus internus. Mov Disord26:1922–1930

Martinez-Torres I, Hariz MI, Zrinzo L, Foltynie T, Limousin P (2009) Improvement of tics aftersubthalamic nucleus deep brain stimulation. Neurology 72:1787–1789

Mink JW (2001) Basal ganglia dysfunction in Tourette’s syndrome: a new hypothesis. PediatrNeurol 25:190–198

Mink JW, Walkup J, Frey KA, Como P, Cath D, DeLong MR, Erenberg G, Juncos J, LeckmanJF, Swerdlow N, Visser-Vandewalle V, Vitek JL, Tourette Syndrome Association, Inc. (2006)Recommended guidelines for deep brain stimulation in Tourette syndrome. Mov Disord21:1831–1838

Muller-Vahl KR, Emrich HM, Dengler R, Munte TF, Dietrich D (2003) Tourette syndrome andobsessive-compulsive disorder: event-related brain potentials show similar mechanisms[correction of mechanisms] of frontal inhibition but dissimilar target evaluation processes.Behav Neurol 14:9–17

Müller-Vahl KR, Cath DC, Cavanna AE, Dehning S, Porta M, Robertson MM (2011) ESSTSGuidelines Group. European clinical guidelines for Tourette syndrome and other tic disorders.Part IV: deep brain stimulation. Eur Child Adolesc Psychiatry 20:209–217

12 Deep Brain Stimulation in Tourette Syndrome 127

Neuner I, Podoll K, Lenartz D, Sturm V, Schneider F (2009) Deep brain stimulation in thenucleus accumbens for intractable Tourette’s syndrome: follow-up report of 36 months. BiolPsychiatry 65:e5–e6

Neuner I, Halfter S, Wollenweber F, Podoll K, Schneider F (2010) Nucleus accumbens deepbrain stimulation did not prevent suicide attempt in Tourette syndrome. Biol Psychiatry68:e19–e20

Porta M, Brambilla A, Cavanna AE, Servello D, Sassi M, Rickards H, Robertson MM (2009a)Thalamic deep brain stimulation for severe treatment-refractory Tourette syndrome: two-yearoutcome. Neurology 73:1375–1380

Porta M, Sassi M, Ali F, Cavanna AE, Servello D (2009b) Neurosurgical treatment for Gilles dela Tourette syndrome: the Italian perspective. J Psychosom Res 67(6):585–590

Rabins P, Appleby BS, Brandt J, DeLong MR, Dunn LB, Gabriëls L, Greenberg BD, Haber SN,Holtzheimer PE 3rd, Mari Z, Mayberg HS, McCann E, Mink SP, Rasmussen S, SchlaepferTE, Vawter DE, Vitek JL, Walkup J, Mathews DJ (2009) Scientific and ethical issues relatedto deep brain stimulation for disorders of mood, behavior, and thought. Arch Gen Psychiatry66:931–937

Robertson MM, Banerjee S, Kurlan R, Cohen DJ, Leckma JF, McMahon W, Pauls DL, Sandor P,van de Wetering BJ (1999) The Tourette syndrome diagnostic confidence index: developmentand clinical associations. Neurology 53:2108–2112

Servello D, Porta M, Sassi M, Brambilla A, Robertson MM (2008) Deep brain stimulation in 18patients with severe Gilles de la Tourette syndrome refractory to treatment; the surgery andstimulation. J Neurol Neurosurg Psychiatry 79:136–142

Servello D, Sassi M, Brambilla A, Porta M, Haq I, Foote KD, Okun MS (2009) De novo andrescue DBS leads for refractory Tourette syndrome patients with severe comorbid OCD: amultiple case report. J Neurol 256:1533–1539

Shahed J, Poysky J, Kenney C, Simpson R, Jankovic J (2007) GPi deep brain stimulation forTourette syndrome improves tics and psychiatric comorbidities. Neurology 68:159–160

Shields DC, Cheng ML, Flaherty AW, Gale JT, Eskandar EN (2008) Microelectrode-guided deepbrain stimulation for Tourette syndrome: within-subject comparison of different stimulationsites. Stereotact Funct Neurosurg 86:87–91

Sturm V, Lenartz D, Koulousakis A, Treuer H, Herholz K, Klein JC, Klosterkotter J (2003) Thenucleus accumbens: a target for deep brain stimulation in obsessive-compulsive- and anxiety-disorders. J Chem Neuroanat 26:293–299

Temel Y, Visser-Vandewalle V (2004) Surgery in Tourette syndrome. Mov Disord 19:3–14Van der Linden C, Colle H, Vandewalle V, Alessi G, Rijckaert D, De Waele L (2002) Successful

treatment of tics with bilateral internal pallidum (GPi) stimulation in a 27-year-old malepatient with Gilles de la Tourette’s syndrome. Mov Disord 17:S341

Vandewalle V, van der Linden C, Groenewegen HJ, Caemaert J (1999) Stereotactic treatment ofGilles de la Tourette syndrome by high frequency stimulation of thalamus. Lancet 353:724

Vernaleken I, Kuhn J, Lenartz D, Raptis M, Huff W, Janouschek H, Neuner I, Schaefer WM,GrŸnder G, Sturm V (2009) Bithalamical deep brain stimulation in Tourette syndrome isassociated with reduction in dopaminergic transmission. Biol Psychiatry 66:e15–e17

Vilela Filho O, Ragazzo PC, Souza JT et al (2010) Bilateral GPe—DBS for Tourette syndrome: adouble-blind prospective controlled study of seven patients. In Abstract Book of the ASSFN(American Society for Stereotactic and Functional Neurosurgery) 2010 Biennial Meeting:Bridging the Future of Neurosurgery. New York, 2010

Visser-Vandewalle V, Temel Y, Boon P, Vreeling F, Colle H, Hoogland G, Groenewegen H, vander Linden C (2003) Chronic bilateral thalamic stimulation: a new therapeutic approach inintractable Tourette syndrome. J Neurosurg 99:1094–1100

Visser-Vandewalle V, Van der Linden Ch, Ackermans L, Temel Y, Tijssen MA, Schruers K,Nederveen P, Kleijers M, Boon P (2006) Deep brain stimulation in Gilles de la Tourette‘ssyndrome. Guidelines of the Dutch-Flemish Tourette Surgery Study Group. Neurosurgery58:E590

128 L. Ackermans et al.

Welter ML, Mallet L, Houeto JL, Karachi C, Czernecki V, Cornu P, Navarro S, Pidoux B,Dormont D, Bardinet E, Yelnik J, Damier P, Agid Y (2008) Internal pallidal and thalamicstimulation in patients with Tourette syndrome. Arch Neurol 65:952–957

Woods DW, Piacentini J, Himle MB, Chang S (2005) Premonitory Urge for Tics Scale (PUTS):initial psychometric results and examination of the premonitory urge phenomenon in youthswith Tic disorders. J Devel Behav Pediatr 26:397–403

Zabek M, Sobstyl M, Koziara H, Dzierzecki S (2008) Deep brain stimulation of the right nucleusaccumbens in a patient with Tourette syndrome. Case report. Neurol Neurochir Pol42:554–559

12 Deep Brain Stimulation in Tourette Syndrome 129

Chapter 13Surgical Treatments for Drug Addictionsin Humans

Bomin Sun and Wei Liu

13.1 Introduction

Drug addiction is a complex illness, characterized by intense and uncontrollable drugcraving, along with compulsive drug seeking and use that persist even in the face ofdisastrous consequences. Addiction can be the consequence of a wide variety of drugs,including nicotine, alcohol, and illicit and prescription drugs (van den Bosch andVerheul 2007). Multiple brain circuits are involved in the disease, such as those inreward and motivation, learning and memory, and inhibitory control over behavior.Because of the changes in the brain’s structure and function, long-term drug depen-dence and addiction usually last for a long time even after drug use has ceased (Volkowet al. 2004, 2008).

Drug addiction is a common problem throughout the world. It is usually com-posed of physical dependence and psychological dependence. Physical dependenceis related to withdrawal syndrome with a noradrenergic hyperactivity in the locuscoeruleus. Physiological detoxification and elimination of withdrawal syndromecould be achieved successfully by substitute therapies or other therapies, such asdopamine transporter blockers, non-dopamine drugs, and cannabinoid antagonists.Psychological dependence correlates with dopaminergic activity in the mesolimbicpathway, especially in the shell of the nucleus accumbens (NAc) (Di Chiara et al.2004). Psychological dependence has a close relationship with drug-seekingbehavior. Eliminating the psychological dependence is very difficult, and there is ahigh relapse rate even several months to 1 year after detoxification.

There are several kinds of treatments for drug addiction, such as drug substitutetherapy, behavioral therapy, and surgical treatment. There is a very high relapse

B. Sun (&) � W. LiuRuijin Hospital, Center for Functional Neurosurgery, Shanghai Jiao Tong University Schoolof Medicine, Shanghai 200025, Chinae-mail: bominsun@sh163.net

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_13, � Springer-Verlag Berlin Heidelberg 2012

131

rate for drug substitute therapy. It is reported that 80–85 % of drug addicts havedrug relapse within 1 month and 97 % of drug addicts have drug relapse withinhalf a year with substitute therapy. Surgery has been used to treat drug addictionoccasionally since the 1940s. In 1978, Kanaka and Balasubramaniam (1978)reported cingulotomy for drug addiction in 60 patients, with excellent results for60–80 %. Medvedev used bilateral cingulotomy in 348 drug addiction patients. Ofthese, 187 patients were followed up for more than 2 years and 45 % of them werecured (complete cessation of use of drug and termination of craving) (Medvedevand Anichkov 2003). Gao reported ablation of the NAc for opiate drug dependencepatients (Gao et al. 2003). The results demonstrated that bilateral NAc lesion hasexcellent effects for opiate drug dependence patients. The relapse rate alsodecreased significantly after 15 months of follow-up. Besides the irreversible NAcablation, deep brain stimulation (DBS) is being applied more and more. As a newtype of surgical method, DBS has dramatically broadened the landscape ofneurosurgery. Increased knowledge of neural circuits and brain imaging methodsled to an expansion of the indications for DBS in various neurologic and neuro-psychiatric disorders (Holtzheimer et al. 2011 Halpern et al. 2011; Kuhn et al.2011). Occasional observations of improvements in alcoholism (Kuhn et al. 2007),nicotine addiction (Neuner et al. 2009; Kuhn et al. 2009), or smoking addiction(Mantione et al. 2010) have been reported for patients treated with DBS foranxiety, obsessive–compulsive disorder, or other psychiatric disorders. Witjaset al. (2005) found that there was a good effect on motor disability and severedopamine addiction in two patients with Parkinson’s disease with severe dopamineaddiction who underwent bilateral subthalamic nucleus DBS for treatment ofdyskinesias and motor fluctuations. Kuhn et al. (2007) reported that a patient’scomorbid alcohol dependence was ameliorated when she was treated for severeanxiety disorder with secondary depressive disorder. Similar results were observedin three long-term, treatment-resistant alcohol-dependent individuals who under-went DBS of the NAc (Heinze et al. 2009). At the same time, animal work alsoprovided support for the application of DBS in addiction (Vassoler et al. 2008).

In this chapter, we briefly outline the optimal surgical target, surgical procedure,perioperative patient management, and surgical results for both lesions and DBS.

13.2 Optimal Surgical Target

With new imaging methods, we can study the function of the brain in real time. We knowthat the drugs activate the reward system. The major neurochemical pathway of thereward system in the brain includes the mesolimbic and mesocortical pathway. Of thesepathways, the mesolimbic pathway plays an important role, and finally links with theNAc, which is the primary release site for the neurotransmitter dopamine. It is widelyaccepted that the initial reinforcing effects of most drugs of abuse rely on the rapidincreased level of dopamine in the NAc. The significant role of the NAc in the drugaddiction mechanism has been demonstrated in many animal studies (Alderson et al.

132 B. Sun and W. Liu

2001). Furthermore, brain-imaging studies in humans have shown a correlation betweena psychostimulant-induced increase of the level of extracellular dopamine in the striatumand self-reported measures of pleasure. Intracranial self-administration studies with D1and D2 receptor agonists also suggest the NAc shell as the critical site of dopaminereward. Stimulation of dopamine transmission in the NAc shell by addictive drugs isshared by a natural reward such as food but lacks its adaptive properties (habituation andinhibition by predictive stimuli). These peculiarities of drug-induced stimulation ofdopamine transmission in the NAc shell result in striking differences in the impact ofdrug-conditioned stimuli on dopamine transmission (Di Chiara et al. 2004). Bothlesioning and DBS are done using the NAc as the surgical target structure.

13.3 Indications and Patient Selection Criteria

The indication and patient selection criteria are the same for lesioning and DBS.Since worldwide experience is limited to a few publications, no definite guidelineson patient selection exist. In our center there is general consensus about theselection criteria for drug addiction surgery:

1. Patients must be consistent with the diagnosis of addiction according to theDiagnostic and Statistical Manual of Mental Disorders, 4th edition, and theInternational Classification of Diseases, tenth revision.

2. Patients must have a history of drug dependence of more than 3 years and musthave undergone at least three ineffective substitute medication therapies.

3. Patients’ craving influences their health and severely affects the quality of lifeof themselves and family members.

4. Patients will seek to stop drug use and termination of craving on their owninitiative without been forced by others.

5. Patients and their families have complete understanding of the surgicalprocedures and have provided signed informed consent, and are able to cooperatewith our surgical team.

6. Patients have a suitable living environment and sufficient postoperative careand they must be able to have follow-up visits at 3, 6, 12, 24, and 36 monthspostoperatively.

13.4 Surgical Procedure

So far nobody has determined the optimal target or procedure for drug addiction;however, minimal invasion of the brain and maximally obtained efficacy are theprinciples of stereotactic neurosurgery. We developed minimally invasive NAcablation and DBS procedures for drug addiction. This procedure using MRI-guidedstereotactic techniques, which is similar to stereotactic capsulotomy, was done aspreviously described (Sun et al. 2005).

13 Surgical Treatments for Drug Addictions 133

13.4.1 Surgical Procedure for DBS

Commercial DBS systems consist of a quadripolar electrode with 1.27-mm diameterelectrode contacts and 1.5-mm length, an extension cable, and an internal pulsegenerator. We use a Soletra internal pulse generator (Medtronic, USA) and 3389DBS electrodes (Medtronic, USA) with an intercontact distance of 0.5 mm.

The head frame should be placed as soon before surgery as possible to minimizethe time before the patient goes to the operating room. A Leksell stereotactic frame ismounted on the patient’s head with the patient under local anesthesia or mildsedation. The base of the frame should be placed approximately parallel to theanterior commissure–posterior commissure line. Once the frame has been placed, thepatient is taken for preoperative MRI targeting. Although MRI, CT, and ventricu-lography can all be used for stereotactic imaging, MRI is necessary for drug addictionsurgery, because the NAc can be recognized directly in both axial and coronal sectionimages (Fig. 13.1) with high-resolution MRI. T2 and inversion recovery images arebeneficial for direct targeting of the NAc and surrounding areas. The bottom of thenucleus is targeted for drug addiction surgery, and is approximately 3 mm anterior tothe anterior commissure, 4 mm from the midline and 6 mm below the anteriorcommissure–posterior commissure level. We measure the entrance trajectory, whichis 18–20� lateral in the coronary plane and 45� anterior in the sagittal plane.

The procedure of electrode implantation is performed with the patient under local orgeneral anesthesia depending on the patient’s cooperation during the surgery. Aftercalculation of stereotactic target coordinates, small bilateral coronal incisions are madeand burr holes are placed bilaterally anterior to the coronal suture and about 3.0–4.0 cmfrom the midline, depending on the predetermined entrance trajectory. After duralopening and cauterization of the pia-arachnoid, we insert the quadripolar electrodes(Medtronic 3389) into the target area. Microelectrode recording is unnecessary for thisprocedure. Impedance measurement is important, because the NAc is located at the

Fig. 13.1 In the T2 inversion recovery MRI image, the nucleus accumbens can seen directly inboth axial (a) and coronal (b) section images

134 B. Sun and W. Liu

bottom of the lateral ventricle, and in our approach the electrode must pass through alower-impedance area (cerebrospinal fluid) before the target is accessed. After theelectrode has reached the target, a high-frequency stimulation (180 Hz, 90 ms, 1–6 V)is applied to observe side effects. Patients should experience serious feelings of heatand mild sweating which can be seen at the face and upper trunk. Meanwhile, heart rateand blood pressure increase significantly. It is very important to see these signs becausethey confirm that the electrode is in the NAc. Then the stimulation generator (Soletra) isimplanted with the patient under general anesthesia. The day after surgery, a postop-erative MRI scan is obtained to document the placement of the electrodes (Fig. 13.2).

The first day after DBS implantation, we start multiple programming sessions toscreen the best stimulation combination, using a fixed pulse width of 90 ls and with thefrequency of stimulation held constant at 145 Hz. The patients are tested individuallyat each lead and each contact (0, 1, 2, 3) utilizing monopolar stimulation. The stim-ulation amplitude is systematically increased in 0.5-V steps in each patient from thestarting value of 0 V in an attempt to obtain an immediate response. If there is noresponse at 6-V stimulation intensity, we increase the pulse width to 120 ls and then150 ls. After only several seconds of stimulation with 2.5–4-V at contact 0 and contact1, most patients feel a transient heart throb and the heart rate increases by about 20–50 % from the baseline. When the stimulation is increased by another 0.5–1 V, thepatient may have the feeling of heat, flushing at the site of stimulation, and evensweating in the trunk. Several minutes after the stimulation has been reduced by 1 V,this feature will fade away and the patient will feel happy and quite relaxed instead.Much higher stimulation intensities (from 4 to 6 V) are used to induce these responsesat contact 2 and contact 3. Some patients also have fear or a feeling of nervousness. Weselect the contact which can induce increasing heart rates and flushing at the loweststimulation threshold, and then set the stimulation intensity 1 V below this thresholdfor chronic stimulation.

Fig. 13.2 MRI follow-up ofbilaterally implanted deepbrain stimulation electrodesin a coronal section image

13 Surgical Treatments for Drug Addictions 135

13.4.2 Surgical Procedures for Lesions

Placement of the head frame and the entrance trajectory are the same as for theDBS procedure. After the opening of dura and cauterization of the pia-arachnoid, astandard thermistor-equipped thermocoagulation electrode (Radionics, Burlington,MA, USA) with a 2-mm uninsulated tip is employed for impedance measurement,followed by stimulation test and actual lesioning. After the position of the elec-trode has been confirmed by a test stimulation, radiofrequency lesions are made bya radiofreqency electrode with heating to 80 �C for 60 s. During lesioning,neurological testing is done to ensure that no impairment of motor or sensoryfunctions occurs. After adequate cooling, the electrode is withdrawn 2 mm and anadditional lesion is made using the same parameters to ensure the completeablation of the target. During the lesioning, severe sweating at the face and uppertrunk of the patient can reappear. The day after surgery, a postoperative MRI scanis obtained to document the placement and extent of the lesions (Fig. 13.3).

13.5 Perioperative Patient Management

Because of their long-term narcotic history, drug addicts are very different fromother neurosurgery patients. The mental status of addictive patients is not stable andthey frequently present with irritation and anxiety. Most patients require long-termor repeated courses of care to achieve the ultimate goal of sustained abstinence.Patients should be allowed to keep their normal lifestyle and habitus, includingcontinuation of use of narcotics after hospitalization. A thorough review of themedical history record and a physical examination must be done by our psycho-surgery team, which consists of three attending psychiatrists, a neurologist, a nurse,and three neurosurgeons to ensure indications for surgical therapy are met. Because

Fig. 13.3 MRI follow-up of bilateral lesions of the nucleus accumbens in axial (a) and coronal(b) section images

136 B. Sun and W. Liu

of long-term substance abuse and use of contaminated syringes, most drug addictshave abnormal liver function, kidney function, etc. So more detailed preoperativescreening such as electrocardiograms and appropriate blood tests is needed to assesspotential medical risks. The specific preoperative psychiatric and psychologicalevaluations are also performed by experienced psychiatrists and clinical psychol-ogists, such as a cognitive performance function test, Wechsler Adult IntelligenceScale IQ and memory test, personality test, Hamilton Anxiety Rating Scale,Hamilton Depression Rating Scale, psychiatric status rating scale and quality-of-lifeassessment.

A formal documentation of each patient including detailed history of drugaddiction, diagnostic and therapeutic history (especially previous detoxificationand abstinence history), the results of the physical, psychiatric, and psychologicalexaminations, the preoperative evaluations, and the surgical plans is given to themedical ethical committee in our medical center for approval. All of the evaluationresults, along with the surgical plan and informed consent, must be explained topatients and their families, and they must agree to cooperate with the surgical teamand participate in a postoperative follow-up program.

To avoid severe and sudden withdrawal symptoms so that the patients canmaintain normal spirit and physical status, patients are allowed to use previouslyused narcotics as usual on the morning of surgery. During stereotactic frameplacement and MRI targeting, a small amount of intravenous sedation is given ifnecessary.

Several hours after surgery, most patients exhibit restlessness, mild orientationdeficit, and confusion, which resolves in a couple of days. Buprenorphine (3 mg) andchlorpromazine (100 mg) can be given intravenously immediately after surgery, andthen the doses can be decreased to half the doses the following day. Three days aftersurgery, buprenorphine and chlorpromazine are withdrawn completely and only asmall dose of anxiolytic can be used in patients with anxiety or insomnia.

After patients have been discharged from hospital, they and their families arerequested to visit an outpatient clinic or take part in a phone interview for evaluationat 3, 6, 12, and 24 months, postoperatively. A follow-up questionnaire includesassessment of desire for narcotics, physical withdrawal symptoms, further preop-erative psychological and psychiatric evaluation, and rating scales for documenta-tion. For suspected relapsing patients, a regular narcotics urinalysis test is necessaryto confirm postoperative use of narcotics.

13.6 Surgical Outcomes

13.6.1 Ablation

So far only a few clinical retrospective studies on surgical treatment in addiction havebeen published. On the basis of these publications, cingulotomy and NAc lesioninghave been used for drug addiction. However, on the basis of the neuropsychiatric

13 Surgical Treatments for Drug Addictions 137

circuit, the orbitofrontal cortex, the frontothalamic pathways, and the limbic systemare potential targets for drug addiction. In fact, targets at any place in the orbito-frontal–striatal–thalamic–limbic–frontal circuit seem to be functionally equipotent,and a lesion or stimulation in any part of the circuit may directly or indirectly affectother parts. In recent years, many centers in China have been trying to use neuro-surgical therapy for drug addiction in humans. Nevertheless, most publications are inChinese and many obstacles have prevented a direct comparison of results acrosscenters, including diagnostic inaccuracies, nonstandard preoperative evaluations,center bias, nonstandard surgical procedures, and different outcome assessmentsystems. Gao et al. (2003) reported radiofrequency lesioning of the NAc for patientswho are addicted to drugs. They found 26.7 % of the patients were cured after15 months with mild complications: two patients had possible personality changesand there were four patients with short-term memory deficit. In our center, ninepatients (one used dolantin intravenously and the rest used heroin intravenously twoto three times per day) underwent bilateral NAc ablation. After surgery, only onepatient with dolantin addiction relapsed within 1 month, whereas eight patients withheroin addiction were drug-free (without any desire and drug-using activity).

13.6.2 Deep Brain Stimulation

There are a few reports on the trial of DBS to treat addiction to nicotine, alcohol, andheroin. Heinze et al. (2009) reported that craving for alcohol and alcohol con-sumption were greatly reduced in three long-term, treatment-resistant alcohol-dependent individuals who underwent DBS of the NAc. Mantione et al. (2010)observed that a 47-year-old woman who had nicotine dependence quit smoking afterchronic DBS of the NAc. In our center, there were two patients with heroindependence who underwent bilateral NAc DBS electrode implantation. One of themhas completely stopped the use of narcotics and has no craving. The other one tookonly a small amount of methadone every day orally without injection of herioin.

13.7 Side Effects and Complications

In all publications, the side effects and complications reported are similar. Nosevere complications such as hemiplegia, aphasia, intracranial hematomas, ordeath directly caused by surgery were reported. In our center, nine patients withlesioning of the NAc experienced short-term side effects on the first day aftersurgery. Similar to anterior capsulotomy patients, most patients had mild transientdeterioration in mental status such as memory deficits and confusion postopera-tively. However, all of these side effects disappeared automatically without anyspecific treatment. Six patients with lesioning of the NAc experienced delayed sideeffects such as mild fatigue, apathy, inactivity, and lack of interest. It is veryinteresting that almost all patients with NAc lesioning had emotional fragility.

138 B. Sun and W. Liu

These side effects resolve within 1–2 years after surgery and do not affect theirquality of life. Only a few of them needed to see a psychiatrist for medicationassistance because of mild anxiety. There were no side effects and complicationsin the two patients with bilateral DBS.

13.8 Conclusion

The NAc is the main source of the initial reinforcing effect of most drug abuse. It islocated at the bottom of the frontal lobe, and there is no important motor or sensoryfunctional area near this location. Lesioning or stimulation of the NAc is considered asafe and effective surgical procedure without long-term severe side effects or com-plications. However, compared with ablation, the microlesion caused by DBS in thetarget area usually disappears in a few weeks following the surgery and the influenceof the local current from stimulator is completely reversible. DBS is an excellentalternative therapy for refractory drug addiction patients. Most patients are com-pletely cured after surgery, and have few complications and side effects. However, itshould be done only by an expert multidisciplinary team with rich experience in thisfield and the patients must be selected strictly according to the inclusion criteria. Wemust keep in mind that surgical therapy should only be considered as a part of thetreatment and must be accompanied by an appropriate psychological rehabilitationplan and a family–social support program.

References

Alderson HL, Parkinson JA, Robbins TW, Everitt BJ (2001) The effects of excitotoxic lesions ofthe nucleus accumbens core or shell regions on intravenous heroin self-administration in rats.Psychopharmacology 153:455–463

Di Chiara G, Bassareo V, Fenu S, De Luca MA, Spina L, Cadoni C, Acquas E, Carboni E,Valentini V, Lecca D (2004) Dopamine and drug addiction: the nucleus accumbens shellconnection. Neuropharmacology 47(Suppl 1):227–241

Gao GD, Wang XL, He SM, Li WX, Wang QF, Liang QC, Zhao YQ, Hou F, Chen L, Li AN (2003)Clinical study for alleviating opiate drug psychological dependence by a method of ablating thenucleus accumbens with stereotactic surgery. Stereotact Funct Neurosurg 81:96–104

Halpern CH, Torres N, Hurtig HI, Wolf JA, Stephen J, Oh MY, Williams NN, Dichter MA, JaggiJL, Caplan AL, Kampman KM, Wadden TA, Whiting DM, Baltuch GH (2011) Expandingapplications of deep brain stimulation: a potential therapeutic role in obesity and addictionmanagement. Acta Neurochir 153:2293–2306

Heinze HJ, Heldmann M, Voges J, Hinrichs H, Marco-Pallares J, Hopf JM, Muller UJ, Galazky I,Sturm V, Bogerts B, Munte TF (2009) Counteracting incentive sensitization in severe alcoholdependence using deep brain stimulation of the nucleus accumbens: clinical and basic scienceaspects. Front Hum Neurosci 3:22

Holtzheimer PE, Mayberg HS (2011) Deep brain stimulation for psychiatric disorders. Ann RevNeurosci 34:289–307

Kanaka TS, Balasubramaniam V (1978) Stereotactic cingulumotomy for drug addiction. ApplNeurophysiol 41:86–92

13 Surgical Treatments for Drug Addictions 139

Kuhn J, Lenartz D, Huff W, Lee S, Koulousakis A, Klosterkoetter J, Sturm V (2007) Remissionof alcohol dependency following deep brain stimulation of the nucleus accumbens: valuabletherapeutic implications? J Neurol Neurosurg Psychiatry 78:1152–1153

Kuhn J, Bauer R, Pohl S, Lenartz D, Huff W, Kim EH, Klosterkoetter J, Sturm V (2009)Observations on unaided smoking cessation after deep brain stimulation of the nucleusaccumbens. Eur Addict Res 15:196–201

Kuhn J, Moller M, Muller U, Bogerts B, Mann K, Grundler TO (2011) Deep brain stimulation forthe treatment of addiction. Addiction 106:1536–1537(1537–1538)

Mantione M, van de Brink W, Schuurman PR, Denys D (2010) Smoking cessation and weightloss after chronic deep brain stimulation of the nucleus accumbens: therapeutic and researchimplications: case report. Neurosurgery 66:218–218

Medvedev SV, Anichkov AD, Polyakov YI (2003) Physiological mechanisms of effectiveness ofbilateral stereotactic cingulotomy against strong psychological dependence in drug addicts.Hum Physio 29(4):492–497

Neuner I, Podoll K, Lenartz D, Sturm V, Schneider F (2009) Deep brain stimulation in thenucleus accumbens for intractable Tourette’s syndrome: follow-up report of 36 months. BiolPsychiatry 65:e5–e6

Sun BM, Krahl SE, Zhan SK, Shen JK (2005) Improved capsulotomy for refractory Tourette’ssyndrome. Stereotact Funct Neurosurg 83:55–56

van den Bosch LMC, Verheul R (2007) Patients with addiction and personality disorder:treatment outcomes and clinical implications. Curr Opin Psychiatr 20:67–71

Vassoler FM, Schmidt HD, Gerard ME, Famous KR, Ciraulo DA, Kornetsky C, Knapp CM,Pierce RC (2008) Deep brain stimulation of the nucleus accumbens shell attenuates cocainepriming-induced reinstatement of drug seeking in rats. J Neurosci 28:8735–8739

Volkow ND, Fowler JS, Wang GJ (2004) The addicted human brain viewed in the light ofimaging studies: brain circuits and treatment strategies. Neuropharmacology 47:3–13

Volkow ND, Wang GJ, Fowler JS, Telang F (2008) Overlapping neuronal circuits in addictionand obesity: evidence of systems pathology. Philos Trans R Soc B Biol Sci 363:3191–3200

Witjas T, Baunez C, Henry JM, Delfini M, Regis J, Cherif AA, Peragut JC, Azulay JP (2005)Addiction in Parkinson’s disease: impact of subthalamic nucleus deep brain stimulation. MovDisord 20:1052–1055

140 B. Sun and W. Liu

Chapter 14Manipulating Addictive Behaviourin Animal Models

Rolinka M. C. Schippers, Tommy Pattijand Taco J. De Vries

14.1 Drug Self-Administration Model

One of the first observations of drug dependence and addictive behaviour in a non-human species was made in morphine-dependent chimpanzees. The chimpanzeeshad to make a choice between food and a morphine injection syringe that theyremembered from earlier morphine infusions given by the experimenter. When thechimpanzees were deprived of morphine, they would choose the morphineinjection syringe, demonstrating their dependence on this drug (Spragg 1940). Inthe 1960s, drug self-administration paradigms with fully automated intravenousinfusions were developed for rats and monkeys. In these instrumental learningprocedures, animals were trained to self-administer drugs of abuse by lever-pressing or nose-poking (Weeks 1962; Thompson and Schuster 1964). It wasshown that laboratory animals readily self-administer the same addictive drugs thatare used by humans, including cocaine, amphetamine, nicotine, heroin and mor-phine. This led to the hypothesis that the rewarding effects of drugs are a phar-macological property, rather than from involvement of psychological and socialprocesses that at time were predominantly thought to predispose to drug addiction.

R. M. C. Schippers � T. Pattij � T. J. De VriesDepartment of Anatomy and Neurosciences, Neuroscience Campus Amsterdam,VU University Medical Center, Amsterdam, The Netherlands

T. J. De VriesDepartment of Molecular and Cellular Neurobiology, Center for Neurogenomics &Cognitive Research, Faculty of Earth and Life Sciences, VU University Amsterdam,Amsterdam, The Netherlands

T. J. De Vries (&)Department of Anatomy and Neurosciences, VU University Medical Center,Neuroscience Campus Amsterdam, Van der Boechorststraat 7 Room B452,1081 BT, Amsterdam, The Netherlandse-mail: tj.devries@vumc.nl

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_14, � Springer-Verlag Berlin Heidelberg 2012

141

It provided the opportunity to develop an animal model to study drug-taking anddrug-seeking and the underlying neural substrates in a well-controlled laboratorysetting.

14.1.1 Self-Administration

Typically, in the animal model of volitional drug self-administration, drugs such ascocaine, heroin and nicotine are administered intravenously or intracerebrally,whereas alcohol, sucrose, water and food pellets are often consumed orally. Self-administration of saline or ‘‘natural reinforcers’’ such as food pellets, sucrose andwater is often used to control for the drug-specific effects of manipulations.

A test chamber usually contains two levers or nose-poke holes. Responding onthe ‘‘active’’ operandum results in delivery of a reward. As such, the animal canself-administer the compound and self-regulate the rate and number of rewards.The second lever or nose-poke hole, often referred to as ‘‘inactive’’, is present, butdoes not result in delivery of a reward. It serves as a readout for the animal’sability to learn the behavioural contingencies and to distinguish between activeand inactive responses. Animals readily acquire and maintain stable operantresponding for a specific reinforcer or drug of abuse (Fig. 14.1). Drug delivery isusually accompanied by presentation of discrete environmental stimuli, such asvisual and auditory cues. These stimuli then become associated with the rewardingeffects of the drug, mimicking the human situation where drug abusers becomeconditioned to stimuli that are associated with drug purchase, preparation, and use.

Typically, a single operant response of the animal results in a single rewarddelivery. This so-called continuous reinforcement schedule can also be transferredinto a fixed-ratio schedule of reinforcement where multiple operant responses arenecessary to obtain a single reward. In this fashion, these schedules of rein-forcement allow the study of simple patterns of drug reinforcement.

14.1.2 Progressive Ratio

To measure motivational aspects of drug-taking, progressive ratio schedules ofreinforcement have been developed (Hodos 1961). During these schedules, thenumber of responses required to obtain a single reward is progressively increasedeither within a session after each reward delivery or between sessions. As such, theeffort for a single drug delivery progressively increases, until the animal stopsresponding within a predefined period. The highest number of operant responsescompleted for one reward delivery is defined as the breaking point and is sensitiveto experimental manipulations, such as dose changes, pharmacological manipu-lations and lesions. Progressive ratio schedules provide information on motiva-tional aspects of drug-taking that cannot be measured using fixed-ratio schedules

142 R. M. C. Schippers et al.

of reinforcement. Importantly, experimental manipulations may differentiallyaffect responding on a fixed-ratio schedule and a progressive ratio schedule,indicating that both schedules provide different types of information on the rein-forcing properties of drugs of abuse.

14.1.3 Dose–Response Relationship

Dose–response relationships can be investigated by changing the concentration ofthe drug in the syringe. Dose–response curves typically display an inverted-Ushape and provide information on individual drug sensitivity. Experimentalmanipulations can shift the dose–response curve and provide information on thereinforcing value of a drug.

Fig. 14.1 Drug self-administration model. Closed circles: active responses. Open circles:inactive responses

14 Manipulating Addictive Behaviour 143

14.1.4 Extinction, Abstinence and Reinstatement

Relapse to drug use after a period of withdrawal is a hallmark of drug addiction.In this regard, reinstatement models are the best validated animal models inaddiction research to study the neurobiological mechanisms of relapse and cravingand to test (pharmacological) interventions as possible relapse-prevention strate-gies (Shaham et al. 2003). In reinstatement models, animals are first trained toself-administer a drug in the presence of drug-associated cues. After a period ofself-administration, drug availability and the drug-associated cues are thenextinguished. During this phase, responses do not result in delivery of a reward andpresentation of the drug-associated cues, consequently leading to decreasedresponding over time. Alternatively, a period of abstinence can be imposed on thesubjects. During this period, animals are not exposed to the test chambers or otherdrug-paired stimuli.

Following extinction, non-reinforced responding on the previously drugassociated operandum can be reinstated by several factors, such as presentationof previously drug associated cues (Davis and Smith 1976), acute re-exposureto the drug by non-contingent drug-priming injections (de Wit 1996) orstressors, such as a brief foot shock exposure (Shaham et al. 2000). Since thesefactors are also known to induce craving and relapse in human addicts, itappears that the predictive and construct validity of the reinstatement model isvery high.

14.2 Conditioned Place Preference

To study conditioned rewarding effects of addictive drugs without instrumentallearning, the conditioned place preference (CPP) paradigm is widely used. Theparadigm is based on classical conditioning learning principles, where contex-tual stimuli are paired with a reward. The CPP apparatus consists of twocompartments that are contextually different, for instance in the visual or tactiledomain. One compartment is paired with a specific drug by non-contingentinjections before the animal is placed in the compartment. The other com-partment is paired with a control substance, usually saline injections. The timespent by the animal in each compartment when given the choice under drug-free conditions is used as an indication for the place preference (Fig. 14.2).Because CPP does not involve instrumental learning to obtain a drug, but isbased on the approach behaviour to a drug-associated context, it allows theinvestigation of the incentive value of the drug-associated environment. Thisparadigm is sensitive to various experimental manipulations, including systemicor local pharmacological interventions and lesions of specific brain regions(Aguilar et al. 2009).

144 R. M. C. Schippers et al.

14.2.1 Extinction and Reinstatement in the CPP Model

The CPP procedure can also be extended with an extinction and reinstatementphase. Extinction is induced either by administering the vehicle in the originaldrug-paired and original vehicle-paired compartments or by repeatedly exposingthe animals to both apparatus compartments without drug administration, until thepreference is no longer observed. Similar to the drug self-administration rein-statement model, extinguished CPP can be reinstated by non-contingent drug-priming injections (Mueller and Stewart 2000) or exposure to stressful stimuli suchas foot shock exposure (Lu et al. 2000). Also, discrete stimuli (such as a tone)previously associated with fear (foot shock) have been shown to reinstate cocaineCPP (Sanchez and Sorg 2001).

14.3 Neurobiology of Addiction: What Have We Learnedfrom Animal Models?

Animal models of drug self-administration have been very useful in the identificationof the neural circuitry underlying motivation, reinforcement and relapse (Fig. 14.3).

The initial site of action for most drugs of abuse is thought to be the ventraltegmental area (VTA). Evidence for this view is derived from observations thatmost drugs of abuse are intracranially self-administered in the VTA (reviewed byO’Brien and Gardner 2005). The VTA sends dopaminergic projections to theamygdala , the nucleus accumbens (NAc) and the medial prefrontal cortex(mPFC) (Koob and Volkow 2010).

14.3.1 Nucleus Accumbens

The NAc is another important primary site of action for the reinforcing properties ofmany drugs. The efferent dopaminergic projections from the VTA to the NAc are

Fig. 14.2 Conditioned place preference

14 Manipulating Addictive Behaviour 145

critical for immediate reward and the initiation of drug-seeking (Koob and Volkow2010). However, CPP studies suggest that the rewarding properties of opiates arepartially independent of the neurotransmitter dopamine, and that non-dopaminergicsystems, such as glutamatergic and GABAergic projections from the VTA to theNAc, are more important for (reinstatement of) opiate CPP (for a review, see Aguilaret al. 2009). The NAc projects, amongst other areas, to the ventral pallidum, which iscritical for further processing of reward signals to the basal ganglia for the conversionof reward into motivational motor actions (Koob and Volkow 2010).

14.3.2 Prefrontal Cortex

The mPFC, a brain region important for planning, executive control and decision-making, plays a role in the acquisition of drug self-administration, primarilythrough efferent glutamatergic projections to the NAc (Tzschentke 2000). Thisstrongly suggests that in addition to rewarding aspects, also cognitive processes areinvolved in the development and persistence of drug addiction. Indeed, this hasbeen shown in several animal studies (for a review, see Jentsch and Taylor 1999).

Fig. 14.3 Brain regions known to be involved in drug addiction and relapse. The areashighlighted with a bold line represent the target regions of the deep brain stimulation studies inanimal models of addiction investigated so far. BLA basolateral amygdala, DS dorsal striatum(Vassoler et al. 2008); LH lateral hypothalamus (Levy et al. 2007); LHb lateral habenula(Friedman et al. 2010, 2011); NAcc core nucleus accumbens core (Liu et al. 2008; Knapp et al.2009); NAcc shell nucleus accumbens shell (Vassoler et al. 2008; Knapp et al. 2009; Hendersonet al. 2010); OFC orbitofrontal cortex; dmPFC dorsal medial prefrontal cortex (Levy et al. 2007);vmPFC ventromedial prefrontal cortex; STN subthalamic nucleus (Rouaud et al. 2009); VPventral pallidum, VTA ventral tegmental area

146 R. M. C. Schippers et al.

Lesioning the mPFC has been shown to disrupt the development of CPP for severaltypes of drugs, including cocaine and morphine. Different subregions of the mPFCseem to be involved in the mediation of the rewarding values of different drugs,suggesting a functional heterogeneity of the mPFC (reviewed by Tzschentke 2000).

Anatomically, the dorsal mPFC sends glutamatergic projections to the coreregion of the NAc (Heidbreder and Groenewegen 2003). Several studiesemploying reinstatement models have shown that this projection is involved inpromoting drug-seeking behaviour (e.g. McFarland and Kalivas 2001). In contrast,glutamatergic projections from the ventral mPFC to the NAc shell have beenshown to be involved in active memory formation processes during extinctiontraining to inhibit drug-seeking when drug-related cues are absent (Peters et al.2008). Collectively, the mPFC seems to determine the intensity of behaviouralresponding by enhancing or inhibiting reward-related brain areas.

The persistent nature of drug-seeking behaviour is hypothesized to be mediatedthrough long-lasting drug-induced alterations in cognitive and motivational net-works, such as the mPFC–NAc connection (Kalivas and Volkow 2005). Forexample, recent studies from our laboratory have shown that reexposure to heroin-associated stimuli induces immediate molecular alterations leading to a decreasedsynaptic strength in the mPFC (Van den Oever et al. 2010).

These and other data point to an impaired top–down cognitive control overdrug-seeking when individuals with a history of drug self-administration areconfronted with drug-associated cues.

14.3.2.1 Amygdala

Functionally, the amygdala establishes connections between motivationallyrewarding stimuli and previously neutral stimuli, and thereby plays an important rolein conditioned reward learning. The basolateral nucleus of the amygdala is stronglyimplicated in cue-induced reinstatement of drug-seeking, indirectly influencing theNAc core via its strong connections with the mPFC (See et al. 2003).

14.3.2.2 Dorsal Striatum

The role of the dorsal striatum in rewarding aspects of drugs of abuse is still underinvestigation. This brain area appears to be particularly involved in habit formationand compulsive drug-seeking. It is hypothesized that the transition from voluntaryto compulsive drug use represents a shift in the involvement of the ventral striatumto involvement of the dorsal striatum (Everitt et al. 2008).

14.3.2.3 Insula

The insular cortex, or insula, has only recently been implicated in addictive behav-iour, in particular in the interoceptive aspects of drug consumption. In support of this

14 Manipulating Addictive Behaviour 147

view, damage to the insular cortex has been reported to profoundly diminish the urgeto smoke in human smokers (Naqvi et al. 2007). In line with these observations,animal studies demonstrated attenuated nicotine-seeking, but also attenuatedcocaine-seeking, after inactivation of the insula (Di Pietro et al. 2008; Forget et al.2010). In particular, the anterior part of the insula is interconnected with the NAc,amygdala and ventral mPFC (Van De Werd and Uylings 2008). Therefore, exposureto drug-related cues possibly leads to the retrieval of an interoceptive memory in theinsula and thereby through its connections with the amygdala and anterior cingulatecortex to drug-craving (Naqvi and Bechara 2010).

14.4 Validity of the Models

14.4.1 Face Validity

As mentioned already, laboratory animals will readily self-administer drugs of abusesimilar to those that are (ab)used by humans. Importantly, the route of administration,intravenous or oral, is analogous to the human situation. In addition, intracerebralroutes of administration can be used in laboratory animals to identify the brainregions critically involved in mediating drug reward and reinforcement.

In contrast to the self-administration model, the CPP model is primarily vali-dated as a behavioural protocol for rodents. Thus far, one article has shown thathumans are also able to display place preference after amphetamine pairing(Childs and de Wit 2009). A confounding factor of the CPP model is that drugs arenon-contingently administered by an experimenter and that the route of adminis-tration is often intraperitoneal.

For both animal addiction models, there are several issues that need to be takeninto account. For instance, during periods of extinction or abstinence and rein-statement testing, the animal is faced with events that are controlled by theexperimenter. In contrast, in clinical settings the subject can voluntarily regulatedrug consumption or is often aware that drugs are not available. In addition, themotivational considerations of a drug-dependent person to quit drug taking cannotbe mimicked in an animal model (Epstein et al. 2006).

14.4.2 Construct Validity

Construct validity refers to a similarity in underlying mechanisms of behaviour inthe animal model and the modelled condition as seen in humans (Geyer andMarkou 1995). Despite the methodological and species differences, it is strikingthat the brain regions that are found to be involved in drug-taking and drug-seekingshow a strong consistency between human and laboratory animals. As such, brain

148 R. M. C. Schippers et al.

imaging studies show activation of the striatum, the orbitofrontal and prefrontalregions and the amygdala after a drug challenge or during craving evoked bypresentation of drug cues in psychostimulant- and opioid-dependent subjects(Goldstein and Volkow 2002).

14.4.3 Predictive Validity

Predictive validity indicates to what degree laboratory-animal behaviour inducedby experimental manipulations predicts the behaviour in humans by an eventanalogous to the modelled condition (Geyer and Markou 1995). A recent reviewextensively discussed the results between preclinical and clinical studies on drugself-administration. An important conclusion emerging from all the studies is thatgenerally the results from rat self-administration studies reliably translate topositive subjective effects (such as ‘‘liking’’, ‘‘high’’ and ‘‘euphoria’’) in clinicalstudies (O’Connor et al. 2011). In addition, many compounds have been tested fortheir ability to reduce drug intake or relapse behaviour. In this regard, the maineffects of these compounds in rodent nicotine, heroin and alcohol self-adminis-tration resemble the effects found in humans (Epstein et al. 2006), indicating alsopredictive validity for pharmacological manipulations. In particular, compoundsthat attenuated reinstatement of drug-seeking in rodents were also found todecrease relapse in humans. For instance, the cannabinoid CB1 antagonist rimo-nabant (SR141716A) has been shown to be effective for relapse prevention inabstinent smokers (Fagerström and Balfour 2006). Likewise, in rats, short-termadministration of rimonabant has been shown to attenuate cue-induced reinstate-ment of nicotine-seeking (De Vries et al. 2005). Similarly, naltrexone has beenshown to reduce relapse to alcohol- and heroin-seeking in both animal studies(Shaham and Stewart 1996) and human studies (Shufman et al. 1994).

Notably, most preclinical studies concentrate on the short-term pharmacologi-cal effects on self-administration and reinstatement, whereas in clinical studieslong-term administration of medication is common practice. In addition, there arerelatively few clinical studies evaluating the efficacy of pharmacotherapies duringa period of abstinence. Rather, the most commonly targeted outcome in humanstudies is to decrease current drug use. Discrepancies between clinical and pre-clinical outcomes may depend on different routes of administration (intravenous,oral), and the timing of administration (during drug use or during abstinence),since there is evidence that distinct neurobiological mechanisms underlie differentstages of drug dependence (Kalivas and Volkow 2005).

Regarding the predictive validity of the CPP paradigm, to date only twocompounds (naloxone and acamprosate) that are approved by the Food and DrugAdministration for treating drug abuse have been evaluated for ethanol CPP. Inaddition, some compounds result in contradictory effects in the CPP and the self-administration paradigm (reviewed by Aguilar et al. 2009). For instance, theN-methyl-D-aspartate receptor agonist memantine blocks reinstatement of cocaine

14 Manipulating Addictive Behaviour 149

CPP, but not cocaine-seeking after cocaine self-administration. Likewise, the D1agonist SKF 81297 reinstates cocaine CPP, but is not able to reinstate cocaine-seeking after cocaine self-administration. Methodological differences or differentaspects of drug dependence that are modelled by both paradigms may account forthe discrepant results.

14.5 DBS in Animal Models of Addiction

Clinical case studies have shown that DBS as a treatment for other psychiatricdisorders decreases co-morbid alcohol or drug addiction (Kuhn et al. 2007;Mantione et al. 2010). Similarly, DBS of the subthalamic nucleus in Parkinsonianpatients dependent on L-dopa (dopamine dysregulation syndrome) reduced theirdrug-seeking (Witjas et al. 2005).

Fuelled by these observations and the potential use of DBS as a treatment fortreatment-resistant addiction (see Sun and Liu, this volume), preclinicalresearchers began to explore the mechanism of action of DBS in animal models ofdrug addiction (Table 14.1).

14.5.1 Nucleus Accumbens

The NAc has received the most attention in this regard, as it is a common target fortreatment-resistant obsessive–compulsive disorder and major depression. In pre-clinical work, DBS of the NAc shell reduced relapse to cocaine-seeking that wasprovoked by a priming injection of cocaine in rats that were trained to self-administer cocaine. Notably, no effects of DBS in the dorsal striatum on cocaine-primed reinstatement were reported, indicating a region-specific effect of DBS.Under the same conditions, NAc shell DBS did not alter seeking for naturalrewards (Vassoler et al. 2008). Local inactivation of the NAc shell with GABAagonists was ineffective in a similar cocaine-primed reinstatement paradigm(McFarland and Kalivas 2001) and in a conditioned cue-induced reinstatementparadigm (Fuchs et al. 2004).

In a separate study, stimulation of the NAc core during the conditioning phaseof morphine-induced CPP reduced place preference (Liu et al. 2008). Recently,two independent preclinical studies evaluated the effects of DBS on voluntaryconsumption of ethanol. Bilateral stimulation of either the NAc core region or theNAc shell region reduced ethanol intake, whereas water consumption remainedunaffected (Knapp et al. 2009; Henderson et al. 2010). This is consistent with casereports showing that NAc DBS remarkably reduces alcohol intake in alcoholics(Müller et al. 2009).

150 R. M. C. Schippers et al.

Tab

le14

.1S

tudi

esin

vest

igat

ing

deep

brai

nst

imul

atio

n(D

BS)

inan

imal

mod

els

ofdr

ugad

dict

ion

Ref

eren

ceB

rain

regi

onS

ubst

ance

Tes

tpr

otoc

olD

BS

Res

ults

Self

-adm

inis

trat

ion

Lev

yet

al.

(200

7)L

H(b

ilat

eral

)C

ocai

ne,

sucr

ose

Ext

inct

ion,

PR

,M

WM

,lo

com

otor

acti

vity

,ps

ycho

mot

orse

nsit

izat

ion

30m

inda

ily,

10da

ys,

inho

me

cage

betw

een

SA

and

test

ing

Dec

reas

eex

tinc

tion

resp

ondi

ngN

oef

fect

onsu

cros

eS

AN

oef

fect

onM

WM

,lo

com

otor

acti

vity

PF

C(b

ilat

eral

)C

ocai

ne,

sucr

ose

Ext

inct

ion,

PR

,M

WM

,lo

com

otor

acti

vity

,ps

ycho

mot

orse

nsit

izat

ion

30m

inda

ily,

10da

ys,

inho

me

cage

betw

een

SA

and

test

ing

Dec

reas

eex

tinc

tion

resp

ondi

ngD

ecre

ase

PR

resp

onse

sN

oef

fect

onsu

cros

eS

A

Vas

sole

ret

al.

(200

8)N

Ac

shel

l(b

ilat

eral

)C

ocai

ne,

sucr

ose

Coc

aine

-ind

uced

rein

stat

emen

tD

urin

g2-

hte

stD

ecre

ase

rein

stat

emen

tN

oef

fect

onsu

cros

eD

S(b

ilat

eral

)C

ocai

neC

ocai

ne-i

nduc

edre

inst

atem

ent

Dur

ing

2-h

test

No

effe

cton

rein

stat

emen

t

Rou

aud

etal

.(2

009)

ST

N(b

ilat

eral

)C

ocai

ne,

sucr

ose

24-h

food

inta

ke,

FR

1,P

R,

dose

resp

onse

Dur

ing

test

No

effe

cton

FR

1re

spon

ses

Dec

reas

eP

Rre

spon

ses

Incr

ease

mot

ivat

ion

for

sucr

ose

Fri

edm

anet

al.

(201

0)L

Hb

(uni

late

ral

righ

the

mis

pher

e)C

ocai

neF

R1,

exti

ncti

on,

rein

stat

emen

t,fo

rced

swim

test

,2-

bott

lech

oice

test

15m

inbe

fore

SA

,15

min

befo

refi

rst

exti

ncti

onse

ssio

n

Dec

reas

eF

R1

resp

onse

sD

ecre

ase

drug

-see

king

Dec

reas

ere

inst

atem

ent

Fri

edm

anet

al.

(201

1)L

Hb

(uni

late

ral

righ

the

mis

pher

e)S

ucro

seF

R1

15m

inbe

fore

SA

,tw

ice

duri

ng12

SA

sess

ions

Dec

reas

eF

R1

resp

onse

s

CP

PL

iuet

al.

(200

8)N

Ac

core

(uni

late

ral

righ

the

mis

pher

e)M

orph

ine

CP

P,

cond

itio

ning

and

drug

-fre

e3-

hin

clud

ing

1-h

CP

Pdu

ring

peri

odof

cond

itio

ning

,7-

10da

ys

Inhi

bit

mor

phin

e-in

duce

dC

PP (con

tinu

ed)

14 Manipulating Addictive Behaviour 151

Tab

le14

.1(c

onti

nued

)

Ref

eren

ceB

rain

regi

onS

ubst

ance

Tes

tpr

otoc

olD

BS

Res

ults

Rou

aud

etal

.(2

009)

ST

N(b

ilat

eral

)C

ocai

ne,

sucr

ose

CP

PA

tea

chst

age

ofth

eC

PP

proc

edur

eIn

hibi

tco

cain

e-in

duce

dC

PP

Inte

nsif

ysu

cros

e-in

duce

dC

PP

Fri

edm

anet

al.

(201

1)L

Hb

(uni

late

ral

righ

the

mis

pher

e)–

CP

PD

urin

gco

ndit

ioni

ngA

vers

ion

toth

eco

ntex

tas

soci

ated

wit

hL

Hb

DB

SC

onsu

mpt

ion

Kna

ppet

al.

(200

9)N

Ac

shel

l(b

ilat

eral

)E

than

olF

ree

etha

nol

inta

keD

urin

g30

min

acce

sspe

riod

Dec

reas

ein

take

No

effe

cton

wat

erin

take

NA

cco

re(b

ilat

eral

)E

than

olF

ree

etha

nol

inta

keD

urin

g30

-min

acce

sspe

riod

Dec

reas

ein

take

No

effe

cton

wat

erin

take

Hen

ders

onet

al.

(201

0)N

Ac

shel

l(b

ilat

eral

)E

than

olF

ree

etha

nol

inta

ke,f

ree

etha

nol

inta

keaf

ter

abst

inen

ce

2da

ys,

2-h

trea

tmen

tcy

cle

(1-h

on/1

-hof

fD

BS

,ra

ndom

lyas

sign

ed)

Dec

reas

ein

take

Dec

reas

ein

take

afte

rab

stin

ence

CP

Pco

ndit

ione

dpl

ace

pref

eren

ce,

DS

dors

alst

riat

um,

FR

1fi

xed-

rati

osc

hedu

lew

ith

rein

forc

emen

taf

ter

ever

yre

spon

se,

LH

late

ral

hypo

thal

amus

,L

Hb

late

ral

habe

nula

,M

WM

Mor

ris

wat

erm

aze,

NA

cnu

cleu

sac

cum

bens

,P

FC

pref

ront

alco

rtex

,P

Rpr

ogre

ssiv

e-ra

tio

sche

dule

,SA

self

-adm

inis

trat

ion,

STN

subt

hala

mic

nucl

eus

152 R. M. C. Schippers et al.

14.5.2 Subthalamic Nucleus

As mentioned earlier, it has been observed that DBS of the STN in Parkinsonianpatients dependent on L-dopa reduced their drug-seeking (Witjas et al. 2005),pinpointing this brain region as a potential target for DBS in drug addicts. Insupport of this idea, in an animal study, STN DBS decreased the reinforcingproperties of cocaine, as indicated by a downshift of the dose–response curve andreduced responding for cocaine on a progressive ratio schedule (Rouaud et al.2009). In contrast, in the same study, responding for sucrose on a progressive ratioschedule was increased. Similarly, STN DBS decreased preference for thecocaine-associated compartment, but increased preference for the food-associatedcompartment (Rouaud et al. 2009). These results suggest that STN DBS specifi-cally attenuates the rewarding and motivational properties of cocaine, but at thesame time enhances the motivational value of ‘‘natural rewards’’ such as sucrose.These findings correspond with decreased motivation for cocaine and increasedmotivation for sucrose following STN lesions (Baunez et al. 2005).

14.5.3 Lateral Habenula

Combined unilateral low-frequency and high-frequency stimulation of the lateralhabenula (LHb) applied before the start of self-administration decreased cocaine-taking on a fixed-ratio schedule with reinforcement after every response. Inaddition, DBS applied 1 day before the first extinction session reduced cocaine-seeking during extinction training and drug-induced reinstatement, showing long-lasting effects of DBS on cocaine-seeking behaviour (Friedman et al. 2010).Additional experiments showed that chemical LHb lesioning increased cocaine-seeking during extinction training, in contrast with the effects of DBS. Moreover,combined LHb stimulation reduced sucrose self-administration, indicating thatDBS in this brain area does not have differential effects on natural and drugreinforcement (Friedman et al. 2011). This diminished interest in natural rewardsas caused by DBS within the LHb would clearly be an unfavourable side effect, inparticular since many drug addicts already experience anhedonic feelings or havewithdrawal-induced anhedonia.

14.5.4 Lateral Hypothalamus and Prefrontal Cortex

A somewhat different approach was used by Levy et al. (2007), who stimulatedeither the lateral hypothalamus (LH) or the PFC during 10 days of abstinence aftercocaine self-administration. On the day following this stimulation protocol, ratsdisplayed reduced responding on the previously cocaine associated lever. The

14 Manipulating Addictive Behaviour 153

second day following stimulation, rats that received PFC DBS displayed reducedmotivation as measured on a progressive ratio schedule (LH DBS was not testedon the second day). These results suggest that the stimulation effect is longer-lasting, and can exceed the stimulation period. This can possibly be explained byplasticity changes induced by DBS. Importantly, Levy et al. did not find evidencefor alterations in sucrose-seeking and the motivation to respond for sucrose on aprogressive ratio schedule following DBS of the LH and DBS of the PFC (Levyet al. 2007).

14.5.5 Insula

The insula could be a potential target for DBS in drug dependence that has notbeen investigated yet. Since lesions in the insula are correlated with decreasedcraving behaviour and not with detrimental effects on motivation for naturalrewards, exploration of DBS in this region would be highly interesting. Inacti-vation of the insula by a mixture of GABA agonists decreased nicotine self-administration under a fixed ratio as well as a progressive ratio. In addition,reinstatement of nicotine-seeking was attenuated (Forget et al. 2010). The largedensity of blood vessels in this brain area (Türe et al. 2000) may, however,complicate electrode implantation.

14.6 Conclusions and Future Directions

Altogether, DBS applied in rats to evaluate possible treatment options for drugaddiction has proven to be useful in exploring the efficacy of different brainregions and in the detection of possible side effects (e.g. effects on natural reward-seeking). The NAc has received the most attention in this regard, as it is a commonbrain region for DBS application in the clinic for treatment-resistant obsessive–compulsive disorder and depression.

Preclinical animal research has shown that mesocorticolimbic loops are involvedin several aspects of drug addiction and that stimulation of these areas can suc-cessfully reduce drug-taking and drug-seeking without influencing the value ofnatural reinforcers. Therefore, future animal studies employing DBS within thiscircuitry may be particularly relevant for our understanding of the underlyingneurobiological mechanisms of DBS. For example, it has been shown that local fieldpotential oscillations in the mPFC, lateral orbitofrontal cortex, mediodorsal thalamusand NAc core are affected by DBS of the NAc (McCracken and Grace 2009). A recentin vivo microdialysis study did not detect alterations in dopamine, serotonin ornoradrenaline release in the NAc core during stimulation in the same area (van Dijket al. 2011). Nonetheless, stimulation of the PFC has been shown to induce alterationsin glutamate receptors in the NAc and VTA (Levy et al. 2007), suggesting that the

154 R. M. C. Schippers et al.

glutamatergic pathway is modulated by DBS. Collectively, these findings suggestthat DBS produces a wide range of immediate and more long-lasting effects atmultiple levels of neuronal communication, initiated by both antidromic andorthodromic activation. A comparison between DBS studies and lesion/inactivationstudies reveals contradictory results. This confirms the notion that the effects of DBSare more widespread and do not solely result from local inhibition of neuronalactivity. For example, inactivation of the NAc shell has opposite effects on cocaine-primed drug-seeking as compared with DBS applied to this region (see earlier andMcFarland and Kalivas 2001; Vassoler et al. 2008).

Clearly, better insight into the effects of DBS on the neuronal connectivity isimportant to improve treatment strategies and to shed further light on the underlyingpathophysiological mechanisms of drug addiction. To this aim, applying DBS inwell-validated animal models can be a very powerful research tool. Targeted(pharmacological) inactivation of different brain regions during DBS might furtherelucidate the involvement of the efferent and afferent pathways of the stimulatedbrain region. Microdialysis, fast-scan cyclic voltammetry and synaptic proteomicsstudies can help to identify local molecular and cellular changes evoked by DBS.

References

Aguilar MA, Rodríguez-Arias M, Miñarro J (2009) Neurobiological mechanisms of thereinstatement of drug-conditioned place preference. Brain Res Rev 59:253–277

Baunez C, Dias C, Cador M, Amalric M (2005) The subthalamic nucleus exerts opposite controlon cocaine and ‘natural’ rewards. Nat Neurosci 8:484–489

Childs E, de Wit H (2009) Amphetamine-induced place preference in humans. Biol Psychiatry65:900–904

Davis WM, Smith SG (1976) Role of conditioned reinforcers in the initiation, maintenance andextinction of drug-seeking behavior. Pavlov J Biol Sci 11:222–236

De Vries TJ, de Vries W, Janssen MC, Schoffelmeer ANM (2005) Suppression of conditionednicotine and sucrose seeking by the cannabinoid-1 receptor antagonist SR141716A. BehavBrain Res 161:164–168

de Wit H (1996) Priming effects with drugs and other reinforcers. Exp Clin Psychopharmacol 4:5–10Di Pietro N, Mashhoon Y, Heaney C, Yager L, Kantak K (2008) Role of dopamine D1 receptors

in the prefrontal dorsal agranular insular cortex in mediating cocaine self-administration inrats. Psychopharmacology 200:81–91

Epstein D, Preston K, Stewart J, Shaham Y (2006) Toward a model of drug relapse: anassessment of the validity of the reinstatement procedure. Psychopharmacology 189:1–16

Everitt BJ, Belin D, Economidou D, Pelloux Y, Dalley JW, Robbins TW (2008) Neuralmechanisms underlying the vulnerability to develop compulsive drug-seeking habits andaddiction. Philos Trans R Soc Lond B Biol Sci 363:3125–3135

Fagerström K, Balfour DJ (2006) Neuropharmacology and potential efficacy of new treatmentsfor tobacco dependence. Expert Opin Investig Drugs 15:107–116

Forget B, Pushparaj A, Le Foll B (2010) Granular insular cortex inactivation as a noveltherapeutic strategy for nicotine addiction. Biol Psychiatry 68:265–271

Friedman A, Lax E, Dikshtein Y, Abraham L, Flaumenhaft Y, Sudai E, Ben-Tzion M, Ami-Ad L,Yaka R, Yadid G (2010) Electrical stimulation of the lateral habenula produces enduringinhibitory effect on cocaine seeking behavior. Neuropharmacology 59:452–459

14 Manipulating Addictive Behaviour 155

Friedman A, Lax E, Dikshtein Y, Abraham L, Flaumenhaft Y, Sudai E, Ben-Tzion M, Yadid G(2011) Electrical stimulation of the lateral habenula produces an inhibitory effect on sucroseself-administration. Neuropharmacology 60:381–387

Fuchs RA, Evans KA, Parker MC, See RE (2004) Differential involvement of the core and shellsubregions of the nucleus accumbens in conditioned cue-induced reinstatement of cocaineseeking in rats. Psychopharmacology (Berl) 176:459–465

Goldstein RZ, Volkow ND (2002) Drug addiction and its underlying neurobiological basis:neuroimaging evidence for the involvement of the frontal cortex. Am J Psychiatry 159:1642–1652

Heidbreder CA, Groenewegen HJ (2003) The medial prefrontal cortex in the rat: evidence for adorso-ventral distinction based upon functional and anatomical characteristics. NeurosciBiobehav Rev 27:555–579

Henderson MB, Green AI, Bradford PS, Chau DT, Roberts DW, Leiter JC (2010) Deep brainstimulation of the nucleus accumbens reduces alcohol intake in alcohol-preferring rats.Neurosurg Focus 29:E12

Hodos W (1961) Progressive ratio as a measure of reward strength. Science 134:943–944Jentsch JD, Taylor JR (1999) Impulsivity resulting from frontostriatal dysfunction in drug abuse:

implications for the control of behavior by reward-related stimuli. Psychopharmacology146:373–390

Kalivas PW, Volkow ND (2005) The neural basis of addiction: a pathology of motivation andchoice. Am J Psychiatry 162:1403–1413

Knapp CM, Tozier L, Pak A, Ciraulo DA, Kornetsky C (2009) Deep brain stimulation of the nucleusaccumbens reduces ethanol consumption in rats. Pharmacol Biochem Behav 92:474–479

Koob GF, Volkow ND (2010) Neurocircuitry of addiction. Neuropsychopharmacology 35:217–238Kuhn J, Lenartz D, Huff W, Lee S, Koulousakis A, Klosterkoetter J, Sturm V (2007) Remission

of alcohol dependency following deep brain stimulation of the nucleus accumbens: valuabletherapeutic implications? J Neurol Neurosurg Psychiatry 78:1152–1153

Levy D, Shabat-Simon M, Shalev U, Barnea-Ygael N, Cooper A, Zangen A (2007) Repeatedelectrical stimulation of reward-related brain regions affects cocaine but not ‘‘natural’’reinforcement. J Neurosci 27:14179–14189

Liu HY, Jin J, Tang JS, Sun WX, Jia H, Yang XP, Cui JM, Wang CG (2008) Chronic deep brainstimulation in the rat nucleus accumbens and its effect on morphine reinforcement. AddictBiol 13:40–46

Lu L, Ceng X, Huang M (2000) Corticotropin-releasing factor receptor type I mediates stress-induced relapse to opiate dependence in rats. NeuroReport 11:2373–2378

Mantione M, van de BW, Schuurman PR, Denys D (2010) Smoking cessation and weight lossafter chronic deep brain stimulation of the nucleus accumbens: therapeutic and researchimplications: case report. Neurosurgery 66:E218

McCracken CB, Grace AA (2009) Nucleus accumbens deep brain stimulation produces region-specific alterations in local field potential oscillations and evoked responses in vivo.J Neurosci 29:5354–5363

McFarland K, Kalivas PW (2001) The circuitry mediating cocaine-induced reinstatement of drug-seeking behavior. J Neurosci 21:8655–8663

Mueller D, Stewart J (2000) Cocaine-induced conditioned place preference: reinstatement bypriming injections of cocaine after extinction. Behav Brain Res 115:39–47

Müller UJ, Sturm V, Voges J, Heinze HJ, Galazky I, Heldmann M, Scheich H, Bogerts B (2009)Successful treatment of chronic resistant alcoholism by deep brain stimulation of nucleusaccumbens: first experience with three cases. Pharmacopsychiatry 42:288–291

Naqvi NH, Bechara A (2010) The insula and drug addiction: an interoceptive view of pleasure,urges, and decision-making. Brain Struct Funct

Naqvi NH, Rudrauf D, Damasio H, Bechara A (2007) Damage to the insula disrupts addiction tocigarette smoking. Science 315:531–534

O’Brien CP, Gardner EL (2005) Critical assessment of how to study addiction and its treatment:Human and non-human animal models. Pharmacol Ther 108:18–58

156 R. M. C. Schippers et al.

O’Connor EC, Chapman K, Butler P, Mead AN (2011) The predictive validity of the rat self-administration model for abuse liability. Neurosci Biobehav Rev 35:912–938

Peters J, LaLumiere RT, Kalivas PW (2008) Infralimbic prefrontal cortex is responsible forinhibiting cocaine seeking in extinguished rats. J Neurosci 28:6046–6053

Rouaud T, Lardeux S, Panayotis N, Paleressompoulle D, Cador M, Baunez C (2009) Reducingthe desire for cocaine with subthalamic nucleus deep brain stimulation. Proc Natl Acad SciU S A 107:1196–1200

Sanchez CJ, Sorg BA (2001) Conditioned fear stimuli reinstate cocaine-induced conditionedplace preference. Brain Res 908:86–92

See RE, Fuchs RA, Ledford CC, McLaughlin J (2003) Drug addiction, relapse, and the amygdala.Ann N Y Acad Sci 985:294–307

Shaham Y, Stewart J (1996) Effects of opioid and dopamine receptor antagonists on relapseinduced by stress and re-exposure to heroin in rats. Psychopharmacology 125:385–391

Shaham Y, Erb S, Stewart J (2000) Stress-induced relapse to heroin and cocaine seeking in rats:a review. Brain Res Rev 33:13–33

Shaham Y, Shalev U, Lu L, de Wit H, Stewart J (2003) The reinstatement model of drug relapse:history, methodology and major findings. Psychopharmacology 168:3–20

Shufman EN, Porat S, Witztum E, Gandacu D, Bar-Hamburger R, Ginath Y (1994) The efficacyof naltrexone in preventing reabuse of heroin after detoxification. Biol Psychiatry 35:935–945

Spragg SDS (1940) Morphine addiction in chimpanzees. Comp Psychol Monogr 15:1–132Thompson T, Schuster CR (1964) Morphine self-administration, food-reinforced, and avoidance

behaviors in rhesus-monkeys. Psychopharmacologia 5:87–94Türe U, Yasargil MG, Al-Mefty O, Yasargil DCH (2000) Arteries of the insula. J Neurosurg

92:676–687Tzschentke TM (2000) The medial prefrontal cortex as a part of the brain reward system. Amino

Acids 19:211–219Van De Werd HJ, Uylings HB (2008) The rat orbital and agranular insular prefrontal cortical

areas: a cytoarchitectonic and chemoarchitectonic study. Brain Struct Funct 212:387–401Van den Oever MC, Spijker S, Smit AB, De Vries TJ (2010) Prefrontal cortex plasticity

mechanisms in drug seeking and relapse. Neurosci Biobehav Rev 35:276–284van Dijk A, Mason O, Klompmakers AA, Feenstra MGP, Denys D (2011) Unilateral deep brain

stimulation in the nucleus accumbens core does not affect local monoamine release.J Neurosci Methods 202:113–118

Vassoler FM, Schmidt HD, Gerard ME, Famous KR, Ciraulo DA, Kornetsky C, Knapp CM,Pierce RC (2008) Deep brain stimulation of the nucleus accumbens shell attenuates cocainepriming-induced reinstatement of drug seeking in rats. J Neurosci 28:8735–8739

Weeks JR (1962) Experimental morphine addiction—method for automatic intravenousinjections in unrestrained rats. Science 138:143–144

Witjas T, Baunez C, Henry JM, Delfini M, Regis J, Cherif AA, Peragut JC, Azulay JP (2005)Addiction in Parkinson’s disease: impact of subthalamic nucleus deep brain stimulation.Mov Disord 20:1052–1055

14 Manipulating Addictive Behaviour 157

Chapter 15Neuropsychiatric Side Effects of DeepBrain Stimulation in Parkinson’s Disease

Christine Daniels and Jens Volkmann

15.1 Psychiatric Symptoms During the Natural Courseof Parkinson’s Disease

Parkinson’s disease (PD) is a neurodegenerative movement disorder characterizedby akinesia, tremor, rigidity, and postural instability. However, nonmotor symp-toms, including olfactory loss, cognitive decline, affective and behavioral disor-ders, and autonomic failure, have resulted in increasing interest of researchers andtherapists over the last few decades since these symptoms have a high impact onquality of life (Gómez-Esteban et al. 2011).

For several decades PD has been considered a model neurodegenerative dis-order, being largely confined to the dopaminergic striatal motor system. This viewhas been challenged by neuropathology studies by Braak et al. (2003) demon-strating a widespread neurodegeneration of cortical and subcortical brain areas inthe late stages of the disease, which develops in a typical temporal and spatialpattern. According to the Braak staging hypothesis, PD is characterized by pro-gressive loss of neurons, neuritis, and pathological protein deposits containing a-synuclein and ubiquitin (Lewy bodies) starting in the brain stem and laterspreading to the diencephalon, archicortex, and neocortex. This staging classifi-cation offers a comprehensive and conclusive concept of PD as an a-synuclein-opathy, which extends beyond the limits of motor pathways or a singleneurotransmitter system and considers nonmotor phenomena as part of the disease.

J. Volkmann (&)Chairman and Professor of Neurology, University of Würzburg,Josef-Schneider-Str.11, 97080 Würzburg, Germanye-mail: volkmann_j@klinik.uni-wuerzburg.de

C. DanielsDepartment of Neurology, University of Würzburg,Josef-Schneider-Str. 11, 97080 Würzburg, Germany

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_15, � Springer-Verlag Berlin Heidelberg 2012

159

However, whether the clinical heterogeneity of PD with a preponderance of certainnonmotor symptoms in some patients or particular motor symptoms in others isparalleled by a distinct distribution of Lewy body disease remains a matter ofdebate (Thobois et al. 2010).

The early (‘‘presymptomatic’’) stage (stages 1 and 2 of Braak) precedes themotoric symptoms by years to decades. It involves the dorsal IX/X motor nuclei,the olfactory bulb, the myenteric plexus, the intermediate reticular zone, caudalraphe nuclei, the gigantocellular reticular nucleus, and the ceruleus–subceruleusnucleus. Psychiatric symptoms may precede parkinsonian motor symptoms in anumber of patients and are believed to reflect the dysregulation of the reticularformation and brain stem serotonergic and noradrenergic systems (Wolters andBraak 2006). They can include sleep disorders (i.e., REM sleep behavior disorder,daytime sleepiness), depression, anhedonia, apathy, anxiety, minimal cognitivedeficits/executive dysfunction, and—in rare cases—psychotic symptoms.

In the intermediate (‘‘symptomatic’’) stage (stages 3 and 4 of Braak) the neuro-degenerative process expands to dopaminergic midbrain structures (particularly thepars compacta of the substantia nigra), the basal prosencephalon, and the mesen-cephalon. Because of the predominant involvement of the dopaminergic system, thecharacteristic motor symptoms occur and lead to the diagnosis PD; however, thepsychiatric symptoms of the early stage may also be exacerbated. The considerabledopamine deficiency may cause a ‘‘hypodopaminergic syndrome’’ characterized bydepression, anxiety, apathy, and anhedonia. This hypodopaminergic syndrome issupposed to reflect the dopaminergic ‘‘understimulation’’ of the mesolimbic systemand associative striatothalamocortical loops, whereas typical parkinsonian motorsymptoms reflect the dopamine deficiency in the motor striatothalamocortical loop.

In the late stage (stages 5 and 6 of Braak) the neurodegenerative process alsoinvolves neocortical sensory association areas and prefrontal, premotor, and pri-mary sensory and motor areas. This stage may be associated with a progressivecognitive decline, which results in dementia.

The neuropathological model of Braak et al. is difficult to reconcile with theclinical evolution of psychiatric symptoms in PD, because disease-related psy-chiatric symptoms in the intermediate and late stages of PD are rarely observed intheir pure form today. Most patients receive a symptomatic drug therapy from anearly motor stage on, and these dopaminergic drugs themselves may cause psy-chiatric problems. Moreover, subtle behavioral changes or even psychiatricsymptoms in the early stage may be linked only retrospectively to PD (as not yetdiagnosed) and are therefore likely underreported.

In practice, psychiatric adverse effects of antiparkinsonian medication or sur-gical treatment (e.g., deep brain stimulation, DBS) are of greater clinical relevanceand will be covered by this chapter. The effects of dopaminergic medication havebeen described elsewhere, so we will focus on the main neuropsychiatric symp-toms in patients with PD that can occur, worsen, or improve after DBS. DBS is ananatomically defined intervention with a reversible mode of action, and has helpedus to study the behavioral effects of a focal neuromodulation of basal gangliacircuits experimentally. These studies have provided better understanding of how

160 C. Daniels and J. Volkmann

neurodegeneration, abnormal functioning of basal ganglia circuits, and exogenousdopaminergic replacement therapy may interact to produce the very heterogeneouspsychiatric presentations of PD. Because of limited data concerning DBS of theinternal segment of the globus pallidus (GPi) in PD, we will focus mainly on DBSof the subthalamic nucleus (STN) (Fig. 15.1) and try to cover symptoms associ-ated with GPi DBS only as far as sufficient evidence is available.

15.2 Psychiatric Symptoms Associated with Deep BrainStimulation and Dopaminergic Drugs

Deep brain stimulation (DBS) of the STN has been established as a highlyeffective treatment option for motor fluctuations and dyskinesia in advanced PD.The neurostimulation effect is closely correlated to the levodopa responsiveness ofmotor symptoms. Therefore, successful STN DBS reduces the need for dopami-nergic drugs by approximately 60–70 %. In a minority of patients, the treatmentwith levodopa or dopamine agonists may be entirely discontinued after STN DBS(Deuschl et al. 2006; Weaver et al. 2009).

Psychiatric side effects after a DBS operation are common. In this context, onehas to distinguish between neuropsychiatric symptoms (1) in the adaption phase,which includes the first 12 months after the DBS operation, and (2) in the long-term treatment phase more than 1 year after surgery. Stimulation parameters and

Fig. 15.1 The STN is subdivided into a large dorsolateral motor territory, a ventromedialassociative territory, and a medial limbic territory. Each territory receives inputs from differentareas of the cerebral cortex and provides output to different target nuclei, including the internalsegment of the globus pallidus (GPi), external segment of the globus pallidus (GPe), substantianigra pars reticulata (SNr), and ventral pallidum. These input–output interactions provide parallelcontrol of motor, oculomotor, cognitive, and emotional functions independently of ‘‘indirect’’pathways via the striatum and GPe (from Benarroch 2008)

15 Neuropsychiatric Side Effects of Deep Brain Stimulation 161

medication need to be adjusted during the first few months after the surgery andpsychiatric symptoms in this period often reflect the complex interaction betweendrug and neurostimulation treatment. In the long term, however, the stimulationparameters and medication remain largely stable, and psychiatric symptoms aremore likely the result of the ongoing neurodegenerative process. However, theassignment to any single cause is often complicated by additional contributingfactors, such as the exact electrode location, preoperative neuropsychiatric status,and individual personality traits. Moreover, the adjustment to a new life situationwith less parkinsonian motor disability itself may cause a psychological burden topatients and contribute to mood or drive problems.

The functional subdivision of the basal ganglia into regions subserving thedifferent basal ganglia–thalamocortical circuits (motor, limbic, associative) is anestablished concept, which is based on evidence derived from primate morpho-logical and functional studies and clinical observations in humans (Temel et al.2005; Hamani et al. 2004). According to this concept, the primate STN is segre-gated into three functional territories: the dorsolateral somatomotor part, theventromedial associative part, and the medial limbic part (Benarroch 2008; Parentand Hazrati 1995). With the hypothesis of the basal ganglia as a global ‘‘go/no-gosystem’’ and the STN as a potent regulator of this system, most of the motor andnonmotor symptoms of PD and the subsequent treatment effects (especially ofhigh-frequency stimulation of the STN) can be explained in an integrative model(Volkmann et al. 2010). On the basis of this model, the tone of the different basalganglia loops is associated with distinct motor, cognitive, or affective symptoms,which can be influenced by either medication (globally) or neurostimulation(focally). For example, underactivity of the limbic loop leads to obsessive–com-pulsive behavior, depression, and apathy at one end (hypodopaminergic syndrome,see earlier) and overactivity to manic/hypomanic states, impulsivity, hyperactivity,and attentional deficit. The dosing function of either medication or stimulationwould follow a U shape, with an intermediate optimal dose range and detrimentaleffects on movements, mood, or behavior with either overdosing or underdosing.Because the dopaminergic denervation may be uneven across the different loops,optimal dosing of a single loop may still be associated with overdosing or un-derdosing of the other circuits.

Dopaminergic neurons operate in either a tonic or a phasic firing mode underphysiological conditions. Phasic release of dopamine in relation to external stimuliis a characteristic feature of neurons in the ventral tegmental area belonging to thedopaminergic reward circuit. Unphysiological chronic pulsatile stimulation byshort-acting dopaminergic drugs may cause clinical sensitization phenomena suchas dyskinesia, impulse control disorders (ICD), or a dopamine dysregulationsyndrome (DDS). Subthalamic neurostimulation may initially exacerbate dyski-nesia, but with long-term treatment dyskinesia disappears and can no longer beinduced by preoperative doses of dopaminergic medication. This has been taken asevidence of desensitization by tonic electrical stimulation (Bejjani et al. 2000).Whether STN DBS has a similar effect on nonmotor sensitization phenomenaremains a matter of intense debate (Castelli et al. 2008).

162 C. Daniels and J. Volkmann

15.2.1 Mania and Depression

15.2.1.1 Mania

Acute manic states can occur immediately after surgery or after the start of high-frequency stimulation. In these cases, a direct effect of surgical microlesioning orstimulation may be assumed. Single cases or small case series have describedacute mania in association with stimulation of electrode contacts that were prob-ably placed not exactly within the motor region of the STN, but in or directlyneighboring the ventral tegmental area, the anteriomedial part of the STN, or thesubstantia nigra. However, a case of acute mania associated with high-frequencystimulation through an optimally placed contact within the STN has also beenreported (Kulisevsky et al. 2002; Mallet et al. 2007; Raucher-Chene et al. 2008;Mandat et al. 2006; Ulla et al. 2011; Herzog et al. 2003).

In the adaptation or long-term phase, euphoric mood is rare and typicallyindicates an imbalance of neurostimulation and dopaminergic medication. In thesecases, the doses of dopaminergic drugs were often not reduced sufficiently(Deuschl et al. 2006; Weaver et al. 2009). A DDS may pose a particular risk forsuch manic episodes, because patients tend to disregard reduced dose prescrip-tions. A preexisting bipolar disorder may be another reason for manic states in thepostoperative period.

15.2.1.2 Depression

Postoperative depression was found in upto 25 % of patients after STN DBS(Berney et al. 2002), and symptoms typically emerge within the first 2 months(Vicente et al. 2009 Jul; Houeto et al. 2002). However, acute and reversibledepression in close association with the onset or change of neurostimulation hasalso been described (Weintraub 2009; Weintraub et al. 2010). As discussed formania, a costimulation of adjacent brain regions (e.g., substantia nigra) via currentspread or due to misplacement of electrodes is discussed alternatively to a directstimulation effect of the motor region of the STN (Bejjani et al. 1999; Tommasiet al. 2008). In the COMPARE trial, 22 patients with PD rated themselves onaverage ‘‘less happy,’’ ‘‘less energetic,’’ and ‘‘more confused’’ when stimulatedthrough a ventral contact (below the optimal motor target) compared with stim-ulation of a more distal contact within the sensorimotor region of the STN. Nopatients in this trial, however, exhibited acute depressive symptoms (Okun et al.2009). In contrast, another study described an increase in positive emotion withventral compared with dorsal contact stimulation (Greenhouse et al. 2011). In linewith these clinical findings in humans, bilateral high-frequency stimulation of theSTN in rats has been found to inhibit the firing of serotonin neurons in the dorsalraphe nucleus, but not of neighboring non-serotonin neurons (Temel et al. 2007).Taken together, the few published cases of acute depression during DBS probably

15 Neuropsychiatric Side Effects of Deep Brain Stimulation 163

represent rare effects of stimulation affecting a still-undetermined region outsidethe STN motor region.

Transient depressive symptoms are common within the first few months after aDBS operation and are often associated with the reduction of dopaminergic medi-cation (Vicente et al. 2009; Houeto et al. 2002). In contrast, during the later adap-tation phase (6 months or longer), clinical trial data strongly argue against a generaldetrimental effect of STN DBS on mood. In two randomized controlled studies aslight antidepressive effect and a relevant anxiolytic effect were found in the STN-DBS-treated group. However, individual patients in both the DBS group and themedication group experienced relevant depression against the group trend, or in rarecases transient mania (Deuschl et al. 2006; Weaver et al. 2009; Witt et al. 2008).

With respect to mood problems in the long-term course after STN DBS, thehigh general risk of depression in PD on the order of 30–40 % needs to be takeninto account (Menza et al. 2009). Therefore, the risk of any PD patient experi-encing disease-related depression has to be separated from an effect of the STNDBS itself. One case–control study and one prospective study with a 3-year fol-low-up suggest that STN DBS does not lead to relevant modifications of mood,anxiety, and personality over 3 years (Castelli et al. 2008; Kaiser et al. 2008).Comparable results have been found after 5 years (Schupbach et al. 2005; Kracket al. 2003). However, in a randomized controlled study comparing STN DBS andGPi DBS a slight worsening of depressive symptoms as assessed by a self-eval-uation scale was found in the STN group, whereas patients with GPi DBS expe-rienced a slight improvement of depressive symptoms (Follett et al. 2010). Ingeneral, depression has been less of a concern after GPi DBS as indicated byseveral larger trials (Volkmann et al. 2001; Anderson et al. 2005).

15.2.2 Apathy

Apathy is a frequent adverse effect of STN DBS, both in the early postoperativeperiod (Houeto et al. 2002; Drapier et al. 2006; Czernecki et al. 2008) and in thelong-term follow-up (Krack et al. 2003; Troster 2009). The precise incidence ofapathy after STN DBS is unknown; however, in one study, the proportion ofpatients with PD who exhibited apathy was documented to be 8.7 % before sur-gery and 24.6 % in the third postoperative year (Funkiewiez et al. 2004). In aprospective study with PD patients undergoing STN DBS and postoperativewithdrawal of dopamine agonists within 2 weeks, apathy occurred after a mean of4.7 (3.3–8.2) months in 34 of 63 patients and was reversible in half of these after12 months (Thobois et al. 2010).

Apathy has been assigned to the spectrum of hypodopaminergic nonmotorsymptoms of PD, because in some patients this symptom responds to dopami-nergic treatment (Chatterjee and Fahn 2002; Marin et al. 1995). It has beenhypothesized that dopaminergic mesolimbic denervation may be the underlyingneuropathological correlate, which clinically manifests itself as apathy during the

164 C. Daniels and J. Volkmann

postoperative reduction of dopaminergic medication. However, it has also beensuggested that STN DBS may directly induce apathy via a limbic side-effectmechanism (Drapier et al. 2006; Temel et al. 2009) as PET data indicate that STNDBS might inadvertently modulate a frontal motivational network that is con-nected to the limbic and associative territories of the STN (Le Jeune et al. 2009). Inpractice, dopaminergic medication is attempted if it is tolerated in those patientsaffected by apathy after surgery. A beneficial response would support a hypo-dopaminergic syndrome. Reprogramming strategies are normally less effective,but may be tried as second-line treatment.

15.2.3 Anxiety

Anxiety is a frequent nonmotor symptom during off-periods in PD patients withlevodopa-associated motor fluctuations. In a controlled study patients with STNDBS had markedly lower Beck Anxiety Inventory scores after 6 months thanpatients receiving the best medical treatment (Witt et al. 2008). A comparablefinding was reported for the extended 2-year follow-up (Houeto et al. 2006).However, caution is warranted in the interpretation of this finding, since the BeckAnxiety Inventory includes several items with a strong somatic connection (suchas an inability to relax and tremor of the hands) that improve considerably afterDBS. Nevertheless, reductions in anxiety following STN DBS have been reportedin studies using other scales and measures of state anxiety during stimulationchallenges (Funkiewiez et al. 2003; Daniele et al. 2003). Whether reductions inanxiety occur secondary to improvements in motor fluctuations or are a genuinenonmotor effect of STN DBS remains unknown. An intrinsic role for the STN infear processing has been suggested by two studies that demonstrated specificdeficits in recognizing fearful facial expressions or film sequences during STNDBS (Biseul et al. 2005; Vicente et al. 2009).

15.2.4 Anger

Acute anger and aggression have been described in two case reports. In one case,acute explosive aggressive behavior was induced by left hemispheric stimulationonly. The left electrode was placed properly within the dorsolateral STN with allfour contacts in this case (Sensi et al. 2004). In another case, intraoperativeaggression was induced by stimulation of the posterior hypothalamus (Bejjaniet al. 2002), indicating that current spread to hypothalamic fiber connections inproximity to the STN could mediate this rare adverse effect of STN DBS.

In a study evaluating 195 patients before and after an STN DBS operation and56 patients before and after a GPi DBS operation for PD, both targets wereassociated with significantly higher Visual Analog Mood Scale anger scores after

15 Neuropsychiatric Side Effects of Deep Brain Stimulation 165

4–6 months as compared with DBS of the ventrolateral thalamus for essentialtremor (Burdick et al. 2011). Anger score changes in STN and GPi patients wereassociated with higher numbers of microelectrode passes; therefore, the authorssuggested a lesional effect rather than a stimulation-induced one. This hypothesiswas supported by the finding that Visual Analog Mood Scale anger scores did notchange irrespective of whether the stimulator was on or off.

15.2.5 Impulse Control Disorders

Impulse control disorders (ICDs) reported in patients with PD comprise patho-logical gambling, compulsive shopping, hypersexuality, and binge eating. Theoverall prevalence of ICDs is estimated to be around 13 % among PD patients(Weintraub 2009). They are more frequent among males, with a young age ofonset and high novelty-seeking personality traits. A cross-sectional study in 3,090patients with PD and dopaminergic treatment showed that ICDs were commoner inpatients treated with dopamine agonists than in those without treatment (Wein-traub et al. 2010). After DBS these pathological behaviors may be exacerbated.Some reports have described the de novo appearance of ICDs after surgery;however, one needs to consider the possibility that these behaviors may have beenconcealed by the patient so as to be eligible for DBS treatment. Two reviewsanalyzed the frequency of ICDs after STN DBS and did not identify a significantdifference in ICDs between patients receiving dopaminergic medication andpatients receiving DBS (Broen et al. 2011; Demetriades et al. 2011). Case reportsand case series of patients with pathological gambling, hypersexuality, and com-pulsive shopping describe patients who dramatically improved after DBS as wellas patients who worsened or developed de novo symptoms. From these publica-tions, neither a clear beneficial effect nor an adverse effect of DBS can be derived.The reasons for these heterogenous outcomes may be found in the complexinteraction of the neurodegenerative pattern within and outside the basal ganglia,individual personality traits, long-term sensitization effects of dopaminergicmedication, and the topographic relation of the stimulating contact to the limbicSTN area in each individual patient.

15.2.6 Dopamine Dysregulation Syndrome

Dopamine dysregulation syndrome (DDS) can be understood as a vicious circle ofan addictive behavior, in which a disturbance of impulse control, craving fordopaminergic drugs, and overdosing of levodopa reinforce themselves. Dataregarding the effect of DBS on DDS are limited. In a case series, DDS remainedunimproved or worsened postoperatively in eight of 14 patients with preoperativeDDS and improved or resolved after bilateral STN DBS in six of 14 patients. In

166 C. Daniels and J. Volkmann

two patients, DDS developed for the first time after bilateral STN DBS (Lim et al.2009). However, individual patients may also improve dramatically after a DBSoperation. DDS is often associated with other ICDs, which may improve (orworsen) after DBS in an analogous manner (Witjas et al. 2005).

15.2.7 Punding

Punding is defined as an intense fascination with excessive, repetitive, and non-goal-oriented behaviors involving acts that can be simple (cleaning, sorting, orordering of objects) or complex (painting, creative writing, or using or repairingcomputers). The manifestations of punding differ across individuals with PD andare often associated with previously learned behaviors.

In a retrospective study, 24 consecutive PD outpatients who underwent STN DBSwere evaluated by structured interview. Five (20.8 %) of the 24 subjects wereidentified as punders (three men, two women). The punders were comparable to thenonpunders in terms of clinical and demographic factors, but differed statisticallywith regard to the length of time from DBS electrode implantation (mean duration ofDBS 3.2 years among punders vs. 5.16 years among nonpunders) (Pallanti et al.2010). In their case series, Lim et al. (2009) reported preoperative punding in 11patients, of which four showed an improvement (STN DBS), whereas seven patientsdid not show improvement of preexistent punding after a DBS operation (STN DBS,one with unilateral right-sided stimulation). As a striking finding, five of ten patientswho had ICD or DDS preoperatively developed de novo punding after a DBSoperation (three STN DBS, two GPi DBS). Despite limited data, the rate of pundersafter DBS seems to be high in view of the rates described for PD patients receivingdopaminergic treatment (1.4–14 %) and suggests that punding might be induced byDBS (Lim et al. 2009; Pallanti et al. 2010; Evans et al. 2004; Miyasaki 2007).However, the target structure remains unknown, particularly with regard to thepatients who developed punding after GPi DBS.

15.2.8 Suicide

A higher than expected rate of suicide has been reported among patients treated bySTN DBS, with a maximum peak in the first year after surgery. In a retrospectivesurvey of patients operated on at 55 movement disorder and surgical centers, the rateof suicides was 0.45 % (24/5,311), and the rate of attempted suicides was 0.90 % (48/5,311) (Voon et al. 2008). In the first postoperative year, the suicide rates (0.26 %)were approximately 12 times higher than the expected (age-, gender-, and country-adjusted) WHO suicide rates and remained slightly elevated at the fourth postop-erative year (0.04 %). This increase in the suicide rate in surgically treated PD is evenmore remarkable considering the up to tenfold lower suicide rate in medically treated

15 Neuropsychiatric Side Effects of Deep Brain Stimulation 167

PD patients compared with the general population (Myslobodsky et al. 2001). Sui-cides or suicide attempts have been observed in PD patients without previous orcurrent psychopathological symptoms, underlining the impulsive nature of thisbehavior (Rodrigues et al. 2010). A tempting presumption is that postoperativesuicide with regard to DBS is part of the ICD spectrum that is caused by neurosti-mulation interfering with the normal inhibitory role of the STN in decision-making.During the postoperative adjustment of stimulation parameters and dopaminergicmedication, the impulse control mechanisms might be particularly instable,reflecting the high incidence of suicides in the first postoperative year.

15.3 Deep Brain Stimulation as a Treatment Optionfor Psychiatric Symptoms in Advanced Parkinson’sDisease?

Owing to the high number of PD patients treated by DBS within the last twodecades, our experience with regard to the nonmotor effects of this therapy hasincreased. There is some evidence that DBS of the STN could be beneficial fordisease- or medication-related psychiatric symptoms in PD patients. Reducing theneed for dopaminergic medication by STN DBS, for example, could be helpful inpatients with drug-induced psychosis, DDS, or ICDs. Individual outcomes, how-ever, are hard to predict and depend on a complex interaction of DBS, medication,postoperative psychosocial (mal-)adaption, the preexisting psychiatric status, andpremorbid personality traits.

The improvement of DDS in individual patients after DBS seems to beunderstandable, as postoperative reduction of the doses of dopaminergic drugsmay help to break the vicious cycle of overdosing and impulsive drug seeking. Acomparable mechanism may be responsible for the ICD spectrum. A number ofcases and small case series have described improvements of ICDs after DBS (Limet al. 2009). However, as mentioned earlier, a comparable proportion of patientswith ICDs did not improve or worsened after implantation of electrodes for DBS.De novo development of pathological gambling, hypersexuality, and compulsiveshopping after STN DBS seem to be rare, but have been described in singlepatients (Lim et al. 2009; Romito et al. 2002; Halbig et al. 2009). In some of thesepatients with persistent (or worsened) ICDs after DBS, the postoperative reductionof dopaminergic medication later led to a marked improvement of the abnormalbehavior (Smeding et al. 2007).

On the other hand, there is some evidence that stereotyped behavior andobsessive–compulsive disorders can be treated by DBS of the ventral striatum orthe anterior limbic part of the STN (Baup et al. 2008; Greenberg et al. 2006) andcases of PD patients with obsessive–compulsive behavior who experienced arelevant improvement after STN DBS have been reported (Fontaine et al. 2004;Mallet et al. 2002).

168 C. Daniels and J. Volkmann

We have previously suggested that impulsivity and compulsive behavior maybe understood as opposite physiological states of the limbic basal ganglia loop,which may be tuned by either medication or neurostimulation in a U-shaped dose–response manner (Fig. 15.2) (Volkmann et al. 2010). However, the physiologicaleffect of DBS may differ from that of dopaminergic medication, as stimulationworks continuously and is not pulsatile, which might help to prevent periodicoverstimulation of the limbic loops and subsequent sensitization phenomena(Castelli et al. 2008). Although DBS is not generally advised for hyperdopamin-ergic psychiatric symptoms in PD, on the basis of this conceptional model, wesuggest that future clinical trials should specifically target this group of patients.

15.4 Practical Recommendations

In conclusion, any physician working with DBS in PD should be aware of the widerange of neuropsychiatric manifestations which can occur at any stage of thetreatment. During the selection process for DBS, a thorough baseline of neuro-psychiatric symptoms associated with either the disease or its medical treatmentshould be established. This will later form an important basis for the interpretationof any postoperative psychiatric adverse effect. The patient should be told aboutthe intention of this preoperative screening, which is not meant to exclude the

Fig. 15.2 Hypothetical U function representing the effects of a modulation of motor, associative,and limbic function symptoms by dopaminergic medication, subthalamic nucleus (STN) deepbrain stimulation, or a combination of both. Excessive as well as insufficient activation states inthe specific loops impair motor functions, mood, and behavior and can induce clinical symptomsassociated with these domains (from Volkmann et al. 2010)

15 Neuropsychiatric Side Effects of Deep Brain Stimulation 169

patient from surgery, but to plan for a close psychiatric follow-up in those patientsat risk. The patient and the caregivers should be sensitized for the development orworsening of depressive symptoms, ICDs, and other psychiatric symptoms espe-cially during the period of postoperative adjustment to medication and stimulation.Controlled trials of stimulation and taking a thorough history of the relationbetween the onset of psychiatric symptoms and medication changes can help todifferentiate the immediate and direct effects of DBS from drug-withdrawal effectsin the postoperative period.

References

Anderson VC, Burchiel KJ, Hogarth P, Favre J, Hammerstad JP (2005) Pallidal vs subthalamicnucleus deep brain stimulation in Parkinson disease. Arch Neurol 62(4):554–560

Baup N, Grabli D, Karachi C, Mounayar S, Francois C, Yelnik J et al (2008) High-frequencystimulation of the anterior subthalamic nucleus reduces stereotyped behaviors in primates.J Neurosci 28(35):8785–8788

Bejjani BP, Damier P, Arnulf I, Thivard L, Bonnet AM, Dormont D et al (1999) Transient acutedepression induced by high-frequency deep-brain stimulation. N Engl J Med 340(19):1476–1480

Bejjani BP, Arnulf I, Demeret S, Damier P, Bonnet AM, Houeto JL et al (2000) Levodopa-induceddyskinesias in Parkinson’s disease: is sensitization reversible? Ann Neurol 47(5):655–658

Bejjani BP, Houeto JL, Hariz M, Yelnik J, Mesnage V, Bonnet AM et al (2002) Aggressivebehavior induced by intraoperative stimulation in the triangle of Sano. Neurology 59(9):1425–1427

Benarroch EE (2008) Subthalamic nucleus and its connections: anatomic substrate for thenetwork effects of deep brain stimulation. Neurology 70(21):1991–1995

Berney A, Vingerhoets F, Perrin A, Guex P, Villemure JG, Burkhard PR et al (2002) Effect onmood of subthalamic DBS for Parkinson’s disease: a consecutive series of 24 patients.Neurology 59(9):1427–1429

Biseul I, Sauleau P, Haegelen C, Trebon P, Drapier D, Raoul S et al (2005) Fear recognition isimpaired by subthalamic nucleus stimulation in Parkinson’s disease. Neuropsychologia43(7):1054–1059

Braak H, Del Tredici K, Rub U et al (2003) Staging of brain pathology related to sporadicParkinson’s disease. Neurobiol Aging 24:197–211

Broen M, Duits A, Visser-Vandewalle V, Temel Y, Winogrodzka A (2011) Impulse control andrelated disorders in Parkinson’s disease patients treated with bilateral subthalamic nucleusstimulation: a review. Parkinsonism Relat Disord 17(6):413–417

Burdick AP, Foote KD, Wu S, Bowers D, Zeilman P, Jacobson CE et al (2011) Do patient’s getangrier following STN, GPi, and thalamic deep brain stimulation. Neuroimage 54(Suppl 1):S227–S232

Castelli L, Zibetti M, Rizzi L, Caglio M, Lanotte M, Lopiano L (2008) Neuropsychiatricsymptoms three years after subthalamic DBS in PD patients: a case-control study. J Neurol255(10):1515–1520

Chatterjee A, Fahn S (2002) Methylphenidate treats apathy in Parkinson’s disease. J Neuropsy-chiatry Clin Neurosci 14(4):461–462

Czernecki V, Schupbach M, Yaici S, Levy R, Bardinet E, Yelnik J et al (2008) Apathy followingsubthalamic stimulation in Parkinson disease: a dopamine responsive symptom. Mov Disord23(7):964–969

Daniele A, Albanese A, Contarino MF, Zinzi P, Barbier A, Gasparini F et al (2003) Cognitive andbehavioural effects of chronic stimulation of the subthalamic nucleus in patients withParkinson’s disease. J Neurol Neurosurg Psychiatry 74(2):175–182

170 C. Daniels and J. Volkmann

Demetriades P, Rickards H, Cavanna AE (2011) Impulse control disorders following deep brainstimulation of the subthalamic nucleus in Parkinson’s disease: clinical aspects. Parkinsons Dis20(2011):658415

Deuschl G, Schade-Brittinger C, Krack P, Volkmann J, Schafer H, Botzel K et al (2006) A randomizedtrial of deep-brain stimulation for Parkinson’s disease. N Engl J Med 355(9):896–908

Drapier D, Drapier S, Sauleau P, Haegelen C, Raoul S, Biseul I et al (2006) Does subthalamicnucleus stimulation induce apathy in Parkinson’s disease? J Neurol 253(8):1083–1091

Evans AH, Katzenschlager R, Paviour D, O’Sullivan JD, Appel S, Lawrence AD et al (2004)Punding in Parkinson’s disease: its relation to the dopamine dysregulation syndrome. MovDisord 19(4):397–405

Follett KA, Weaver FM, Stern M, Hur K, Harris CL, Luo P et al (2010) Pallidal versus subthalamicdeep-brain stimulation for Parkinson’s disease. N Engl J Med 362(22):2077–2091

Fontaine D, Mattei V, Borg M, von Langsdorff D, Magnie MN, Chanalet S et al (2004) Effect ofsubthalamic nucleus stimulation on obsessive-compulsive disorder in a patient with Parkinsondisease. Case report. J Neurosurg 100(6):1084–1086

Funkiewiez A, Ardouin C, Krack P, Fraix V, Van Blercom N, Xie J et al (2003) Acutepsychotropic effects of bilateral subthalamic nucleus stimulation and levodopa in Parkinson’sdisease. Mov Disord 18(5):524–530

Funkiewiez A, Ardouin C, Caputo E, Krack P, Fraix V, Klinger H et al (2004) Long term effectsof bilateral subthalamic nucleus stimulation on cognitive function, mood, and behaviour inParkinson’s disease. J Neurol Neurosurg Psychiatry 75(6):834–839

Gómez-Esteban JC, Tijero B, Somme J, Ciordia R, Berganzo K, Rouco I, Bustos JL, Valle MA,Lezcano E, Zarranz JJ (2011) Impact of psychiatric symptoms and sleep disorders on thequality of life of patients with Parkinson’s disease. J Neurol 258(3):494–499

Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF et al (2006) Three-yearoutcomes in deep brain stimulation for highly resistant obsessive-compulsive disorder.Neuropsychopharmacology 31(11):2384–2393

Greenhouse I, Gould S, Houser M, Hicks G, Gross J, Aron AR (2011) Stimulation at dorsal andventral electrode contacts targeted at the subthalamic nucleus has different effects on motorand emotion functions in Parkinson’s disease. Neuropsychologia 49(3):528–534

Halbig TD, Tse W, Frisina PG, Baker BR, Hollander E, Shapiro H et al (2009) Subthalamic deepbrain stimulation and impulse control in Parkinson’s disease. Eur J Neurol 16(4):493–497

Hamani C, Saint-Cyr JA, Fraser J, Kaplitt M, Lozano AM (2004) The subthalamic nucleus in thecontext of movement disorders. Brain 127(Pt 1):4–20

Herzog J, Reiff J, Krack P, Witt K, Schrader B, Muller D et al (2003) Manic episode withpsychotic symptoms induced by subthalamic nucleus stimulation in a patient with Parkinson’sdisease. Mov Disord 18(11):1382–1384

Houeto JL, Mesnage V, Mallet L, Pillon B, Gargiulo M, du Moncel ST et al (2002) Behaviouraldisorders, Parkinson’s disease and subthalamic stimulation. J Neurol Neurosurg Psychiatry72(6):701–707

Houeto JL, Mallet L, Mesnage V, du Tezenas MS, Behar C, Gargiulo M et al (2006) Subthalamicstimulation in Parkinson disease: behavior and social adaptation. Arch Neurol 63(8):1090–1095

Kaiser I, Kryspin-Exner I, Brucke T, Volc D, Alesch F (2008) Long-term effects of STN DBS onmood: psychosocial profiles remain stable in a 3-year follow-up. BMC Neurol 8:43

Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C et al (2003) Five-yearfollow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease.N Engl J Med 349(20):1925–1934

Kulisevsky J, Berthier ML, Gironell A, Pascual-Sedano B, Molet J, Pares P (2002) Maniafollowing deep brain stimulation for Parkinson’s disease. Neurology 59(9):1421–1424

Le Jeune F, Drapier D, Bourguignon A, Peron J, Mesbah H, Drapier S et al (2009) Subthalamic nucleusstimulation in Parkinson disease induces apathy: a PET study. Neurology 73(21):1746–1751

Lim SY, O’Sullivan SS, Kotschet K, Gallagher DA, Lacey C, Lawrence AD et al (2009)Dopamine dysregulation syndrome, impulse control disorders and punding after deep brainstimulation surgery for Parkinson’s disease. J Clin Neurosci 16(9):1148–1152

15 Neuropsychiatric Side Effects of Deep Brain Stimulation 171

Mallet L, Mesnage V, Houeto JL, Pelissolo A, Yelnik J, Behar C et al (2002) Compulsions,Parkinson’s disease, and stimulation. Lancet 360(9342):1302–1304

Mallet L, Schupbach M, N’Diaye K, Remy P, Bardinet E, Czernecki V et al (2007) Stimulation ofsubterritories of the subthalamic nucleus reveals its role in the integration of the emotionaland motor aspects of behavior. Proc Natl Acad Sci U S A 104(25):10661–10666

Mandat TS, Hurwitz T, Honey CR (2006) Hypomania as an adverse effect of subthalamic nucleusstimulation: report of two cases. Acta Neurochir (Wien) 148(8):895–897; discussion 8

Marin RS, Fogel BS, Hawkins J, Duffy J, Krupp B (1995) Apathy: a treatable syndrome.J Neuropsychiatry Clin Neurosci 7(1):23–30

Menza M, Dobkin RD, Marin H, Mark MH, Gara M, Buyske S et al (2009) A controlled trial ofantidepressants in patients with Parkinson disease and depression. Neurology 72(10):886–892

Miyasaki JM, Al Hassan K, Lang AE, Voon V (2007) Punding prevalence in Parkinson’s disease.Mov Disord 22(8):1179–1181

Myslobodsky M, Lalonde FM, Hicks L (2001) Are patients with Parkinson’s disease suicidal? JGeriatr Psychiatry Neurol 14(3):120–124

Okun MS, Fernandez HH, Wu SS, Kirsch-Darrow L, Bowers D, Bova F et al (2009) Cognitionand mood in Parkinson’s disease in subthalamic nucleus versus globus pallidus interna deepbrain stimulation: the COMPARE trial. Ann Neurol 65(5):586–595

Pallanti S, Bernardi S, Raglione LM, Marini P, Ammannati F, Sorbi S et al (2010) Complexrepetitive behavior: punding after bilateral subthalamic nucleus stimulation in Parkinson’sdisease. Parkinsonism Relat Disord 16(6):376–380

Parent A, Hazrati LN (1995) Functional anatomy of the basal ganglia. II. The place ofsubthalamic nucleus and external pallidum in basal ganglia circuitry. Brain Res Brain Res Rev20(1):128–154

Raucher-Chene D, Charrel CL, de Maindreville AD, Limosin F (2008) Manic episode withpsychotic symptoms in a patient with Parkinson’s disease treated by subthalamic nucleusstimulation: improvement on switching the target. J Neurol Sci 273(1–2):116–117

Rodrigues AM, Rosas MJ, Gago MF, Sousa C, Fonseca R, Linhares P et al (2010) Suicide attemptsafter subthalamic nucleus stimulation for Parkinson’s disease. Eur Neurol 63(3):176–179

Romito LM, Raja M, Daniele A, Contarino MF, Bentivoglio AR, Barbier A et al (2002) Transientmania with hypersexuality after surgery for high frequency stimulation of the subthalamicnucleus in Parkinson’s disease. Mov Disord 17(6):1371–1374

Schupbach WM, Chastan N, Welter ML, Houeto JL, Mesnage V, Bonnet AM et al (2005)Stimulation of the subthalamic nucleus in Parkinson’s disease: a 5 year follow up. J NeurolNeurosurg Psychiatry 76(12):1640–1644

Sensi M, Eleopra R, Cavallo MA, Sette E, Milani P, Quatrale R et al (2004) Explosive-aggressivebehavior related to bilateral subthalamic stimulation. Parkinsonism Relat Disord10(4):247–251

Smeding HM, Goudriaan AE, Foncke EM, Schuurman PR, Speelman JD, Schmand B (2007)Pathological gambling after bilateral subthalamic nucleus stimulation in Parkinson disease.J Neurol Neurosurg Psychiatry 78(5):517–519

Temel Y, Visser-Vandewalle V, Aendekerk B, Rutten B, Tan S, Scholtissen B et al (2005) Acuteand separate modulation of motor and cognitive performance in parkinsonian rats by bilateralstimulation of the subthalamic nucleus. Exp Neurol 193(1):43–52

Temel Y, Boothman LJ, Blokland A, Magill PJ, Steinbusch HW, Visser-Vandewalle V et al (2007)Inhibition of 5-HT neuron activity and induction of depressive-like behavior by high-frequencystimulation of the subthalamic nucleus. Proc Natl Acad Sci U S A 104(43):17087–17092

Temel Y, Tan S, Vlamings R, Sesia T, Lim LW, Lardeux S et al (2009) Cognitive and limbiceffects of deep brain stimulation in preclinical studies. Front Biosci 14:1891–1901

Thobois S, Ardouin C, Lhommee E, Klinger H, Lagrange C, Xie J et al (2010) Non-motordopamine withdrawal syndrome after surgery for Parkinson’s disease: predictors andunderlying mesolimbic denervation. Brain 133(Pt 4):1111–1127

Tommasi G, Lanotte M, Albert U, Zibetti M, Castelli L, Maina G et al (2008) Transient acutedepressive state induced by subthalamic region stimulation. J Neurol Sci 273(1–2):135–138

172 C. Daniels and J. Volkmann

Troster AI (2009) Neuropsychology of deep brain stimulation in neurology and psychiatry. FrontBiosci 14:1857–1879

Ulla M, Thobois S, Llorca PM, Derost P, Lemaire JJ, Chereau-Boudet I et al (2011) Contact dependentreproducible hypomania induced by deep brain stimulation in Parkinson’s disease: clinical,anatomical and functional imaging study. J Neurol Neurosurg Psychiatry 82(6):607–614

Vicente S, Biseul I, Peron J, Philippot P, Drapier S, Drapier D et al (2009) Subthalamic nucleusstimulation affects subjective emotional experience in Parkinson’s disease patients. Neuro-psychologia 47(8–9):1928–1937

Volkmann J, Allert N, Voges J, Weiss PH, Freund HJ, Sturm V (2001) Safety and efficacy ofpallidal or subthalamic nucleus stimulation in advanced PD. Neurology 56(4):548–551

Volkmann J, Daniels C, Witt K (2010) Neuropsychiatric effects of subthalamic neurostimulationin Parkinson disease. Nat Rev Neurol 6(9):487–498

Voon V, Krack P, Lang AE, Lozano AM, Dujardin K, Schupbach M et al (2008) A multicentrestudy on suicide outcomes following subthalamic stimulation for Parkinson’s disease. Brain131(Pt 10):2720–2728

Weaver FM, Follett K, Stern M, Hur K, Harris C, Marks WJ Jr et al (2009) Bilateral deep brainstimulation vs best medical therapy for patients with advanced Parkinson disease: arandomized controlled trial. JAMA 301(1):63–73

Weintraub D (2009) Impulse control disorders in Parkinson’s disease: prevalence and possiblerisk factors. Parkinsonism Relat Disord 15(Suppl 3):S110–S113

Weintraub D, Koester J, Potenza MN, Siderowf AD, Stacy M, Voon V et al (2010) Impulsecontrol disorders in Parkinson disease: a cross-sectional study of 3090 patients. Arch Neurol67(5):589–595

Witjas T, Baunez C, Henry JM, Delfini M, Regis J, Cherif AA et al (2005) Addiction inParkinson’s disease: impact of subthalamic nucleus deep brain stimulation. Mov Disord20(8):1052–1055

Witt K, Daniels C, Reiff J, Krack P, Volkmann J, Pinsker MO et al (2008) Neuropsychologicaland psychiatric changes after deep brain stimulation for Parkinson’s disease: a randomised,multicentre study. Lancet Neurol 7(7):605–614

Wolters EC, Braak H (2006) Parkinson’s disease: premotor clinico-pathological correlations.J Neural Transm Suppl 70:309–319

15 Neuropsychiatric Side Effects of Deep Brain Stimulation 173

Chapter 16Psychiatric Aspects of Parkinson’s Diseasein Animal Models of Deep BrainStimulation of the Subthalamic Nucleus

S. K. H. Tan, H. Hartung, V. Visser-Vandewalle, T. Sharpand Y. Temel

16.1 Introduction

In the early 1990s, deep brain stimulation (DBS) of the subthalamic nucleus (STN)was introduced to treat motor symptoms of Parkinson’s disease (PD) (Limousinet al. 1995; Pollak et al. 1993). Benabid and colleagues reported substantialimprovement in motor function, which encouraged other clinical centres to use thistreatment option shortly after (Limousin et al. 1995; Pollak et al. 1993). A fewyears after the introduction of STN DBS, various clinical centres noticed theappearance of psychiatric side effects. One of the first reports was by Rodriguezet al. (1998), who describe that in their series of 12 STN DBS treated patients, onedeveloped severe depressive symptoms, and more reports of patients experiencingdepressive symptoms followed (Houeto et al. 2000; Kumar et al. 1999). It has beenestimated that depression occurs in 2–33 % of STN DBS treated patients (Applebyet al. 2007; Takeshita et al. 2005; Temel et al. 2006). Furthermore, the risk ofsuicide in the first few years after surgery was significantly increased (Voon et al.2008). Most authors explained psychiatric side effects by non STN related causessuch as tapering of dopaminergic medication or changes in psychosocial context.Although these might apply to a subgroup of patients experiencing psychiatric sideeffects, it was surprising that a mechanism involving the STN itself was notconsidered, especially since the STN was already well known to have a limbic

S. K. H. Tan � V. Visser-Vandewalle � Y. TemelDepartment of Neuroscience, Maastricht University, Maastricht, The Netherlands

S. K. H. Tan � V. Visser-Vandewalle � Y. Temel (&)Department of Neurosurgery, Maastricht University Medical Centre,PO Box 5800, 6202 AZ Maastricht, The Netherlandse-mail: y.temel@maastrichtuniversity.nl

H. Hartung � T. SharpDepartment of Pharmacology, University of Oxford, Oxford, UK

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_16, � Springer-Verlag Berlin Heidelberg 2012

175

function (Temel et al. 2005). A case of acute depression during DBS through thecontact which had the best motor effect supported the idea of an STN-mediatedpsychotropic effect (Kumar et al. 1999). Later others reconstructed stimulationcontacts that cause depression to be located within the STN boundaries (Tommasiet al. 2008).

Interestingly, various functional imaging studies have demonstrated that STNDBS induced activity changes in the prefrontal cortex, an area implicated in mood.Although it has been suggested that these alterations are a consequence of directmodulation of upstream cortical projections, it should be stressed that prefrontalactivity may also depend on monoamine neurotransmission. Therefore, we proposeda somewhat counterintuitive mechanism where STN DBS alters downstream pro-jections towards brainstem monoamine systems. In particular, the serotonin (5-hydroxytryptamine; 5-HT) system has been related to mood regulation. There is alarge body of evidence supporting a dysfunctional 5-HT system be associated withdepression. For example, early observations reported low levels of 5-hydroxyin-doleacetic acid, the major 5-HT metabolite, in cerebrospinal fluid of depressedpatients, and more recent imaging studies showed abnormal 5-HT transmission incortical and subcortical regions with various radioactive tracers (Asberg et al. 1976;Cannon et al. 2007). However, the most convincing evidence for a dysfunctional 5-HT system in depression is the clinical effectiveness of inhibitors of 5-HT reuptake toimprove depression (Taylor et al. 2006). Moreover, regarding depression as a sideeffect of STN DBS, several anecdotal reports described 5-HT reuptake inhibitors toeffectively treat depressive symptoms in STN DBS treated patients. This supports thehypothesis of an STN DBS mediated alteration in 5-HT transmission that may be akey factor in the development of depressive side effects.

How STN DBS can influence 5-HT transmission and 5-HT-related behaviourswas unknown until recently. With recent developments in animal models, variousstudies combined STN DBS with in vivo electrophysiological, neurochemical andbehavioural techniques to obtain detailed information on how STN DBS modulatesthe 5-HT system. Studies mainly focused on the 5-HT system of the dorsal raphenucleus (DRN), which contains most of the 5-HT neurons of the central nervoussystem (Steinbusch 1981). In the following sections, we will discuss recent findingsin animal studies investigating the effect of STN DBS on the DRN 5-HT system,which we hold responsible for the development of psychiatric side effects.

16.2 The Effect of STN DBS on 5-HT Neuronal Firing

Temel et al. (2007) were the first to investigate the effect of bilateral STN DBS onthe 5-HT system. Using in vivo extracellular single unit recording techniques, theyevaluated the effect of STN DBS on the activity of putative DRN 5-HT neurons.Interestingly, a brief period of STN DBS (2 min) caused an immediate and dra-matic decrease in the firing rate (-45 %) of putative 5-HT neurons in anaesthe-tized rats (Fig. 16.1a). Also, the vast majority of recorded 5-HT neurons (91 %)

176 S. K. H. Tan et al.

responded with an inhibition. Importantly, STN DBS only caused inhibition of5-HT neurons when stimulation with high frequency (100 Hz or more) andamplitudes between 30 and 150 lA was applied (Temel et al. 2007). These aresettings comparable to clinical stimulation paradigms (Tan et al. 2010). Theinhibition of 5-HT neurons seemed to be STN DBS dependent since 5-HT neu-ronal activity returned to baseline values shortly after cessation of the stimulus.It remained undetermined whether the inhibitory effect of STN DBS on 5-HTneurons was dependent on the integrity of the dopamine system. It is wellestablished that in PD degenerative changes occur in the DRN 5-HT system andthat the 6-hydroxydopamine (6-OHDA) PD model is often accompanied by 5-HTalterations (Tan et al. 2011a). STN DBS in PD models utilizing 6-OHDA orreserpine caused the same inhibition of 5-HT neuronal activity in comparison withrats with an intact dopamine system (Temel et al. 2007). Importantly, this inhib-itory effect was STN-specific. Stimulation of remote and neighbouring structuresdid not cause an inhibition of 5-HT neurons (Temel et al. 2007). In addition,

0

200 10 50 100 130

( : Hz, at 30 µA)

-60

-40

-20

% C

hang

e fr

omba

selin

e fir

ing

rate

*

**

*

*

0 3 30 100 150

( : µA, at 100 Hz)

NB 5-HT

1.5

2.0

2.5

StimSpike train

5 10 1500.0

0.5

1.0

Time (min)

Mea

n f

irin

rat

e (H

z)g

merge

(a)

(b)

(c)

Fig. 16.1 The inhibitoryeffect of subthalamic nucleusdeep brain stimulation (STNDBS) on neuronal activity ofDRN serotonin (5-HT)neurons. a The effect of STNstimulation frequency andamplitude on 5-HT neuronalfiring rate. Only high-frequency stimulation of100 Hz or greater inhibited5-HT neuronal firing. Dataare presented as themean ± the standard error ofthe mean (n = 6).b Photomicrograph of aneurobiotin (NB)-labelleddorsal raphe nucleus (DRN)neuron which was positivefor 5-HTimmunocytochemistry(arrow) (scale bar 10 lm).c The corresponding meanfiring rate (15-s bins) of theDRN 5-HT neuron inb before, during and after5 min STN DBS (a fromTemel et al. 2007; b, c fromHartung et al. 2011)

16 Psychiatric Aspects of Parkinson’s Disease 177

intra-STN injections of muscimol, a GABA-A agonist, mimicking the action ofDBS resulted in a similar inhibition of 5-HT neurons (Temel et al. 2007).

A recent electrophysiological study confirmed the inhibitory effect of bilateralSTN DBS on 5-HT neuronal activity. With the same electrodes and stimulationparadigm (130 Hz, 60 ls, 100–200 lA), a longer stimulation duration of 5 mincaused a significant decrease in firing rate (-26 %) in half of the recorded putative5-HT neurons (37 of 74 neurons; 50 %) (Hartung et al. 2011). In this study, teninhibited neurons were juxtacellularly labelled with neurobiotin and confirmed tocontain 5-HT (Fig. 16.1b). Interestingly, most of the inhibited neurons (74 %)remained inhibited for 5 min after stimulation had been stopped. The persistinginhibition beyond the stimulation may suggest neuroplastic changes were inducedwhen relatively long periods of STN DBS were applied. In this study someputative 5-HT neurons did not respond (18 of 74 neurons; 24 %) and others wereexcited (19 of 74 neurons; 26 %).

16.3 The Effect of STN DBS on 5-HT Release

Although STN DBS seemed to have a mainly inhibitory but heterogeneous effecton neuronal activity, it was questioned whether 5-HT release was also changed. Toassess this issue, microdialysis experiments followed. These studies particularlyfocused on 5-HT release in the prefrontal cortex and hippocampus, which receivedense 5-HT innervation from the DRN.

In line with the afore-mentioned electrophysiological findings, Navailles et al.(2010) found a substantial decrease in prefrontal cortical and hippocampal 5-HTrelease after unilateral STN DBS in anaesthetized rats. Bilateral STN DBS, which isclinically preferred, also caused a significant decrease in 5-HT release in the pre-frontal cortex (Tan et al. 2012). We observed this in experiments with anaesthetizedand freely moving animals. Similar to the effect of STN DBS on 5-HT neuronalactivity, it was established that 5-HT release is also independent of dopaminergicintegrity (Navailles et al. 2010, Tan et al. 2012). Generally, 5-HT release in theprefrontal cortex and hippocampus has been related to mood regulation. Therefore,this decrease in forebrain 5-HT release may be involved in the development ofdepressive and other psychiatric symptoms in STN DBS treated patients.

Interestingly, we also found bilateral STN DBS inhibits 5-HT release in thestriatum (Tan et al. 2012), which receives extensive 5-HT innervation from theDRN (Steinbusch 1981). Striatal 5-HT has mainly been linked to motor functionand more recently to levodopa-induced dyskinesias (Carta et al. 2007; Rylanderet al. 2010). Pharmacological inhibition of 5-HT transmission alleviated levodopa-induced dyskinesias (Carta et al. 2007). Interestingly, clinical data also showed areduction of levodopa-induced dyskinesias after STN DBS in PD patients (Deuschlet al. 2006; Krack et al. 2003). Although this has mainly been related to areduction in the dopaminergic medication, stimulation-induced decrease in striatal5-HT release may contribute to this beneficial effect as well.

178 S. K. H. Tan et al.

16.4 Anatomical Pathway of STN DBS Related Inhibitionof the 5-HT System

Although STN DBS has a significant impact on 5-HT transmission, the anatomicalpathway is unknown. A direct projection from the STN to the DRN does not seemto exist (Peyron et al. 1998). A multisynaptic pathway connecting both structuresis most likely. Potential anatomical relay areas can be identified on the basis ofinformation obtained from tracing experiments. From studies mapping STN outputregions and those brain areas projecting to the DRN, the prefrontal cortex, sub-stantia nigra, ventral pallidum, and lateral habenula can be considered as potentialrelay stations (Groenewegen and Berendse 1990; Peyron et al. 1998). To furtheraddress this question, we recently used c-Fos, a marker of neuronal activity, tomap activity changes in these areas. The expression of c-Fos was significantlyincreased in the prelimbic region of the prefrontal cortex and the lateral habenula(Tan et al. 2011b) (Fig. 16.2). These two structures are known to have a stronginhibitory influence on DRN 5-HT neurons mediated via activation of local DRNGABA neurons (Sharp et al. 2007). The latter subpopulation of DRN neurons ismainly located in the lateral DRN subdivisions. Interestingly, we found STN DBSsignificantly increases c-Fos expression predominately in these lateral subdivisions(Tan et al. 2011b). In addition, c-Fos neurons in this region double-labelled with amarker for putative GABA neurons. Together these data suggest STN DBS acti-vates the prefrontal cortex and lateral habenula, which in turn inhibit the DRN 5-HT system. However, other brain regions, e.g. ventral pallidum and substantianigra, may also be involved and may respond with a reduction or change in localactivity which cannot be determined by c-Fos expression measurements.

16.5 The Effect of STN DBS on 5-HT-Related Behaviour

Although the previously described experiments support a strong inhibition of 5-HTneurotransmission, it has to be determined whether this actually accounts for STNDBS induced psychiatric symptoms. Various behavioural paradigms are availableto evaluate mood related functions in animal models. One of the most commonlyused models is the forced swim test (FST), where learned helplessness behaviour isassessed when a rat is exposed to an inescapable stressor (Cryan et al. 2002).Bilateral STN DBS in 6-OHDA treated rats induced increased immobilitybehaviour in the FST, which reflects increased learned helplessness and thedevelopment of depressive-like behaviour (Tan et al. 2011b; Temel et al. 2007).Arguably, Parkinsonian motor deficits may interfere with the FST. However, thebilateral partial 6-OHDA model used in our experiments has been shown to induceonly subtle motor deficits (Temel et al. 2007). STN DBS also increased interactionin the social interaction test (Tan et al. 2011b). Both behavioural alterations havebeen observed in low 5-HT conditions.

16 Psychiatric Aspects of Parkinson’s Disease 179

Interestingly, pretreatment with the 5-HT reuptake inhibitor citalopram pre-vented STN DBS from inducing depressive-like behaviour in the FST (Temel et al.2007). This finding confirmed a 5-HT-dependent mechanism of STN DBS relatedpsychiatric symptoms and may have potential clinical implications. Firstly,treatment with a 5-HT reuptake inhibitor may be appropriate for PD patients whohave postoperative depressive symptoms. Secondly, it is tempting to speculate thatPD patients with increased risk of psychiatric symptoms after STN DBS maybenefit from drug pretreatment to enhance 5-HT function. This could also meanthat pretreatment may enable PD patients previously not eligible for DBS treat-ment because of psychiatric vulnerability to undergo this surgical therapy.

16.6 Conclusion and Future Perspectives

Altogether there is compelling evidence from animal studies that STN DBS has asignificant impact on the DRN 5-HT system. Not only did STN DBS inhibit 5-HTneuronal activity, it also decreased 5-HT release in the forebrain. Anatomical datasuggest the prefrontal cortex and lateral habenula may mediate this inhibitory

LH c-Fos expression

0

5

10

15 Control

Non-StimStim

*

#

c-F

os

Rat

io (

cells

/ m

m2 )3rd V

LHmLHl

MH

FMIPrL

IL

CG1mPFC c-Fos expression

0

2

4

6

8 ControlNon-StimStim

PrL IL

*

c-F

os

Rat

io (

cells

/mm

2 )

(a) (b) (c)

(d) (e) (f)

Fig. 16.2 The effect of STN DBS on c-Fos expression in the lateral habenula (LH) and medialprefrontal cortex in stimulated rats (b, e) and non-stimulated controls (a, d). Note the particularincrease of c-Fos expression in the medial aspects of the LH and the prelimbic part of the medialprefrontal cortex. Cumulative data are presented as the mean ± the standard error of the mean forthe LH (c) and medial prefrontal cortex (f). CG1 cingulate cortex 1, FMI forceps minor of corpuscallosum, IL infralimbic cortex, LHl laterolateral habenula, LHm mediolateral habenula, MHmedial habenula, PrL prelimbic cortex, 3rd V third ventricle, bar 200 lm (from Tan et al. 2011b)

180 S. K. H. Tan et al.

effect of stimulation. Moreover, STN DBS induced depressive-like behaviour thatcould be prevented by antidepressant drugs elevating 5-HT levels. Thus, these datastress the role of the STN not only in motor regulation but also in mood regulation.In addition, psychiatric symptoms after STN DBS most probably relate to adysfunctional 5-HT system. Currently available data support a strong link betweenthe STN and the DRN 5-HT system. However, future studies may need to focus onthe median raphe 5-HT system and on changes in 5-HT receptors as these havealso been implicated in mood regulation.

References

Appleby BS et al (2007) Psychiatric and neuropsychiatric adverse events associated with deepbrain stimulation: a meta-analysis of ten years’ experience. Mov Disord 22:1722–1728

Asberg M et al (1976) ‘‘Serotonin depression’’—a biochemical subgroup within the affectivedisorders? Science 191:478–480

Cannon DM et al (2007) Elevated serotonin transporter binding in major depressive disorderassessed using positron emission tomography and [11C]DASB; comparison with bipolardisorder. Biol Psychiatry 62:870–877

Carta M et al (2007) Dopamine released from 5-HT terminals is the cause of L-DOPA-induceddyskinesia in parkinsonian rats. Brain 130:1819–1833

Cryan JF et al (2002) Assessing antidepressant activity in rodents: recent developments andfuture needs. Trends Pharmacol Sci 23:238–245

Deuschl G et al (2006) A randomized trial of deep-brain stimulation for Parkinson’s disease.N Engl J Med 355:896–908

Groenewegen HJ, Berendse HW (1990) Connections of the subthalamic nucleus with ventralstriatopallidal parts of the basal ganglia in the rat. J Comp Neurol 294:607–622

Hartung H et al (2011) High-frequency stimulation of the subthalamic nucleus inhibits the firingof juxtacellular labelled 5-HT-containing neurones. Neuroscience 186:135–145

Houeto JL et al (2000) Subthalamic stimulation in Parkinson disease: a multidisciplinaryapproach. Arch Neurol 57:461–465

Krack P et al (2003) Five-year follow-up of bilateral stimulation of the subthalamic nucleus inadvanced Parkinson’s disease. N Engl J Med 349:1925–1934

Kumar R et al (1999) Comparative effects of unilateral and bilateral subthalamic nucleus deepbrain stimulation. Neurology 53:561–566

Limousin P et al (1995) Effect of parkinsonian signs and symptoms of bilateral subthalamicnucleus stimulation. Lancet 345:91–95

Navailles S et al (2010) High-frequency stimulation of the subthalamic nucleus and L-3,4-dihydroxyphenylalanine inhibit in vivo serotonin release in the prefrontal cortex andhippocampus in a rat model of Parkinson’s disease. J Neurosci 30:2356–2364

Peyron C et al (1998) Forebrain afferents to the rat dorsal raphe nucleus demonstrated byretrograde and anterograde tracing methods. Neuroscience 82:443–468

Pollak P et al (1993) Effects of the stimulation of the subthalamic nucleus in Parkinson disease.Rev Neurol (Paris) 149:175–176

Rodriguez MC et al (1998) The subthalamic nucleus and tremor in Parkinson’s disease. MovDisord 13(Suppl 3):111–118

Rylander D et al (2010) Maladaptive plasticity of serotonin axon terminals in levodopa-induceddyskinesia. Ann Neurol 68:619–628

Sharp T et al (2007) Important messages in the ‘post’: recent discoveries in 5-HT neuronefeedback control. Trends Pharmacol Sci 28:629–636

16 Psychiatric Aspects of Parkinson’s Disease 181

Steinbusch HW (1981) Distribution of serotonin-immunoreactivity in the central nervous systemof the rat-cell bodies and terminals. Neuroscience 6:557–618

Takeshita S et al (2005) Effect of subthalamic stimulation on mood state in Parkinson’s disease:evaluation of previous facts and problems. Neurosurg Rev 28:179–86 (discussion 187)

Tan SK et al (2012) A combined in vivo neurochemical and electrophysiological analysis of theeffect of high-frequency stimulation of the subthalamic nucleus on 5-HT transmission. ExpNeurol 233:145–53

Tan S et al (2010) Experimental deep brain stimulation in animal models. Neurosurgery67:1073–1079 (discussion 1080)

Tan SK et al (2011a) Serotonin-dependent depression in Parkinson’s disease: a role for thesubthalamic nucleus? Neuropharmacology 61:387–399

Tan SK et al (2011b) High frequency stimulation of the subthalamic nucleus increases c-fosimmunoreactivity in the dorsal raphe nucleus and afferent brain regions. J Psychiatr Res45:1307–1315

Taylor MJ et al (2006) Early onset of selective serotonin reuptake inhibitor antidepressant action:systematic review and meta-analysis. Arch Gen Psychiatry 63:1217–1223

Temel Y et al (2005) The functional role of the subthalamic nucleus in cognitive and limbiccircuits. Prog Neurobiol 76:393–413

Temel Y et al (2006) Behavioural changes after bilateral subthalamic stimulation in advancedParkinson disease: a systematic review. Parkinsonism Relat Disord 12:265–272

Temel Y et al (2007) Inhibition of 5-HT neuron activity and induction of depressive-like behaviorby high-frequency stimulation of the subthalamic nucleus. Proc Natl Acad Sci U S A104:17087–17092

Tommasi G et al (2008) Transient acute depressive state induced by subthalamic regionstimulation. J Neurol Sci 273:135–138

Voon V et al (2008) A multicentre study on suicide outcomes following subthalamic stimulationfor Parkinson’s disease. Brain 131:2720–2728

182 S. K. H. Tan et al.

Chapter 17Scientific Recordings in Deep BrainStimulation

Michael X. Cohen

17.1 ‘‘Piggy-Backing’’ on a Clinical Procedure for ScientificPurposes

The primary purpose of deep brain stimulation (DBS) is for treatment of a clinicaldisorder. DBS is most widely used for treatment of Parkinson’s disease and othermotor impairments (Bronstein et al. 2010; Flora et al. 2010), but the use of DBSfor treating psychiatric disorders is increasingly being evaluated, with positiveinitial reports. Psychiatric uses of DBS include treatment of major depression(Mayberg et al. 2005; Schlaepfer et al. 2008), obsessive–compulsive-disorder(Mian et al. 2010; de Koning et al. 2011), and Tourette’s syndrome (Hariz andRobertson 2010). Thus far, it seems that DBS may be a targeted, safe, and effectivetreatment option for many ailments of the brain and mind. The next 10 years willlikely see a large increase in the prevalence of DBS and diseases for which it isused.

This is great news for patients and their families, medical practitioners,insurance companies, and society in general (the latter owing to DBS decreasingdisease-related loss of productivity). This is also great news for neuroscientists,psychologists, and clinical researchers. The DBS surgery provides an opportunityto conduct cutting-edge research measuring electrical activity directly from a partof the brain that cannot be measured using noninvasive techniques, with little or noadded cost or adverse side effects for the patients, surgical team, or psychiatrists.Human neuroscientists have long dreamed of doing the kinds of invasiverecordings that are done in nonhuman animals, and the increase in the prevalence

M. X. Cohen (&)Department of Psychology, University of Amsterdam, Amsterdam, The Netherlande-mail: mikexcohen@gmail.com

M. X. CohenDepartment of Physiology, University of Arizona, Tucson, USA

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_17, � Springer-Verlag Berlin Heidelberg 2012

183

of DBS will facilitate significant advances in our understanding of the complexelectrophysiological dynamics that supports human cognitive, emotional, andperceptual processes. Even better, this research will contribute not only to theneuroscience community, but also to the clinical community by facilitating a betterunderstanding of the mechanisms and effective targets of DBS.

17.2 Methods of Recording Electrophysiological Activityin Deep Brain Structures

There are two possibilities to record electrophysiological activity from DBS. Thefirst is during the surgical implantation of the DBS electrodes. In these intraop-erative recordings, the surgical procedure is paused for a few minutes after one orboth DBS electrodes have been implanted. Before the wires that connect the DBSelectrodes to the stimulator are placed under the skin, they can be attached to anelectroencephalogram (EEG) amplifier, and activity from the DBS target can berecorded. Cognitive experiments can be done by setting up a computer monitor inthe operating room, through which the patient can perform simple tasks. Becauseof the size of these electrodes (cylindrical leads with 1.27-mm diameter and 0.5 to1.5-mm interelectrode spacing), activity from individual neurons cannot beresolved. These electrodes thus measure local field potentials (the summed den-dritic activity of populations of neurons), and sampling rates of approximately500–2,000 Hz are sufficient. The main disadvantages of this approach are that thepatient may be sedated or may have just come out of anesthesia, and that themiddle of invasive brain surgery may not be an optimal environment for patients tofocus on computer-based cognitive tasks.

The second possibility to record electrophysiological activity from the deepbrain structures is through postoperative recordings. In this case, the surgery takesplace over two separate sessions: In the first session, the DBS electrodes areimplanted; in the second session, the battery pack/stimulator is implanted in thechest. During the time between surgical procedures, which can be hours to days,the electrode leads are left externalized (typically at the top or back of the head)before the stimulator is implanted. Through a special adaptor, the leads can beplugged into a standard EEG amplifier, and the DBS electrodes can be used tomonitor electrophysiological activity.

The main advantages of this approach are that the patients are not in surgeryand therefore can sit comfortably while performing cognitive tasks, are notanesthetized, and can take breaks if they become tired. This translates to improveddata quality and confidence that the patients are focused on the task. It is alsopossible to record simultaneous scalp EEG, although it is possible that not allelectrodes can be used, depending on where the implantation was. For example, ifthe implantation had a dorsal entry, it may not be possible to use electrodes overthe entry point areas owing to stitches and sensitive skin.

184 M. X. Cohen

For some operations, before the DBS electrodes are implanted, microwires aretemporarily implanted to confirm localization on the basis of electrophysiologicalproperties of spiking activity. These microwires are removed prior to implantingthe DBS macrocontacts. In this case, it is possible to use these electrodes tomonitor activity from individual neurons during cognitive tasks (Zaghloul et al.2009). At the time of this writing, permanently implanted microwires (e.g., at thetips of the macrocontacts) are not available or used, although such equipmentwould provide extremely valuable data regarding the relation between single-unitor multiunit recordings and local field potentials. The microwires are too small forstimulation to provide a clinical benefit, which is why macrocontacts are used.Thus, microwires attached to the end of the macrocontacts would provide usefulscientific data but may be of limited or no clinical use in terms of treatmentefficacy. Whether such microwires would cause any additional damage or wouldinterfere with the stimulation is not clear.

There is a third way to conduct clinical and neuroscientific research with DBS,although it does not involve direct recordings. This is by turning the stimulator offand on while patients perform cognitive tasks. Although the DBS electrodes are notaccessible for recording, behavior and scalp EEG could be monitored. In this case,one can assess the causal contribution of the stimulated region. The advantages ofthis approach are that one can address issues of causality, the research is not limitedto a small time window during surgery, and turning the stimulator off for a few hoursis tolerable by most patients. The two main disadvantages are that the results may notbe straightforward to interpret because the effect of DBS extends well beyond theregion stimulated (McIntyre and Hahn 2010) owing to complex orthodromic andantidromic effects on other regions (McCracken and Grace 2007, 2009) and that, inmost cases, experimenters and patients are not blind to the manipulation (i.e., theyknow whether the stimulator is on or off).

17.3 What We Have Learned from DBS Recordings

There are too many scientific reports of DBS recordings to review all of them here.Instead, two key findings from our research will be highlighted. Postoperativerecordings from the nucleus accumbens in patients undergoing treatment fordepression or obsessive–compulsive disorder were taken during simple reward-guided decision-making tasks, in combination with surface EEG activity tomonitor cortical dynamics. Over several studies, we observed top–down signalingfrom medial frontal areas (Cohen et al. 2009a), which became stronger duringreward anticipation (Cohen et al. 2012). These findings provide the first evidencefor rapid electrophysiological communication between the medial frontal cortexand ventral striatum, and suggest that the medial frontal cortex may bias reward-and motivation-related processing in subcortical structures. Second, we have alsoconsistently observed high-frequency gamma oscillations that are time-locked toalpha phase (Cohen et al. 2009b). Cross-frequency coupling (e.g., alpha–gamma

17 Scientific Recordings in Deep Brain Stimulation 185

coupling) is thought to be a substrate for coordinating the activities of multipleneural networks (Lisman 2005). Indeed, we have observed that alpha–gammacoupling timing is modulated by reward (Cohen et al. 2009b), and that the strengthof this coupling decreases during cues that indicate that behavioral switches arenecessary (Cohen et al. 2009a). These findings suggest that the striatum may usetemporal coding schemes based on temporally precise interactions among activityin multiple frequency bands. These kinds of observations are inaccessible withnoninvasive techniques such scalp electroencephalography and functional mag-netic resonance imaging, and help link human nucleus accumbens processing totheoretical and empirical work in nonhuman animals (Canolty and Knight 2010).

17.4 What the Future of DBS Recordings Might Bring

With the increasing use and acceptance of DBS as a viable treatment option forseveral disorders, DBS recordings will become more widely used and accepted inboth fundamental and clinical human neuroscience. There are at least four cate-gories of improvement/future possibilities.

First, better experimental paradigms should be developed and used. Many ofthe paradigms used in current DBS research are simple and straightforward,comprise few conditions, and tap into basic motor/motivation processes. Ofcourse, simple paradigms maintain the advantages of being easy for patients tocomplete, and the findings being easy for researchers to interpret. Issues of gen-eralizability become more important as paradigms become more complex, but abetter characterization of the processes that are deficient and intact in differentpatient groups will facilitate interpretation of the results (this point is discussedmore below). Future DBS research might better utilize the rich repertoire ofcognitive, perceptual, emotional, and social experimental paradigms that havebeen developed within psychology over the past 60 years.

Second, more sophisticated and physiologically inspired data analyses shouldbe performed. The high signal to noise and spatial precision of DBS recordingsmeans that analyses that are difficult to perform with scalp electroencephalographyor magnetoencephalography (MEG) can be done with DBS. These includeinvestigation of high-frequency oscillations, synchronization, and cross-frequencycoupling. As our understanding of the physiological mechanisms and cognitiveimplications of brain electrical dynamics recorded by electroencephalographyincreases, so will the sophistication of neurobiologically inspired mathematicaland statistical data analyses.

Third, multimodal imaging should be performed to understand how the DBSregion interacts with other brain systems to form large-scale subcortical–corticalnetworks. The most accessible tool is to record simultaneous scalp EEG, withwhich it is possible to examine the millisecond-resolution temporal interactionsbetween cortical and subcortical dynamics. MEG is also possible, although thispresents a bigger challenge because many hospitals do not have MEG scanners,

186 M. X. Cohen

and because MEG is more sensitive to magnetic interference from the DBSelectrodes and stimulator (if implanted during the MEG recording). Simultaneousfunctional magnetic resonance imaging and DBS recordings will also providepowerful insights, as well an opportunity to test fundamental questions regardingthe relationship between electrophysiological and hemodynamic activity. To myknowledge, these simultaneous recordings have not been performed, and the safetyof DBS inside a strong and rapidly fluctuating magnetic field remains to bedetermined. Multimodal imaging may also include computational models, theoutputs of which can be tested against the observed output of the DBS activity.These recordings may prove useful for testing and constraining neurobiologicallyinspired computational models.

Fourth, new electrode technology might permit better-quality recordings or thecapability to continue recording after implantation is complete. Better-qualityrecordings might be obtained, for example, by use of microwires attached to themacrocontacts that would allow simultaneous single-unit or multiunit recordingsalongside the field potential recordings. This would allow direct comparison ofmicroscopic and mesoscopic neural activity. The capability to continue recordingelectrophysiological data after implantation is complete would allow both scien-tific and clinical data to be collected during long-term treatment.

Finally, future DBS studies may target multiple brain regions to stimulate anetwork rather than a single brain area. This may prove important if the effects ofDBS occur via regulating network-level dynamical activity. For example, bothsubgenual cingulate and ventral striatal stimulation seem effective at alleviatingsymptoms of major depression (Mayberg et al. 2005; Schlaepfer et al. 2008),implying that it is stimulation of this interconnected limbic circuit, rather thanstimulation of either specific brain area, that mediates the clinical improvement.One can hypothesize that stimulating both regions in a delayed fashion, such thatthe ventral striatum receiving a pulse shortly after the cingulate has received apulse, may provide a clinical benefit beyond the effects of stimulating either target.To my knowledge, this has not been tested. In addition to potential clinical ben-efits, this would facilitate more detailed scientific investigations.

17.5 Why Psychiatrists and Surgeons Should Be Interestedin DBS Recordings

The scientific benefits of DBS research extend beyond increasing our basicunderstanding of brain electrophysiology. A deeper understanding of the func-tioning of DBS targets may facilitate clinical benefits. For example, recordingsmay help optimize stimulation parameters by localizing disease-relevant activitychanges to specific contacts. That is, many DBS contacts contain four electrodes; ifrecordings can localize disease-relevant activity to one or two electrodes, theclinical effects of stimulation may be most robust at those electrodes. So far, this

17 Scientific Recordings in Deep Brain Stimulation 187

has not been extensively investigated, likely because of small sample sizes in eachstudy. However, with increasing prevalence and clinical benefits of DBS, studieswith large enough sample sizes should begin to address whether electrophysio-logical activity recorded from intraoperative or postoperative DBS recordings canpredict treatment success. Ideally, of course, it would be best to have a predictor oftreatment success before DBS electrode implantation, but because so much isunknown about the mechanisms of DBS improvement, even postsurgical predic-tors would be important. Further, because DBS is often used in combination withother (e.g., pharmacological) treatments, such predictability might be also usefulfor determining, e.g., how much medication is needed after DBS treatment.

17.6 Why Non-Human-Animal Physiologists Should CareAbout DBS Recordings

Animal models allow more detailed and invasive investigations of the electro-physiological and chemical effects of DBS, and the mechanisms by which DBSmay alleviate diseases (Tan et al. 2010). Thus, from a clinical and scientificperspective, animal models of DBS may prove important for better DBS targetingand stimulation protocols.

However, from a scientific perspective, human DBS recordings offer a rareopportunity to test whether the functional dynamics and electrophysiologicalprofiles of animal deep brain regions are similar to those of humans. There areanatomical/functional differences between species, and it is important to knowwhich functional properties are conserved, and which are different, betweenhumans and other animals. Typically, such functional cross-species investigationsare done by comparing single-unit spiking activity in animals with the hemody-namic response in humans. However, the neurophysiological dynamics that drivehemodynamic activity are complex and imperfectly understood. Therefore, com-paring field potentials from humans and nonhuman animals facilitates a moredirect cross-species comparison.

17.7 Limitations and Problems with Generalizations

Needless to say, any results from patients are suspect when it comes to general-izing about normal-functioning brains. This is particularly the case for DBS,because the region of the recordings is the target of DBS, and it is the targetbecause it is believed that region is pathological.

Effective treatment resulting from DBS of a region does not necessarily meanthat region is dysfunctional; it is possible that the larger circuit in which this regionis a node is dysfunctional, and stimulating any node within the circuit will regulate

188 M. X. Cohen

the entire system into a more normal state. But because pathological activity hasbeen reported in many DBS targets, a better approach to this problem is byaddressing it empirically. Unfortunately, the best empirical approach would be toimplant DBS electrodes into healthy controls, but this is not ethically permissible.This leaves two appropriate approaches.

First, the same experiment paradigm can be tested in matched control subjectswith scalp electroencephalography. Nonsignificant group differences in scalp EEGand task performance would support the idea that that function is relatively intactin the DBS patient group. However, it is possible that the task-relevant corticaldynamics and behavior are intact, but the DBS target region is still dysfunctional,so this null result cannot be unambiguously interpreted.

Second, multiple patient groups with DBS of the same target region (e.g.,depression, obsessive–compulsive disorder, substance abuse, all with DBS in thenucleus accumbens) can be tested on a variety of tasks. Then, patterns of activitycan be compared across patients to examine what are the disease-specific anddisease-independent patterns of activity. However, this approach requires havingaccess to many patients with different diseases but with the same DBS target.

17.8 Ethical Considerations

For the clinical application of DBS, there are important ethical considerations,including which patients are appropriate for DBS, how the consent process shouldwork, and who can give consent under what circumstances. Many publicationsaddress clinical ethical considerations; these considerations will not be addressedhere. Once ethical approval for the surgery and clinical application of DBS hasbeen obtained, and the psychiatrists and scientists have the opportunity to conductscientific recordings, additional ethical approval should be obtained for theserecordings.

For most psychological paradigms (e.g., simple computer tasks), there shouldbe no major ethical concerns. More serious concerns may arise if the studyinvolves additional invasive procedures such as testing the effects of medication ondeep brain activity. Some ethical considerations include whether the recordingswill require an extended stay in the hospital, whether there is increased risk ofinfection, and whether the recordings will delay the start of the DBS treatment.

In my experience, most patients are cooperative and happy to perform thesestudies and provide data. They believe that these results will be valuable to theclinical and scientific communities. Further, the testing provides a welcome dis-traction between surgical procedures.

Researchers must remain cognizant of the mental and physical condition oftheir patients, and must be aware that the data, although rare and important, are notso valuable that they should be obtained at the expense of the comfort and will-ingness of the patients. Patients should be aware of the scientific value of therecordings and what could happen with their data (e.g., published in scientific

17 Scientific Recordings in Deep Brain Stimulation 189

papers), and should be aware that the scientific recordings are not necessarilyrelated to the clinical application of DBS. Of course, if recordings begin to be usedto optimize clinical recordings and this improves treatment success, participationin these experiments may indeed have a direct clinical benefit to the patients.Whether participation will potentially impact their treatment course should bemade clear prior to the experiments, and, if the experiments are of a scientific andnot clinical nature, patients must be aware that their participation is voluntary, theycan quit the research at any time, and this has no bearing on their clinicalprogression.

17.9 Conclusions

DBS holds great promise as a treatment option for a growing number of clinicaldisorders, from motion disorders to emotion disorders. A fortuitous and seren-dipitous by-product of DBS is the ability to conduct basic neuroscientific research.Recording electrical activity directly from deep brain structures in awake humansprovides an important and rare opportunity to test hypotheses, link human toanimal research, and perform both scientifically and clinically relevant research.Although recordings from DBS electrodes have limitations (limited generaliz-ability and lack of a healthy control group), they also have considerable advan-tages (spatial localization, temporal resolution, and high signal-to-noise ratio) overnoninvasive human neuroimaging techniques. Future DBS research using moresophisticated experimental paradigms and data analysis techniques, and comparingresults across multiple patient groups, will continue to provide insights into basicneurocognitive mechanisms and how these mechanisms become disrupted inpatient populations.

References

Bronstein JM, Tagliati M, Alterman RL, Lozano AM, Volkmann J, Stefani A, Horak FB, OkunMS, Foote KD, Krack P, Pahwa R, Henderson JM, Hariz MI, Bakay RA, Rezai A, Marks WJJr, Moro E, Vitek JL, Weaver FM, Gross RE, DeLong MR (2010) Deep brain stimulation forParkinson disease: an expert consensus and review of key issues. Arch Neurol 68:165

Canolty RT, Knight RT (2010) The functional role of cross-frequency coupling. Trends Cogn Sci14:506–515

Cohen MX, Axmacher N, Lenartz D, Elger CE, Sturm V, Schlaepfer TE (2009a) Nucleiaccumbens phase synchrony predicts decision-making reversals following negative feedback.J Neurosci 29:7591–7598

Cohen MX, Axmacher N, Lenartz D, Elger CE, Sturm V, Schlaepfer TE (2009b) Goodvibrations: cross-frequency coupling in the human nucleus accumbens during rewardprocessing. J Cogn Neurosci 21:875–889

190 M. X. Cohen

Cohen MX, Bour L, Mantione M, Figee M, Vink M, Tijssen MA, Rootselaar AF, Munckhof PV,Richard Schuurman P, Denys D (2012) Top–down-directed synchrony from medial frontalcortex to nucleus accumbens during reward anticipation. Hum Brain Mapp 33:246–252

de Koning PP, Figee M, van den Munckhof P, Schuurman PR, Denys D (2011) Current status ofdeep brain stimulation for obsessive-compulsive disorder: a clinical review of differenttargets. Curr Psychiatry Rep 13:274–282

Flora ED, Perera CL, Cameron AL, Maddern GJ (2010) Deep brain stimulation for essentialtremor: a systematic review. Mov Disord 25:1550–1559

Hariz MI, Robertson MM (2010) Gilles de la Tourette syndrome and deep brain stimulation. EurJ Neurosci 32:1128–1134

Lisman J (2005) The theta/gamma discrete phase code occurring during the hippocampal phaseprecession may be a more general brain coding scheme. Hippocampus 15:913–922

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM,Kennedy SH (2005) Deep brain stimulation for treatment-resistant depression. Neuron45:651–660

McCracken CB, Grace AA (2007) High-frequency deep brain stimulation of the nucleusaccumbens region suppresses neuronal activity and selectively modulates afferent drive in ratorbitofrontal cortex in vivo. J Neurosci 27:12601–12610

McCracken CB, Grace AA (2009) Nucleus accumbens deep brain stimulation produces region-specific alterations in local field potential oscillations and evoked responses in vivo.J Neurosci 29:5354–5363

McIntyre CC, Hahn PJ (2010) Network perspectives on the mechanisms of deep brainstimulation. Neurobiol Dis 38:329–337

Mian MK, Campos M, Sheth SA, Eskandar EN (2010) Deep brain stimulation for obsessive-compulsive disorder: past, present, and future. Neurosurg Focus 29:E10

Schlaepfer TE, Cohen MX, Frick C, Kosel M, Brodesser D, Axmacher N, Joe AY, Kreft M,Lenartz D, Sturm V (2008) Deep brain stimulation to reward circuitry alleviates anhedonia inrefractory major depression. Neuropsychopharmacology 33:368–377

Tan S, Vlamings R, Lim L, Sesia T, Janssen ML, Steinbusch HW, Visser-Vandewalle V, TemelY (2010) Experimental deep brain stimulation in animal models. Neurosurgery 67:1073–1079(discussion 1080)

Zaghloul KA, Blanco JA, Weidemann CT, McGill K, Jaggi JL, Baltuch GH, Kahana MJ (2009)Human substantia nigra neurons encode unexpected financial rewards. Science323:1496–1499

17 Scientific Recordings in Deep Brain Stimulation 191

Chapter 18Neurotransmitter Release During DeepBrain Stimulation

Osama A. Abulseoud, Emily J. Knight and Kendall H. Lee

18.1 Introduction

Although patients have benefited significantly from the development of newpharmacological treatments for psychiatric disorders, many of these therapies havebeen either not completely effective or not well tolerated over the long course ofthe disease. In response to these shortcomings, over the past 15 years there havebeen significant advances in stereotactic and functional neurosurgical techniquesthat have led to new strategies in the treatment of psychiatric disorders (Rempleet al. 2008; Poewe 2009). Among these newer surgical therapies, electrical stim-ulation of specific brain nuclei, known commonly as deep brain stimulation (DBS),has become a promising alternative to conventional pharmacological management.In particular, DBS is now FDA-approved for treatment of obsessive–compulsivedisorder (OCD) (Greenberg et al. 2006), and is under investigation for treatment oftreatment-resistant depression (Mayberg et al. 2005) and Tourette’s syndrome(Maciunas et al. 2007).

Despite its well-established clinical efficacy, the mechanism of DBS action isincompletely understood. Because DBS and ablative surgery (i.e., subthalamotomy)are similarly effective for treating Parkinson’s disease and essential tremor, thestimulation-evoked silencing of pathologically hyperactive neurons was initially

O. A. AbulseoudDepartment of Psychiatry and Psychology, Mayo Clinic, Rochester, MN, USA

E. J. KnightMayo Graduate School, Rochester, MN, USA

K. H. Lee (&)Department of Neurosurgery and Physiology, Mayo Clinic, Rochester, MN, USAe-mail: lee.kendall@mayo.edu

K. H. LeeNeuroengineering Laboratory, Mayo Clinic, Rochester, MN, USA

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_18, � Springer-Verlag Berlin Heidelberg 2012

193

postulated as the primary mechanism (Benabid et al. 1987, 2000). This notion wasfurther supported by early work measuring electrophysiological activity duringDBS (Beurrier et al. 2001; Magarinos-Ascone et al. 2002). However, more recentstudies have reported the activation of output nuclei (see Garcia et al. 2005 and thefollowing sections). This confounding paradox has apparently been resolved bymathematical models suggesting that, because of dissimilar excitability of neuralelements, soma inhibition and axonal activation are both expected at the DBSelectrode site (McIntyre and Grill 1998; McIntyre et al. 2004a). The axonal acti-vation hypothesis (McIntyre et al. 2004b, c; Johnson et al. 2008) has enormousimplications for the DBS mechanism of action in psychiatric disorders. Indeed, DBSshould evoke changes in neural activity and neurochemical transmission in inter-connected structures within the neural circuit that ultimately underlie clinicalbenefit. Nevertheless, our understanding of these local and distal effects of DBSremains far from complete, in large part because of the technical difficulties incombining measurement modalities for global assessment of neural activity andchemical-specific sensing.

18.2 DBS Evokes Changes in Distal Neural Activity

Electrophysiological studies have clearly demonstrated modulation of activity intarget neurons, consistent with axonal activation during DBS. For example, thesetarget neuronal effects have been recorded in the internal and external segments ofthe globus pallidus following subthalamic nucleus (STN) stimulation (Hashimotoet al. 2003; Kita et al. 2005; Miocinovic et al. 2006) and in the substantia nigrareticulata and substantia nigra compacta (Smith and Grace 1992; Benazzouz et al.2000; Maurice et al. 2003). Additionally, nucleus accumbens (NAc) DBS inhibitsfiring in orbitofrontal neurons in the rat model (McCracken and Grace 2007).Although definitive, the downside of the electrophysiological approach is thattargets must be selected a priori and few targets can be evaluated concurrently.As an alternative, for simultaneous global assessment of neural activity, brainimaging techniques are preferable.

As such, both PET and functional MRI (fMRI) brain imaging protocols havebeen used to assess the effects of DBS at the global-network level. For example,several clinical studies utilizing PET, H2

15O PET, and 18F-fluorodeoxyglucose(18F-FDG) PET support the axonal activation hypothesis of DBS (Ceballos-Bau-mann 2003). PET and H2

15O PET record changes in regional cerebral blood flow(CBF) (Hershey et al. 2003; Sestini et al. 2005), whereas 18F-FDG PET measuresregional cerebral glucose metabolism (Eidelberg and Edwards 2000), with bothconsidered to reflect altered local neuronal activity or changed input into theregion of measurement (Grafton and DeLong 1997). PET CBF studies haverevealed increased CBF in the subgenual cingulate and decreased CBF in theprefrontal cortex and anterior cingulate in patients with treatment-resistantdepression as compared with normal controls (Mayberg et al. 2005). Following

194 O. A. Abulseoud et al.

subgenual cingulate DBS, CBF decreased in the subgenual cingulate and increasedin the prefrontal cotrex and dorsal cingulate in these patients (Mayberg et al.2005). With use of 18F-FDG PET, nucleus accumbens (NAc) DBS has also beenshown to result in decreased metabolism in the subgenual cingulate and in pre-frontal regions in patients with treatment-resistant depression (Bewernick et al.2010) and OCD (Nuttin et al. 2003; Abelson et al. 2005; Van Laere et al. 2006).Taken together, these PET results suggest that the net effect of DBS is to modulatethe activity of output circuitry supporting the axonal activation hypothesis.

Clinical studies utilizing fMRI have also supported the axonal activationhypothesis. The fMRI protocol measures blood-oxygenation-level-dependent(BOLD) contrast (Ogawa et al. 1990), which provides in vivo real-time anatomicmaps of blood oxygenation in the brain under normal physiological conditions(Babiloni et al. 2009; van Eijsden et al. 2009). In the first attempt to utilize 1.5-TfMRI in four Parkinson’s disease patients during STN DBS, Jech et al. (2001)showed BOLD signal activation in structures of the basal ganglia complex such asthe globus pallidus, thalamus, substantia nigra, premotor cortex, and dorsolateralprefrontal cortex. In a more recent fMRI study examining the effects of DBS,Philips et al. (2006) implanted DBS electrodes bilaterally in five Parkinson’sdisease patients. They reported that BOLD signal activation was seen in theipsilateral basal ganglia in all subjects and ipsilateral thalamus in six of theelectrodes tested. Importantly, in one case report (Baker et al. 2007), stimulation ofthe anterior limb of the internal capsule and ventral striatum for the treatment ofOCD resulted in BOLD signal activation in the ipsilateral middle frontal gyrus,dorsomedial thalamus, putamen, anterior cingulate cortex, head of the caudate,globus pallidus, and contralateral cerebellum. In another study of four patients withOCD, fMRI was performed in one patient following bilateral DBS of the anteriorlimb of the internal capsule and revealed activation in the pons, striatum, rightfrontal cortex, and left superior temporal gyrus (Nuttin et al. 2003). In conjunctionwith the PET results, these fMRI studies provide further support for the axonalactivation hypothesis.

18.3 DBS Elicits Local and Distal NeurotransmitterRelease

Given the strong electrophysiological and imaging evidence for the axonal acti-vation hypothesis as described already, studies have been performed usingmicrodialysis to test the hypothesis that neurotransmitters are released in variousefferent targets during DBS (Bruet et al. 2001; Windels et al. 2003; Hamani et al.2010). For example, Hamani et al. (2010), using microdialysis, demonstratedserotonin release in the hippocampus following DBS of the ventromedial pre-frontal cortex in a rat model of depression. However, the relatively large size ofmicrodialysis probes has been shown to disrupt tissue in the immediate vicinity of

18 Neurotransmitter Release During Deep Brain Stimulation 195

the probe, resulting in underestimations of extracellular neurotransmitter levelscompared with alternative measurement techniques that utilize chemical micro-sensors (Clapp-Lilly et al. 1999; Borland et al. 2005). Thus, approaches alternativeto microdialysis will be necessary to assess neurotransmitter release during DBS.Indeed, chemical microsensors, which offer a smaller probe (5–10-lm diameter vs.200–400-lm diameter for microdialysis probes), have already shown dopaminerelease in the striatum evoked by STN DBS in the intact and parkinsonian rat 6-hydroxydopamine model (Lee et al. 2006; Blaha et al. 2008; Covey and Garris2009). These latter findings are important on several levels. For example, neuro-transmitter release during DBS has been difficult to establish with microdialysis(Paul et al. 2000; Bruet et al. 2001; Meissner et al. 2003), a result that underscoresthe need for a small probe size in chemical recordings.

Basic knowledge of the mechanism of DBS will ultimately be critical to thefurther development of the technology and surgical procedures to produce asignificant improvement in patient outcome. To examine the neurochemical effectsof DBS, our laboratory has developed a novel device called Wireless InstantaneousNeurotransmitter Concentration Sensor System (WINCS) specifically designed tomonitor neurochemical release during both experimental and clinical DBS surgicalprocedures (Fig. 18.1). As such, research subject safety, signal fidelity, and inte-gration with existing DBS surgical procedures were key priorities during thedevelopment of WINCS. This device, designed in compliance with FDA-recog-nized consensus standards for medical electrical device safety, consists of a rela-tively small, wireless, sterilizable battery-powered unit that can interface withcarbon fiber microelectrode (CFM) or enzyme-based microsensors for real-timemonitoring of neurotransmitter release in the brain (Agnesi et al. 2009; Bledsoe et al.2009; Shon et al. 2010a). For neurotransmitter release studies, the WINCS devicehas significant advantages over other commercially available wireless recordingsystems as it offers (1) an advanced microprocessor with superior analog-to-digitalconversion, greater internal memory, and faster clock speed, (2) wirelesslyprogrammable waveform parameters (scan bias, range, and rate) using an advancedbluetooth module for wireless communication, (3) a higher precision voltagereference for the microprocessor, (4) a low-power mode to preserve battery life,voltage sensing, and low-power alert, and most importantly (5) proven compatibilityand functionality in the bore of an MRI scanner during image acquisition.

A putative neurochemical that may be of importance to DBS mechanisms isadenosine. Using a bioluminescence technique, Bekar et al. (2008) showed inthalamic slices and exposed mouse cortices in vivo that high-frequency stimulationtriggered an abrupt increase in the level of extracellular ATP around the stimulationelectrode. Removal of extracellular Ca2+ from the bath solution in the sliceexperiments (to prevent synaptic release of ATP) resulted in enhanced ATPbioluminescence, indicating that ATP release was primarily nonsynaptic andprobably resulted from an efflux of cytosolic ATP. Furthermore, this group showedthat extracellular hydrolysis of released ATP is associated with increases in the levelof extracellular adenosine and that adenosine A1 receptor activation depresses

196 O. A. Abulseoud et al.

excitatory transmission in the thalamus and reduces both tremor- and DBS-inducedside effects.

Cechova and Venton (2008) measured adenosine release in the striatum withchemical microsensors during electrical stimulation in the vicinity of the nigro-striatal dopaminergic tract. Our group (Agnesi et al. 2009) used an adenosinebiosensor coupled with WINCS-based fixed-potential amperometry in small-animalmodels to record local release of adenosine following stimulation of the ventro-lateral thalamus with various frequencies. Extracellular adenosine concentrationsincreased proportionately with increasing current intensity or frequency andreturned back to prestimulation baseline levels between stimulations. Importantly,increases in the concentrations of extracellular adenosine appear to match elevationsin CBF that result from increases in neural activity (Brundege and Dunwiddie 1997;Phillis 2004).

In addition to adenosine, we used WINCS-based fixed-potential amperometryrecordings to elicit evoked local glutamate release following high-frequencystimulation of a large-animal (pig) motor cortex. Similar to adenosine, extracel-lular glutamate concentrations increased proportionately with increasing currentintensities and returned back to prestimulation levels between stimulations (Agnesiet al. 2009). Furthermore, we have demonstrated (Griessenauer et al. 2010) thecapability of WINCS to detect serotonin release in a slice preparation of the dorsalraphe nucleus.

We have shown that STN DBS elicits dopamine and adenosine release distallyin the caudate nucleus of pigs (Shon et al. 2010b), and recently we established themeasurement of electrically evoked dopamine levels in the anesthetized maleSprague–Dawley rat using fast-scan cyclic voltammetry at a carbon-fiber micro-electrode (CFM). For our small-animal experiment, we implanted a CFM into thecore of the NAc of a urethane-anesthetized rat and fast-scan cyclic voltammetryrecordings were taken during brief (2-s) electrical stimulation with an electrodeimplanted in the ipsilateral medial forebrain bundle. With high-frequency

Fig. 18.1 Pig MRI (a) and experimental setup (b) for neurochemical recordings of dopamineand adenosine release evoked by site-specific electrical stimulation. Two potentiometers areshown: a commercially available hardwired potentiostat and, in the blowup, the wirelesspotentiostat attached to the head frame that we have developed and that is MRI-compatible

18 Neurotransmitter Release During Deep Brain Stimulation 197

stimulation (100 Hz), dopamine was clearly released immediately during and afterstimulation (Fig. 18.2). The color plot (Fig. 18.2, panel A) and the background-subtracted voltammogram (Fig. 18.2, panel C) demonstrate the appearance ofdopamine release from the NAc immediately during and after stimulation of themedial forebrain bundle (Fig. 18.2, panel D), with the voltammogram peak atapproximately +0.6 V indicating dopamine oxidation (Fig. 18.2, panel B).

Taken together, studies monitoring neurotransmitter release strongly suggestthat chemical microsensors are well suited for establishing neurochemical corre-lates of DBS locally and within the neural network. Combined with brain imagingevidence that neural activity in specific brain areas is increased, the available dataalso suggest that release of neurotransmitters may be the mediator of DBS efficacy.However, establishing a causal relationship between functional activation and

Fig. 18.2 In vivo dopamine release measured with Wireless Instantaneous NeurotransmitterConcentration Sensor System (WINCS)-based fast-scan cyclic voltammetry at carbon-fibermicroelectrodes in the nucleus accumbens of the anesthetized rat. a Electrical stimulation(100 Hz, 0.5-ms pulse width, for 2 s) of the medial forebrain bundle evoked dopamine release inthe nucleus accumbens. The color plot shows the appearance of dopamine release immediatelyduring and after stimulation. b Current versus time plot at +0.6 V. c Background-subtractedvoltammogram for dopamine demonstrates measurement of dopamine release (red line in plot a).The black line indicates the current generated by forward-going potentials, and the red lineindicates the current generated by the reverse-going potentials. d Photograph of the WINCSdevice

198 O. A. Abulseoud et al.

neurotransmitter release in the same brain region measured simultaneously byfMRI and chemical microsensors is yet to be accomplished.

18.4 Current Understanding of the Pathogenesisand Treatment of Psychiatric Disorders:Implications for Mechanisms of DBS

Despite the complexity of the neurological conditions treated with DBS, theneurobiological underpinnings of psychiatric disorders remain far more complex.Psychotropic drugs, in general, regulate the availability of a particular neuro-transmitter or serve as ligands for particular neurotransmitter receptors (Nestlerand Duman 2002). However, the fact that mental illnesses tend to be chronic innature, and a substantial proportion of symptom improvement occurs only afterseveral months of long-term stimulation (Greenberg et al. 2006), argues thattherapeutic benefits may result from underlying long-term changes (plasticity)occurring within the neuronal network (Lujan et al. 2008). Thus, it is not clear howDBS’s ability to modulate neurotransmitters, which presumably occurs immedi-ately, can also improve psychiatric disorders. Yet, in refractory depression,symptom improvement as a result of stimulating the subgenual cingulate cortex(Mayberg et al. 2005) or the NAc (Schlaepfer et al. 2008; Malone et al. 2009) hasbeen well documented.

Although establishing the necessary link between symptom change and a cor-responding change in neurotransmitters still requires further investigation, studieshave implicated several neurotransmitters, including serotonin (Willner 1985;Nagayama et al. 1991), dopamine (Dunlop and Nemeroff 2007), and glutamate(Mitchell and Baker 2010), in the pathogenesis and treatment of psychiatric dis-orders. These same neurotransmitters may also underlie the therapeutic effect ofDBS. For example, as a hallmark of antidepressant medications, serotonin (Nic-hols and Nichols 2008) has been well studied in animal models of depression(Willner 1985; Nagayama et al. 1991). With use of microdialysis techniques,central serotonin levels have been shown to increase following long-term treat-ment with antidepressant medications such as fluoxetine, paroxetine, and citalo-pram (Dawson et al. 2002). Although microdialysis technology couldunderestimate extracellular neurotransmitter levels, the rise in serotonin levelscould not completely explain the mechanism of action of these antidepressantsgiven the delayed effect of medications on depressive symptoms in the face of animmediate rise of serotonin levels. However, serotonin may be one importantneurotransmitter in the mechanism of DBS, as demonstrated by Hamani et al.(2010). They found a rise in serotonin level in rat hippocampus following stim-ulation of the ventromedial prefrontal cortex. Additionally, we (Griessenauer et al.2010) have reported DBS-related modulation of serotonin-transmission in thedorsal raphe nucleus in rats by WINCS.

18 Neurotransmitter Release During Deep Brain Stimulation 199

The role of dopamine dysregulation in depression has also become a focus ofrecent attention (Dunlop and Nemeroff 2007), and studies suggest that dopaminemay be involved in the therapeutic effect of DBS (Vernaleken et al. 2009; Walkeret al. 2009; Falowski et al. 2011). Indeed, using a Western blot analysis ofextracted tissue, Falowski et al. (2011) have shown in a rat model of depressionthat NAc DBS decreases expression of tyrosine hydroxylase in the prefrontalcortex, suggesting modulation of the monoamine system with DBS. Further,concomitant decreases in prefrontal cortex norepinephrine and dopamine levelswere detected using high-pressure liquid chromatography, and rats receiving theNAc DBS demonstrated a reduction in depressive and anxious behaviors (Fa-lowski et al. 2011). Also, in subjects with Tourette’s syndrome, where dopaminefunction alteration has been studied over the past four decades (Steeves et al.2010), the results remain controversial. However, one study found that bilateralthalamic stimulation caused a 16.3 % decrease in dopamine binding potential asmeasured by 18F-PET from the on to off condition, which suggests an increase indopamine release due to thalamic stimulation (Vernaleken et al. 2009).

In addition to the monoamines, glutamate is another critical neurotransmitterthat has been shown to be both implicated in the neurobiological processesinvolved in depression (Mitchell and Baker 2010) and modulated by DBS (Agnesiet al. 2010). Indeed, our laboratory has shown that local thalamic stimulationresults in glutamate release in rats (Agnesi et al. 2010). Taken together, theseresults are beginning to support a role for several neurotransmitter systems in thetherapeutic effect of DBS for psychiatric conditions.

18.5 Conclusion

The axonal activation hypothesis has important implications for how we approachmechanisms of action of DBS for treatment of psychiatric disorders. As opposed toa predominant effect of local inhibition at the stimulation site, the prevailing effectappears to be excitation of efferent target neurons and subsequent changes inneural network activity. There is mounting evidence that these changes includedneurotransmitter release locally and at various nodes within the neural network.Despite what appears to be growing acceptance of this general scheme, however,which regions are affected, how they are affected, and what neurotransmittersmediate these changes remain largely unanswered. Answers to these questionsmay allow improvement in the use of DBS for treatment of psychiatric disorders.

200 O. A. Abulseoud et al.

References

Abelson JL, Curtis GC, Sagher O, Albucher RC, Harrigan M, Taylor SF, Martis B, Giordani B(2005) Deep brain stimulation for refractory obsessive-compulsive disorder. Biol Psychiatry57(5):510–516

Agnesi F, Tye SJ, Bledsoe JM, Griessenauer CJ, Kimble CJ, Sieck GC, Bennet KE, Garris PA,Blaha CD, Lee KH (2009) Wireless instantaneous neurotransmitter concentration system-based amperometric detection of dopamine, adenosine, and glutamate for intraoperativeneurochemical monitoring. J Neurosurg 111(4):701–711

Agnesi F, Blaha CD, Lin J, Lee KH (2010) Local glutamate release in the rat ventral lateralthalamus evoked by high-frequency stimulation. J Neural Eng 7(2):26009

Babiloni C, Pizzella V, Gratta CD, Ferretti A, Romani GL (2009) Fundamentals ofelectroencefalography, magnetoencefalography, and functional magnetic resonance imaging.Int Rev Neurobiol 86:67–80

Baker KB, Kopell BH, Malone D, Horenstein C, Lowe M, Phillips MD, Rezai AR (2007) Deepbrain stimulation for obsessive-compulsive disorder: using functional magnetic resonanceimaging and electrophysiological techniques: technical case report. Neurosurgery 61(5 Suppl2):E367–E368; discussion E368

Bekar L, Libionka W, Tian GF, Xu Q, Torres A, Wang X, Lovatt D, Williams E, Takano T,Schnermann J, Bakos R, Nedergaard M (2008) Adenosine is crucial for deep brainstimulation-mediated attenuation of tremor. Nat Med 14(1):75–80

Benabid AL, Pollak P, Louveau A, Henry S, de Rougemont J (1987) Combined (thalamotomyand stimulation) stereotactic surgery of the VIM thalamic nucleus for bilateral Parkinsondisease. Appl Neurophysiol 50(1–6):344–346

Benabid AL, Koudsie A, Benazzouz A, Fraix V, Ashraf A, Le Bas JF, Chabardes S, Pollak P(2000) Subthalamic stimulation for Parkinson’s disease. Arch Med Res 31(3):282–289

Benazzouz A, Gao DM, Ni ZG, Piallat B, Bouali-Benazzouz R, Benabid AL (2000) Effect ofhigh-frequency stimulation of the subthalamic nucleus on the neuronal activities of thesubstantia nigra pars reticulata and ventrolateral nucleus of the thalamus in the rat.Neuroscience 99(2):289–295

Beurrier C, Bioulac B, Audin J, Hammond C (2001) High-frequency stimulation produces atransient blockade of voltage-gated currents in subthalamic neurons. J Neurophysiol85(4):1351–1356

Bewernick BH, Hurlemann R, Matusch A, Kayser S, Grubert C, Hadrysiewicz B, Axmacher N,Lemke M, Cooper-Mahkorn D, Cohen MX, Brockmann H, Lenartz D, Sturm V, SchlaepferTE (2010) Nucleus accumbens deep brain stimulation decreases ratings of depression andanxiety in treatment-resistant depression. Biol Psychiatry 67(2):110–116

Blaha CD, Lester DB, Ramsson ES, Lee KH, Garris PA (2008) Striatal dopamine release evokedby subthalamic stimulation in intact and 6-OHDA-lesioned rats: relevance to deep brainstimulation in Parkinson’s disease. In: Proceedings of the 12th international conference on InVivo methods, University of British Columbia, Vancouver, Canada, pp 395–397

Bledsoe JM, Kimble CJ, Covey DP, Blaha CD, Agnesi F, Mohseni P, Whitlock S, Johnson DM,Horne A, Bennet KE, Lee KH, Garris PA (2009) Development of the wireless instantaneousneurotransmitter concentration system for intraoperative neurochemical monitoring usingfast-scan cyclic voltammetry. J Neurosurg 111(4):712–723

Borland LM, Shi G, Yang H, Michael AC (2005) Voltammetric study of extracellular dopaminenear microdialysis probes acutely implanted in the striatum of the anesthetized rat. J NeurosciMethods 146(2):149–158

Bruet N, Windels F, Bertrand A, Feuerstein C, Poupard A, Savasta M (2001) High frequencystimulation of the subthalamic nucleus increases the extracellular contents of striataldopamine in normal and partially dopaminergic denervated rats. J Neuropathol Exp Neurol60(1):15–24

18 Neurotransmitter Release During Deep Brain Stimulation 201

Brundege JM, Dunwiddie TV (1997) Role of adenosine as a modulator of synaptic activity in thecentral nervous system. Adv Pharmacol 39:353–391

Ceballos-Baumann AO (2003) Functional imaging in Parkinson’s disease: activation studies withPET, fMRI and SPECT. J Neurol 250(Suppl 1):I15–I23

Cechova S, Venton BJ (2008) Transient adenosine efflux in the rat caudate-putamen. J Neurochem105(4):1253–1263

Clapp-Lilly KL, Roberts RC, Duffy LK, Irons KP, Hu Y, Drew KL (1999) An ultrastructuralanalysis of tissue surrounding a microdialysis probe. J Neurosci Methods 90(2):129–142

Covey DP, Garris PA (2009) Using fast-scan cyclic voltammetry to evaluate striatal dopaminerelease elicited by subthalamic nucleus stimulation. Conf Proc IEEE Eng Med Biol Soc2009:3306–3309

Dawson LA, Nguyen HQ, Smith DL, Schechter LE (2002) Effect of chronic fluoxetine andWAY-100635 treatment on serotonergic neurotransmission in the frontal cortex. J Psycho-pharmacol 16(2):145–152

Dunlop BW, Nemeroff CB (2007) The role of dopamine in the pathophysiology of depression.Arch Gen Psychiatry 64(3):327–337

Eidelberg D, Edwards C (2000) Functional brain imaging of movement disorders. Neurol Res22(3):305–312

Falowski SM, Sharan A, Reyes BA, Sikkema C, Szot P, Van Bockstaele EJ (2011) An Evaluationof Neuroplasticity and Behavior Following Deep Brain Stimulation of the NucleusAccumbens in an Animal Model of Depression. Neurosurgery 69(6):1281–1290

Garcia L, D’Alessandro G, Bioulac B, Hammond C (2005) High-frequency stimulation inParkinson’s disease: more or less? Trends Neurosci 28(4):209–216

Grafton ST, DeLong M (1997) Tracing the brain’s circuitry with functional imaging. Nat Med3(6):602–603

Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, Salloway SP, Okun MS,Goodman WK, Rasmussen SA (2006) Three-year outcomes in deep brain stimulation for highlyresistant obsessive-compulsive disorder. Neuropsychopharmacology 31(11):2384–2393

Griessenauer CJ, Chang SY, Tye SJ, Kimble CJ, Bennet KE, Garris PA, Lee KH (2010) WirelessInstantaneous Neurotransmitter Concentration System: electrochemical monitoring of serotoninusing fast-scan cyclic voltammetry—a proof-of-principle study. J Neurosurg 113(3):656–665

Hamani C, Diwan M, Macedo CE, Brandao ML, Shumake J, Gonzalez-Lima F, Raymond R,Lozano AM, Fletcher PJ, Nobrega JN (2010) Antidepressant-like effects of medial prefrontalcortex deep brain stimulation in rats. Biol Psychiatry 67(2):117–124

Hashimoto T, Elder CM, Okun MS, Patrick SK, Vitek JL (2003) Stimulation of the subthalamicnucleus changes the firing pattern of pallidal neurons. J Neurosci 23(5):1916–1923

Hershey T, Revilla FJ, Wernle AR, McGee-Minnich L, Antenor JV, Videen TO, Dowling JL,Mink JW, Perlmutter JS (2003) Cortical and subcortical blood flow effects of subthalamicnucleus stimulation in PD. Neurology 61(6):816–821

Jech R, Urgosik D, Tintera J, Nebuzelsky A, Krasensky J, Liscak R, Roth J, Ruzicka E (2001)Functional magnetic resonance imaging during deep brain stimulation: a pilot study in fourpatients with Parkinson’s disease. Mov Disord 16(6):1126–1132

Johnson MD, Miocinovic S, McIntyre CC, Vitek JL (2008) Mechanisms and targets of deep brainstimulation in movement disorders. Neurotherapeutics 5(2):294–308

Kita H, Tachibana Y, Nambu A, Chiken S (2005) Balance of monosynaptic excitatory anddisynaptic inhibitory responses of the globus pallidus induced after stimulation of thesubthalamic nucleus in the monkey. J Neurosci 25(38):8611–8619

Lee KH, Blaha CD, Harris BT, Cooper S, Hitti FL, Leiter JC, Roberts DW, Kim U (2006)Dopamine efflux in the rat striatum evoked by electrical stimulation of the subthalamic nucleus:potential mechanism of action in Parkinson’s disease. Eur J Neurosci 23(4):1005–1014

Lujan JL, Chaturvedi A, McIntyre CC (2008) Tracking the mechanisms of deep brain stimulationfor neuropsychiatric disorders. Front Biosci 13:5892–5904

202 O. A. Abulseoud et al.

Maciunas RJ, Maddux BN, Riley DE, Whitney CM, Schoenberg MR, Ogrocki PJ, Albert JM,Gould DJ (2007) Prospective randomized double-blind trial of bilateral thalamic deep brainstimulation in adults with Tourette syndrome. J Neurosurg 107(5):1004–1014

Magarinos-Ascone C, Pazo JH, Macadar O, Buno W (2002) High-frequency stimulation of thesubthalamic nucleus silences subthalamic neurons: a possible cellular mechanism inParkinson’s disease. Neuroscience 115(4):1109–1117

Malone DA Jr, Dougherty DD, Rezai AR, Carpenter LL, Friehs GM, Eskandar EN, Rauch SL,Rasmussen SA, Machado AG, Kubu CS, Tyrka AR, Price LH, Stypulkowski PH, Giftakis JE,Rise MT, Malloy PF, Salloway SP, Greenberg BD (2009) Deep brain stimulation of the ventralcapsule/ventral striatum for treatment-resistant depression. Biol Psychiatry 65(4):267–275

Maurice N, Thierry AM, Glowinski J, Deniau JM (2003) Spontaneous and evoked activity ofsubstantia nigra pars reticulata neurons during high-frequency stimulation of the subthalamicnucleus. J Neurosci 23(30):9929–9936

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM,Kennedy SH (2005) Deep brain stimulation for treatment-resistant depression. Neuron45(5):651–660

McCracken CB, Grace AA (2007) High-frequency deep brain stimulation of the nucleusaccumbens region suppresses neuronal activity and selectively modulates afferent drive in ratorbitofrontal cortex in vivo. J Neurosci 27(46):12601–12610

McIntyre CC, Grill WM (1998) Sensitivity analysis of a model of mammalian neural membrane.Biol Cybern 79(1):29–37

McIntyre CC, Grill WM, Sherman DL, Thakor NV (2004a) Cellular effects of deep brainstimulation: model-based analysis of activation and inhibition. J Neurophysiol 91(4):1457–1469

McIntyre CC, Mori S, Sherman DL, Thakor NV, Vitek JL (2004b) Electric field and stimulatinginfluence generated by deep brain stimulation of the subthalamic nucleus. Clin Neurophysiol115(3):589–595

McIntyre CC, Savasta M, Kerkerian-Le Goff L, Vitek JL (2004c) Uncovering the mechanism(s)of action of deep brain stimulation: activation, inhibition, or both. Clin Neurophysiol115(6):1239–1248

Meissner W, Harnack D, Reese R, Paul G, Reum T, Ansorge M, Kusserow H, Winter C,Morgenstern R, Kupsch A (2003) High-frequency stimulation of the subthalamic nucleusenhances striatal dopamine release and metabolism in rats. J Neurochem 85(3):601–609

Miocinovic S, Parent M, Butson CR, Hahn PJ, Russo GS, Vitek JL, McIntyre CC (2006)Computational analysis of subthalamic nucleus and lenticular fasciculus activation duringtherapeutic deep brain stimulation. J Neurophysiol 96(3):1569–1580

Mitchell ND, Baker GB (2010) An update on the role of glutamate in the pathophysiology ofdepression. Acta Psychiatr Scand 122(3):192–210

Nagayama H, Tsuchiyama K, Yamada K, Akiyoshi J (1991) Animal study on the role ofserotonin in depression. Prog Neuropsychopharmacol Biol Psychiatry 15(6):735–744

Nestler E, Duman R (2002) Neuropsychopharmacology: the fifth generation of progress: anofficial publication of the American College of Neuropsychopharmacology. K. Davis,Lippincott Williams and Wilkins, Philadelphia

Nichols DE, Nichols CD (2008) Serotonin receptors. Chem Rev 108(5):1614–1641Nuttin BJ, Gabriels LA, Cosyns PR, Meyerson BA, Andreewitch S, Sunaert SG, Maes AF,

Dupont PJ, Gybels JM, Gielen F, Demeulemeester HG (2003) Long-term electrical capsularstimulation in patients with obsessive-compulsive disorder. Neurosurgery 52(6):1263–1272;discussion 1272–1264

Ogawa S, Lee TM, Nayak AS, Glynn P (1990) Oxygenation-sensitive contrast in magneticresonance image of rodent brain at high magnetic fields. Magn Reson Med 14(1):68–78

Paul G, Reum T, Meissner W, Marburger A, Sohr R, Morgenstern R, Kupsch A (2000) Highfrequency stimulation of the subthalamic nucleus influences striatal dopaminergic metabolismin the naive rat. NeuroReport 11(3):441–444

18 Neurotransmitter Release During Deep Brain Stimulation 203

Phillips MD, Baker KB, Lowe MJ, Tkach JA, Cooper SE, Kopell BH, Rezai AR (2006)Parkinson disease: pattern of functional MR imaging activation during deep brain stimulationof subthalamic nucleus–initial experience. Radiology 239(1):209–216

Phillis JW (2004) Adenosine and adenine nucleotides as regulators of cerebral blood flow: rolesof acidosis, cell swelling, and KATP channels. Crit Rev Neurobiol 16(4):237–270

Poewe W (2009) Treatments for Parkinson disease–past achievements and current clinical needs.Neurology 72(7 Suppl):S65–S73

Remple MS, Sarpong Y, Neimat JS (2008) Frontiers in the surgical treatment of Parkinson’sdisease. Expert Rev Neurother 8(6):897–906

Schlaepfer TE, Cohen MX, Frick C, Kosel M, Brodesser D, Axmacher N, Joe AY, Kreft M,Lenartz D, Sturm V (2008) Deep brain stimulation to reward circuitry alleviates anhedonia inrefractory major depression. Neuropsychopharmacology 33(2):368–377

Sestini S, Ramat S, Formiconi AR, Ammannati F, Sorbi S, Pupi A (2005) Brain networksunderlying the clinical effects of long-term subthalamic stimulation for Parkinson’s disease: a4-year follow-up study with rCBF SPECT. J Nucl Med 46(9):1444–1454

Shon YM, Chang SY, Tye SJ, Kimble CJ, Bennet KE, Blaha CD, Lee KH (2010a) Comonitoringof adenosine and dopamine using the Wireless Instantaneous Neurotransmitter ConcentrationSystem: proof of principle. J Neurosurg 112(3):539–548

Shon YM, Lee KH, Goerss SJ, Kim IY, Kimble C, Van Gompel JJ, Bennet K, Blaha CD, ChangSY (2010b) High frequency stimulation of the subthalamic nucleus evokes striatal dopaminerelease in a large animal model of human DBS neurosurgery. Neurosci Lett 475(3):136–140

Smith ID, Grace AA (1992) Role of the subthalamic nucleus in the regulation of nigral dopamineneuron activity. Synapse 12(4):287–303

Steeves TD, Ko JH, Kideckel DM, Rusjan P, Houle S, Sandor P, Lang AE, Strafella AP (2010)Extrastriatal dopaminergic dysfunction in Tourette syndrome. Ann Neurol 67(2):170–181

van Eijsden P, Hyder F, Rothman DL, Shulman RG (2009) Neurophysiology of functionalimaging. Neuroimage 45(4):1047–1054

Van Laere K, Nuttin B, Gabriels L, Dupont P, Rasmussen S, Greenberg BD, Cosyns P (2006)Metabolic imaging of anterior capsular stimulation in refractory obsessive-compulsivedisorder: a key role for the subgenual anterior cingulate and ventral striatum. J Nucl Med47(5):740–747

Vernaleken I, Kuhn J, Lenartz D, Raptis M, Huff W, Janouschek H, Neuner I, Schaefer WM,Grunder G, Sturm V (2009) Bithalamical deep brain stimulation in Tourette syndrome isassociated with reduction in dopaminergic transmission. Biol Psychiatry 66(10):e15–e17

Walker RH, Koch RJ, Moore C, Meshul CK (2009) Subthalamic nucleus stimulation andlesioning have distinct state-dependent effects upon striatal dopamine metabolism. Synapse63(2):136–146

Willner P (1985) Antidepressants and serotonergic neurotransmission: an integrative review.Psychopharmacology (Berl) 85(4):387–404

Windels F, Bruet N, Poupard A, Feuerstein C, Bertrand A, Savasta M (2003) Influence of thefrequency parameter on extracellular glutamate and gamma-aminobutyric acid in substantianigra and globus pallidus during electrical stimulation of subthalamic nucleus in rats.J Neurosci Res 72(2):259–267

204 O. A. Abulseoud et al.

Chapter 19The Potential Role of Nonneuronal Cellsin the Deep Brain Stimulation Mechanism

What Are Glia? What Are Their Functions?Could They Be Players in Deep BrainStimulation?

Vinata Vedam-Mai, Michael S. Okun and Elly M. Hol

19.1 Introduction

Glial cells, often referred to as neuroglia or glia, are nonneuronal cells in theperipheral nervous system and the central nervous system (CNS). They are capableof maintaining homeostasis, providing support and protection for the brain’sneurons, and are responsible for forming myelin. The different classes of glial cellsare astrocytes, oligodendrocytes, microglia, and oligodendrocyte precursor cells,or NG2 cells. Glial cells are typically categorized on the basis of their anatomy,physiological features, and the markers they express. True to the Greek origin ofthe word, glia act as the glue of the nervous system. However, they also function to(1) surround the neurons and hold them in place, (2) supply nutrients to neurons,(3) act as insulation, and (4) assist in phagocytosis. It has been thought for manyyears that glia do not play an active role in neurotransmission; however, it isbecoming clearer that astrocytes can modulate neurotransmission, and the mech-anisms by which these cells exert their action are becoming better characterized(Panatier et al. 2011; Volterra et al. 2005; Wang et al. 2008).

V. Vedam-Mai (&) � M. S. OkunDepartment of Neurology, Movement Disorders Center, McKnight Brain Institute,University of Florida College of Medicine, 100 S. Newell Drive,Gainesville, FL 32610, USAe-mail: vinved@neurosurgery.ufl.edu

E. M. HolNetherlands Institute for Neuroscience, Royal Netherlands Academy of Arts and Sciences,Amsterdam, The Netherlands

E. M. HolCenter for Neuroscience, Swammerdam Institute for Life Sciences,University of Amsterdam, Amsterdam, The Netherlands

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_19, � Springer-Verlag Berlin Heidelberg 2012

205

In the rodent CNS each astrocyte supports and modulates about 100,000 syn-apses; this number is even higher in the human CNS, where about two millionsynapses are supported by a single astrocyte (Oberheim et al. 2006). This factimplies that a single astrocyte might control the function of an incredibly largenumber of neurons. It has been known for a long time that astrocytes are criticallyinvolved in the glutamate–glutamine cycle primarily because of glutamate uptake.The glutamine synthesized is utilized by neurons for the de novo synthesis ofglutamate and GABA. Because of their unique morphology, astrocytes projecttheir processes towards blood vessels, and the terminating ‘‘endfeet’’ line the wallsof the blood vessels (Wang et al. 2008). Hence, it is surmised that astrocytesprovide metabolic support for neurons, although it is still rather unclear as to howexactly this is achieved (Allaman et al. 2011). Astrocytes are able to sense neu-ronal activity by ion channels, receptors, and transporters and causing localincreases in calcium levels (within the glial microdomain), as well as a moregeneral increase in levels of intracellular calcium (Grosche et al. 1999). Hence it ispossible that astrocytes can process and thus respond to neuronal activity(Perea et al. 2005). NG2 cells are present throughout the human brain, both in thegray matter region and in the white matter region (Dawson et al. 2003). Quite likeastrocytes, NG2 glial cells also express receptors for neurotransmitters(Papay et al. 2004; Lin and Bergles 2004). This is indicative that neuronal activityinfluences their behavior. Recent research suggests that NG2 cells are involved inrapid signaling with neighboring neurons, which is achieved through direct syn-apses (Paukert et al. 2006). The actual mechanism of this signaling has beendescribed in distinctly different brain regions, such as the hippocampus, cortex,and cerebellum. Through the synapses, there is Ca2+ signaling within the NG2cells, and the Ca2+ influx is modifiable through neuronal activity (Paukert et al.2006). It seems probable that the NG2 cells are capable of detecting neighboringneuronal activity, and that they achieve this through neurotransmitters. Receptoractivation in vitro has been shown to induce Ca2+ influx, resulting in early geneexpression, proliferation, and lineage progression (Kirchhoff and Kettenmann1992; Gallo et al. 1994; Knutson et al. 1997; Velez-Fort et al. 2010; Nishiyamaet al. 2009). Therefore, we can conclude that (1) glia express voltage-sensitivechannels and metabotropic and ionotropic neurotransmitter receptors channelsakin to neurons, and this property makes them capable of transmitting andreceiving neuroactive signals, (2) glia communicate between each other throughcalcium waves, and (3) the dynamic and bidirectional communication betweenneurons and glia has led to our defining a ‘‘tripartite synapse’’ which includes gliaas an active player in communication with the neuron (Eroglu et al. 2010; Sakryet al. 2011).

206 V. Vedam-Mai et al.

19.2 What Is Deep Brain Stimulation?

Deep brain stimulation (DBS) was established as an alternative to ablative stere-otaxy by the modern French neurosurgeon Benabid and his colleagues in Grenoblein 1987 (although many scientists prior to this long-lasting procedure hadperformed brain stimulation). Since lesioning was accomplished by burning thetarget brain tissue using an electrode, the surgeons needed to tweak the physio-logical frequencies, as well as attempt to stimulate surrounding areas adjacent tothe target to ensure that the best site had been targeted, and adverse events limited.It was during such testing procedures that Benabid discovered he could suppresstremor by using high-frequency stimulation (HFS). By the late 1990s, the sub-thalamic nucleus (STN) and the internal globus pallidus had been established asvaluable targets for patients with Parkinson’s disease. Today, DBS is an extremelypowerful tool and may be utilized in the treatment of many neurological as well aspsychiatric diseases. Even though it has been several decades since its first use, andits clinical application has grown exponentially, the underlying mechanisms andits overall impact on the neural network remain elusive.

Although there are several hypotheses and reviews regarding the topic, it stillremains largely unclear as to how DBS actually works. Early hypotheses postu-lated that DBS activates neuronal elements in the surrounding and stimulatedregion (Windels et al. 2000; Matsunaga et al. 2001); however, it has subsequentlybeen argued that DBS actually inhibits neuronal activity at the site of stimulation,thereby leading to a total decreased input from the stimulated structure/nucleus,but DBS also seems to excite fibers (Vitek 2002; Boraud et al. 1996; Wu et al.2001; Dostrovsky et al. 2000; Beurrier et al. 2001). Although neurons near thestimulation electrode can be inhibited during HFS, those farther away can actuallybe activated, making the mechanism of action quite complex (Vitek 2002). Fur-thermore, given the complexity of the nervous system, and the proximity andproperties of the different cell types present surrounding the site of actual stimu-lation, it is conceivable that DBS is effected by glial cells working through neu-ronal networks (Vedam-Mai et al. 2011). Considering their proximity and role theyplay in detecting neuronal activity as described in the previous section, it is quitepossible that glial cells are involved in modulating DBS-induced neuronalfunction.

19.3 Cellular Hypotheses of High-Frequency DBS

19.3.1 Can Glia Be Electrically Stimulated?

Typically, neural cells react to their environment, and specifically respond whenthere are changes in their environment. These cellular responses can be eithermolecular or physical in nature. It is known that electrical stimulation affects

19 The Potential Role of Nonneuronal Cells 207

neuronal bodies and axons, but it has also been shown that glial cells can react toelectrical stimulation by a molecular response (Yanagida et al. 2000). Hence, eventhough it is known in which brain areas DBS has its effects, it is hard to say whichcells being affected by DBS cause the resulting therapeutic effect.

Kojima et al. (1992) and others have reported that physical stress such aselectrical stimulation can result in regulation of cellular function (Yanagida et al.2000). They described in vitro experiments where the expression of specific genes(nerve growth factor, c-fos, c-jun) was upregulated as a result of low-frequencyelectrical stimulation of cells (Koyama et al. 1997). Furthermore, they discussedthe possibility that the electrical stimulation due to a Ca2+ influx could activate thedifferentiation of PC12 cells into neurons as a consequence of activation of dif-ferent second-messenger systems (Kimura et al. 1998). There are several suchreports of electrically induced cellular and molecular effects (Kojima et al. 1992;Koyama et al. 1997; Kimura et al. 1998); however, the mechanisms of downstreamand long-term effects require further careful study.

It has been proposed that astrocytic membrane responses can mirror synapticevents under an HFS paradigm such as that shown in vitro in primary hippocampalastrocyte cultures (Bekar et al. 2005). This capability to mirror neuronal activitymay help astrocytes react to sudden high-frequency signals and may perhapssynchronize glial functioning to neuronal activity (Bekar et al. 2005).

Most studies investigating the mechanism of action of DBS tend to focus onchanges in neuronal activity in the target area of stimulation (inhibition), and theexcitation of targets and circuits. It is possible, however, that DBS causes an effectin both neurons and glial cells in the network, and the combination may contributeto its therapeutic effect.

The leading hypothesis regarding the mechanism of DBS is that excitatory cellsin the STN are inhibited by the electrical stimulation, but an alternative hypothesisis that the astroglia or NG2 cells are activated, resulting in the secretion of agliotransmitter that in turn inhibits excitatory cells in the STN. To attempt to teasethese different possibilities apart, several investigators approached the questionusing different methods including optogenetics in hemiparkinsonian rats:Gradinaru et al. (2009) concluded that direct activation (using light) of STNastrocytes could result in neuronal inhibition of the STN.

19.4 STN DBS, Neurotransmitters, and Glia

Glial cells could be one of the perfect candidates to be involved in DBS. Thesecells are actively involved in neural signaling as discussed previously, and canpropagate calcium waves upon stimulation through an astroglial network (Giaumeet al. 2010). Astrocytes can be directly stimulated by high frequencies, but theyalso likely react to the implantation of the stimulation electrode (Tawfik et al.2010). The presence of these reactive astrocytes near the implanted electrode maylead to an altered modulation of neural signaling. Significantly, a subset of

208 V. Vedam-Mai et al.

astrocytes in specific neurogenic niches are neural stem cells in the adult brain, andit has been shown that injury may induce stem cell properties in cortical astrocytes,outside the neurogenic niches. Taken together, this leads to an appealinghypothesis that the effect of DBS on brain function can be directly induced by theeffect of astrocytes on neuronal networks, or by astrocyte-like neural stem cellsthat are capable of division and genesis of more astrocytes, neurons, and oligo-dendrocytes in response to HFS (for a review, see Vedam-Mai et al. 2011).

Early studies of astrocyte–neuron interactions used electrical stimulation toevoke long-distance Ca2+ signaling (Nedergaard 1994). It is also known that elec-trical stimulation of brain tissue results in an activation of glia, leading to an increasein intracellular cytosolic Ca2+ concentrations (Schipke and Kettenmann 2004).Local glial activation can lead to propagating Ca2+ waves through gap junctionsbetween adjacent astrocytes which can travel through glial networks in the brain (asmuch as a few centimeters) (Schipke and Kettenmann 2004; Zahs and Newman1997; Bowser et al. 2004; Charles et al. 2005; Hamann et al. 2005). The increasingCa2+ concentration can elicit a release of gliotransmitters, such as ATP/adenosine,glutamate, D-serine, and prostaglandin E2. Gliotransmitter release, in turn, canregulate neuronal excitability, and modulate synaptic transmission and plasticity(Nedergaard 1994; Bezzi and Volterra 2001; Newman 2003; Dani et al. 1992).

A neurochemical that may be of significance to the effect of STN DBS isadenosine. Adenosine is a neuromodulator that is found throughout the brain, andis thought to exert its postsynaptic effects through G-protein-coupled receptors(Fukumitsu et al. 2005; Jacobson and Gao 2006). Of particular interest is the classof A1 adenosine receptors, which have the tendency to decrease the activity ofadenylyl cyclase, thereby opening up potassium channels, which leads to thehyperpolarization of neurons, which in turn makes them less active. Bekar et al.(2008) demonstrated that adenosine is a product of thalamic DBS, and thereby iscapable of inhibiting tremor. Also, adenosine has been suggested as a potentialmediator of thalamic DBS for the treatment of essential tremor (Shah et al. 2010).The release of adenosine can be measured in the striatum using chemical micro-sensors during electrical stimulation around the nigrostriatal dopaminergic tract(Shah et al. 2010). Importantly, the reported increases in the levels of extracellularadenosine seem to match increases in cerebral blood flow, which result fromincreased neural activity (Shah et al. 2010). Chang et al. (2009) have shown thatSTN DBS elicits adenosine release in the striatum as measured by chemicalmicrosensors. Furthermore, adenosine is known to play a role in astrocyte sig-naling, and this may become even more pertinent given the recent investigations ofthe local effects of DBS on glial cells (Tawfik et al. 2010; Bekar et al. 2005;Haydon and Carmignoto 2006).

The procedure of inserting a DBS electrode into the brain can itself result in therelease of adenosine and glutamate, owing to the effect of astrocytic activation inresponse to the mechanical stimulation (Newman and Zahs 1997; Kozlov et al.2006). This effect, termed the microthalamotomy effect, caused by the insertionof DBS electrodes in some cases is sufficient to improve the patient’s clinicalsymptoms.

19 The Potential Role of Nonneuronal Cells 209

Steiner et al. (2008) investigated the effects on cellular plasticity and prolif-eration in the substantia nigra (SN) as a result of a unilateral STN lesion, used as amodel of DBS. They demonstrated that the lesion was capable of inducing theproliferation of microglial cells, and also NG2 cells that coexpressed the astrocyticmarker S100ß, showing their glial origin. In other experiments, Steiner et al.(2006) performed 6-hydroxydopamine lesioning in the SN of healthy rats, andsubjected the animals to an enriched environment with physical exercise. Theydemonstrated that the enriched environment in combination with physical activityresults in increased cell proliferation in the rat SN following the lesion (Steineret al. 2006). They also showed that in the lesioned animals, the enriched envi-ronment with physical activity increased the numbers of newborn NG2 cellssignificantly when compared with lesioned animals maintained in standard con-ditions. They did not, however, elucidate the molecular mechanisms of the pro-liferation induction and they did not study the final fate of these newly born NG2cells in the SN. This and other studies provide some evidence for the presence ofthese NG2 cells in the basal ganglia (Halassa and Haydon 2010) in the intact brainas well as upon injury (Buffo et al. 2008), thereby indicating a latent reparativepotential that can potentially be tapped for therapeutic strategies.

Since it has been demonstrated that NG2 cells are able to form synapticjunctions with neurons, there is evidence that through these junctions the NG2cells can have direct communication with neurons through calcium signaling andgliotransmitter release (Bergles et al. 2000; Kulik et al. 1999).

The results from the above-described experiments imply that certain stimuli arecapable of resulting in the release of neurotransmitters and gliotransmitters,thereby modifying the spatiotemporal characteristics of the resulting neuralresponses. Thus, glutamate and adenosine have been suggested as key mediators ofglia-to-neuron signaling. Critical to this is the fact that glia respond to neuronalactivity through Ca2+ wave influxes, which then elicit glutamate release. Tawfiket al. (2010) showed that when HFS is applied to thalamic slices (from ferrets), itresults in an immediate release of both glutamate and adenosine, and they return tothe normal levels as soon as stimulation is stopped. They concluded from theirexperiments that the neurotransmitter/gliotransmitter is at least in part releasedfrom a nonneuronal source. They further suggested that the glutamate released islikely released from astrocytes in a Ca2+-dependent fashion. Hence, HFS-mediatedglutamate release and adenosine release are perhaps significant in abolishingsynchronized neural network oscillations as seen in tremors and seizures.

19.5 Neuropsychiatric Disorders and DBS

DBS is slowly emerging as a therapeutic option for the treatment of medicallyintractable neuropsychiatric diseases. The road to identification of appropriatetargets for neuropsychiatric disorders, however, has been difficult because thebrain networks for explaining these ailments are somewhat complicated.

210 V. Vedam-Mai et al.

Furthermore, the limited availability of animal models has made it difficult totarget the brain for neuropsychiatric DBS. The currently available surgical targetshave been chosen on the basis of the pathophysiology of the disease. The ventralanterior internal capsule and the nucleus accumbens are currently being(Van Laere et al. 2006) considered as targets for the treatment of obsessive–compulsive disorder (OCD) and depression, and the subgenual cingulate cortexwhite matter is being considered as a target for the treatment of depression (Lujanet al. 2008). Mayberg et al. (2005) and Greenberg et al. (2006) have performedclinical trials for the treatment of clinical depression. Mayberg et al. (2005)demonstrated that DBS of the subgenual cingulate (Brodmann area 25) whitematter results in a sustained improvement of patients with treatment-resistantdepression. They performed PET scans, which showed that the cerebral networkswere affected, implying antidepressant benefits. Their findings were consistentwith the resulting suppression of abnormally elevated baseline subgenual cingulateactivity. They propose that the efficacy of stimulation could be a result of theactivation of the inhibitory GABAergic afferents. They further suggest that therecould be long-term changes in the properties of the neural network even after thestimulation has stopped as a result of a prolonged stimulation paradigm. Greenberget al. chose the ventral capsule/ventral striatum (VC/VS) as a target for DBS inOCD, on the basis of preliminary results obtained from lesioning experiments, aswell as observations from neuroimaging research using OCD models (Rauch et al.2006). Rauch et al. (2006) conducted a series of 15O PET imaging experiments inpatients, and observed that acute, high-frequency DBS resulted in an increase inthe perfusion of orbitofrontal cortex, anterior cingulate cortex, striatum, pallidum,and thalamus compared with control conditions. They concluded that acute DBS atthis target (VC/VS) was somehow related to the activation of the neural circuitryresponsible for OCD. Van Laere et al. (2006) proposed that preimplantationmetabolism in the subgenual cingulate cortex in a series of OCD patients asobserved using 18F-fluorodeoxyglucose PET was positive, and correlated with thetherapeutic outcome of VC/VS DBS for the treatment of OCD. A larger samplesize and confirmation of these results would be helpful for patient selection. Theseas well as further investigations are crucial for determining appropriate targetnuclei for other neuropsychiatric disorders, and for the proper comprehension ofthe neural and molecular mechanisms underlying DBS.

19.6 An Emerging General Scheme

It is currently known that the glial cells in the brain are actively involved insynaptic communication (Bezzi et al. 1016). The tripartite synapse hypothesis(which includes presynaptic and postsynaptic neuronal elements, and glia) haschanged our approach to the study of neurotransmitters and their effects on neuralnetworks (Perea et al. 2005). There is now evidence indicating that DBS candirectly activate glial cells to elicit gliotransmitter release. This, in turn, has a

19 The Potential Role of Nonneuronal Cells 211

global effect on the tripartite synapse as well as the glial and neuronal network(Perea et al. 2005). Rather than the previously proposed local inhibition at the siteof stimulation as a result of DBS, the effect seems to be excitatory in nature,including both glial and neuronal elements, and downstream changes in neuralnetwork activity (Shah et al. 2010).

Although there is acceptance of this proposed general scheme, several questionsremain unanswered: What elements are affected by DBS? How are they affected?Which neurotransmitters are responsible for mediating these changes? Thus,characterizing the glial effects of DBS on neurotransmission will provide us with abetter in-depth comprehension of its mechanism of action and effect, therebyproviding us with better tools for patient care. Specifically, we will require aconsolidated approach to tease out and define the intimate relationships ofDBS-mediated glial activation and neural network activity.

References

Allaman I, Belanger M, Magistretti PJ (2011) Astrocyte-neuron metabolic relationships: forbetter and for worse. Trends Neurosci 34:76–87. doi:10.1016/j.tins.2010.12.001

Bekar LK et al (2005) Complex expression and localization of inactivating Kv channels incultured hippocampal astrocytes. J Neurophysiol 93:1699–1709. doi:10.1152/jn.00850.2004

Bekar L et al (2008) Adenosine is crucial for deep brain stimulation-mediated attenuation oftremor. Nat Med 14:75–80. doi:10.1038/nm1693

Bergles DE, RobertsJD, Somogyi P, Jahr CE (2000) Glutamatergic synapses on oligodendrocyteprecursor cells in the hippocampus. Nature 405:187–191. doi:10.1038/35012083

Beurrier C, Bioulac B, Audin J, Hammond C (2001) High-frequency stimulation produces atransient blockade of voltage-gated currents in subthalamic neurons. J Neurophysiol85:1351–1356

Bezzi P, Volterra A (2001) A neuron-glia signalling network in the active brain. Curr OpinNeurobiol 11:387–394

Bezzi P, Domercq M, Vesce S, Volterra A (2001) Neuron-astrocyte cross-talk during synaptictransmission: physiological and neuropathological implications. Prog Brain Res 132:255–265.doi:10.1016/S0079-6123(01)32081-2

Boraud T, Bezard E, Bioulac B, Gross C (1996) High frequency stimulation of the internal globuspallidus (GPi) simultaneously improves parkinsonian symptoms and reduces the firingfrequency of GPi neurons in the MPTP-treated monkey. Neurosci Lett 215:17–20

Bowser DN, Khakh BS (2004) ATP excites interneurons and astrocytes to increase synapticinhibition in neuronal networks. J Neurosci 24:8606–8620. doi:10.1523/JNEUROSCI.2660-04.2004

Buffo A et al (2008) Origin and progeny of reactive gliosis: a source of multipotent cells in theinjured brain. Proc Natl Acad Sci U S A 105:3581–3586. doi:10.1073/pnas.0709002105

Chang SY, Shon YM, Agnesi F, Lee KH (2006) Microthalamotomy effect during deep brainstimulation: potential involvement of adenosine and glutamate efflux. Conf Proc IEEE EngMed Biol Soc 2009:3294–3297. doi:10.1109/IEMBS.2009.5333735

Charles A (2005) Reaching out beyond the synapse: glial intercellular waves coordinatemetabolism. Sci STKE 2005:pe6. doi:10.1126/stke.2702005pe6

Dani JW, Chernjavsky A, Smith SJ (1992) Neuronal activity triggers calcium waves inhippocampal astrocyte networks. Neuron 8:429–440

212 V. Vedam-Mai et al.

Dawson MR, Polito A, Levine JM, Reynolds R (2003) NG2-expressing glial progenitor cells: anabundant and widespread population of cycling cells in the adult rat CNS. Mol Cell Neurosci24:476–488

Dostrovsky JO et al (2000) Microstimulation-induced inhibition of neuronal firing in humanglobus pallidus. J Neurophysiol 84:570–574

Eroglu C, Barres BA (2010) Regulation of synaptic connectivity by glia. Nature 468:223–231.doi:10.1038/nature09612

Fukumitsu N et al (2005) Adenosine A1 receptor mapping of the human brain by PET with 8-dicyclopropylmethyl-1-11C-methyl-3-propylxanthine. J Nucl Med 46:32–37

Gallo V, Patneau DK, Mayer ML, Vaccarino FM (1994) Excitatory amino acid receptors in glialprogenitor cells: molecular and functional properties. Glia 11:94–101. doi:10.1002/glia.440110204

Giaume C, Koulakoff A, Roux L, Holcman D, Rouach N (2010) Astroglial networks: a stepfurther in neuroglial and gliovascular interactions. Nat Rev Neurosci 11:87–99. doi:10.1038/nrn2757

Gradinaru V, Mogri M, Thompson KR, Henderson JM, Deisseroth K (2009) Opticaldeconstruction of parkinsonian neural circuitry. Science 324:354–359. doi:10.1126/science.1167093

Greenberg BD et al (2006) Three-year outcomes in deep brain stimulation for highly resistantobsessive-compulsive disorder. Neuropsychopharmacology 31:2384–2393. doi:10.1038/sj.npp.1301165

Grosche J et al (1999) Microdomains for neuron-glia interaction: parallel fiber signaling toBergmann glial cells. Nat Neurosci 2:139–143. doi:10.1038/5692

Halassa MM, Haydon PG (2010) Integrated brain circuits: astrocytic networks modulate neuronalactivity and behavior. Annu Rev Physiol 72:335–355. doi:10.1146/annurev-physiol-021909-135843

Hamann M, Rossi DJ, Mohr C, Andrade AL, Attwell D (2005) The electrical response ofcerebellar Purkinje neurons to simulated ischaemia. Brain 128:2408–2420. doi:10.1093/brain/awh619

Haydon PG, Carmignoto G (2006) Astrocyte control of synaptic transmission and neurovascularcoupling. Physiol Rev 86:1009–1031. doi:10.1152/physrev.00049.2005

Jacobson KA, Gao ZG (2006) Adenosine receptors as therapeutic targets. Nat Rev Drug Discov5:247–264. doi:10.1038/nrd1983

Kimura K, Yanagida Y, Haruyama T, Kobatake E, Aizawa M (1998) Gene expression in theelectrically stimulated differentiation of PC12 cells. J Biotechnol 63:55–65

Kirchhoff F, Kettenmann H (1992) GABA triggers a [Ca2+]i increase in murine precursor cells ofthe oligodendrocyte lineage. Eur J Neurosci 4:1049–1058

Knutson P, Ghiani CA, Zhou JM, Gallo V, McBain CJ (1997) K+ channel expression and cellproliferation are regulated by intracellular sodium and membrane depolarization inoligodendrocyte progenitor cells. J Neurosci 17:2669–2682

Kojima J et al (1992) Electrically promoted protein production by mammalian cells cultured onthe electrode surface. Biotechnol Bioeng 39:27–32. doi:10.1002/bit.260390106

Koyama S, Haruyama T, Kobatake E, Aizawa M (1997) Electrically induced NGF production byastroglial cells. Nat Biotechnol 15:164–166. doi:10.1038/nbt0297–164

Kozlov AS, AnguloMC, Audinat E, Charpak S (2006) Target cell-specific modulation ofneuronal activity by astrocytes. Proc Natl Acad Sci USA 103:10058–10063. doi:10.1073/pnas.0603741103

Kulik A, Haentzsch A, Luckermann M, Reichelt W, Ballanyi K (1999) Neuron-glia signaling viaalpha(1) adrenoceptor-mediated Ca(2 +) release in Bergmann glial cells in situ. J Neurosci19:8401–8408

Lin SC, Bergles DE (2004) Synaptic signaling between neurons and glia. Glia 47:290–298.doi:10.1002/glia.20060

Lujan JL, Chaturvedi A, McIntyre CC (2008) Tracking the mechanisms of deep brain stimulationfor neuropsychiatric disorders. Front Biosci 13:5892–5904

19 The Potential Role of Nonneuronal Cells 213

Matsunaga K, Uozumi T, Hashimoto T, Tsuji S (2001) Cerebellar stimulation in acute cerebellarataxia. Clin Neurophysiol 112:619–622

Mayberg HS et al (2005) Deep brain stimulation for treatment-resistant depression. Neuron45:651–660. doi:10.1016/j.neuron.2005.02.014

Nedergaard M (1994) Direct signaling from astrocytes to neurons in cultures of mammalian braincells. Science 263:1768–1771

Newman EA (2003) Glial cell inhibition of neurons by release of ATP. J Neurosci 23:1659–1666Newman EA, Zahs KR (1997) Calcium waves in retinal glial cells. Science 275:844–847Nishiyama A, Komitova M, Suzuki R, Zhu X (2009) Polydendrocytes (NG2 cells):

multifunctional cells with lineage plasticity. Nat Rev Neurosci 10:9–22. doi:10.1038/nrn2495

Oberheim NA, Wang X, Goldman S, Nedergaard M (2006) Astrocytic complexity distinguishesthe human brain. Trends Neurosci 29:547–553. doi:10.1016/j.tins.2006.08.004

Panatier A et al (2011) Astrocytes are endogenous regulators of basal transmission at centralsynapses. Cell. doi:10.1016/j.cell.2011.07.022

Papay R et al (2004) Mouse alpha1B-adrenergic receptor is expressed in neurons and NG2oligodendrocytes. J Comp Neurol 478:1–10. doi:10.1002/cne.20215 (2004)

Paukert M, Bergles DE (2006) Synaptic communication between neurons and NG2+ cells. CurrOpin Neurobiol 16:515–521. doi:10.1016/j.conb.2006.08.009

Perea G, Araque A (2005) Properties of synaptically evoked astrocyte calcium signal revealsynaptic information processing by astrocytes. J Neurosci 25:2192–2203. doi:10.1523/JNEUROSCI.3965-04.2005

Perea G, Araque A (2005) Synaptic regulation of the astrocyte calcium signal. Journal of neuraltransmission 112:127–135. doi:10.1007/s00702-004-0170-7

Rauch SL (2003) Neuroimaging and neurocircuitry models pertaining to the neurosurgicaltreatment of psychiatric disorders. Neurosurg Clin N Am 14:213–223, vii–viii

Rauch SL et al (2006) A functional neuroimaging investigation of deep brain stimulation inpatients with obsessive-compulsive disorder. Journal of Neurosurgery 104:558–565.doi:10.3171/jns.2006.104.4.558

Sakry D, Karram K, Trotter J (2011) Synapses between NG2 glia and neurons. J Anat 219:2–7.doi:10.1111/j.1469-7580.2011.01359.x

Schipke CG, Kettenmann H (2004) Astrocyte responses to neuronal activity. Glia 47:226–232.doi:10.1002/glia.20029

Shah RS et al (2010) Deep brain stimulation: technology at the cutting edge. J Clin Neurol6:167–182. doi:10.3988/jcn.2010.6.4.167

Steiner B et al (2006) Enriched environment induces cellular plasticity in the adult substantianigra and improves motor behavior function in the 6-OHDA rat model of Parkinson’s disease.Exp Neurol 199:291–300. doi:10.1016/j.expneurol.2005.11.004

Steiner B et al (2008) Unilateral lesion of the subthalamic nucleus transiently provokes bilateralsubacute glial cell proliferation in the adult rat substantia nigra. Neurosci Lett 430:103–108.doi:10.1016/j.neulet.2007.10.045

Tawfik VL et al (2010) Deep brain stimulation results in local glutamate and adenosine release:investigation into the role of astrocytes. Neurosurgery 67:367–375. doi:10.1227/01.NEU.0000371988.73620.4C

Van Laere K et al (2006) Metabolic imaging of anterior capsular stimulation in refractoryobsessive-compulsive disorder: a key role for the subgenual anterior cingulate and ventralstriatum. J Nucl Med 47:740–747

Vedam-Mai V et al (2011) Deep brain stimulation and the role of astrocytes. Mol Psychiatry.doi:10.1038/mp.2011.61

Velez-Fort M, Maldonado PP, Butt AM, Audinat E, Angulo MC (2010) Postnatal switch fromsynaptic to extrasynaptic transmission between interneurons and NG2 cells. J Neurosci30:6921–6929. doi:10.1523/JNEUROSCI.0238-10.2010

Vitek JL (2002) Mechanisms of deep brain stimulation: excitation or inhibition. Mov Disord17(Suppl 3):S69–S72

214 V. Vedam-Mai et al.

Volterra A, Meldolesi J (2005) Astrocytes, from brain glue to communication elements: therevolution continues. Nat Rev Neurosci 6:626–640. doi:10.1038/nrn1722

Wang DD, Bordey A (2008) The astrocyte odyssey. Progress in neurobiology 86:342–367.doi:10.1016/j.pneurobio.2008.09.015

Windels F et al (2000) Effects of high frequency stimulation of subthalamic nucleus onextracellular glutamate and GABA in substantia nigra and globus pallidus in the normal rat.Eur J Neurosci 12:4141–4146

Wu YR, Levy R, Ashby P, Tasker RR, Dostrovsky JO (2001) Does stimulation of the GPi controldyskinesia by activating inhibitory axons? Mov Disord 16:208–216

Yanagida Y, Mizuno A, Motegi T, Kobatake E, Aizawa M (2000) Electrically stimulatedinduction of hsp70 gene expression in mouse astroglia and fibroblast cells. J Biotechnol79:53–61

Zahs KR, Newman EA (1997) Asymmetric gap junctional coupling between glial cells in the ratretina. Glia 20:10–22

19 The Potential Role of Nonneuronal Cells 215

Chapter 20Animal Studies in Deep BrainStimulation Research

Matthijs G. P. Feenstra and Damiaan Denys

20.1 Goals for Animal Research in Deep Brain Stimulation

‘‘The mechanism of action is not well understood’’—no phrase is repeated moreoften in reports on deep brain stimulation (DBS) in psychiatry and neurology. Allresearchers agree that the mechanism of action of DBS needs to be elucidated inorder to establish its full potential. Although this is one of the primary goals ofDBS research, the chance that it may be reached with clinical studies alone isminimal. In this chapter, we will discuss whether animal research may helpprovide the answer.

Translational research is characterized by a vigorous interaction betweenclinical and preclinical scientists and is aimed at the implementation of noveltreatments or the improvement of existing clinical therapies. In the case of DBS inpsychiatry, experimental clinical trials for obsessive–compulsive disorder (OCD),Tourette syndrome, and major depressive disorder (MDD) were started beforepreclinical studies. The initial targets were chosen on the basis of historicallyestablished neurosurgical interventions. Preclinical research using animal modelsmay provide clinicians with evidence-based feedback validating their brain targetsand stimulation parameters. Although the main goal is to offer better insight intothe mechanism of action of DBS, animal research is also necessary to provide an

M. G. P. Feenstra (&) � D. DenysNetherlands Institute for Neuroscience, Royal Netherlands Academyof Arts and Sciences, Amsterdam, The Netherlandse-mail: m.feenstra@nin.knaw.nl

M. G. P. Feenstra � D. DenysDepartment of Psychiatry, Academic Medical Center,University of Amsterdam, Amsterdam, The Netherlands

D. Denyse-mail: ddenys@gmail.com

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_20, � Springer-Verlag Berlin Heidelberg 2012

217

experimental basis for new therapeutic applications, for novel targets, and foroptimizing stimulation parameters in psychiatric disorders.

20.2 Strengths and Weaknesses of Animal Research in DBS

Why would DBS research in animals be in a better position than clinical research tofulfill these aims? There are three essential advantages. Firstly, in animals, com-parisons can be made between stimulation of normal and pathological brains. Sinceobviously, because of ethical reasons, there are no clinical data available on theeffects of DBS on the normal brain, the exploration of the impact of DBS on normalphysiological function and behavior depends entirely on animal studies. As appliesto any novel treatment, testing in normal, healthy controls is indispensable for theevaluation of targeted, invasive brain stimulation. Knowledge of the effects of DBSin healthy animals is needed not only to elucidate the mechanisms of action of DBS,but also to obtain an index of side effects and safety measures. Another mainadvantage is the possibility to perform invasive measurements to determine theimpact of DBS on cellular and molecular processes. Despite the enormous progressin noninvasive imaging and electrophysiological recordings of human brain activ-ity, invasive measurements of cellular activity and plasticity, such as geneexpression, neurotransmitter release, and cellular proliferation, are mandatory toprovide deeper insight into the mechanism of action of DBS. These techniques arerestricted to experiments in living animals or the postmortem evaluation of theirbrains. Finally, animal research allows the time-consuming and meticulous testingof a wide range of stimulation electrodes and parameters, including novel approa-ches. Clinical studies in this direction are limited by the availability of patients,safety aspects, and high costs. Clinical research will, therefore, by its nature berather conservative, whereas animal research will be more innovative by testingnovel targets, electrodes, or parameters.

On the other hand, there are intrinsic limitations to DBS research in animals. Toachieve the above-described goals, studies in well-established animal models ofspecific psychiatric disorders are needed in addition to tests in control animals.Although numerous animal models of psychiatric disorders are currently availablein rodents, they all have ‘‘substantial limitations,’’ paradoxically reflecting the lackof knowledge of the neurobiological mechanisms underlying the clinical states(Nestler and Hyman 2010). Animal models based on causes similar to those inhumans, i.e., having construct validity, are therefore scarce in rodents and innonhuman primates. Apart from this, animal models rarely emulate all clinicalsymptoms of a particular psychiatric condition, and some symptoms, because oftheir subjective, typically human nature, are impossible to mimic or study, even innonhuman primates.

A second problem is that even though the organization of the rodent brainresembles that of the human brain in many aspects, anatomical differences hamperthe translation of some of the clinically used stimulation targets to the rodent brain.

218 M. G. P. Feenstra and D. Denys

A case in point is the different anatomy in rodents of the anterior limb of the internalcapsule, one of the most important targets for DBS in psychiatric patients. In thehuman and nonhuman primate brain, a broad band of white matter separates thecaudate nucleus from the putamen. In rodents, this band is absent, althoughnumerous isolated white matter fascicules pass through all striatal structures.Therefore, either much larger volumes of rodent striatal tissue would need to bestimulated to mimic the effects of stimulation of the anterior limb of the internalcapsule of primates or the internal capsule would have to be stimulated at a different,more caudal position. Studies in nonhuman primates certainly may provide a betteranswer to specific anatomical or functional questions, but ethical, practical, andfinancial limitations preclude extensive use of primate species.

A third major problem in relating animal findings to clinical research is thestimulation procedure itself. It is often observed that DBS electrodes used foranimal research show large variations and are generally different in constructionand application from their human counterparts (see also Gubellini et al. 2009). Allclinical studies in psychiatry have been done with three to five electrode designs,differing only in their dimensions. In sharp contrast, for rodent studies, 10–20designs have been used, differing in all possible aspects, such as material, thick-ness, length, and number of contact points. A major lack of current animal researchis that no studies are available reporting a direct comparison of different electrodedesigns used in behavioral or physiological experiments. In addition, stimulationin animals is almost never applied continuously as in the clinic—rodents are oftenstimulated only during the experiment or for a few hours per day for 1–2 weeksbefore the actual experiment.

20.3 Prospects for Animal Research in DBS

Considering the previous points, we feel that the advantages of animal researchhave not yet been exploited to their full extent. Remarkably few studies haveexamined basic aspects of neuronal function or behavior after DBS in targets usedin psychiatric disorders. However, one is also bound to conclude that because ofthe limitations, DBS research in rodents cannot be expected to answer completelyall questions raised by clinicians. Although the problem of different electrodescould be solved readily, the anatomical differences are inherent to the use ofrodents, and the development of improved animal models of psychiatric disease isa key problem inherent to translational research, not just for DBS research.

We propose that a careful selection of stimulation targets and specific brainfunctions for examination in healthy controls and in models of disease may help toovercome these limitations—at least partially. Psychiatric disorders are catego-rized on the basis of phenomenological and clinical symptoms with observationalrating scales to measure severity of symptoms such as social phobia and obses-sions. These symptoms do not lend themselves to studies in experimental animalsbecause of their subjective, typically human nature. This hampers the translation

20 Animal Studies in Deep Brain Stimulation Research 219

of human behavioral changes to animal research and the elucidation of themechanisms of action. A different approach in which the clinical condition isdeconstructed into more basic, broad-dimensional or behavioral manifestationssuch as anxiety or cognitive inflexibility is better suited for translation to animals.Anxiety is an example of a disease-independent neurobehavioral domain for whichtranslational methods are available (Davis et al. 2010; Aupperle and Paulus 2010).Another example is attention, which was successfully explored in controls and inparkinsonian animals, stimulated in the subthalamic nucleus (Baunez 2011; Temelet al. 2009). Only a domain that can be translated from humans to animals and viceversa offers a basis for successful studies. In recent years, similar approaches havebeen advocated both in the study of endophenotypes in psychiatry (Gottesman andGould 2003; Chamberlain and Menzies 2009) and as a basis to devise animalmodels of psychiatric disease (Gould and Gottesman 2006; Kellendonk et al. 2009;Fernando and Robbins 2011).

In the following, we will discuss two domains that are highly relevant for thepsychiatric diseases that are successfully treated with DBS, i.e., anxiety andreward processing.

20.4 Anxiety

Anxiety is a core symptom of a number of psychiatric diseases. Recent studiesshow that the anxiolytic effect of DBS in treatment-resistant OCD patients isimpressive. After turning on stimulation, anxiety ratings decrease rapidly withinminutes and massively (Denys et al. 2010). It is puzzling, though, how stimulationof the nucleus accumbens and ventral anterior internal capsule results in such arapid reduction of anxiety symptoms, and whether DBS affects a specific type ofanxiety. Recent animal research has allowed the classification of well-definedsubtypes of anxiety that can be differentiated on the basis of specific neurobio-logical substrates in brain areas and neuronal circuits.

An important classification of anxiety is based on the distinction betweenconditioned and unconditioned or innate anxiety (see Millan 2003). In conditionedanxiety, an originally neutral cue predicts a fear-inducing situation through con-tingent association with that situation (in rodents this generally refers to paincaused by electric shocks). Unconditioned anxiety relates to situations which bythemselves induce fear or aversion. Conditioned fear is strongly based on circuitsinvolving the amygdala, with additional involvement of the hippocampus when acontext (an environment such as a box or a room) is the predicting cue. Thesecircuits were originally defined in rodent research, but are preserved in humans(Alvarez et al. 2008), showing the translational potential of this approach. Whereascue conditioning may be a model for phobias and posttraumatic stress disorder,context conditioning has been proposed as a model for general anxiety disorder(Luyten et al. 2011). A somewhat different classification was offered by Daviset al. (2010), who described phasic fear as a response to a short-term threat and

220 M. G. P. Feenstra and D. Denys

sustained fear or ‘‘anxiety’’ as a slow-onset, long-lasting response to a sustainedthreat. In their view, sustained fear is mediated by the bed nucleus of the striaterminalis (BNST), whereas phasic fear depends on the central nucleus of theamygdala. Interestingly, of these brain areas, only the BNST may be implicated inthe DBS responses as the nucleus accumbens/ventral anterior internal capsuleborders the BNST in the human brain (see also Chap. 4). Recent animal research,however, suggests that stimulation of other nearby targets may have selectiveanxiolytic effects as well. An improved extinction of fear conditioning wasobserved after stimulation of an area on the border of the nucleus accumbens andthe caudate nucleus (Rodriguez-Romaguera et al. 2012), whereas stimulation ofthe anterior internal capsule produced a differential effect on conditioned andunconditioned anxiety (van Dijk et al. 2012).

These examples show that measurements based on specific neurobehavioralconstructs may offer new directions for clinical studies. It now seems important toknow whether the anxiolytic effects induced by DBS in psychiatric patients may becategorized as enhanced extinction of conditioned fear or as reduction of primarymeasures of unconditioned or conditioned fear. Once similar effects are detected inanimals and humans, the real search for the mechanism of action can begin—how doelectrical pulses in these targets lead to a decrease in a specific subtype of anxiety?

20.5 Reward

Reward processing is impaired in OCD, MDD, and addiction. Altered brainactivation during reward anticipation was recently reported in OCD patients,similar to previous findings in MDD and substance addiction (Figee et al. 2011).No effects of DBS on reward processing have been reported in clinical or animalstudies related to OCD. Yet, various aspects of reward processing, e.g., anticipa-tion, can be translated between rodents and humans and may therefore be of use asa dimensional approach to examine the effects of DBS in animal models andhumans. In drug addiction, DBS has been tested on reward-seeking paradigms inrodents and has provided examples of altered motivation for food (see Chap. 14).

A particularly interesting paradigm to study the reward system is intracranialself-stimulation or brain stimulation reward (Olds 1958; Milner 1991; Carlezonand Chartoff 2007). Operant responding is reinforced by contingent short (e.g.,0.5 s) trains of stimulation of the medial forebrain bundle (MFB) in the lateralhypothalamus. A wide range of different animal species are motivated to press alever or perform another action to receive these stimulations. This paradigm maybe used to test the rewarding effects of drugs. In the presence of a rewarding drug,stimulation becomes reinforcing at lower current intensities or lower frequencies(Kornetsky and Bain 1992). Self-stimulation is effective in a wide variety of brainareas centered around or connected to the MFB. The prevailing hypothesis is thatstimulation of myelinated fibers of the MFB is responsible for the reinforcingeffect, and that this may involve a direct or indirect synaptic path impinging on the

20 Animal Studies in Deep Brain Stimulation Research 221

dopamine neurons of the ventral tegmental area (Milner 1991). Interestingly,the targets that are now in use for DBS in psychiatry are close to these pathways.So, the question arises whether the beneficial effects of DBS are due to activationof the MFB providing an immediate rewarding effect (Oshima and Katayama2010). It is remarkable that early studies of human brain stimulation in the 1950sexplicitly used self-stimulation paradigms as therapeutic treatments for psychiatricconditions (Bishop et al. 1963; Heath 1963). These studies, which fell short ofexisting ethical standards (Baumeister 2000; Hariz et al. 2010), suggested thatbrain stimulation might be rewarding in humans as it is in rodents. A recent reportshows that nonhuman primates will work for several hours to receive brainstimulation in the nucleus accumbens/ventral anterior internal capsule area (Bichotet al. 2011). Similar procedures have been developed in rats (Rokosik and Napier2011). The stimulation parameters in all these studies fall in the range of theparameters used for therapeutic DBS in psychiatry. Yet, there are two importantdifferences between self-stimulation and DBS. Firstly, the active versus passiverole of the subject in the stimulation procedure, and secondly, the use of shortperiods versus the continuous presentation of electrical pulses. Therefore, thecrucial question is whether passively receiving continuous stimulation hasrewarding effects similar to those of actively initiating short-term stimulation.Given the clear translatable paradigms involved, this would be an ideal question toanswer in rodent studies. Previous rodent results point to some possible differencesbetween short-term and continuous stimulation. When stimulation is active as longas a rat holds the lever, most rats keep the stimulation to less than 0.5 s (Milner1991). Other experimenters observed that rats learn to escape from longer-lastingstimulation, and it was suggested that this may indicate the presence of a slowlyincreasing aversive, nociceptive effect (Pollock and Kornetsky 1990). Another lineof evidence suggests that the effects of (experimenter-induced) brain stimulationreward on dopaminergic activity strongly depend on the temporal density of thestimulation. Effects on dopamine release are important as recent studies suggestthat activation of dopaminergic neuronal activity is sufficient to support self-stimulation (Witten et al. 2011). Stimulation trains of 0.5 s once every 12 s led tocontinuously increased dopamine efflux, whereas similar trains once every 1.5 sproduced a higher initial peak, but a subsequent decrease to baseline levels(Hernandez et al. 2006). These data suggest a fundamental difference betweenshort-lasting and (almost) continuous stimulation, as the latter may initially haveeffects that resemble brain stimulation reward, but cannot endure. These possibleaversive and nonenduring effects are difficult to reconcile with the successfulclinical use of DBS in psychiatry. It is therefore still an open question if themechanism of action of clinical DBS involves reinforcing effects such as brainstimulation reward. Translational research could clarify this issue.

222 M. G. P. Feenstra and D. Denys

20.6 Conclusion

What has animal research taught us about DBS so far? Several authors provided aproof of principle that DBS in rodents is capable of changing pathological conditionsthat selectively model symptoms of psychiatric disorders (see Chaps. 7, 11, 14).Others identified possible mechanisms of action of both the therapeutic effects andthe side effects of clinical DBS (see Chaps. 11, 16). Still others suggested novelanatomical targets or stimulation parameters to treat compulsivity (see Chap. 7) oraddiction (see Chap. 14). Although it is too early to conclude whether these willresult in successful translational research, i.e., if they will lead to improvement ofclinical therapies, they do show that translational studies of DBS in psychiatry holdgreat promise to offer better explanations for the mechanism of action of DBS andnovel targets for its application. To fulfill this promise, it is essential to concentrate onthe use of neurobehavioral domains such as anxiety and reward that have proventranslational value instead of disorders. Only the use of equivalent experimentalparadigms in both clinical and animal studies will make it possible to translate humanbehavior to animal behavior and vice versa. Unfortunately, results from differentlaboratories, using different electrodes and stimulation parameters, hamper theinterpretation and comparison of results. The use of more standardized methodsshould contribute to the value of these studies and accelerate their progress.

References

Alvarez RP, Biggs A, Chen G, Pine DS, Grillon C (2008) Contextual fear conditioning inhumans: cortical-hippocampal and amygdala contributions. J Neurosci 28:6211–6219

Aupperle RL, Paulus MP (2010) Neural systems underlying approach and avoidance in anxietydisorders. Dialogues Clin Neurosci 12:517–531

Baumeister AA (2000) The Tulane electrical brain stimulation program a historical case study inmedical ethics. J Hist Neurosci 9:262–278

Baunez C (2011) A few examples of the contribution of animal research in rodents for clinicalapplication of deep brain stimulation. Prog Brain Res 194:105–116

Bichot NP, Heard MT, Desimone R (2011) Stimulation of the nucleus accumbens as behavioralreward in awake behaving monkeys. J Neurosci Methods 199:265–272

Bishop MP, Elder ST, Heath RG (1963) Intracranial self-stimulation in man. Science 140:394–396Carlezon WA Jr, Chartoff EH (2007) Intracranial self-stimulation (ICSS) in rodents to study the

neurobiology of motivation. Nat Protoc 2:2987–2995Chamberlain SR, Menzies L (2009) Endophenotypes of obsessive–compulsive disorder:

rationale, evidence and future potential. Expert Rev Neurother 9:1133–1146Davis M, Walker DL, Miles L, Grillon C (2010) Phasic vs sustained fear in rats and humans: role

of the extended amygdala in fear vs anxiety. Neuropsychopharmacology 35:105–135Denys D, Mantione M, Figee M, van den Munckhof P, Koerselman F, Westenberg H, Bosch A,

Schuurman R (2010) Deep brain stimulation of the nucleus accumbens for treatment-refractory obsessive-compulsive disorder. Arch Gen Psychiatry 67:1061–1068

Fernando AB, Robbins TW (2011) Animal models of neuropsychiatric disorders. Annu Rev ClinPsychol 7:39–61

20 Animal Studies in Deep Brain Stimulation Research 223

Figee M, Vink M, de Geus F, Vulink N, Veltman DJ, Westenberg H, Denys D (2011)Dysfunctional reward circuitry in obsessive-compulsive disorder. Biol Psychiatry 69:867–874

Gould TD, Gottesman II (2006) Psychiatric endophenotypes and the development of valid animalmodels. Genes Brain Behav 5:113–119

Gottesman II, Gould TD (2003) The endophenotype concept in psychiatry: etymology andstrategic intentions. Am J Psychiatry 160:636–645

Gubellini P, Salin P, Kerkerian-Le Goff L, Baunez C (2009) Deep brain stimulation inneurological diseases and experimental models: from molecule to complex behavior. ProgNeurobiol 89:79–123

Hariz MI, Blomstedt P, Zrinzo L (2010) Deep brain stimulation between 1947 and 1987: theuntold story. Neurosurg Focus 29:E1

Heath RG (1963) Electrical self-stimulation of the brain in man. Am J Psychiatry 120:571–577Hernandez G, Hamdani S, Rajabi H, Conover K, Stewart J, Arvanitogiannis A, Shizgal P (2006)

Prolonged rewarding stimulation of the rat medial forebrain bundle: neurochemical andbehavioral consequences. Behav Neurosci 120:888–904

Kellendonk C, Simpson EH, Kandel ER (2009) Modeling cognitive endophenotypes ofschizophrenia in mice. Trends Neurosci 32:347–358

Kornetsky C, Bain G (1992) Brain-stimulation reward: a model for the study of the rewardingeffects of abused drugs. NIDA Res Monogr 124:73–93

Luyten L, Vansteenwegen D, van Kuyck K, Gabriëls L, Nuttin B (2011) Contextual conditioningin rats as an animal model for generalized anxiety disorder. Cogn Affect Behav Neurosci11:228–244

Millan MJ (2003) The neurobiology and control of anxious states. Prog Neurobiol 70:83–244Milner PM (1991) Brain-stimulation reward: a review. Can J Psychol 45:1–36Nestler EJ, Hyman SE (2010) Animal models of neuropsychiatric disorders. Nat Neurosci

13:1161–1169Olds J (1958) Self-stimulation of the brain; its use to study local effects of hunger, sex, and drugs.

Science 127:315–324Oshima H, Katayama Y (2010) Neuroethics of deep brain stimulation for mental disorders: brain

stimulation reward in humans. Neurol Med Chir (Tokyo) 50:845–852Pollock J, Kornetsky C (1990) Pharmacologic evidence for nociception resulting from

noncontingent ‘‘rewarding’’ brain stimulation. Physiol Behav 47:761–765Rodriguez-Romaguera J, Do Monte FH, Quirk GJ (2012) Deep brain stimulation of the ventral

striatum enhances extinction of conditioned fear. Proc Natl Acad Sci USA 109:8764–8769Rokosik SL, Napier TC (2011) Intracranial self-stimulation as a positive reinforcer to study

impulsivity in a probability discounting paradigm. J Neurosci Methods 198:260–269Temel Y, Tan S, Vlamings R, Sesia T, Lim LW, Lardeux S, Visser-Vandewalle V, Baunez C

(2009) Cognitive and limbic effects of deep brain stimulation in preclinical studies. FrontBiosci 14:1891–1901

van Dijk A, Klanker M, Hamelink R, Feenstra M, Denys D (2012) Differential anxiolytic effectsduring deep brain stimulation in striatal areas and the internal capsule. FENS abstr 2630

Witten IB, Steinberg EE, Lee SY, Davidson TJ, Zalocusky KA, Brodsky M, Yizhar O, Cho SL,Gong S, Ramakrishnan C, Stuber GD, Tye KM, Janak PH, Deisseroth K (2011) Recombinase-driver rat lines: tools, techniques, and optogenetic application to dopamine-mediatedreinforcement. Neuron 72:721–733

224 M. G. P. Feenstra and D. Denys

Chapter 21Neuroimaging Deep Brain Stimulationin Psychiatric Disorders

Martijn Figee, Pepijn van den Munckhof, Rick Schuurmanand Damiaan Denys

21.1 Introduction

Until the 1990s, psychosurgery almost exclusively employed ablative lesions. Tar-geting was based on anatomic studies and animal experiments, and was furtherdeveloped by correlating clinical effects to autopsy findings (Moniz 1936; Talairachet al. 1949). In 1999, Vandewalle et al. (1999) and Nuttin et al. (1999) introduced deepbrain stimulation (DBS) as an experimental treatment for Tourette syndrome (TS) andobsessive–compulsive disorder (OCD), respectively. The TS target in the thalamus wasbased on the thalamotomy target from Hassler and Dieckman (1970), whereas the OCDtarget in the anterior limb of the internal capsule (ALIC) was based on the capsulotomytarget for treatment-resistant OCD (Bingley et al. 1977). Meanwhile, neuroimaging ofpsychiatric disorders evolved from basic structural computer tomography and magneticresonance imaging (MRI) techniques to more sophisticated functional modalities,including positron emission tomography (PET), functional MRI (fMRI), and diffusiontensor imaging (DTI). These techniques have greatly expanded our knowledge of thepathogenesis of psychiatric disorders, and have helped us understand the therapeutics ofDBS. Neuroimaging may also serve to identify potential new DBS targets for psychi-atric disorders. This chapter discusses neuroimaging studies of OCD, major depressivedisorder (MDD), TS, and addiction, and DBS-related brain changes in these disorders.

M. Figee (&) � D. DenysDepartment of Psychiatry and Neurosurgery, Academic Medical Center,Amsterdam, The Netherlandse-mail: m.figee@amc.uva.nl

P. van den Munckhof � R. SchuurmanDepartment of Neurosurgery, Academic Medical Center, Amsterdam, The Netherlands

D. DenysNetherlands Institute for Neuroscience, Royal Netherlands Academy of Arts and Sciences,Amsterdam, The Netherlands

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_21, � Springer-Verlag Berlin Heidelberg 2012

225

21.2 Obsessive–Compulsive Disorder

21.2.1 Neuroimaging of OCD Disease

Numerous structural and functional imaging studies have related OCD to diseaseof the cortical–striatal–thalamic–cortical network (CSTC) (Whiteside et al. 2004;Menzies et al. 2008; Radua et al. 2010). The most consistent structural imagingfindings are increased gray matter volume of the basal ganglia, particularly thecaudate nucleus and putamen, in association with decreased gray matter volume ofthe anterior cingulate cortex (ACC) and orbitofrontal cortex (OFC) (Radua et al.2010; Menzies et al. 2008). DTI revealed white matter tract abnormalities in themedial frontal cortex and in the corpus callosum (Bora et al. 2011), and in theACC and ALIC, suggesting disrupted cortical–cortical connections (Lehman et al.2011), and frontal cortical–ventral striatal connections, including the nucleus ac-cumbens (NAc), thalamus, and brainstem (Lehman et al. 2011). Functionalimaging studies identified hyperactivity in the head of the caudate nucleus andOFC in the resting state (Whiteside et al. 2004) and during OCD symptomprovocation, along with hyperactivity of the thalamus, dorsolateral prefrontalcortex, parietal cortex, ACC, and limbic areas (Rotge et al. 2008). Improvement ofOCD symptoms following treatment with selective serotonin reuptake inhibitors orcognitive behavioral therapy is related to a decrease of hyperactivity of the OFCand caudate nucleus, and the functional correlation between these structures(Saxena and Rauch 2000). This latter finding suggests that OCD is related not onlyto hyperactivity of CSTC nodes, but also to increased functional coupling betweenthese nodes. This view is supported by more recent evidence from resting-statefMRI studies that revealed excessive coupling between CSTC nodes, especiallybetween the ventral striatum and the OFC (Harrison et al. 2009; Sakai et al. 2010),of which the latter is correlated with symptom severity (Harrison et al. 2009).Increased coupling was found also between the dorsal striatum and the ventralstriatum, which is thought to underlie compulsive drug seeking as well (Belin andEveritt 2008). Both OCD and addiction have been related to NAc dysfunctionduring reward processing (Figee et al. 2011; Hommer et al. 2011). Activation ofthe prefrontal cortex and ventral striatum is related to healthy reward processing(Knutson et al. 2001), whereas the dorsal striatum contributes to habitual control ofbehavior (Tricomi et al. 2009). Increased coupling between frontal and striatalregions and between the ventral striatum and the dorsal striatum may thus reflect ashift from healthy goal-directed behavior towards compulsive habits.

In summary, neuroimaging studies in OCD confirm dysfunction in all CSTCnodes: structural abnormalities and hyperactivity in the OFC, ACC, basal ganglia,and thalamus, in association with dysfunctional white matter connections andfunctional connectivity between these nodes.

226 M. Figee et al.

21.2.2 Neuroimaging and DBS for the Treatment of OCD

Current DBS targets for treatment-resistant OCD are all located within the CSTCnetwork (de Koning et al. 2011): the ALIC, the ventral striatum/ventral internalcapsule (VS/VC), the NAc, the subthalamic nucleus (STN), and the inferior tha-lamic peduncle (ITP). How effective are these targets, and how does DBS mod-ulate CSTC brain dysfunction in OCD?

21.2.2.1 ALIC and VC/VS

Bilateral DBS targeted at the ALIC and VC/VS results in 45 % symptomimprovement, with a responder rate (defined as more than 35 % symptomimprovement) of 19 of 31 patients (de Koning et al. 2011). Two studies mappedfunctional brain changes related to acute ALIC stimulation when no clinical effectshad occurred yet. High-frequency ALIC stimulation 10 days after electrodeimplantation in one OCD patient induced blood-oxygen-dependent activation of thebilateral striatum, pons, and frontal, temporal, and occipital cortex (Nuttin et al.2003). High-frequency VC/VS stimulation 2 weeks after electrode implantation insix OCD patients induced activation of the dorsal striatum (putamen), ventral globuspallidus, thalamus, subgenual ACC (sgACC), and medial OFC as evidenced by 15O-CO2 PET (Rauch et al. 2006). Low-frequency stimulation elicited no activationpatterns that differed from nonstimulation, supporting the widely held hypothesisthat only high-frequency DBS is effective for the treatment of psychiatric disorders.Of note, the immediate changes in putamen and OFC following bilateral VC/VSstimulation were right-sided, which is puzzling because the right side of the ALICcontains fewer and wider bundles than the left side (Axer et al. 1999). The exactstimulation locations were not mentioned in these two acute DBS imaging studies,but it could be inferred from related clinical data (Greenberg et al. 2006) that ventralstriatal/NAc as well as more dorsal internal capsule electrode contacts were stimu-lated, in both unipolar and bipolar modes. The clinical response to chronic ALICDBS was related to decreased PET activity in the OFC in two OCD patients after3–6 weeks of stimulation (Abelson et al. 2005) and in two OCD patients after3 months of continuous stimulation (Nuttin et al. 2003). In the latter study, OFCdeactivation was also noted in a nonresponder and seemed to occur irrespective ofmonopolar or bipolar stimulation, high voltage (9 V) or low voltage (4 and 5.5 V),and white matter stimulation only (two patients) or stimulation of NAc gray matter aswell (one patient). Six OCD patients were scanned with glucose PET preoperativelyand after 3–26 months of continuous ALIC stimulation (van Laere et al. 2006).Chronic ALIC DBS decreased activity in the sgACC (Brodmann area 32), rightdorsolateral prefrontal cortex, and right anterior insula. The ALIC connects frontalcortical areas with basal ganglia, thalamus, and brainstem, which are all activated byALIC stimulation initially. However, chronic and therapeutic ALIC DBS seems tospecifically inhibit frontal cortical areas and normalize OFC hyperactivity.

21 Neuroimaging Deep Brain Stimulation 227

21.2.2.2 Nucleus Accumbens

Bilateral DBS targeted at the NAc resulted in 51 % symptom improvement with aresponder rate of 11 of 19 patients, whereas unilateral DBS targeted at the rightNAc resulted in only 21 % symptom improvement, with a responder rate of onlyone of ten patients (de Koning et al. 2011). Functional brain changes related toDBS directly targeted at the NAc have not been reported yet. However, ALICstimulation in Belgian studies (Nuttin et al. 2003; van Laere et al. 2006) likelyinvolved NAc stimulation as well, since they used large quadripolar electrodes(Pisces Quad 388 from Medtronic with contact points 3 mm long and separatedfrom adjacent contacts by 6 mm) and high voltages (up to 10.5 V) at severalelectrode contact points, including the most ventral contact located in the NAc.These patients displayed bilateral NAc hyperactivity before implantation of ALICelectrodes, which normalized after chronic stimulation (van Laere et al. 2006).

21.2.2.3 Subthalamic Nucleus

Bilateral DBS targeted at the STN resulted in 31 % symptom improvement with aresponder rate of 12 of 16 patients (Mallet et al. 2008), but the responder criterionwas less stringent than in the ALIC, VC/VS, and NAc studies (25 vs. 35 %symptom improvement, respectively). Although there is no neuroimagingevidence for direct STN involvement in OCD, STN DBS may be effective for thetreatment of OCD by normalizing frontal hyperactivity through the indirectinhibitory CSTC pathway. Indeed, Le Jeune et al. (2010) found decreased ACCactivity during STN DBS in ten OCD patients, as measured with glucose PET and,similar to the ALIC findings, therapeutic effects correlated with a decrease of OFChyperactivity.

21.2.2.4 Inferior Thalamic Peduncle

Bilateral ITP DBS resulted in 49 % symptom improvement, with a responder rateof five of five patients (Jiménez-Ponce et al. 2008). ITP DBS is likely to alter OFCactivity because the ITP is a major connecting point between the thalamus and theOFC (Axer et al. 1999), although there are no functional imaging data of ITP DBSavailable in OCD.

21.2.2.5 Summary

In summary, the current DBS targets for treatment-resistant OCD, which are alllocated within the CSTC network, result in 44 % symptom improvement, with aresponder rate of 47 of 71 patients. DBS targeted at the ALIC, VC/VS, and STNinduces local and global functional changes within the connected CSTC network,

228 M. Figee et al.

and the clinical responses are related to normalization of OFC hyperactivity. DBStargeted at the NAc seems to normalize disease-related NAc hyperactivity, whichrestores its local reward function and decreases excessive frontostriatal coupling.

21.3 Major Depressive Disorder

21.3.1 Neuroimaging of MDD Disease

The symptoms in MDD and the underlying brain circuitry are less consistent andmore heterogeneous compared with OCD, involving brain systems that regulatemood and emotions, reward processing and motivation, attention and memory, stressresponses, energy, sleep, appetite, and libido (Drevets et al. 2008). Corticolimbicsystems subserving these various processes in MDD have been characterized by adorsal motor and sensory circuit that involves premotor, temporal, and sensorycortices, and a ventral reward and emotion regulating circuit that includes the ventralACC, the OFC, and the ventral striatum, hippocampus, and amygdala, extending intoa visceral network with the hypothalamus and brainstem (Drevets et al. 2008).

Structural MRI studies have consistently related MDD to volumetric deficitswithin the corticolimbic circuit, e.g., reduced volume of the OFC, sgACC, superiortemporal gyrus, and basal ganglia, and volumetric reductions of the amygdala andhippocampus that may have developed secondary to the illness (Lorenzetti et al.2009; Drevets et al. 2008). Resting-state fMRI and PET findings reveal hypoac-tivity of dorsomedial and dorsolateral prefrontal cortices, along with hyperactivityof the ventral corticolimbic circuit. Hyperactivity is particularly found in thesgACC, which is related to depression severity and can be reversed by pharma-cotherapy (Sacher et al. 2011). Both increases and decreases of functionalconnectivity have been found between limbic and cortical structures (Hasler andNorthoff 2011). DTI studies report dysfunctional white matter tracts betweenfrontal and temporal cortices, as well as in ACC fibers (Maller et al. 2010).

The anhedonia of depression can be viewed as a state of reduced motivation toseek rewards and to engage with all aspects of the world (Alcaro and Panksepp2011). In accordance, attenuated ventral striatal responses to reward anddysfunctional mesolimbic dopaminergic neurotransmission have often beenobserved in MDD (Pizzagalli et al. 2009; Robinson et al. 2011; Nestler and Carlezon2006). Dysfunction of the brainstem, ventral striatum, and limbic system may lead tothe characteristic hedonic-emotional disturbances of MDD, with hyperactivesgACC reflecting impaired modulatory control over pathological limbic responses.

In summary, neuroimaging studies of MDD reveal disease in the corticolimbicnetwork: reduced volumes and hyperactivity in the sgACC, OFC, temporal cortex,amygdala, and hippocampus, and hypoactivity in dorsal cortical regions.Dysfunctional white matter connections and functional connectivity are foundbetween cortical and limbic structures and between frontal and temporal cortices.

21 Neuroimaging Deep Brain Stimulation 229

21.3.2 Neuroimaging and DBS for the Treatment of MDD

Two of the three currently used DBS targets for MDD are located in the ventralcorticolimbic network; the NAc and the VC/VS. The subcallosal cingulate gyrus(SCG) connects with both dorsal and ventral networks.

21.3.2.1 Nucleus Accumbens

Bilateral DBS targeted at the NAc resulted in a 36 % MDD symptom improvementafter 1 year, with a responder rate (50 % reduction) of five of ten patients (Bewernicket al. 2010). Similar to OCD patients, MDD patients have blunted NAc responsesduring reward processing (Pizzagalli et al. 2009), which may reflect anhedonia. Oneweek of acute stimulation at the lowest two contacts in the NAc core and shell did notproduce subjective antidepressive effects; however, in a descriptive case study, onepatient reported a sudden urge to visit Cologne Cathedral, and another patient wishedto take up her old bowling hobby again (Schlaepfer et al. 2008). These immediatehedonic improvements were accompanied by increased metabolism in the NAccompared with the level before surgery as evidenced by PET. Acute NAc DBS alsoincreased metabolism in the connected amygdala and decreased activity in medial anddorsal cortical areas. Following 6 months of chronic NAc stimulation in seven MDDpatients, local changes of NAc metabolism were no longer observed. However,metabolism across its connected ventral network seemed to have normalized, withdecreased metabolism in the OFC, sgACC, thalamus, and amygdala in responderscompared with nonresponders (Bewernick et al. 2010). Metabolic decreases werefound in the posterior ACC and caudate nucleus, and increases were found in theprecentral gyrus. In parallel with the neuroimaging findings of DBS for the treatment ofOCD, acute stimulation of the NAc seems to restore its local function, seeminglyreducing excessive activity in the connected frontostriatal and limbic network.

21.3.2.2 Ventral Striatum/Ventral Internal Capsule

In an open study, bilateral VC/VS DBS resulted in 47 % symptom improvement after1 year, with a responder rate of seven of 15 MDD patients (Malone et al. 2009). Sincetherapeutic VC/VS stimulation in OCD modulates frontal cortical areas, basal ganglia,thalamus, and brainstem and also improves mood, it may affect similar brain regions inMDD. However, there are no functional imaging data for VC/VS DBS in MDD.

21.3.2.3 Subcallosal Cingulate Gyrus

Bilateral DBS targeted at the SCG (which includes the sgACC) resulted in a 49 %symptom improvement after 1 year, with a responder rate of 22 of 49 MDD patients

230 M. Figee et al.

(Lozano et al. 2008, 2011; Puigdemont et al. 2011). The SCG connects with allcorticolimbic network nodes that are involved in MDD disease, including thesgACC, ACC, OFC, NAc, hypothalamus, amygdala, and brainstem. In addition, theSCG contains cortico-cortical fibers which connect with the dorsolateral prefrontal,temporal, and parietal cortices. Accordingly, SCG DBS was found to modulate brainactivity in all of these nodes. Mayberg et al. (2005) confirmed hyperactivity in thesgACC and decreased activity in dorsal cortical regions in five MDD patientscompared with matched healthy controls by 15O-H2O PET. In three patients in whomelectrodes were subsequently implanted in the SCG, therapeutic DBS decreasedsgACC hyperactivity and increased dorsal prefrontal cortex hypoactivity. Theauthors replicated this finding with 18F-fluorodeoxyglucose PET in a second sampleof eight SCG DBS responders (Lozano et al. 2008). However, in this study, DBS notonly decreased SCG gray matter metabolism, but also increased activity in theadjacent white matter, which suggests that SCG DBS may either inhibit or excitegray matter network nodes through direct activation of local white matter. SCG DBSalso reduces activity in the OFC, medial frontal cortex, anterior insula, and hypo-thalamus, and increases metabolism in anterior and posterior cingulate, premotor,and parietal regions (Mayberg et al. 2005; Lozano et al. 2008).

21.3.2.4 Medial Forebrain Bundle

Coenen et al. (2009) used DTI to explore the mechanism of transient hypomaniaafter STN stimulation in Parkinson’s disease. Hypomania was related to stimu-lation of the STN electrode contact that had white matter connections with themedial forebrain bundle (MFB). On the basis of this observation, and the fact thatthe MFB is connected to the dopaminergic ventral tegmental area and to alleffective DBS targets for MDD, Coenen et al. (2011) recently proposed the MFBas a potential DBS target for MDD treatment.

21.3.2.5 Summary

In summary, DBS for treatment-resistant MDD has been targeted at the ventralcorticolimbic network (NAc and VC/VS), which resulted in a 42 % symptomimprovement and a responder rate of 12 of 25 patients. DBS at both dorsal and ventralnetworks (SCG) induced 48 % symptom improvement, with a responder rate of 22 of49 patients. NAc DBS may induce immediate hedonic improvement in MDD byrestoring NAc function, which is followed by normalization of excessive activity inthe OFC, sgACC, thalamus, and amygdala. Similar to NAc DBS, stimulation of SGCwhite matter normalizes MDD hyperactivity of the OFC and sgACC; however,additional changes are found in the hypothalamus. Furthermore, SCG DBS uniquelystimulates cortico-cortical fibers, which normalizes hypoactivity in the dorsolateralprefrontal cortex and parietal cortex.

21 Neuroimaging Deep Brain Stimulation 231

21.4 Addiction

Imaging studies in addicted subjects suggest that excessive drug use is associatedwith increased dopaminergic activity in the NAc and ventral tegmental area (Kooband Volkow 2010). Decreased activity in these reward regions may be responsible forthe anhedonic withdrawal effects that drive compulsive drug taking, in associationwith disrupted activity of the dorsolateral prefrontal cortex, OFC and ACC reflectingimpaired inhibitory control and impulsivity. Although all of these brain structures

Fig. 21.1 Example of implantation sites of electrodes and stimulator

Fig. 21.2 Nucleus accumbens/ventral striatum/anterior limb of the internal capsule target. RSFGright superior frontal gyrus, RDC right dorsal cortex, RINS right insula, OFC orbitofrontal cortex,NAc nucleus accumbens, SCG subcallosal gyrus, AMY amygdala, LT left thalamus

232 M. Figee et al.

could be potential DBS targets for addiction, only the ALIC/NAc has actually beentargeted in a total of seven addicted humans (Luigjes et al. 2012; Sun and Liu, thisvolume). In addition, beneficial effects of STN DBS on addictive behaviors arereported in the treatment of Parkinson’s disease (Luigjes et al. 2012). No imagingstudies investigating the mechanism of therapeutic DBS in addicted patients areavailable. As reported elsewhere in this chapter, NAc DBS may normalize dys-functional activity in the NAc and in connected frontal areas (Bewernick et al. 2010),which may be therapeutic for addiction by reducing craving, increasing salience of

Fig. 21.3 Subthalamic nucleus target. CING cingulum, OFC orbitofrontal cortex

Fig. 21.4 Subcallosal gyrus target. DLPFC dorsolateral prefrontal cortexprefrontal cortex, PMCpremotor cortex, CING cingulum, PC parietal cortex, SCG subcallosal gyrus, OFC orbitofrontalcortex, INS insula, HT hypothalamus, BS brainstem

21 Neuroimaging Deep Brain Stimulation 233

natural reinforcers, and improving inhibitory control. Finally, SCG DBS might beeffective for addiction because of its role in emotional control and findings ofincreased sgACC metabolism after intravenous administration of methylphenidatein addicted subjects compared with controls (Volkow et al. 2005).

21.5 Tourette Syndrome

21.5.1 Neuroimaging of TS Disease

TS is hypothesized to be caused by a failure of inhibition of the somatosensory‘‘premonitory urges’’ and associated motor enactments that constitute tics (Mink2001). Structural MRI studies reported reduced caudate nucleus volumes in chil-dren and adults with TS, and reduced putamen and globus pallidus volumes in TSpatients with comorbid OCD (Peterson et al. 2003; Bloch et al. 2005). However,high-precision surface-based diffeomorphic MRI techniques in drug-naïve TSadults failed to show volume differences in the basal ganglia or thalamus (Wanget al. 2007). Recent voxel-based brain morphometry in adult TS patients showedreduced gray matter volumes in the medial OFC, ACC, ventrolateral prefrontalcortex, operculum, amygdala, and hippocampus, whereas the volumes of theprimary somatosensory cortex, putamen, and right dorsal premotor cortex wereincreased (Draganski et al. 2010). Although these cortical gray matter changeswere not associated with comorbid OCD, a negative correlation was detectedbetween ventral striatal volume (NAc) and OCD symptom severity. DTI analysisrevealed white matter tract abnormalities in the corpus callosum, the anterior andposterior limb of the internal capsule, and long association fiber pathways such asthe superior longitudinal fascicle (Draganski et al. 2010; Neuner et al. 2010).

Functional MRI (fMRI) studies identified TS-related hyperactivity in brainregions that are thought to represent features of the premonitory urges, such as thesomatosensory and posterior parietal cortices, putamen, amygdala, and hippocampus(Wang et al. 2011). Furthermore, hyperactivity was observed throughout the motorpathway, including the primary motor cortex, prefrontal cortex, supplemental motorarea, posterior part of the ACC, putamen, globus pallidus, thalamus, and substantianigra (Bohlhalter et al. 2006; Wang et al. 2011). In contrast, CSTC network nodesthat exert top–down control over motor pathways, such as the caudate nucleus andthe anterior part of the ACC, were shown to be hypoactive (Wang et al. 2011).

In summary, recent fMRI studies in TS patients have identified pathologicalactivity in the CSTC circuits, the preceding sensory premonitory urge, and thefailed inhibition of both phenomena. The conflicting structural MRI results forbasal ganglia volumes may be explained by the use of different measuring tech-niques or, alternatively, may reflect the parallel existence of both hyperactivepremonitory urge/tics and hypoactive inhibitory CSTC circuits.

234 M. Figee et al.

21.5.2 Neuroimaging and DBS for the Treatment of TS

To date, four DBS targets have been used for treatment-resistant TS: the medialpart of the thalamus, the internal part of the globus pallidus, the external part of theglobus pallidus, and the ALIC/NAc (for a review, see Ackermans et al., thisvolume). Reported symptom improvement ranged between 24 and 95 %. Thus far,only one group has used neuroimaging to investigate brain changes caused bytherapeutic DBS in TS (Vernaleken et al. 2009; Kuhn et al. 2012). Three TSpatients who showed good responses to 6 months of medial thalamic DBS werescanned while DBS was on and off with [18F]fallypride PET to measure striataland extrastriatal dopamine D2/3-receptor binding. With DBS on, D2/3-receptoravailability of TS patients was higher than for matched healthy controls in thethalamus, temporal cortex, caudate nucleus, and putamen, which could reflect D2/

3-receptor upregulation being a result of chronic DBS. When DBS was switchedoff for 1 h, thalamus D2/3-receptor availability decreased by 7–18 % in twobilaterally stimulated patients, suggesting increased dopaminergic transmissionafter discontinuation of DBS. D2/3-receptor binding in one left-side-stimulatedpatient decreased by 6.4 % in the left thalamus but increased by 28 % in thecontralateral right thalamus. Conversely, putamen D2/3-receptor availabilityincreased when DBS was off in the patients with bithalamic DBS, but decreased inthe patients with unilateral thalamic stimulation. These results suggest that ther-apeutic thalamic DBS in TS modulates dopaminergic transmission in the motorstriatal circuit: bilateral thalamic stimulation causes a local dopamine leveldecrease in the thalamus and an increase in the putamen, whereas opposite changesare reported after unilateral stimulation.

21.6 Conclusions and Future Perspectives

Structural and functional neuroimaging studies have revealed dysfunction ofcortical–striatal–limbic networks in OCD, MDD, addiction, and TS. Functionalimaging studies that have investigated the mechanism of action of therapeuticDBS are limited, and have mainly focused on OCD and MDD (Figs. 21.1, 21.2,21.3, 21.4). DBS of the ALIC, VC/VS, and NAc normalizes frontostriatal couplingand excessive activity in the OFC and ACC, which may be therapeutic for OCDand MDD by restoring goal-directed behavior and improving emotional, cognitive,and behavioral control. DBS-induced restoration of local NAc activity seemsrelated to immediate hedonic and motivational changes in OCD and MDD. Similarto these ventral striatal DBS targets, SCG DBS in MDD normalizes OFC andsgACC hyperactivity, although its specific antidepressant effects may also belinked to reversal of hypoactivity in dorsal and parietal cortical areas and to itseffects on hypothalamus metabolism. Thalamic DBS for the treatment of TSmodulates dopaminergic transmission in motor striatal areas. Although there are

21 Neuroimaging Deep Brain Stimulation 235

no imaging data for DBS for addiction, NAc DBS may be the best choice becauseit normalizes activity in the ventral striatal reward system and in frontal inhibitorycontrol areas. Stimulation of the SCG has never been tried for treatment ofaddiction but might normalize sgACC hyperactivity in response to compulsivedrug taking.

Despite the recent advances in the field of psychiatric neuroimaging, mostcurrent psychiatric DBS targets have not been based on neuroimaging results. Onlythe SCG was defined from fMRI and PET findings in MDD patients. Recently, theMFB was proposed as a new MDD target on the basis of fiber-tracking imaging ofmood changes following STN DBS in Parkinson’s disease. The elucidation of themechanism of action of current DBS targets and the search for better brain targetswould profit from more studies reporting on baseline/preoperative activity of thediseased network, the exact neuroanatomical location of active electrode contactsand stimulation parameters, and DBS-induced changes of both the diseasednetwork and concomitantly modulated networks (which may explain DBS-relatedside effects). The combination of neuroimaging and DBS offers a unique researchtool to understand brain networks of psychiatric diseases and how to effectivelymodulate them.

Acknowledgments We would like to thank Rob Kreuger for his work on the illustrations.

References

Abelson JL, Curtis GC, Sagher O, Albucher RC, Harrigan M, Taylor SF, Martis B et al (2005) Deepbrain stimulation for refractory obsessive–compulsive disorder. Biol Psychiatry 57(5):510–516

Alcaro A, Panksepp J (2011) The SEEKING mind: primal neuro-affective substrates forappetitive incentive states and their pathological dynamics in addictions and depression.Neurosci Biobehav Rev 35(9):1805–1820

Axer H, Lippitz BE, Von Keyserlingk DG (1999) Morphological asymmetry in anterior limb ofhuman internal capsule revealed by confocal laser and polarized light microscopy. PsychiatryRes 91(3):141–154

Belin D, Everitt BJ (2008) Cocaine seeking habits depend upon dopamine-dependent serialconnectivity linking the ventral with the dorsal striatum. Neuron 57(3):432–441

Bewernick BH, Hurlemann R, Matusch A, Kayser S, Grubert C, Hadrysiewicz B, Axmacher Net al (2010) Nucleus accumbens deep brain stimulation decreases ratings of depression andanxiety in treatment-resistant depression. Biol Psychiatry 67(2):110–116

Bingley T, Leksell L, Meyerson BA (1977) Long-term results of stereotactic capsulotomy inchronic obsessive–compulsive neurosis. In: Sweet WH, Obrador S, Martin-Rodrigues JG(eds) Neurosurgical treatment in psychiatry, pain and epilepsy. University Park Press,Baltimore, pp 287–289

Bloch MH, Leckman JF, Zhu H, Peterson BS (2005) Caudate volumes in childhood predictsymptom severity in adults with Tourette syndrome. Neurology 65(8):1253–1258

Bohlhalter S, Goldfine A, Matteson S, Garraux G, Hanakawa T, Kansaku K et al (2006) Neuralcorrelates of tic generation in Tourette syndrome: an event-related functional MRI study.Brain 129(Pt 8):2029–2037

236 M. Figee et al.

Bora E, Harrison BJ, Fornito A, Cocchi L, Pujol J, Fontenelle LF, Velakoulis D et al (2011)White matter microstructure in patients with obsessive–compulsive disorder. J PsychiatryNeurosci 36(1):42–46

Coenen VA, Honey CR, Hurwitz T, Rahman AA, McMaster J, Burgel U, Madler B (2009)Medial forebrain bundle stimulation as a pathophysiological mechanism for hypomania insubthalamic nucleus deep brain stimulation for Parkinson’s disease. Neurosurgery 64:1106–1114 (1105–1114)

Coenen VA, Schlaepfer TE, Maedler B, Panksepp J (2011) Cross-species affective functions ofthe medial forebrain bundle-Implications for the treatment of affective pain and depression inhumans. Neurosci Biobehav Rev 35:1971–1981

de Koning PP, Figee M, van den Munckhof P, Schuurman PR, Denys D (2011) Current status ofdeep brain stimulation for obsessive–compulsive disorder: a clinical review of differenttargets. Curr Psychiatry Rep 13(4):274–282

Draganski B, Martino D, Cavanna AE, Hutton C, Orth M, Robertson MM et al (2010)Multispectral brain morphometry in Tourette syndrome persisting into adulthood. Brain133(Pt 12):3661–3675

Drevets WC, Price JL, Furey ML (2008) Brain structural and functional abnormalities in mooddisorders: implications for neurocircuitry models of depression. Brain Struct Funct 213(1–2):93–118

Figee M, Vink M, de Geus F, Vulink N, Veltman DJ, Westenberg H, Denys D (2011) Dysfunctionalreward circuitry in obsessive–compulsive disorder. Biol Psychiatry 69(9):867–874

Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, Salloway SP et al(2006) Three-year outcomes in deep brain stimulation for highly resistant obsessive–compulsive disorder. Neuropsychopharmacology 31(11):2384–2393

Harrison BJ, Soriano-Mas C, Pujol J, Ortiz H, López-Solà M, Hernández-Ribas R, Deus J et al(2009) Altered corticostriatal functional connectivity in obsessive–compulsive disorder. ArchGen Psychiatry 66(11):1189–1200

Hassler R, Dieckmann G (1970) Traitement stéréotaxique des tics et cris inarticulés oucopralalique considérés comme phénomène d’obsession motrice au cour de la maladies deGilles de la Tourette. Rev Neurol 123(2):89–100

Hasler G, Northoff G (2011) Discovering imaging endophenotypes for major depression. MolPsychiatry 16(6):604–619

Hommer DW, Bjork JM, Gilman JM (2011) Imaging brain response to reward in addictivedisorders. Ann N Y Acad Sci 1216: 50–61

Jiménez-Ponce F, Velasco-Campos F, Castro-Farfán G, Nicolini H, Velasco AL, Salín-Pascual R,Trejo D et al (2008) Preliminary study in patients with obsessive–compulsive disorder treatedwith electrical stimulation in the inferior thalamic peduncle. Neurosurgery 65(2):203–209

Knutson B, Fong GW, Adams CM, Varner JL, Hommer D (2001) Dissociation of rewardanticipation and outcome with event-related fMRI. NeuroReport 12:3683–3687

Koob GF, Volkow ND (2010) Neuropsychopharmacology 35(1):217–238Kuhn J, Janouschek H, Raptis M, Rex S, Lenartz D, Neuner I, Mottaghy FM et al (2012) In Vivo

evidence of deep brain stimulation-induced dopaminergic modulation in Tourette’s syndrome.Biol Psychiatry 71(5):e11–e13

Le Jeune F, Vérin M, N’Diaye K, Drapier D, Leray E, Du Montcel ST, Baup N et al (2010)Decrease of prefrontal metabolism after subthalamic stimulation in obsessive–compulsivedisorder: a positron emission tomography study. Biol Psychiatry 68(11):1016–1022

Lehman JF, Greenberg BD, McIntyre CC, Rasmussen SA, Haber SN (2011) Rules ventral prefrontalcortical axons use to reach their targets: implications for diffusion tensor imaging tractographyand deep brain stimulation for psychiatric illness. J Neurosci 31(28):10392–10402

Lorenzetti V, Allen NB, Fornito A, Yucel M (2009) Structural brain abnormalities in majordepressive disorder: a selective review of recent MRI studies. J Affect Disord 117(1–2): 1–17

Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH (2008)Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. BiolPsychiatry 64(6):461–467

21 Neuroimaging Deep Brain Stimulation 237

Lozano AM, Giacobbe P, Hamani C, Rizvi SJ, Kennedy SH, Kolivakis TT, Debonnel G et al(2012) A multicenter pilot study of subcallosal cingulate area deep brain stimulation fortreatment-resistant depression. J Neurosurg 116(2):315–322

Luigjes J, van den Brink W, Feenstra M, van den Munckhof P, Schuurman PR, Schippers R,Mazaheri A, De Vries TJ, Denys D (2012) Deep brain stimulation in addiction: a review ofpotential brain targets. Mol Psychiatry 17(6):572–583

Maller JJ, Thomson RH, Lewis PM, Rose SE, Pannek K, Fitzgerald PB (2010) Traumatic braininjury, major depression, and diffusion tensor imaging: making connections. Brain Res Rev64(1):213–240

Mallet L, Polosan M, Jaafari N, et al (2008) Subthalamic nucleus stimulation in severe obsessive-compulsive disorder. N Engl J Med 359(20):2121–2134

Malone DA, Dougherty DD, Rezai AR, Carpenter LL, Friehs GM, Eskandar EN, Rauch SL et al(2009) Deep brain stimulation of the ventral capsule/ventral striatum for treatment-resistantdepression. Biol Psychiatry 65(4):267–275

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, et al(2005). Deep brain stimulation for treatment-resistant depression. Neuron 45(5):651–660

Menzies L, Chamberlain SR, Laird AR, Thelen SM, Sahakian BJ, Bullmore ET (2008) Integratingevidence from neuroimaging and neuropsychological studies of obsessive–compulsive disorder:the orbitofronto-striatal model revisited. Neurosci Biobehav Rev 32(3):525–549

Mink JW (2001) Neurobiology of basal ganglia circuits in Tourette syndrome: faulty inhibition ofunwanted motor patterns? Adv Neurol 85:113–122

Moniz AE (1936) Essai d’un traitement chirurgical de certaines psychoses. Bulletin del’Academie de Médecine (Paris) 115:385–392

Nestler EJ, Carlezon WA (2006) The mesolimbic dopamine reward circuit in depression. BiolPsychiatry 59(12):1151–1159

Neuner I, Kupriyanova Y, Stöcker T, Huang R, Posnansky O, Schneider F, Tittgemeyer M, Shah NJ(2010) White-matter abnormalities in Tourette syndrome extend beyond motor pathways.Neuroimage 51(3):1184–1193

Nuttin B, Cosyns P, Demeulemeester H, Gybels J, Meyerson B (1999) Electrical stimulation in anteriorlimbs of internal capsules in patients with obsessive–compulsive disorder. Lancet 354(9189):1526

Nuttin BJ, Gabriëls LA, Cosyns PR, Meyerson BA, Andréewitch S, Sunaert SG, Maes AF et al(2003) Long-term electrical capsular stimulation in patients with obsessive–compulsivedisorder. Neurosurgery 52(6):1263–1274

Peterson BS, Thomas P, Kane MJ, Scahill L, Zhang H, Bronen R et al (2003) Basal gangliavolumes in patients with Gilles de la Tourette syndrome. Arch Gen Psychiatry 60(4):415–424

Pizzagalli DA, Holmes AJ, Dillon DG, Goetz EL, Birk JL, Bogdan R, Dougherty DD, IosifescuDV, Rauch SL, Fava M (2009) Reduced caudate and nucleus accumbens response to rewardsin unmedicated individuals with major depressive disorder. Am J Psychiatry 166:702–710

Puigdemont D, Perez-Egea R, Portella MJ, Molet J, de Diego-Adelino J, Gironell A, Radua J,Gomez-Anson B, Rodriguez R, Serra M, de Quintana C, Artigas F, Alvarez E, Perez V (2012)Deep brain stimulation of the subcallosal cingulate gyrus: further evidence in treatment-resistant major depression. Int J Neuropsychopharmacol 15(1):121–133

Radua J, van den Heuvel OA, Surguladze S, Mataix-Cols D (2010) Meta-analytical comparisonof voxel-based morphometry studies in obsessive–compulsive disorder vs other anxietydisorders. Arch Gen Psychiatry 67(7):701–711

Rauch SL, Dougherty DD, Malone D, Rezai A, Friehs G, Fischman AJ, Alpert NM et al (2006) Afunctional neuroimaging investigation of deep brain stimulation in patients with obsessive–compulsive disorder. J Neurosurg 104(4):558–565

Robinson OJ, Cools R, Carlisi CO, Sahakian BJ, Drevets WC (2011) Ventral striatum responseduring reward and punishment reversal learning in unmedicated major depressive disorder.Am J Psychiatry 2011:1–8

Rotge J, Guehl D, Dilharreguy B, Cuny E, Tignol J, Bioulac B, Allard M et al (2008) Examencritique Provocation of obsessive–compulsive symptoms : a quantitative voxel-based meta-analysis of functional neuroimaging studies. J Psychiatry Neurosci 33(33):405–412

238 M. Figee et al.

Sacher J, Neumann, J, Fünfstück T, Soliman A, Villringer A, Schroeter ML (2011). Mapping thedepressed brain: A meta-analysis of structural and functional alterations in major depressivedisorder. J affect disord 140(2):142–148

Sakai Y, Narumoto J, Nishida S, Nakamae T, Yamada K, Nishimura T, Fukui K (2010)Corticostriatal functional connectivity in non-medicated patients with obsessive–compulsivedisorder. Eur Psychiatry

Saxena S, Rauch SL (2000) Functional neuroimaging and the neuroanatomy of obsessive–compulsive disorder. Psychiatric Clin N Am 57(Suppl 8(3)):26–35

Schlaepfer TE, Cohen MX, Frick C, Kosel M, Brodesser D, Axmacher N, Joe AY et al (2008)Deep brain stimulation to reward circuitry alleviates anhedonia in refractory majordepression. Neuropsychopharmacology 33(2):368–377

Talairach J, Hécaen H, David M (1949) Lobotomie préfrontale limitée par électrocoagulation desfibres thalamo-forntales à leur émergence du bras antérieur de la capsule interne. In:Proceedings of the 4th Congrès Neurologique International, Masson, Paris, p 141

Tricomi E, Balleine BW, O’Doherty JP (2009) A specific role for posterior dorsolateral striatumin human habit learning. Eur J Neurosci 29(11):2225–2232

van der Vandewalle V, Linden C, Groenewegen HJ, Caemaert J (1999) Stereotactic treatment of Gillesde la Tourette syndrome by high frequency stimulation of thalamus. Lancet 353(9154):724

Van Laere K, Nuttin B, Gabriels L, Dupont P, Rasmussen S, Greenberg BD, Cosyns P (2006)Metabolic imaging of anterior capsular stimulation in refractory obsessive–compulsive disorder: akey role for the subgenual anterior cingulate and ventral striatum. J Nucl Med 47(5):740–747

Vernaleken I, Kuhn J, Lenartz D, Raptis M, Huff W, Janouschek et al (2009) Bithalamic deepbrain stimulation in Tourette syndrome is associated with reduction in dopaminergictransmission. Biol Psychiatry 66(10):e15–e17

Volkow ND, Wang GJ, Ma Y, Fowler JS, Wong C, Ding YS et al (2005) Activation of orbital andmedial prefrontal cortex by methylphenidate in cocaine-addicted subjects but not in controls:relevance to addiction. J Neurosci 25:3932–3939

Wang L, Lee DY, Bailey E, Hartlein JM, Gado MH, Miller MI et al (2007) Validity of large-defomration high dimensional brain mapping of the basal ganglia in adults in Tourettesyndrome. Psychiatry Res 154(2):181–190

Wang Z, Maia TV, Marsh R, Colibazzi T, Gerber A, Peterson BS (2011) The neural circuits thatgenerate tics in Tourette’s syndrome. Am J Psychiatry 168(12):1326–1337

Whiteside SP, Port JD, Abramowitz JS (2004) A meta–analysis of functional neuroimaging inobsessive–compulsive disorder. Psychiatry Interpers Biol Process 132:69–79

21 Neuroimaging Deep Brain Stimulation 239

Chapter 22Optogenetic Strategies for the Treatmentof Neuropsychiatric Disorders: Circuit-Function Analysis and ClinicalImplications

Daniel L. Albaugh and Garret D. Stuber

22.1 Introduction to Optogenetic Tools

Optogenetics involves the introduction of foreign opsin proteins into geneticallydefined cell populations to manipulate or report their electrical excitability orintracellular signaling in response to specific wavelengths of light (Fenno et al.2011; Yizhar et al. 2011; Zhang et al. 2010). Among the growing list of opsinproteins utilized for these purposes are the light-gated cation channel channel-rhodopsin-2 (ChR2) and the chloride pump halorhodopsin (NpHR), as well as therecently developed OptoXR family of light-gated G-protein-coupled receptors(allowing longer timescale control of intracellular signaling cascades) (Airan et al.2009; Zhang et al. 2007; Boyden et al. 2005). Although the implementation ofthese techniques within systems neuroscience is still in its infancy, there hasalready been great success in manipulating neural circuits with unprecedentedspecificity (Stuber et al. 2010, 2011; Tye et al. 2011; Ciocchi et al. 2010;Haubensak et al. 2010; Tecuapetla et al. 2010; Tsai et al. 2009). Opsin proteins canbe genetically introduced into neurons in a variety of fashions, such as expressingthem in transgenic animals or virally mediated gene delivery (Yizhar et al. 2011;Zhang et al. 2010). In addition, with use of Cre recombinase technologies orspecific gene promoters, virally mediated opsin expression can be experimentallylimited to distinct cell subpopulations near the viral vector injection site. Thus,

D. L. AlbaughCurriculum in Neurobiology, University of North Carolina at Chapel Hill,Chapel Hill, NC, USA

G. D. Stuber (&)Departments of Psychiatry & Cell and Molecular Physiology, UNC Neuroscience Center,University of North Carolina at Chapel Hill, Chapel Hill, NC, USAe-mail: gstuber@med.unc.eduhttp://www.stuberlab.org

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_22, � Springer-Verlag Berlin Heidelberg 2012

241

even within highly heterogeneous neural tissue, a single cell type can be manip-ulated on a physiologically relevant timescale in relative isolation. Additionalcircuit specificity arises from the targeting of optical fibers, which are used todeliver light to deep brain structures in vivo to activate opsin proteins expressed byneurons. These optical fibers, which can be chronically implanted into brain tissue,can be placed at the virus injection site or within afferent fibers to selectivelystimulate or inhibit either the somata or distal axon terminals of opsin-expressingcells (see Yizhar et al. 2011; Fig. 22.1). When combined, the molecular andneuroanatomical specificity provided by optogenetic tools can be quite powerful.

Given the high temporal precision of optogenetic activation/inactivation as wellas the cellular targeting specificity, the advantages of optogenetic strategies overthe traditional tools of pharmacology, electrical stimulation, and lesioning shouldbe apparent. Consequently, optogenetics is poised to rapidly broaden our under-standing of the neural circuitry guiding the physiological processes and behavior inhealth and disease. In this chapter, we review some of the advancements in sys-tems neuroscience made possible by optogenetic tools, focusing on studiesexamining the neurocircuity of reward and anxiety. We also discuss the promise ofoptogenetics as a technique to complement electrical deep brain stimulation (DBS)therapy for neuropsychiatric diseases.

Optogenetic modulationof distal axonal fibers

Optogenetic modulation of genetically-defined neurons

(a) (b)

Fig. 22.1 Optogenetic modulation of neural circuit elements. a Genetically defined neuronsexpressing opsin proteins, such as channelrhodopsin-2 (ChR2), are shown in orange, andneighboring neurons that are genetically distinct are shown in green. When light of theappropriate wavelength is introduced into the neural tissue (shown in blue), only neuronsexpressing ChR2 are directly excited, whereas neurons not expressing ChR2 (shown in green)show no direct change in firing due to light exposure. b Optical stimulation fibers can be placeddirectly above terminal fields of opsin-expressing neurons to selectively activate or inactive fibersthat originate from genetically targeted neurons, while leaving other fibers in the regionunaffected

242 D. L. Albaugh and G. D. Stuber

22.2 Reward

Although optogenetic strategies have only recently been developed, they haveallowed rapid advancements in our understanding of the neural circuitry under-lying reward processing and various neuropsychiatric phenotypes characterized bydysregulation of these circuits (e.g., depression, addiction). The application ofoptogenetic tools in this area of research has focused largely on the ventral stri-atum of the basal ganglia as well as midbrain dopaminergic systems that projectthroughout the forebrain. The nucleus accumbens (NAc) is a heterogeneousstructure comprising multiple cell types (mostly medium spiny neurons selectivelyexpressing either the D1 or the D2 dopamine receptor subtype), receiving afferentinnervation from multiple sources, e.g., ventral tegmental area (VTA), amygdala,hippocampus, thalamus, and prefrontal cortex (Sesack and Grace 2010). Given thiscomplex neuroconnectivity, teasing out the behavioral phenotypes mediated byspecific cell types and circuits in disorders characterized by dysfunctional rewardsystems has been a daunting task. However, recent applications of optogeneticswithin various nodes of these neural circuits have demonstrated important func-tional roles of neural circuit elements in controlling reward-seeking behavior.

In a recent study, we applied optogenetic tools to study the role of neuro-transmission from the basolateral nucleus of the amygdala (BLA) to the NAc incontrolling reward-seeking behaviors. Previous studies have shown that electricalstimulation of the BLA can alter NAc dopamine release, and that inactivation ofthe BLA reduces cue-induced behavioral responding (Floresco et al. 1998; Joneset al. 2010; Ambroggi et al. 2008); however, a mechanistic understanding of howthe BLA-to-NAc projection influences reward-seeking behavior has been elusive.Because principal neurons of the BLA project to diverse target regions, includingcentral nucleus of the amygdala (CeA) targets involved in fear and anxiety (seeSect. 22.3), nonselective manipulations of the BLA are likely to perturb multiplecircuits underlying many different neurobehavioral actions. By selectively acti-vating opsin-expressing glutamatergic fibers in the NAc originating in the BLA,we were able to evaluate the motivational properties of selective activation of theBLA-to-NAc projection. Using an operant task in which brief optical stimulationof this pathway (using ChR2) was contingent upon a nose-poke behavior, wefound that optical stimulation of the BLA-to-NAc pathway readily reinforcedbehavioral responding, but activation of another glutamate input to the NAc fromthe medial prefrontal cortex did not. Additionally, the processing of naturalrewards mediated by this pathway was explored using a Pavlovian conditioningparadigm in which a multimodal discrete cue predicted the availability of sucrose,infused into the well of an operant chamber. With training, normal mice displayedanticipatory licking behavior in response to the reward-predictive cue as well asconsummatory licking following reward delivery. Notably, optogenetic inhibitionof the BLA-to-NAc pathway using NpHR time-locked to the cue-presentationperiod abolished the development of both anticipatory and reward consumptionlicking. Thus, neural activity within the BLA-to-NAc pathway exerts powerful

22 Optogenetic Strategies for the Treatment of Neuropsychiatric Disorders 243

control over reward-seeking behaviors, with important roles in processing bothunconditioned and cue-mediated reward seeking. Although this study was the firstto show the role of anatomically distinct glutamate input to the NAc, additionalstudies have used optogenetic tools to further delineate the role of dopaminergicinput in mediating physiological responses and reward-related behaviors.

A unique advantage of optogenetic tools lies in the synthesis of unparalleledspatial specificity in regional or circuit stimulation with temporal resolution onphysiological scales (or faster). Previous work has demonstrated that dopaminergicneurons exhibit two firing patterns: a low-frequency tonic activity (approximately3–8 Hz) and a higher-frequency phasic ‘‘bursting’’ firing pattern (15–20 Hz)(Grace and Bunney 1984a, b). Although correlative analyses using single-unit andmultiunit electrode arrays have demonstrated that burst firing of dopamine neuronsin the VTA are time-locked to rewards and cues that predict them (for reviews, seeSchultz et al. 1997; Wanat et al. 2009), the ability to mimic this firing pattern invivo selectively in dopaminergic neurons of the VTA required an optogeneticapproach (Tsai et al. 2009). Expression of ChR2 selectively in dopaminergicneurons of the VTA allowed optical control of burst firing in vivo and concurrentlymeasurement of any affective responses to this stimulation using a conditionedplace preference paradigm. Over several training sessions, mice experiencedpairings of stimulation-induced burst firing with a single section of a compart-mentalized conditioning chamber. Following this training and in the absence ofany stimulation, mice given access to the entire chamber displayed a significantconditioned place preference for the burst-firing-paired compartment, reflective ofthe rewarding nature of burst firing and demonstrating the sufficiency of this firingfor behavioral conditioning in the absence of any additional reinforcer. Notably, asimilar place conditioning paradigm using tonic stimulation revealed an inabilityof this firing pattern to condition a place preference. Additional studies have alsodemonstrated that direct optogenetic activation of VTA dopaminergic neurons canalso support operant self-stimulation behavior (Adamantidis et al. 2011; Wittenet al. 2011). Taken together, these studies have demonstrated a definitive role fordirect activation of dopaminergic neuronal activity in controlling reward-relatedbehaviors.

Optogenetic activation of dopamine-producing neurons has also yieldedimportant information on postsynaptic consequences in the striatum followingstimulation of dopaminergic fibers. For example, although suggested by indirectevidence, the ability of dopaminergic terminals arising from the VTA to coreleaseglutamate as a neurotransmitter required a strategy to selectively stimulate axonalfibers from dopamine-producing neurons, while not stimulating other glutama-tergic afferents that were in close proximity (Stuber et al. 2010; Tecuapetla et al.2010). With use of transgenic mice coupled with recombinase-driven expression ofChR2, it was possible to selectively introduce ChR2 into midbrain dopaminergicneurons as described above. Over time, ChR2 is trafficked along axonal fibers ofVTA dopamine-producing neurons that innervate the ventral and dorsal striatum.Whole-cell patch-clamp recordings from postsynaptic medium spiny neurons inthese regions revealed that optical stimulation of dopaminergic fibers resulted in

244 D. L. Albaugh and G. D. Stuber

detectable excitatory postsynaptic currents, which were pharmacologicallyblocked by a-amino-3-hydroxy-5-methyl-4-isoxazolepropionate receptor antago-nist. Thus, the use of optogenetic strategies in this case provided the first directevidence of glutamate corelease from dopaminergic terminals that project to theventral striatum.

In addition to aiding our understanding of the basic circuit mechanisms ofreward, recent work using optogenetic tools has hinted at the potential clinicalbenefits of optically stimulating brain circuits for the treatment of reward dys-function disorders. Using a model of depression induced by social-defeat stress,Covington et al. (2010) optically evoked patterns of burst firing in the prefrontalcortex during behavioral testing in chronically defeated mice. Such mice typicallydisplay a variety of reward-related deficits, including a reduction in sucrosepreference and social interaction behaviors. Notably, the optical stimulation wasobserved to block these depressive-like behaviors, with stressed mice receivingsuch stimulation performing comparably to nonstressed controls on both tests. Thisfinding complements preclinical work by others using standard electrodes (e.g.,Hamani et al. 2012), and suggests that the prefrontal cortex may be an efficacioustarget for DBS for the treatment of reward-related disorders (see also Luigjes et al.2012). However, it is important to note that the cellular and circuit mechanismsunderlying the aforementioned antidepressant effect remain poorly understood,and thus the advantage of target specificity provided by optogenetic tools cannotpresently be realized in this preclinical therapeutic context. Indeed, the vector usedby Covington et al. was panneuronal, stimulating both glutamatergic pyramidalcells and GABAergic interneurons, and no analysis of circuit-specific stimulationeffects was undertaken. Further translational studies are sorely needed to elucidatethe optimal temporal stimulation patterns and relevant targets underlying suchtherapeutic effects, which likely critically rely on stimulation of specific circuits.

22.3 Anxiety-Related Disorders

Anxiety-related disorders represent one of the commonest classes of neuropsy-chiatric disorders, and have thus received substantial attention within the behav-ioral neurosciences. A variety of animal models have been developed to dissect theneurocircuitry of anxiety, often including the application of fear-conditioningparadigms in rodents, which are thought be analogous to disorders in humans (Shinand Liberzon 2010). Much research in this area has focused on the amygdaloidcomplex, which consists of several anatomically and functionally distinct nuclei,including the BLA and CeA, as well as a mesh-like sheet of intercalated cellmasses spanning the rostrocaudal regions (Pape and Pare 2010). Slice electro-physiology analyses have demonstrated strong modulatory input from both theBLA and the intercalated cell masses converging upon the CeA (the putativebehavioral output center in fear expression), including an unreciprocated excit-atory projection from the BLA to the CeA (Pape and Pare 2010; Likhtik et al.

22 Optogenetic Strategies for the Treatment of Neuropsychiatric Disorders 245

2008). Attempts to link such amygdalar circuits to behavior have been complicatedby the high degree of internuclear connectivity within the amygdala, as well as theclose spatial proximity of distinct nuclei and their subdivisions (e.g., both the BLAand the CeA have lateral and medial subnuclei) (Pape and Pare 2010). Forexample, although lesioning studies have identified both the BLA and the CeA asimportant mediators in the expression of conditioned fear (Goosens and Maren2001; Campeau and Davis 1995), the data cannot specify if any distinct circuits orsubnuclei are differentially involved in the behavior. Optogenetic tools provide afar greater degree of circuit-level specificity, and have provided circuit-functionanalyses of amygdalar microcircuits within the context of anxiety and fear-basedlearning (Tye et al. 2011; Ciocchi et al. 2010; Haubensak et al. 2010; Johansenet al. 2010).

An optogenetic approach has recently been used to determine the role of theexcitatory projection from the BLA to the CeA in unconditioned anxiety (Tye et al.2011). Mice were exposed to either an open-field or an elevated-plus-maze par-adigm, in which anxiety is measured as time spent near the walls or in the closedarms of the field and maze, respectively (see Ramos 2008). Targeted expression ofopsins within the BLA, coupled with optical stimulation of BLA fibers thatinnervate the CeA, allowed the authors to assess any anxiety-altering properties ofcircuit stimulation or inhibition solely within the BLA-to-CeA projection. Inter-estingly, excitation of this pathway was found to be anxiolytic, a behavioral effectcorrelated with reduced c-fos expression (a gross histological measure of neuralactivity) in the medial CeA (CeM). Coupled with slice electrophysiology datademonstrating light-evoked feed-forward inhibition of the lateral CeA (CeL; orCeM) arising from excitatory BLA input to inhibitory neurons of the medial CeAnuclei (CeL), the data suggest that excitation of the CeL from the BLA serves todisinhibit the CeM via an inhibitory CeL-to-CeM projection, an effect that mayunderlie unconditioned anxiety. Further strengthening this argument, NpHR-mediated inhibition of the BLA–CeA projection was shown to be anxiogenic andpromote activity in the CeM region (again measured by c-fos activation).

In addition to unconditioned anxiety, the roles of various amygdalar nuclei infear conditioning have been optogenetically dissected. Similar to the results of Tyeet al. (2011), optogenetic manipulations of the CeA have demonstrated this regionto be an integral component in the acquisition and expression of conditioned fear,furthering the idea of overlap between neural circuits mediating fear and anxiety.In the report by Ciocchi et al. (Ciocchi et al. 2010), stimulation of the CeM withChR2 was shown to result in a robust and reversible unconditioned freezingbehavior in mice, likely attributable to activation of amygdalar fear circuitry. Thisfinding was followed up with pharmacological analyses, including inactivation ofthe CeL with a pharmacological cocktail during fear conditioning. Notably,bilateral CeL inactivation during the conditioning period suppressed the acquisi-tion of cue-induced freezing behavior in these mice, suggesting a role for thisregion in conditioned fear learning. Moreover, in vivo electrophysiological anal-yses identified two distinct groups of CeL neurons, defined by their opposingresponses (i.e., inhibition or excitation) to conditioned fear cues. Furthermore,

246 D. L. Albaugh and G. D. Stuber

another report using optogenetic tools demonstrated that the CeL neurons that areinhibited during presentation of a cue previously paired with exposure to aversivestimuli (termed CeLoff cells) selectively express protein kinase C (PKC)-d, amolecular marker for roughly half of the GABAergic CeL neurons (Haubensaket al. 2010). Pharmacological inactivation of the CeLoff cells resulted in dramat-ically increased conditioned and unconditioned freezing behavior, suggesting thatthe CeLoff cells oppose the PKC-d-negative CeLon cells to limit the activation ofthe CeM during fear learning and expression. Taken together, these reportsidentify inhibitory input from the CeL to the CeM as a powerful mediator ofconditioned fear expression, and show that a subset of these cells, geneticallydefined by PKC-d expression, are inhibited by fear-conditioned cues to gatefreezing behavior. This work also highlights the powerful capability of optogenetictools to selectively manipulate genetically defined intranuclear cell populations, anapproach that holds much promise for both comprehensive circuit mapping and thedevelopment of more sophisticated DBS targeting.

22.4 Translational Approaches

In addition to its applications in basic biomedical research, optogenetic tools holdmuch promise for the clinic, specifically as an improvement over traditional DBStherapy. In DBS, chronic electrode implants are stereotactically directed to atargeted brain region, whereby electrical stimulation of that region allows the rapidamelioration of neurological symptoms. This method of DBS has met with greatsuccess in treating the motor symptoms of Parkinson’s disease, particularly whenhigh-frequency electrical stimulation is directed at the subthalamic nucleus, animportant component of the basal ganglia motor circuit (Volkmann et al. 2009).Evidence for the utility of electrical stimulation methods for the treatment ofneuropsychiatric disorders has also been obtained, including small-sample casestudies using DBS (commonly in the ventral striatum) to ameliorate symptoms ofobsessive–compulsive and major depressive disorders (reviewed in de Koninget al. 2011; Holtzheimer and Mayberg 2011; Krack et al. 2010). Despite theeffectiveness of these therapies, the inherent regional nonspecificity of traditionalDBS leaves much to be desired, as electrical diffusion to neighboring brain areasand stimulation of fibers of passage are both likely. The consequences of thisnonspecificity are not trivial. Indeed, with DBS directed at the subthalamic nucleusfor treatment of motor symptoms in Parkinson’s disease, patients may report manyaffective side effects, including both mania and dysphoria; such side effects areoften attributed to the nontargeted stimulation of neighboring limbic areas (Kracket al. 2010). If optogenetic tools are to supplant electrodes as the optimal methodfor DBS, the added specificity of spatially restricted and cell-type-specific opsinexpression and stimulation will be its prime advantage.

Although optogenetically mediated DBS holds many potential benefits, includingthe aforementioned anatomical and cell-type specificity, reversibility of stimulation,

22 Optogenetic Strategies for the Treatment of Neuropsychiatric Disorders 247

and the availability of an increasingly diverse library of light-activated membraneion channels and receptors (allowing both rapid channel activation and directstimulation of intracellular signaling cascades) (Zhang et al. 2011), much needs to belearned before this technique can be brought to the clinic. Among the importantissues are the safety and long-term efficacy of both the viral vectors and theimplantable optical fibers necessary for the expression and stimulation of opsins,respectively. With regard to viral vectors, adeno-associated virus (AAV) is the mostlikely candidate for gene therapy, owing to its high and long-term expression effi-ciency, as well as low immunogenicity (Monahan and Samulski 2000). In humanclinical trials, AAV vectors have shown promise for gene therapy in neurologicaldisorders. For example, in a phase I clinical trial, the dopamine precursor enzymearomatic L-amino acid decarboxylase was safely and successfully delivered into theputamen region of six Parkinson’s disease patients, with relatively minor postsur-gical complications. Increases in the levels of aromatic L-amino acid decarboxylaseabove presurgery levels were verified by PET at multiple time points, and persistedfor at least 96 weeks in two patients (Muramatsu et al. 2010). However promisingthese results, optogenetic DBS therapy would have a considerable added compli-cation over such gene therapy treatments in that it will require the chronic implan-tation of optical fibers to deliver light into deep brain structures. In contrast to the longhistory of experimental and clinical work demonstrating the safety of chronicallyimplanted electrode components (Coffey 2009), little or nothing is known regardingthe tolerability of optical fibers in human nervous tissue. Possible complicationsinclude the susceptibility of the foreign fiber material to generate an immunogenicresponse, mechanical tissue damage during implantation, thermal damage due to theheat generated from the laser (Cardin et al. 2010), and fiber breakage followingimplantation.

Presently, the best evidence that optogenetic DBS may be safe in humans comesfrom two reports of virally mediated opsin expression and stimulation in nonhumanprimate brain (Han et al. 2009; Diester et al. 2011). In both studies, targetedexpression of functional opsin proteins within the macaque cortex was achieved bylentivirus (Diester et al. 2011; Han et al. 2009) or AAV (Diester et al. 2011) injection.Notably, high transduction efficiency at the virus injection site (over 50 % of neu-rons) was achievable with both vectors and using multiple gene promoters (hThy1,hSyn, CaMKIIa), was specific to the targeted cell type and brain region, and gen-erated only minimal immune response. Given the paramount importance of long-duration protein expression to any future clinical trials, it is also encouraging thatopsins were found to be highly expressed and functional after several months (at least8 months in the report by Han et al. 2009), and retained a normal cellular morphology(although abnormal dendritic opsin aggregates were noted with viral overexpression;see Diester et al. 2011). Collectively, these reports provide initial evidence thatoptogenetic tools may be suitably applied to primate brain, although extensivecharacterization will be required before any work with humans can commence.Among other research trajectories, it will be important to determine the safety andlong-term efficacy (with time courses of years rather than months) of optical DBS

248 D. L. Albaugh and G. D. Stuber

therapy in well-characterized primate models of neurodegenerative and psychiatricdiseases.

As mentioned previously, further work will also be needed to identify the optimaltargets and stimulation patterns for optogenetic DBS for the treatment of neuro-psychiatric disorders. Such analyses will be critical for translating results in animalmodels to the clinic for a variety of reasons. Compared with DBS therapy formovement disorders, the efficacy of the stimulation parameters for the treatment ofneuropsychiatric illnesses will be difficult to discern, especially during intraoperativetesting (i.e., therapeutic outcomes will not be immediate; although see Haq et al.2011). Similarly, interindividual differences in the efficacy of specific stimulationparameters may be difficult to resolve, again owing to the absence of immediatetherapeutic results. Moreover, current DBS treatments for neuropsychiatric disordersusing standard electrodes typically apply continuous stimulation protocols(Goodman and Alterman 2012), further highlighting a need to discern the efficacy ofalternative (e.g., cycled or on-demand) patterns of stimulation. Related to theseefficacy concerns, adverse consequences of DBS may, in some cases, be related to thestimulation parameters. For example, recent case studies of DBS for treatment ofobsessive–compulsive disorder (a major candidate neuropsychiatric disorder forDBS therapy; see de Koning et al. 2011) have described manic episodes andincreased impulsivity that appear to result from the supratherapeutic stimulationintensity (Luigjes et al. 2011; Haq et al. 2010). Indeed, in both reports, reducing thevoltage or field of stimulation was shown to effectively reverse the described adverseeffects. Although stimulation using optogenetic DBS will likely be less diffuse (andthus untargeted circuits are less likely to be stimulated), the possibility of adverseconsequences arising from stimulation remain, and preclinical and clinical work willbe needed to determine the optimal stimulation protocols for each neuroanatomicaltarget and disorder to be treated.

Optogenetic strategies have rapidly emerged as one of the most powerful toolsin systems neuroscience, and have guided many important discoveries with rele-vance for neuropsychiatric disorders. The benefits of optogenetic tools for basicresearch are beginning to be realized, with increasingly elegant examples of op-togenetically guided experiments arising in the literature. In contrast, the value ofoptogenetic tools in a clinical setting, including their use in DBS therapy, is muchless clear. The initial characterization of opsin expression and function in primatebrain tissue is promising, and it will be of great interest to see further extension ofthese preliminary translational findings.

References

Adamantidis AR, Tsai HC, Boutrel B, Zhang F, Stuber GD, Budygin EA, Touriño C, Bonci A,Deisseroth K, de Lecea L (2011) Optogenetic interrogation of dopaminergic modulation ofthe multiple phases of reward-seeking behavior. J Neurosci 31:10829–10835

Airan RD, Thompson KR, Fenno LE, Bernstein H, Deisseroth K (2009) Temporally precise invivo control of intracellular signaling. Nature 458:1025–1029

22 Optogenetic Strategies for the Treatment of Neuropsychiatric Disorders 249

Ambroggi F, Ishikawa A, Fields HL, Nicola SM (2008) Basolateral amygdala neurons facilitatereward-seeking behavior by exciting nucleus accumbens neurons. Neuron 59:648–661

Boyden ES, Zhang F, Bamberg E, Nagel G, Deisseroth K (2005) Millisecond-timescale,genetically targeted optical control of neural activity. Nat Neurosci 8:1263–1268

Campeau S, Davis M (1995) Involvement of the central nucleus and basolateral complex of theamygdala in fear conditioning measured with fear-potentiated startle in rats trainedconcurrently with auditory and visual conditioned stimuli. J Neurosci 15:2301–2311

Cardin JA, Carlén M, Meletis K, Knoblich U, Zhang F, Deisseroth K, Tsai LH, Moore CI (2010)Targeted optogenetic stimulation and recording of neurons in vivo using cell-type-specificexpression of channelrhodopsin-2. Nat Protoc 5:247–254

Ciocchi S, Herry C, Grenier F, Wolff SB, Letzkus JJ, Vlachos I, Ehrlich I, Sprengel R, Deisseroth K,Stadler MB, Muller C, Luthi A (2010) Encoding of conditioned fear in central amygdalainhibitory circuits. Nature 468:277–282

Coffey RJ (2009) Deep brain stimulation devices: a brief technical history and review. ArtifOrgans 33:208–220

Covington HE 3rd, Lobo MK, Maze I, Vialou V, Hyman JM, Zaman S, LaPlant Q, Mouzon E,Ghose S, Tamminga CA, Neve RL, Deisseroth K, Nestler EJ (2010) Antidepressant effect ofoptogenetic stimulation of the medial prefrontal cortex. J Neurosci 30:16082–16090

De Koning PP, Figee M, van den Munchof P, Schuurman PR, Denys D (2011) Current status ofdeep brain stimulation for obsessive-compulsive disorder. Curr Psychiatry Rep 13:274–282

Diester I, Kaufman MT, Mogri M, Pashaie R, Goo W, Yizhar O, Ramakrishnan C, Deisseroth K,Shenoy KV (2011) An optogenetic toolbox designed for primates. Nat Neurosci 14:387–397

Fenno L, Yizhar O, Deisseroth K (2011) The development and applications of optogenetics. AnnRev Neurosci 34:289–412

Floresco SB, Yang CR, Phillips AG, Blaha CD (1998) Basolateral amygdala stimulation evokesglutamate receptor-dependent dopamine efflux in the nucleus accumbens of the anaesthetizedrat. Eur J Neurosci 10:1241–1251

Goodman WK, Alterman RM (2012) Deep brain stimulation for intractable psychiatric disorders.Ann Rev Med 63:511–524

Goosens KA, Maren S (2001) Contextual and auditory fear conditioning are mediated by lateral,basal, and central amygdaloid nuclei in rats. Learn Mem 8:148–155

Grace AA, Bunney BS (1984a) The control of firing pattern in nigral dopamine neurons: singlespike firing. J Neurosci 4:2866–2876

Grace AA, Bunney BS (1984b) The control of firing pattern in nigral dopamine neurons: burstfiring. J Neurosci 4:2877–2890

Hamani C, Machado DC, Hipólide DC, Dubiela FP, Suchecki D, Macedo CE, Tescarollo F,Martins U, Covolan L, Nobrega JN (2012) Deep brain stimulation reverses anhedonic-likebehavior in a chronic model of depression: role of serotonin and brain derived neurotrophicfactor. Biol Psychiatry 71:30–35

Han X, Qian X, Bernstein JG, Zhou HH, Franzesi GT, Stern P, Bronson RT, Graybiel AM,Desimone R, Boyden ES (2009) Millisecond-timescale optical control of neural dynamics inthe nonhuman primate brain. Neuron 62:191–198

Haq IU, Foote KD, Goodman WK, Ricciuti N, Ward H, Sudhyadhom A, Jacobson CE, Siddiqui MS,Okun MS (2010) A case of mania following deep brain stimulation for obsessive compulsivedisorder. Stereotact Funct Neurosurg 88:322–328

Haq IU, Foote KD, Goodman WG, Wu SS, Sudhyadhom A, Ricciuti N, Siddiqui MS, Bowers D,Jacobson CE, Ward H, Okun MS (2011) Smile and laughter induction and intraoperativepredictors of response to deep brain stimulation for obsessive-compulsive disorder.Neuroimage 54(Suppl 1):S247–S255

Haubensak W, Kunwar PS, Cai H, Ciocchi S, Wall NR, Ponnusamy R, Biag J, Dong H-W,Deisseroth K, Callaway EM, Fanselow MS, Luthi A, Anderson DJ (2010) Genetic dissectionof an amygdala microcircuit that gates conditioned fear. Nature 468:270–276

Holtzheimer PE, Mayberg HS (2011) Deep brain stimulation for psychiatric disorders. Annu RevNeurosci 34:289–307

250 D. L. Albaugh and G. D. Stuber

Johansen JP, Hamnaka H, Monafils MH, Behnia R, Deisseroth K, Blair HT, LeDoux JE (2010)Optical activation of lateral amygdala pyramidal cells instructs associative fear learning. ProcNatl Acad Sci U S A 107:12692–12697

Jones JL, Day JJ, Aragona BJ, Wheeler RA, Wightman RM, Carelli RM (2010) Basolateralamygdala modulates terminal dopamine release in the nucleus accumbens and conditionedresponding. Biol Psychiatry 67:737–744

Krack P, Hariz MI, Baunez C, Guridi J, Obeso JA (2010) Deep brain stimulation: from neurologyto psychiatry? Trends Neurosci 33:474–484

Likhtik E, Popa D, Apergis-Schoute L, Fidacaro GA, Pare D (2008) Amygdala intercalatedneurons are required for expression of fear extinction. Nature 454:642–645

Luigjes J, Mantione M, van den Brink W, Schuurman PR, van den Munckhof P, Denys D (2011)Deep brain stimulation increases impulsivity in two patients with obsessive–compulsivedisorder. Int Clin Psychopharmacol 26:338–340

Luigjes J, van den Brink W, Feenstra M, van den Munckhof P, Schuurman PR, Schippers R,Mazaheri A, De Vries TJ, Denys D (2012) Deep brain stimulation in addiction: a review ofpotential brain targets. Mol Psychiatry. doi:10.1038/mp.2011.114

Monahan PE, Samulski RJ (2000) Adeno-associated virus vectors for gene therapy: more prosthan cons? Mol Med Today 6:433–440

Muramatsu S, Fujimoto K, Kato S, Mizukami H, Asari S, Ikeguchi K, Kawakami T, Urabe M,Kume A, Sato T, Watanabe E, Ozawa K, Nakano I (2010) A phase I study of aromatic L-amino acid decarboxylase gene therapy for Parkinson’s disease. Mol Ther 18:1731–1735

Pape HC, Pare D (2010) Plastic synaptic networks of the amygdala for the acquisition,expression, and extinction of conditioned fear. Physiol Rev 90:419–463

Ramos A (2008) Animal models of anxiety: do I need multiple tests? Trends Pharmacol Sci29:493–498

Schultz W, Dayan P, Montague PR (1997) A neural substrate of prediction and reward. Science275:1593–1599

Sesack SR, Grace AA (2010) Cortico-basal ganglia reward network: Microcircuitry. Neuropsy-chopharmacology 15:27–47

Shin LM, Liberzon I (2010) The neurocircuitry of fear, stress, and anxiety disorders.Neuropsychopharmacology 35:169–191

Stuber GD, Hnasko TS, Britt JP, Edwards RH, Bonci A (2010) Dopaminergic terminals in thenucleus accumbens but not the dorsal striatum corelease glutamate. J Neurosci 30:8229–8233

Stuber GD, Sparta DR, Stamatakis AM, van Leeuwen WA, Hardjoprajitno JE, Cho S, Tye KM,Kempadoo KA, Zhang F, Deisseroth K, Bonci A (2011) Excitatory transmission from theamygdala to nucleus accumbens facilitates reward seeking. Nature 475:377–380

Tecuapetla F, Patel JC, Xenias H, English D, Tadros I, Shah F, Berlin J, Deisseroth K, Rice ME,Tepper JM, Koos T (2010) Glutamatergic signaling by mesolimbic dopamine neurons in thenucleus accumbens. J Neurosci 30:7105–7110

Tsai HC, Zhang F, Adamantidis A, Stuber GD, Bonic A, de Lecea L, Deisseroth K (2009) Phasicfiring in dopaminergic neurons is sufficient for behavioral conditioning. Science 324:1080–1084

Tye KM, Prakash R, Kim S-Y, Fenno LE, Grosenick L, Zarabi H, Thompson KR, Gradinaru V,Ramakrishnan C, Deisseroth K (2011) Amygdala circuitry mediating reversible andbidirectional control of anxiety. Nature 471:358–362

Volkmann J, Albanese A, Kulisevsky J, Tornqvist AL, Houeto JL, Pidoux B, Bonnet AM, Mendes A,Benabid AL, Fraix V, Van Blercom N, Xie J, Obeso J, Rodriguez-Oroz MC, Guridi J, Schnitzler A,Timmermann L, Gironell AA, Molet J, Pascual-Sedano B, Rehncrona S, Moro E, Lang AC,Lozano AM, Bentivoglio AR, Scerrati M, Contarino MF, Romito L, Janssens M, Agid Y (2009)Long-term effects of pallidal or subthalamic deep brain stimulation on quality of life in Parkinson’sdisease. Mov Disord 24:1154–1161

Wanat MJ, Willuhn I, Clark JJ, Phillips PEM (2009) Phasic dopamine release in appetitivebehaviors and drug abuse. Curr Drug Abuse Rev 2:195–213

22 Optogenetic Strategies for the Treatment of Neuropsychiatric Disorders 251

Witten IB, Steinberg EE, Lee SY, Davidson TJ, Zalocusky KA, Brodsky M, Yizhar O, Cho SL,Gong S, Ramakrishnan C, Stuber GD, Tye KM, Janak PH, Deisseroth K (2011) Recombinase-driver rat lines: tools, techniques, and optogenetic application to dopamine-mediatedreinforcement. Neuron 72:721–733

Yizhar O, Fenno LE, Davidson TJ, Mogri M, Deisseroth K (2011) Optogenetics in neuralsystems. Neuron 14:9–34

Zhang F, Wang LP, Brauner M, Liewald JF, Kay K, Watzke N, Wood PG, Bamberg E, Nagel G,Gottschalk A, Deisseroth K (2007) Multimodal fast optical interrogation of neural circuitry.Nature 446:633–639

Zhang F, Gradinaru V, Adamantidis AR, Durand R, Airan RD, de Lecea L, Deisseroth K (2010)Optogenetic interrogation of neural circuits: technology for probing mammalian brainstructures. Nat Proc 5:439–456

Zhang F, Vierock J, Yizhar O, Fenno LE, Tsunoda S, Kianianmomeni A, Prigge M, Berndt A,Cushman J, Polle J, Magnuson J, Hegemann P, Deisseroth K (2011) The microbial opsinfamily of optogenetic tools. Cell 147:1446–1457

252 D. L. Albaugh and G. D. Stuber

Chapter 23Next-Generation Electrodes for SteeringBrain Stimulation

H. C. F. Martens, M. M. J. Decré and E. Toader

23.1 Introduction

During the 1990s, deep brain stimulation (DBS) emerged as a treatment option formovement disorders such as essential tremor, Parkinson’s disease, and dystonia(Benabid et al. 1991; Kumar et al. 1998; Krauss et al. 1999). Long-term experiencehas demonstrated that DBS is safe and its side effects are reversible and may leadto marked quality-of-life improvements in neurological patients who cannot beadequately treated with medication (Diamond and Jankovic 2005). To date, manythousands of patients have benefited from DBS therapy and it is expected thatapplication of DBS in the treatment of movement disorders will grow further in theyears to come. These clinical successes in the neurological field have spurredinterest in application of DBS for treatment of other brain disorders, includingdrug-refractory psychiatric indications such as obsessive–compulsive disorder andclinical depression (Nuttin et al. 2003; Mayberg et al. 2005; Larson 2008). Giventhe high prevalence and long-term nature of psychiatric disorders, successfuldevelopment of DBS as a treatment option would have a potentially significanthealth-economic impact.

Despite its clinical successes, DBS technology is still in its infancy. Device- andprocedure-related adverse event rates are still relatively high in comparison withthose of other active implantable medical devices and there is a need for deviceimprovement. In a recent survey by US neurologists it was found that a prime reason

H. C. F. Martens (&) � M. M. J. DecréSapiens Steering Brain Stimulation B.V., High Tech Campus 41,5656 AE, Eindhoven, The Netherlandse-mail: hubert.martens@sapiensneuro.com

E. ToaderPhilips Research Laboratories, High Tech Campus 34,5656 AE, Eindhoven, The Netherlands

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_23, � Springer-Verlag Berlin Heidelberg 2012

253

for not referring patients for DBS was concern about the frequent occurrence ofadverse effects (Shih and Tarsy 2011). One of the issues with existing DBS systemsis stimulation-induced adverse events, in 15–30 % of patients (Burdick et al. 2010).Stimulation spreading outside intended target regions is generally considered acause of side effects. Indeed, computer simulations have demonstrated that slightmisplacements of DBS electrodes may easily lead to the unwanted stimulation ofstructures adjacent to the target area (McIntyre et al. 2004). With currently availablesystems that employ relatively large annular electrodes, this can only be preventedby lowering the stimulation amplitude. Thus, although DBS side effects arereversible, reducing them may only be possible at the cost of potentially reducedtherapeutic benefits. Therefore, in addition to novel clinical development, techno-logical advances are needed to further improve the therapy and make it available to alarger patient population. In particular, the development of DBS systems with moreand smaller electrodes would enable us to target stimulation more selectively—‘‘steering brain stimulation’’—with the expected additional benefit of making thetreatment less critically dependent on lead placement.

23.2 Electrode Designs for Future DBS Systems

Existing DBS electrodes are based on annular platinum–iridium rings assembledaround a flexible polymer carrier. For example, Medtronic DBS lead models 3389,3387, and 3391 (Coffey 2008), having electrode sizes of 1.5, 1.5, and 3.0 mm,respectively, and electrode pitches of 2.0, 3.0, and 7.0 mm, respectively, arecurrently available electrodes. These four-contact leads can address a relativelylarge volume of tissue in the brain, which provides some flexibility for the post-surgical optimization of the therapy. However, by design, the resolution attainablefor delivering stimulation is limited to several millimeters, which is not optimal inview of typical DBS target dimensions and stereotactic placement accuracies(Zylka et al. 1999; D’Haese et al. 2010).

23.2.1 Requirements for Next-Generation DBS Electrodes

In designing next-generation DBS electrodes, we can build on the experiencegained from two decades of clinical application of DBS and our improved theo-retical knowledge regarding targets and the primary mechanisms of action. It isplausible that in order to improve the efficacy of DBS and reduce DBS side effects,ideally, in addition to being safe and reliable, the next-generation DBS electrodesshould meet the following global characteristics:

1. Resolution sufficient to resolve the smallest clinically relevant target sizes(about 2 mm)

2. Ability to compensate for typical surgical accuracy limits (about 1 mm)3. Field-steering for shaping the stimulation to target anatomy (‘‘selectivity’’)

254 H. C. F. Martens et al.

In the following sections we will present various improved DBS electrode designsthat are currently in development stages and discuss how well they meet theserequirements.

23.2.2 Model-Based DBS Electrode Design

To move towards improved electrode designs, some basic understanding is requiredof the dependence between the geometry and the layout of electrodes, on the onehand, and the volume of activation (VOA) around such electrodes, on the otherhand. To this end, bioelectric computational models can be applied, which allow oneto study large parameter spaces and derive engineering rules. Basically, the mod-eling of VOAs consists of two steps. Firstly, one computes the three-dimensionaldistribution of the electrical field in the tissue surrounding the electrode for a givenstimulation setting by (numerically) solving the mathematical equations that governthe distribution of electrical currents. Commonly, this is done using finite-element-method computer modeling (Edsberg 2008). Secondly, the activation of neuronalelements within the electrical field is estimated. To this end, the ‘‘activation func-tion’’ (AF) is often computed (Rattay 1999; McIntyre et al. 2004); it is related to thesecond spatial derivative of the electrical potential. The AF gives a measure of thedepolarizing forces exerted on neuronal elements by the stimulation fields and givesan indication of the likeliness of activation of neuronal elements. More elaborateneurocomputational models can be used for the detailed study of activation effects(Miocinovic et al. 2006). The application of such bioelectric models generatesimportant insights into the direct effects of stimulation around a electrode—a fieldthat has been pioneered by Holsheimer and colleagues for optimization of spinalcord stimulation (Holsheimer and Wesselink 1997) and by McIntyre and colleaguesfor DBS (McIntyre et al. 2004; Miocinovic et al. 2006).

23.2.3 Conventional Annular DBS Electrodes

The volumes of activated tissue generated around the FDA-approved Medtronic3387 and 3389 annular electrode designs are more or less spherically shaped andcentered on the activated electrode; Butson and McIntyre (2006) investigated therole of the annular electrode geometry on the shape of the VOA and concluded thatthe aspect ratio (height divided by diameter) of the VOA scales with the electrode’saspect ratio, albeit not in a one-to-one manner. High electrode aspect ratios elicitedmore elongated VOAs, which may be attractive for certain anatomic targets. Westudied in detail the dependence of the VOA on the geometry (height and diameter)of annular electrodes (Fig. 23.1). Our investigations indicated the relation betweenelectrode geometry and the shape of the VOA. We quantified the aspect ratio of aVOA by dividing the height of the AF = 20 mV contour by its width. As expected,the electrode’s aspect ratio influences the shape of the VOA, and this might be usedto design electrodes specific for a certain target anatomy (Butson and McIntyre

23 Next-Generation Electrodes for Steering Brain Stimulation 255

2006). However, the effect reduces with increasing stimulus intensity, and veryextreme (and unpractical) electrode geometries are needed to significantly impactthe VOA. In short, with such single-ring or few-ring electrode designs the issue ofresolution and placement accuracy remains unsolved.

Given these observations, it appears to be a logical step to move to multi-electrode designs, which should provide the user with additional flexibility in bothshaping and positioning the VOA axially along the lead. We modeled a 12-ringDBS lead employing multiple annular contacts of low aspect ratio and which areclosely placed together to enable smooth axial control over VOAs. Our simula-tions indicate that such leads would allow finer axial positioning of the VOA(Fig. 23.2), with an accuracy that directly corresponds to the pitch of individualring electrodes. Such electrode designs could be useful for finely adapting the axialposition of the VOA, for example, for correction after brain-shift recovery.

Using annular DBS electrodes as discussed in this section has the primaryadvantage for device manufacturers that conventional lead fabrication techniquescan be employed. Multielectrodes can be fabricated by mechanically assembling

Fig. 23.1 Influence of electrode shape on volume of activated tissue (VOA). Top The distributionof the activation function (AF) for two extreme electrode geometries; the stimulation amplitude is-3.6 V in both cases and the contour line shows the AF = 20 mV boundaries. A small electrodeheight (aspect ratio 1:4) leads to a low aspect ratio of the VOA, whereas a long electrode (aspectratio 2) translates into more elongated VOAs. Bottom The relation between the aspect ratio of theVOAs and those of the electrode for various electrode diameters for two stimulus intensities(1 and 3 mA)

256 H. C. F. Martens et al.

multiple rings in a row on a carrier. The resolution is essentially limited byhandling constraints of the individual contacts giving a minimum electrode sizeand by the mechanical assembly accuracy, which limits the attainable pitch.However, the user’s benefit from such an electrode design is limited to onlyadditional axial control over VOAs; correction for lead misplacements and shapingthe stimulation to target areas are not achievable.

23.2.4 Electrodes Capable of Steering Stimulation

Clearly, innovative electrode designs are required to provide the true steeringfunctionality needed for correction of lead misplacements and shaping the stim-ulation to target anatomy. Generally speaking, two main approaches are beingpursued in the neural-engineering field: (1) segmented DBS electrodes based onmechanically assembled leads (Hegland 2010) and (2) DBS electrode arrays(Martens et al. 2011) based on microfabrication technologies and lithographicpatterning. The images on the right in Fig. 23.3 show illustrative examples of bothlead technologies. Pushing mechanical lead assembly techniques, individualelectrode contacts can be combined into segmented ring electrodes (Hegland2010). Mechanical integrity requirements, however, likely limit this kind ofarrangement to relatively large segments (two or three segments per ring) and alsoconstrain the total number of electrodes that can be combined. Using lithographicpatterning techniques derived from semiconductor and display fabrication methodsenables the realization of DBS leads carrying precisely defined electrode arrays.The resolution of these techniques is more than sufficient for the needs in clinicalneurostimulation and goes far beyond what is possible with mechanical assemblyof discrete elements. A first version employs a 64-electrode-array having a 0.75-mm axial pitch and 45o circumferential resolution owing to its staggered electrodearrangement (see Fig. 23.3, plot d, image h). Computer simulations (not shown)

radius [mm]

2

-2

0

-2 20

150

100

50

0

-50

-100

-150

AF (mV)

radius [mm]

2

-2

0

-2 20radius [mm]

heig

ht [m

m]

heig

ht [m

m]

heig

ht [m

m]2

-2

0

-2 20

Fig. 23.2 AF profiles around a 12-ring annular electrode. The stimulation amplitude is -3.6 Vin all cases. The contour line corresponds to the AF = 20 mV boundary. The left plot shows theactivation of electrodes 4–7 and the middle plot shows the activation of electrodes 6–9. The rightplot displays the pattern for activation of electrodes 4–9. As demonstrated by the horizontaldashed lines, the activation profile can be shifted axially with a resolution corresponding to theelectrode pitch and its aspect ratio can be controlled by grouping adjacent electrode contacts

23 Next-Generation Electrodes for Steering Brain Stimulation 257

demonstrated that the array-based DBS electrode design is capable of displacingstimulation fields over a distance of 1–2 mm from the lead’s central axis (Martenset al. 2011), which is sufficient to compensate for typical stereotactic targetingaccuracy limitations encountered during DBS electrode implantations (Zylka et al.1999; D’Haese et al. 2010).

We further evaluated the steering capacity of both designs by means of acomputational analysis. We propose that the benefit of steering can be evaluatedtheoretically by both the target coverage, i.e., the fraction of the target area that lieswithin the VOA, and the target selectivity, i.e., the fraction of the VOA that is effectivein covering intended target tissue and for which stimulation does not leak outside toadjacent structures. Let us briefly illustrate the meaning of these two parameters: a hightarget coverage combined with low target selectivity implies that despite good overlapof stimulation with the intended target region, a large fraction of stimulation leaksoutside the target to adjacent structures, thereby potentially inducing side effects.Ideally, both target coverage and target selectivity should be maximized for the optimaltherapeutic benefit and minimized risk of stimulation side effects.

We simulated field distributions for both lead designs while incrementallyactivating more electrodes. The theoretically achievable selectivity of steering wasevaluated by assuming that beneficial effects and side effects related to DBSresided in different ‘‘quadrants’’ around the lead. In all field plots in Fig. 23.3, thetop-left quadrant, indicated by green, corresponds to a localized target area,whereas the remaining three quadrants (in orange) illustrate adjacent areas that arelikely to cause effects if stimulated at a sufficiently high level. It is important torealize that a lead cannot be rotated physically once it has been implanted, so theorientation of the electrodes with respect to the target is fixed once they have beenimplanted. A simple comparative analysis demonstrates that the segmented lead isless able to selectively stimulate the desired target area of the top-left quadrantwithout also stimulating adjacent areas. Whereas the theoretical selectivity of thesegmented lead drops when aiming for better target coverage, the high-resolutionDBS array can achieve both high target selectivity and high target coverage.

23.3 Discussion

After more than two decades of clinical application of chronic DBS, there isabundant evidence that for well-selected patients with movement disorders, DBSprovides a powerful treatment option. More recently, clinical research into the useof DBS for treatment of severe psychiatric diseases has yielded promising results(Nuttin et al. 2003; Mayberg et al. 2005; Larson 2008), and it can be hoped that fora subset of these patients in DBS may provide equally good levels of symptomrelief. Despite these clear successes, the technology is still in its infancy as isreflected among other factors in the relatively high rates of device-related andstimulation-related adverse effects associated with DBS today. The high rate ofstimulation-induced adverse effects reported in the literature (Benabid et al. 2009;

258 H. C. F. Martens et al.

Burdick et al. 2010) may be attributable in part to the fact that existing DBSelectrodes lack the precision to accurately stimulate the tiny target regions ofinterest (McIntyre et al. 2004). As a result, stimulation may leak outside intoadjacent structures, where it induces unwanted effects (Benabid et al. 2009).Generally, stimulation-induced side effects are reversible, and if the stimulationintensity in nontarget areas is reduced, such side effects could be prevented.

It is intuitively understood that the use of smaller DBS electrodes would enablemore precise delivery of stimulation to target areas. We used computational modelsto study several improved designs—(1) multiring electrodes, (2) segmented rings,

50%

Field:

Configuration:

Target Coverage:Target Selectivity: 25%37%

Field:

Configuration:

Target Coverage:Target Selectivity:

25% 71%58%

64 electrode array

2x annular electrodes

2x3 segmented electrodes

50% 56%62%

25% 61%61%

(a)

(e) (f) (g) (h)

(b) (c) (d)

Fig. 23.3 Two DBS lead technologies with steering capability and evaluation of the steeringefficacy for both designs. The plots a–c correspond to a design based on mechanically assembledelectrode segments shown on the right (d). Plots e–g correspond to a lead design that makes useof microfabrication techniques where an accurately defined electrode array is provided around thefull circumference of the lead’s distal end (h). Plots a–c and e–g display cross-sectional views ofsimulated stimulation field shapes and demonstrate the steering functionality of the novel leaddesigns. The black circle in the center represents the lead body, and the blue outline depicts thefield of stimulation. All simulations were conducted assuming constant voltage input. Greenregions in the top-left quadrants correspond to target areas and the orange regions depictstructures that may lead to side effects. The corresponding electrode configurations and resultingtheoretical target coverage and target selectivity are depicted below each field-distribution plot.Plots a–c demonstrate the simulated field distributions for the segmented lead for incrementalactivation of individual electrodes. Plots e–g demonstrate the computed field distributions for the64-electrode lead. Clearly, the higher-resolution lead is able to provide superior target selectivityat the same or higher level of target coverage

23 Next-Generation Electrodes for Steering Brain Stimulation 259

and (3) high-resolution electrode arrays—and introduced the two parameters targetcoverage (quantifying the theoretical beneficial effect) and target selectivity(quantifying the theoretical risk of side effects), which allow one to theoreticallyevaluate the benefits of steering brain stimulation. Our computational analysesconfirm the ability of these improved designs to provide stimulation more accuratelyto target regions as compared with conventional DBS leads.

Table 23.1 summarizes the performance of the three technologies reviewedhere with respect to the requirements that are believed to enable better therapeuticefficacy: resolution sufficient to address the smallest targets, ability to correct forslight misplacements, and field-steering function to selectively shape stimulationvolumes to target regions. Our computational results suggest that course electrodedesigns will always require a careful balancing of target coverage versus targetselectivity. However, owing to the fine distribution of electrodes, our analysisindicates that the high-resolution electrode array is able to simultaneously opti-mize target selectivity and target coverage. Thus, we believe that the requirementsfor next-generation DBS systems can only be achieved through the use of novelmicrofabrication technologies enabling the fabrication of DBS electrode arrays.

The first prototypes using novel microfabrication lead technology have beenrealized recently and were evaluated during acute implantation in a nonhumanprimate. Indeed, it was demonstrated that side-effect thresholds are highly depen-dent on the stimulation direction, indicating the need for steering (Martens et al.2011). Notwithstanding these initial findings supporting the superior selectivestimulation ability of these new electrode designs, their actual therapeutic benefithas yet to be demonstrated clinically. Clearly, lead technology is progressing and itis expected that studies on the benefits of new DBS systems will be able to start soon.

23.4 Conclusions

Several next-generation DBS lead technologies are in development. These all havein common that they employ more and smaller electrodes providing more degreesof freedom in DBS programming. Ultimately, this will allow clinicians to moreaccurately and selectively target stimulation of the brain, which could translateinto better clinical outcomes.

Table 23.1 Overview of the performance of potential next-generation DBS electrode technologywith respect to requirements

DBS technology Requirement

Resolution Placement correction Selectivity

Multiring Yes No NoSegmented No Yes NoElectrode array Yes Yes Yes

260 H. C. F. Martens et al.

References

Benabid AL, Pollak P et al (1991) Long-term suppression of tremor by chronic stimulation of theventral intermediate thalamic nucleus. The Lancet 337(8738):403–406

Benabid AL, Chabardes S et al (2009) Deep brain stimulation of the subthalamic nucleus for thetreatment of Parkinson’s disease. Lancet Neurol 8:67–81

Burdick AP, Fernandez HH et al (2010) Relationship between higher rates of adverse events indeep brain stimulation using standardized prospective recording and patient outcomes.Neurosurg Focus 29(2):E4

Butson CR, McIntyre CC (2006) Role of electrode design on the volume of tissue activatedduring deep brain stimulation. J Neural Eng 3:1–8

Coffey R (2008) Deep brain stimulation devices: a brief technical history and review. ArtifOrgans 33(3):208–220

D’Haese PF, Pallavaram S et al (2010) Clinical accuracy of a customized stereotactic platform fordeep brain stimulation after accounting for brain shift. Stereotact Funct Neurosurg88(2):81–87

Diamond A, Jankovic J (2005) The effect of deep brain stimulation on quality of life in movementdisorders. J Neurol Neurosurg Psychiatr 76(9):1188–1193

Edsberg L (2008) The finite element method. Introduction to computation and modeling fordifferential equations. Wiley-Interscience, Hoboken, pp 140–146

Hegland M (2010) Implantable medical lead with multiple electrode configurations. Fridley,Medtronic

Holsheimer J, Wesselink WA (1997) Optimum electrode geometry for spinal cord stimulation:the narrow bipole and tripole. Med Biol Eng Comput 35:493–497

Krauss JK, Pohle T et al (1999) Bilateral stimulation of globus pallidus internus for treatment ofcervical dystonia. Lancet 354(9181):837–838

Kumar R, Lozano AM et al (1998) Pallidotomy and deep brain stimulation of the pallidum andsubthalamic nucleus in advanced Parkinson’s disease. Mov Disord 13(S1):73–82

Larson PS (2008) Deep brain stimulation for psychiatric disorders. Neurotherapeutics 5(1):50–58Martens HCF, Toader E et al (2011) Spatial steering of deep brain stimulation volumes using a

novel lead design. Clin Neurophysiol 211:558–566Mayberg HS, Lozano AM et al (2005) Deep brain stimulation for treatment-resistant depression.

Neuron 45(5):651–660McIntyre CC, Mori S et al (2004) Electric field and stimulating influence generated by deep brain

stimulation of the subthalamic nucleus. Clin Neurophysiol 115(3):589–595Miocinovic S, Parent M et al (2006) Computational analysis of subthalamic nucleus and

lenticular fasciculus activation during therapeutic deep brain stimulation. J Neurophysiol96(3):1569–1580

Nuttin B, Gabriels LA et al (2003) Long-term electrical capsular stimulation in patients withobsessive–compulsive disorder. Neurosurgery 52:1263–1274

Rattay F (1999) The basic mechanism for the electrical stimulation of the nervous system.Neuroscience 89(2):335–346

Shih LC, Tarsy D. Survey of U.S. neurologists’ attitudes towards deep brain stimulation forParkinson’s disease. Neuromodulation Technol Neural Interface 14(3):208–213

Zylka W, Sabczynski J et al (1999) A Gaussian approach for the calculation of the accuracy ofstereotactic frame systems. Med Phys 26(3):381–391

23 Next-Generation Electrodes for Steering Brain Stimulation 261

Chapter 24Future Applications: Nanotechniques

Russell J. Andrews, Jessica E. Koehne and Meyya Meyyappan

24.1 Introduction

The brain’s cells—neurons and glia—communicate at the micron to submicronlevel. This communication is both electrical and chemical. As deep brain stimu-lation (DBS) becomes more refined, it will become increasingly important to havetechniques for interacting with the brain on a more precise and efficacious elec-trical and chemical (neurotransmitter) basis. Reducing the size of electrodes to thatof the brain’s cells will allow precision electrical recording and stimulation. It isalso possible to enhance greatly the efficacy of electrical stimulation and recording(i.e., improved charge transfer) by reducing the electrode size. Nanoelectrodes arealso candidates to improve neurotransmitter recording—raising the possibility ofmultifunction DBS electrodes that can monitor both electrical and chemicalactivity as well as stimulate electrically—all with cellular-level precision.

24.2 Nanoscale Advantages for DBS

Reducing the size of electrodes (radius r) from macro (1 mm or more) or micro(100 lm or more) to nano (less than 1 lm) sizes results in dramatic (orders ofmagnitude) improvement in spatial resolution, temporal resolution, and sensitivity(signal-to-noise ratio):

1. Spatial resolution: defined by r2. Temporal resolution: cell time constant t = RuCd = rCd

0/4k

R. J. Andrews (&) � J. E. Koehne � M. MeyyappanCenter for Nanotechnology, NASA Ames Research Center,Moffett Field, CA 94035, USAe-mail: rja@russelljandrews.org

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_24, � Springer-Verlag Berlin Heidelberg 2012

263

3. Sensitivity: signal-to-noise ratio is/in � nFC0D0/r

The ability to interact with the brain depends on the bioimpedance of thedevice—which in turn depends on the characteristics of the brain tissue (whichcannot be altered) and the characteristics of the electrode (which can be altered).A major issue in DBS, limiting its safety and efficacy, is the electrolysis of waterand the resulting pH change, which is toxic to neural tissue. Charge transferimproves when the impedance is decreased and the capacitance is increased.Appropriately configured nanoscale electrodes can decrease impedance (measuredby electrochemical impedance spectroscopy) and increase capacitance (measuredby cyclic voltammetry) by orders of magnitude over traditional metal electrodes(e.g., platinum, tungsten).

Nanoelectrodes, by virtue of their increased signal-to-noise ratio when appro-priately configured, offer the possibility of improving the detection of neuro-transmitters over the standard achieved with carbon fiber microelectrodes (CFMs;as described by Abulseoud et al., this volume).

24.3 Nanoelectrodes for DBS: Fabrication

Carbon nanotubes (CNTs) were first described by Iijima (1991) 20 years ago. ForDBS electrodes, the very similar carbon nanofibers (CNFs) are more appropriatesince they are easier to fabricate in a vertically aligned fashion (Cruden et al.2003). It has been shown that coating CNF nanoelectrode arrays with polypyrrole(PPy) decreases impedance and increases capacitance by several orders ofmagnitude over uncoated CNF (or noble metal) electrodes (Nguyen-Vu et al.2006). The PPy coating also prevents the irreversible clumping which occurs whenuncoated CNFs are exposed to biological fluids. To sustain healthy networks ofPC12 neurons (useful for neurotransmitter analysis in Parkinson’s disease researchsince PC12 cells can produce dopamine under appropriate conditions), CNFelectrodes were found to require treating not only with PPy but also with collagentype IV and nerve growth factor (Nguyen-Vu et al. 2007).

Scanning electron microscope (SEM) images of CNF electrodes not coated withPPy on which a network of PC12 cells has been grown are presented at the top ofFig. 24.1 (Nguyen-Vu et al. 2007). The clumping of the CNFs into teepee-likestructures is apparent, and the two enlargements show the extensive neural fibrils(thought to be a response of the PC12 cells to stress) which appear when the PC12cells are grown on CNFs not coated with PPy. SEM images of PPy-coated CNFarrays on which a network of PC12 cells has been grown are shown in the middleof Fig. 24.1. The individual (as opposed to clumped) structure of the CNFs isapparent, and the neural fibrils are not seen (likely owing to less stress on the PC12cell network when grown on the PPy-coated, unclumped CNFs). A high-magni-fication SEM image of the structural interaction between a PC12 cell and the PPy-coated CNFs is presented at the bottom of Fig. 24.1. The PPy-coated CNFs

264 R. J. Andrews et al.

maintain vertical alignment yet are sufficiently flexible to bend under the weight ofthe PC12 cell network. Individual CNFs may penetrate the PC12 cell membrane(raising the possibility of intracellular recording/stimulation).

24.4 Nanoelectrodes for DBS: Results

It was noted (Sect. 24.3) that in vitro studies have demonstrated markedlyimproved charge transfer characteristics for PPy-coated CNF nanoelectrodes incomparison with standard metal microelectrodes. Two reports with potentialclinical implications illustrate the benefits of nanoelectrode techniques.

In an elegant series of experiments, standard metal microelectrodes (stainlesssteel, tungsten, and indium tin oxide) were coated with CNTs (with or withoutfurther coating with PPy) and compared with the uncoated electrodes both in vitroand in vivo (Keefer et al. 2008). The impedance was decreased, and the chargetransfer increased, more than tenfold and 40-fold, respectively, for the CNT-coatedelectrodes versus the uncoated electrodes. Electrodes coated with PPy-coatedCNTs resulted in a 1,600-fold increase in charge transfer. The CNT-coated elec-trodes and the uncoated electrodes were compared in vivo in recordings of ratmotor cortex and primate visual cortex. Over all frequencies evaluated (rat1–1,000 Hz; primate 1–300 Hz), the CNT-coated electrodes showed a 7.4–15.5-dB power increase (depending upon the animal model and frequency band) onpower spectral density analysis in comparison with the uncoated electrodes.

The benefits of PPy-coated CNF nanoelectrodes for stimulation/recording in therat hippocampal brain slice have been demonstrated recently (de Asis et al. 2009).Stimulation of a Schaffer collateral and recording from the striatum pyramidale ofthe CA1 region was compared for tungsten wire electrodes, platinum microelec-trode arrays, CNF electrodes, and PPy-coated CNF electrodes (Fig. 24.2). Theexperimental setup is shown at the top of Fig. 24.2; relevant results are given in atthe bottom of Fig. 24.2. From the lower graphs at the bottom of Fig. 24.2, it isapparent that only the PPy-coated CNF electrode was able to stimulate the rathippocampal slice at a current pulse less than 1 mA and an electrode voltage lessthan 1 V (levels that are sufficiently low to preclude the electrolysis of water).Thus, only the PPy-coated CNF electrode is able to safely stimulate the rat hip-pocampus from the Schaffer collateral region to the striatum pyramidale region.

Incorporation of PPy-coated nanoelectrodes into DBS devices will permitgreater sensitivity in recording brain electrical activity and greater safety andefficacy in stimulating brain tissue. Additionally, the greater precision of micron-and nano-sized electrodes will present the opportunity to ‘‘sculpt’’ DBS in waysimpossible with the macroelectrodes presently in use (or even the presentlyavailable microelectrodes).

24 Future Applications: Nanotechniques 265

266 R. J. Andrews et al.

24.5 Nanoelectrodes for Neurotransmitter Recording

The regions of the brain communicate using both electrical (axon and dendrite)and chemical (synapse) techniques. DBS and other forms of neuromodulation haverelied primarily on interacting with the brain electrically (hence the neural—electrical interface). Many brain disorders, such as Parkinson’s disease and majordepression, appear to be primarily disorders of neurotransmitters rather than dis-orders of electrical activity. Moreover, since neurons constitute only 10 % of thebrain, it is reasonable to expect the major component of the brain—glial cells—toplay a significant role in many if not most brain disorders (Ni et al. 2007). Thedevelopment of real-time telemetric in vivo monitoring of neurotransmitters hascreated the ability to add neurotransmitter monitoring to brain electrical moni-toring and modulation (stimulation) for DBS (see Abulseoud et al., this volume).

The standard technique for real-time neurotransmitter detection and monitoringis fast-scan cyclic voltammetry using CFMs (see Abulseoud et al., this volume).However, the same principles for using nanoarrays to enhance electrical recordingand stimulation noted in Sect. 24.2 apply to the detection and monitoring ofneurotransmitters. Unpublished observations by the NASA Ames NanotechnologyGroup more than 5 years ago showed that a CNF electrode can detect the neu-rotransmitter dopamine with a faster response time and a lower detection thresholdthan a CFM.

Recently, a collaboration between the Mayo Clinic (Rochester, MN, USA)neurosurgery and bioengineering departments and the NASA Ames Center forNanotechnology (Moffett Field, CA, USA) has evolved to adapt nanoelectrodes forenhanced monitoring of neurotransmitters such as dopamine, adenosine, glutamate,and serotonin. A 3 9 3 nanoelectrode array similar to that described in Sect. 24.4has been compared with a CFM for the fast-scan cyclic voltammetry detection ofdopamine using the Wireless Instantaneous Neurotransmitter Concentration System(Koehne et al. 2011). The surface areas of the CFM and the CNF electrode were

Fig. 24.1 Top Scanning electron microscope (SEM) images of a network of PC12 neurons oncollapsed microbundles of uncoated carbon nanofibers (CNFs). Capillary forces duringpreparation prior to cell culture irreversibly pull the CNFs into microbundles. Upper imagePC12 cells form an extensive neural network on the uncoated CNF electrode. Lower imagesHigher magnification reveals the neural fibril growth (believed to be a stress response)particularly at the points where a neurite anchors to a CNF microbundle (left); neural fibrilsbridge the submicron-diameter neurites (right). Note the similarity in diameter of the neuralfibrils and the CNFs. Middle: SEM images of a network of PC12 neurons on polypyrrole (PPy)-coated CNFs. Note the lack of neural fibrils seen with PC12 networks on uncoated CNFs. LeftPC12 cells develop neural extensions when nerve growth factor (NGF) is added to the growthmedium. The inset shows PC12 cells under similar conditions without NGF. Right Highermagnification reveals the PC12 neurites settle on the PPy-coated CNF electrode without thestress-response neural fibrils seen with the uncoated CNF electrode. Bottom PPy-coated CNFs aresufficiently rigid to maintain vertical alignment but are able to bend under the force exerted by thePC12 cell body. It appears that some CNFs may penetrate the cell membrane (from Nguyen-Vuet al. 2007 with permission)

b

24 Future Applications: Nanotechniques 267

268 R. J. Andrews et al.

made as equivalent as possible and characterized using scanning electron micros-copy and atomic force microscopy. The results are presented in Fig. 24.3. The CNFelectrode is at least as effective as the CFM in detecting dopamine. However, theCNF electrode can be readily fabricated into a multiplexed array to permit multipleneurotransmitter monitoring sites within the brain. The benefits of CNFs in terms ofboth understanding brain disorders involving neurotransmitters and improving theclinical efficacy of DBS should be substantial.

24.6 Conclusions and Future Nanotechniques for DBS

In addition to the nanotechniques described herein for improved brain electricaland chemical monitoring and modulation, several other nanotechniques couldenhance DBS in the future. A particular problem for DBS is the invasive nature ofthe electrode and pulse generator (battery plus microprocessor) implantation—thisis a major source of morbidity both in terms of infection (given the large size andlong subcutaneous course of the connecting leads and pulse generator/battery) andintracranial hemorrhage from electrode placement through the brain. This willbecome an even greater problem as smaller and more precise DBS devices (asdescribed earlier) raise the feasibility of multiple DBS devices being implanted inseveral regions of the brain to enhance efficacy.

Great advances have been made over the past several decades by interventionalneuroradiologists in placing catheters via the arterial system into increasinglysmall blood vessels throughout the brain. Vascular malformations and tumorslocated very distal to the major arteries of the circle of Willis, which were pre-viously inaccessible, are now catheterized routinely. Indeed, the brain’s capillarysystem is a ‘‘highway system’’ reaching even the most remote or deep-seatedregion of the brain—because all brain tissue needs the oxygen and glucose that themicron-diameter capillaries provide.

Fig. 24.2 Top Setup for stimulating rat hippocampal slices. Upper row 3 9 3 nanoelectrodearray, with electrode numbering convention on the right. Stimulus is applied between electrodes 8and 9. Lower row Rat hippocampal slice with tungsten electrode stimulation site (Schaffercollateral—cross) and recording site (striatum pyramidale—filled circle) (left); rat hippocampalslice on an array with stimulation between electrodes 8 and 9 (Schaffer collateral region) andrecording site as with tungsten electrode stimulation (striatum pyramidale—filled circle) (right).DG dentate gyrus, SC Schaffer collateral, CA1 cornus ammonis 1, CA3 cornus ammonis 3, SPstriatum pyramidale, SR striatum radiatum, SLM striatum lacunosum moleculare. Bottom:Electrode voltage (a) and response amplitude for field potential (b) versus stimulation current,and enlarged plots at low currents (c and d, respectively). Error bars ± standard deviation, filleddiamonds PPy-coated CNF electrode (in b and d the amplitude of the short-duration fieldpotential), crosses PPy-coated CNF electrode amplitude of the long-duration field potential (inb and d), filled circles tungsten electrodes, filled squares uncoated CNF electrode, filled trianglesplatinum array (from de Asis et al. 2009 with permission)

b

24 Future Applications: Nanotechniques 269

It has been shown that electrodes within blood vessels supplying nervous systemtissue can record and stimulate electrical activity as effectively as electrodes in theparenchyma adjacent to the blood vessels (Llinas et al. 2005). This opens up thebrain’s capillary system as a minimally invasive route for the precise placement ofDBS electrodes (dozens or hundreds, if desired) anywhere in the brain. As electrodesare reduced in size to the micron or submicron level, the risk of bleeding fromcapillary wall puncture disappears (since red blood cells are greater than 5 lm in

Fig. 24.3 Dopamine detection by a carbon fiber microelectrode (CFM) (left) and a carbonnanofiber (CNF) electrode (right) using the Wireless Instantaneous Neurotransmitter Concen-tration System. a, b Three-dimensional color plots for a 2.5 lM dopamine injection. c,d Background-subtracted (BGS) cyclic voltammograms (CV) for a 2.5 lM dopamine injection. e,f Measured current densities at various dopamine concentrations (from Koehne et al. 2011 withpermission)

270 R. J. Andrews et al.

diameter). Thus the nano-sized electrodes can be pushed through the capillary wallto sample, e.g., neurotransmitter levels in the brain parenchyma (Kendall Lee,personal communication, 2010).

Another nanotechnique for DBS in the future could obviate the need for asubcutaneous battery altogether. As the power needed to drive a DBS devicedecreases—thanks to improved charge transfer as described earlier plus moreefficient stimulation protocols using computational analysis for ‘‘antikindling’’(Hauptmann et al. 2007)—the battery becomes less massive. A recent articledescribed the development of zinc oxide (ZnO) nanowire arrays that function as apiezoelectric nanogenerator (Xu et al. 2010). Similar to photovoltaic panel arraysto generate substantial electricity for residential or industrial use, arrays of ZnOnanowires can utilize the pressure changes within the intracranial compartment(e.g., blood vessel pulsations, brain movement with body movement) to generateelectricity to drive a nanoarray DBS device.

The incorporation of nanotechniques at various levels of DBS—from nanoac-cess via the brain capillaries, to nanoarrays to enhance precision and efficacy ofbrain electrical and chemical monitoring and modulation, to nanogenerators for aself-contained DBS power supply—will radically change the way DBS is per-formed. Given the wide variety of disorders in psychiatry for which effectivepharmacologic or other treatments are not available for many patients, the pros-pects for DBS in psychiatry are much brighter than they would be without suchnanotechniques.

References

Cruden BA, Cassell AM, Ye Q, Meyyappan M (2003) Reactor design considerations in the hotfilament/direct current plasma synthesis of carbon nanofibers. J Appl Phys 94:4070–4078

de Asis ED, Nguyen-Vu TDB, Arumugam PU, Chen H, Cassell AM, Andrews RJ, Yang CY, Li J(2009) High efficient electrical stimulation of hippocampal slices with vertically alignedcarbon nanofiber microbrush array. Biomed Microdevices 11:801–808

Hauptmann C, Popovych O, Tass PA (2007) Desynchronizing the abnormally synchronizedneural activity in the subthalamic nucleus: a modeling study. Expert Rev Med Devices4:633–650

Iijima S (1991) Helical microtubules of graphitic carbon. Nature 354:56–57Keefer EW, Botterman BR, Romero MI, Rossi AF, Gross GW (2008) Carbon nanotube coating

improves neuronal recordings. Nat Nanotechnol 3:434–439Koehne JE, Marsh M, Boakye A et al (2011) Carbon nanofiber electrode array for

electrochemical detection of dopamine using fast scan cyclic voltammetry. Analyst136:1802–1805

Llinas RR, Walton KD, Nakao M, Hunter I, Anquetil PA (2005) Neuro-vascular central nervousrecording/stimulating system: using nanotechnology probes. J Nanopart Res 7:111–127

Nguyen-Vu TDB, Chen H, Cassell AM, Andrews R, Meyyappan M, Li J (2006) Verticallyaligned carbon nanofiber arrays: an advance toward electrical-neural interfaces. Small2:89–94

24 Future Applications: Nanotechniques 271

Nguyen-Vu TDB, Chen H, Cassell AM, Andrews RJ, Meyyappan M, Li J (2007) Verticallyaligned carbon nanofiber architecture as a multifunctional 3-D neural electrical interface.IEEE Trans Biomed Eng 54:1121–1128

Ni Y, Malarkey EB, Parpura V (2007) Vesicular release of glutamate mediates bidirectionalsignaling between astrocytes and neurons. J Neurochem 103:1273–1284

Xu S, Qin Y, Xu C et al (2010) Self-powered nanowire devices. Nat Nanotech 5:366–373

272 R. J. Andrews et al.

Chapter 25Ethical Guidance for the Use of DeepBrain Stimulation in Psychiatric Trialsand Emerging Uses: Reviewand Reflections

Emily Bell and Eric Racine

25.1 Introduction

There are important reasons to consider ethical and social issues in deep brainstimulation (DBS) for the treatment of psychiatric disorders. Promising researchdemonstrates the potential for DBS to improve the symptoms of some patients withpsychiatric illnesses. But there is a need to proactively assess the ethical landscapeand values reflected within current practices, to address ethical and social challenges,and to apply evidence-based ethics practices in guiding new research. Even in theapproved uses of DBS in movement disorders such as Parkinson’s disease (which aremuch more widely practiced than psychiatric DBS) practitioners face challengesrelated to resource allocation, appropriate patient selection, and the management ofpsychosocial factors before, during, and after DBS (Bell et al. 2011). In this chapter,we review and reflect on existing guidance for the use of DBS in psychiatric trials.First, we briefly review the scope of the academic discussion regarding ethical andsocial challenges in psychiatric DBS. Second, we examine the practical ethicalguidance that exists on this topic that is available to practitioners and researchers.Third, we step back and discuss justifications for the emergence of specific ethicalguidance for DBS, gaps in current ethical guidance, and suggest ways to support thetranslation of ethical deliberation and scholarship into practice.

E. Bell (&) � E. RacineNeuroethics Research Unit, Institut de Recherches Cliniques de Montréal,110 Avenue des Pins Ouest, Montreal, QC H2W 1R7, Canadae-mail: emily.bell@ircm.qc.ca

E. RacineUniversité de Montréal, 2900 Boul Édouard-Montpetit, Montreal,QC H3T 1J4, Canada

E. RacineMcGill University, 845 Sherbrooke St, Montreal, QC H3A0G4, Canada

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_25, � Springer-Verlag Berlin Heidelberg 2012

273

25.2 Ethical and Social Issues Raised by the Use of DBSin Psychiatric Conditions

DBS is distinct in many ways from both historically undesirable interventions suchas lobotomy and more recently accepted ablative neurosurgical approaches ofpsychosurgery. The ethical concerns of this new technology are still beingunderstood and appreciated by scholars across disciplines and through differentcollaborative efforts. Venturing to acknowledge and attend to the ethical and socialissues raised by neurosurgical approaches for treating psychiatric illness is notnew. Prompted by earlier negative public opinion and invasive psychosurgeryapproaches such as lobotomy, the National Commission for the Protection ofHuman Subjects in Biomedical and Behavioral Research issued a report in 1977 onethics in the practice and research of psychosurgery (National Commission for theProtection of Human Subjects in Biomedical and Behavioral Research 1977). Thisreport and the Commission’s subsequent recommendations that, in some cases,psychosurgery is ethically and scientifically appropriate when governed by a set ofstrict ethical standards, established that there was both public and academic desireto attend to specific ethical concerns brought about by neurosurgical interventionin the most vulnerable psychiatric patients. Almost 30 years later, encouragingevidence from small clinical trials of DBS in obsessive–compulsive disorder(OCD) and refractory depression have led to new, and sometimes renewed, dis-cussions of the ethical implications of neurosurgical interventions in psychiatricdisorders. Indeed, a wide range of ethical and social issues have been raised byexperts in neurosurgery, psychiatry, philosophy, and bioethics; and a number ofmultidisciplinary research groups and collaborative efforts have set out to tacklethese issues, through empirical research and interdisciplinary discussion (e.g., theEuropean Academy research group Deep Brain Stimulation in Psychiatry: Guid-ance for Responsible Research and Application; the Canadian Institutes of HealthResearch funded States of Mind Network; the Berman Institute of Bioethics Ethicsin Brain Science program; the Cleveland Clinic NeuroEthics program). The scopeof the ethical issues in DBS for the treatment of psychiatric disorders is broad (Bellet al. 2009). In psychiatric DBS, there is literature discussing the potentialimplications, good or bad, for personality, and personal and narrative identity(Hildt 2006; Synofzik and Schlaepfer 2008; Schechtman 2010) as well as forautonomy, decisional capacity, and informed consent (Glannon 2008; Dunn et al.2011). Synofzik and Schlaepfer (2011) have also discussed the application of thestandard ethical principles of beneficence, nonmaleficence, and autonomy inguiding research and care for psychiatric DBS patients (Synofzik and Schlaepfer2011). This is not an exhaustive list of the wealth of ethical discussion on the topicof ethics and psychiatric DBS, and many of these authors and others have sug-gested practical solutions to protect research subjects and patients, and for con-ducting ethically sound clinical trials of DBS in psychiatric disorders. Recently,more specific concerns have emerged related to publication bias (Schlaepfer andFins 2010) and application of the FDA humanitarian device exemption in DBS for

274 E. Bell and E. Racine

the treatment of OCD (Fins et al. 2011). In addition, evidence of the modulation ofmood using DBS has sprouted discussion about the ethical criteria for a potentialuse in healthy human enhancement (Synofzik and Schlaepfer 2008), although arecent qualitative study reveals that neurosurgeons themselves may find the use ofDBS for the treatment of nonpathological traits unwarranted (Mendelsohn et al.2010). Other empirical investigations have been published or are under way,including a recent examination of the perspectives of North American functionalneurosurgeons on psychiatric surgery, in which most of the participants demon-strated a positive attitude towards neurosurgery for the treatment of psychiatricdisorders provided that there is ethical and psychiatric oversight (Lipsman et al.2011). Our own qualitative research on the perspectives of Canadian healthcareproviders working in DBS has demonstrated that resource allocation, the trainingof qualified personnel, and the psychosocial context of mental illness are importantfactors which are not well characterized but may play an important role in theethical landscape of psychiatric DBS (Bell et al. 2011).

25.3 Reviewing ethical guidance for DBS in PsychiatricConditions

As witnessed in the previous section, there are many potential ethical and socialissues that have been raised in the academic literature with regard to the extensionof DBS in psychiatric disorders. In some cases, suggestions have been formulatedto tackle the ethical challenges. Although we acknowledge this to be the case, weintend to focus in this review on exploring the explicit ethical guidance that isavailable in the academic literature for practitioners and researchers as it relates toDBS in psychiatric disorders. We have chosen to emphasize explicit ethicalguidance because we feel that this type of guidance, appearing generally as lists ofcriteria or tables detailing guidance, is most accessible, offers the clearest per-spectives on how to proceed, and can be most easily compared and contrasted.Readers should note that our review stresses the general areas of convergence ofthe guidance and less the divergences that may exist between pieces of guidance.We acknowledge that our strategy also does not touch on recommendations madein discussion-oriented literature. However, as part of our reflections on the ethicalguidance (Sect. 25.4), we consider this literature and discuss gaps which may existin the ethical guidance as a whole.

Our review of the literature on ethics in psychiatric DBS revealed eight articles(see Table 25.1) containing explicit ethical guidance (see Table 25.2). The earliestof these were 2002–2003 letters to the editor written by Nuttin et al. (2002, 2003)of the DBS-OCD collaborative group, with American and European authors, andthe most recent were several articles written in 2010 and 2011 by Canadian,American, and Australian authors (Lipsman et al. 2010; Mian et al. 2010; Carteret al. 2011; Dunn et al. 2011). What each of these guidance articles has in common

25 Ethical Guidance for the Use of Deep Brain Stimulation 275

is a series of recommendations or criteria (four to 16 items) to be considered ormet in the conduct of DBS in psychiatric disorders. The articles differ in how theyarrived at these recommendations (e.g., consensus workshop, expert guidance,multidisciplinary collaboration) and these pieces of guidance are also describeddifferently, e.g., ‘‘[m]inimum ethical requirements for trials of DBS for addiction’’(Carter et al. 2011) and ‘‘[r]equirements for therapeutic research in the field ofdeep brain stimulation in patients with mental disorders’’ (Kuhn et al. 2009).Sometimes the guidance deals with a specific psychiatric disorder (e.g., addiction,OCD, or treatment-resistant depression) or topic (i.e., informed consent). SeeTable 25.1 for more information on these guidance articles.

We identified seven major themes represented in the guidance: (1) ethicaloversight, (2) experienced and interdisciplinary teams, (3) patient selection, (4)obligations to research subjects (and independence), (5) conditions for informedconsent, (6) social responsibilities, and (7) scientific practices. Table 25.2 explainsthe ethical significance and the ethical principles encompassed by each of thesethemes and summarizes in more detail what is contained in the guidance. In thenext section, we discuss the guidance on these seven topics.

Table 25.1 Explicit ethical guidance for psychiatric deep brain stimulation (DBS) found in theliterature

DBS for the treatment of psychiatric disorders(Nuttin et al. 2002, 2003)

Nine points of guidance for all psychiatric DBSon behalf of the OCD-DBS CollaborativeGroup, North American and Europeanparticipants

Scientific and ethical issues related to DBS forthe treatment of disorders of mood,behavior, and thought (Rabins et al. 2009)

Sixteen points of consensus for DBS in mood,behavior, and thought disorders developedthrough an NIH–Dana Foundation fundedworkshop, majority US participants withsome European contribution

DBS as a new therapeutic approach in therapy-resistant mental disorders: ethical aspects ofinvestigational treatment (Kuhn et al. 2009)

Five new recommendations for DBS in‘‘therapy-resistant mental disorders’’ andrestatement of previous guidelines by Nuttinet al., Germany

DBS for OCD: past, present, and future (Mianet al. 2010)

Synthesis of relevant guidelines for patientselection, and exclusion for psychiatricneurosurgery (specifically DBS in OCD)from the report of the National Commissionfor the Protection of Human Subjects ofBiomedical and Behavioral Research, USA

Criteria for the ethical conduct of psychiatricneurosurgery clinical trials (Lipsman et al.2010)

Seven criteria for the conduct of clinical trialsin psychiatric neurosurgery, Canada

Ethical issues raised by proposals to treataddiction using DBS (Carter et al. 2011)

Eleven minimum ethical requirements for theconduct of trials in DBS for the treatment ofaddiction, Canada and Australia

Ethical issues in DBS research for treatment-resistant depression: focus on risk andconsent (Dunn et al. 2011)

Four preliminary recommendations forinformed consent in DBS for the treatmentof treatment-resistant depression, USA

OCD obsessive–compulsive disorder

276 E. Bell and E. Racine

Tab

le25

.2S

umm

ary

ofth

eex

plic

itet

hica

lgu

idan

cesu

rrou

ndin

gps

ychi

atri

cD

BS

Gen

eral

them

esof

the

guid

ance

Eth

ical

prin

cipl

esen

com

pass

edby

the

them

eE

xam

ples

ofw

hat

isco

ntai

ned

inth

egu

idan

ce

Eth

ical

over

sigh

tP

rote

ctio

nof

the

rese

arch

subj

ect

thro

ugh

adhe

renc

eto

clea

rgu

idel

ines

for

rese

arch

and

foll

owin

gpr

inci

ples

ofre

sear

chet

hics

.Ens

urin

gre

spec

tfor

rese

arch

subj

ects

thro

ugh

bene

fice

nce

and

upho

ldin

gno

nmal

efice

nce,

such

that

they

are

not

expo

sed

toun

due

leve

lsof

risk

wit

hout

pote

ntia

lbe

nefi

ts

Ove

rsig

ht(a

ppro

val)

byan

ethi

csco

mm

itte

e/IR

Ba,

b,c

,e

Ong

oing

over

sigh

ta,c

and

sche

dule

dev

alua

tion

sof

the

stud

yan

ddi

ssem

inat

ion

offi

ndin

gse

Eth

ics

revi

ewfo

rhu

man

itar

ian

uses

b

Exp

erie

nced

and

inte

rdis

cipl

inar

yte

amP

rom

otes

just

ice

thro

ugh

equi

tabl

etr

eatm

ent

byre

duci

ngth

epo

ssib

ilit

yof

mis

unde

rsta

ndin

gsbe

twee

nth

epa

tien

tan

din

divi

dual

team

mem

bers

.Thi

sal

soen

sure

sth

atpa

tien

tsar

ese

enth

roug

hdi

ffer

ent

lens

esw

ith

rela

tion

toth

eir

oppo

rtun

ity

topa

rtic

ipat

ean

dm

itig

ates

som

eof

the

pote

ntia

lef

fect

sof

confl

icts

ofin

tere

stpr

esen

tw

ithi

nth

ete

amif

cons

ensu

sis

uphe

ldas

the

stan

dard

Mul

tidi

scip

lina

ryte

ampa

rtic

ipat

iona,

b,c

,d,e

Clo

seco

llab

orat

ion

betw

een

func

tion

alne

uros

urge

ons,

psyc

hiat

rist

sa,b,d

,cas

epa

nelc ,a

ndps

ycho

logi

stsd

Tea

ms

wit

hex

peri

ence

inne

uros

urge

ryfo

rps

ychi

atri

cdi

sord

ers

pref

erre

da,b

Par

tici

pati

onof

expe

rtor

clin

ical

rese

arch

cent

ersa,

b

Pat

ient

sele

ctio

nR

igor

ous

pati

ent

sele

ctio

nen

sure

sth

atth

ere

sear

chsu

bjec

tsw

ith

the

mos

tto

gain

are

the

ones

who

rece

ive

the

mos

tbe

nefi

tfr

omri

skie

rin

terv

enti

ons.

Pre

cise

sele

ctio

nup

hold

sfu

ndam

enta

lte

nant

sof

ethi

csin

rese

arch

byen

suri

ngth

atth

esc

ient

ific

data

are

vali

dan

dre

liab

le.

The

reli

abil

ity

ofth

ese

data

iscr

ucia

lfo

rth

etr

ansl

atio

nof

this

rese

arch

tobe

nefi

tth

em

ost

vuln

erab

lepa

tien

ts

Dis

orde

rca

uses

sign

ifica

ntim

pact

onqu

alit

yof

life

c ,pe

rson

alha

rm,

dise

ase

burd

en,

ordi

sabi

lity

b,d

,e,f

Res

ista

ntto

trea

tmen

ta,b,d

,e,f

Mul

tidi

scip

lina

rypa

rtic

ipat

ion

inse

lect

iona,

b,d

,e

(con

tinu

ed)

25 Ethical Guidance for the Use of Deep Brain Stimulation 277

Tab

le25

.2(c

onti

nued

)

Gen

eral

them

esof

the

guid

ance

Eth

ical

prin

cipl

esen

com

pass

edby

the

them

eE

xam

ples

ofw

hat

isco

ntai

ned

inth

egu

idan

ce

Obl

igat

ions

tore

sear

chsu

bjec

tsT

here

sear

chte

amha

ssp

ecifi

cre

spon

sibi

liti

esto

rese

arch

subj

ects

topr

omot

eth

eir

wel

fare

.The

seob

liga

tion

sin

clud

edi

sclo

sing

impo

rtan

tin

form

atio

nth

atth

esu

bjec

tsne

edto

give

info

rmed

cons

ent,

prov

idin

gca

reth

roug

hth

ele

ngth

ofth

est

udy,

allo

win

gsu

bjec

tsto

wit

hdra

wat

any

tim

efr

omth

est

udy,

and

ensu

ring

that

subj

ects

inth

etr

ial

are

not

wor

seof

fth

anpa

tien

tsno

tin

the

tria

l(i

mpa

cts

nega

tive

lyth

eir

curr

ent

care

)

Dis

clos

ean

ypo

tent

ial

confl

icts

ofin

tere

sta

Ens

ure

that

pati

ents

dono

tsu

ffer

any

nega

tive

fina

ncia

lef

fect

sby

wit

hdra

win

gfr

omth

etr

ialb

Ena

ble

pati

ents

topa

rtic

ipat

ean

dw

ithd

raw

free

lyfr

omth

est

udya,

b

Com

mit

tolo

ng-t

erm

foll

ow-u

pof

pati

ents

b,c

,f

Con

diti

ons

for

info

rmed

cons

ent

Res

pect

for

pers

ons

and

thei

rau

tono

my

isre

cogn

ized

and

addr

esse

dby

obta

inin

gin

form

edco

nsen

t.T

his

tena

ntof

ethi

calc

ondu

ctpr

escr

ibes

that

the

part

icip

ant

invo

lved

inre

sear

chha

sth

eca

paci

tyto

cons

ent,

unde

rsta

nds

the

natu

reof

the

rese

arch

,th

epo

tent

ial

bene

fits

,an

dri

sks,

isw

illi

ngto

part

icip

ate,

and

offe

rsco

nsen

tfr

eeof

coer

cion

.T

ruly

info

rmed

cons

ent

also

expl

ores

the

subj

ects

’m

otiv

atio

nsto

part

icip

ate,

poss

ible

ther

apeu

tic

mis

conc

epti

ons,

and

wil

ling

ness

for

long

-ter

mpa

rtic

ipat

ion

Par

tici

pant

ssh

ould

poss

ess

deci

sion

alca

paci

tya,

b,c

,f

Inco

rpor

ate

disc

losu

reof

pote

ntia

llo

ng-t

erm

cons

eque

nces

ofD

BS

b

Und

erst

and

that

depr

esse

dpa

tien

ts(p

sych

iatr

icpa

tien

ts)

may

cont

inue

topo

sses

sde

cisi

onal

capa

city

desp

ite

the

seve

rity

and

intr

acta

bili

tyof

the

dise

aseb

,g

Inco

rpor

atio

nof

care

give

rsin

the

info

rmed

cons

ent

proc

essb

,e

Use

ofa

stan

dard

ized

cons

ent

docu

men

tban

dde

mon

stra

tion

ofth

ew

orth

ofm

etho

dsap

plie

din

the

info

rmed

cons

ent

proc

esse

Att

enti

onpa

idto

the

pote

ntia

lef

fect

sof

desp

erat

ion

onin

form

edco

nsen

tgan

dth

epo

tent

ial

occu

rren

ceof

ther

apeu

tic

mis

conc

epti

onb

Soc

ial

resp

onsi

bili

ties

Soc

ialr

espo

nsib

ilit

ies

that

gobe

yond

the

purv

iew

ofst

anda

rdre

sear

chet

hics

take

into

cons

ider

atio

nth

epu

blic

atio

nan

dco

mm

unic

atio

n(k

now

ledg

etr

ansl

atio

n)of

find

ings

,th

eso

cial

cons

eque

nces

ofre

sear

ch,

and

resp

onsi

bili

ties

ofth

ete

amto

broa

der

part

icip

ants

than

the

rese

arch

subj

ects

them

selv

es

Fai

ran

dba

lanc

edpu

blic

atio

nof

resu

lts

and

prev

enti

onof

posi

tive

bias

inpu

blis

hed

find

ings

b,f

Tho

ught

ful

cons

ider

atio

nas

tow

hich

diso

rder

ssh

ould

bein

vest

igat

edc

Res

tric

tth

eus

eof

DB

Sfo

rth

erap

euti

cpu

rpos

eson

lyan

dno

tfo

ren

hanc

emen

tor

othe

rm

otiv

esa,

f

(con

tinu

ed)

278 E. Bell and E. Racine

Tab

le25

.2(c

onti

nued

)

Gen

eral

them

esof

the

guid

ance

Eth

ical

prin

cipl

esen

com

pass

edby

the

them

eE

xam

ples

ofw

hat

isco

ntai

ned

inth

egu

idan

ce

Sci

enti

fic

prac

tice

sT

heet

hica

lco

nduc

tof

scie

ntifi

cre

sear

chre

quir

esth

eim

plem

enta

tion

ofbe

stsc

ient

ific

prac

tice

s.In

this

cont

ext,

ensu

ring

soun

dra

tion

ale

for

targ

etse

lect

ion

and

the

avai

labi

lity

oflo

ng-t

erm

data

onal

ltr

eate

dpa

tien

tsre

info

rces

aco

mm

itm

ent

tofu

rthe

ring

soun

dap

proa

ches

tops

ychi

atri

cD

BS

Dem

onst

rate

the

effe

ctiv

enes

sbe

yond

that

ofcu

rren

tab

lati

vene

uros

urgi

cal

ther

apie

sb

Dev

elop

thor

ough

scie

ntifi

cra

tion

ale

for

targ

etsi

tese

lect

ionb

,c,f

Tra

ckan

dfo

llow

allc

ases

ofD

BS

for

MB

Tth

roug

ha

regi

stry

b

The

tabl

eid

enti

fies

the

gene

ral

area

sof

conv

erge

nce

ofth

eex

plic

itet

hica

lgu

idan

cein

DB

Sfo

rps

ychi

atri

cdi

sord

ers

(firs

tco

lum

n),

and

disc

usse

sth

epo

ssib

leet

hica

lrea

sons

why

thes

ear

eim

port

ant(

seco

ndco

lum

n).I

nth

eth

ird

colu

mn

we

refe

renc

eth

eei

ghta

rtic

les

cite

din

Tab

le25

.1w

ith

expl

icit

ethi

cal

guid

ance

tode

mon

stra

teex

ampl

esof

how

this

them

eem

erge

sin

the

guid

ance

IRB

inst

itut

iona

lre

view

boar

d,M

BT

moo

d,be

havi

or,

orth

ough

ta

Nut

tin

etal

.(2

002,

2003

)b

Rab

ins

etal

.(2

009)

cK

uhn

etal

.(2

009)

dM

ian

etal

.(2

010)

eL

ipsm

anet

al.

(201

0)f

Car

ter

etal

.(2

011)

gD

unn

etal

.(2

011)

25 Ethical Guidance for the Use of Deep Brain Stimulation 279

25.3.1 Ethical Oversight: The Protection of Research Subjects

All pieces of general guidance acknowledge the importance of research ethicsapproval and oversight in psychiatric DBS trials (Nuttin et al. 2003; Kuhn et al. 2009;Rabins et al. 2009; Lipsman et al. 2010). For example, as a minimum requirement forstudies in psychiatric DBS, Nuttin et al. (2003) recommend that an ethics committeeshould approve and provide ‘‘ongoing oversight’’ of the investigational protocol.Similarly, Lipsman et al. (2010) argue for ‘‘regulated, dispassionate oversight gov-erning the ethical conduct of clinicians and researchers’’ in psychiatric neurosurgery.Consensus reached by Rabins et al. (2009) extends the recommendation in favor ofethical oversight (‘‘patients should not undergo DBS for disorders of MBT [mood,behavior, or thought] without participating in an established, duly constituted,independently reviewed research protocol’’) to innovative or humanitarian appli-cations. They write, ‘‘deep brain stimulation performed for compassionate orhumanitarian use in single or small groups of patients should not be exempted fromindependent ethical review and oversight’’ (Rabins et al. 2009). Likely due in part tovariations in where the guidance originates, most recommendations do not alwaysmake explicit the mechanism by which studies should undergo ethical review,although it can be surmised that institutional review boards or comparable com-mittees responsible for implementing and upholding research ethics policies in othercountries would serve this function.

25.3.2 Experienced and Interdisciplinary Teams: Fosteringa Global Perspective

An interdisciplinary team, including specialists from functional neurosurgery andpsychiatry, is identified as a minimum requirement for conducting trials of DBS inpsychiatric disorders (Nuttin et al. 2003). The reasons for this are as follows: (1) togather an assessment of the patient from many different clinical perspectives,‘‘decision for candidacy should be made by a multidisciplinary team composed of:qualified psychiatrist, neurologist, and neurosurgeon for patient evaluation; qual-ified psychologists for psychometric testing’’ (Mian et al. 2010); and (2) to ensurea balanced and agreed upon final decision about the patient as a research subject,‘‘[u]nanimous approval should be obtained before proceeding w/op [with theoperation]’’ (Mian et al. 2010). Moreover, multidisciplinary involvement is rec-ommended for the long-term follow-up of patients, as DBS ‘‘necessitates a teamapproach to the evaluation of potential recipients; the implantation procedure;programming, and adjustment of concomitant medications; and ongoing moni-toring’’ (Rabins et al. 2009). Kuhn et al. (2009) also propose the incorporation of‘‘case-advisory panel[s],’’ which would add psychosocial, ethical, and legalexpertise to these evaluations. A multidisciplinary team is not necessarily per-ceived as being a sufficient requirement, as some recommendations also stress the

280 E. Bell and E. Racine

involvement of or close collaboration with experts or experienced centers inneurosurgery for psychiatric disorders (Nuttin et al. 2003; Rabins et al. 2009).

25.3.3 Patient Selection: Selection of Subjects Who Meetthe Criteria and Will Provide the Most Usable Knowledgefor the Treatment of Future Patients

On the topic of patient selection, the guidance for psychiatric DBS specifies severalimportant points. The guidance consistently identifies the need to establish severityand impact of illness, disability due to illness, and refractoriness in potential can-didates for DBS (Nuttin et al. 2003; Kuhn et al. 2009; Rabins et al. 2009; Lipsmanet al. 2010; Mian et al. 2010). It is recommended that the team members conduct athorough review of treatment failures to determine adequate dosing and applicationof known therapies (Rabins et al. 2009; Lipsman et al. 2010), including psycho-therapeutic approaches shown to be effective in the specific disorder of the patient(Mian et al. 2010). In addition, some of the guidance suggests integrating a broader,social support into the patient selection process. For instance, Kuhn et al. (2009)recommend focusing on the quality of life of the patient, Mian et al. (2010) suggestconsidering a lack of personal support systems, including family or friends as apotential exclusion factor, and Rabins et al. (2009) stress the evaluation of thepatient’s social situation and ‘‘potential for meaningful recovery.’’ Consensusachieved by Rabins et al. (2009) demonstrates how high the level of involvementrequired to meet the recommendations of appropriate patient selection can be.

‘‘… potential subjects in studies of DBS should be evaluated carefully andthoroughly to include: a review of all available records; information from thepatient’s clinicians to establish a baseline assessment of disease severity; docu-mentation of comorbidities; documentation in the patient’s history of the failure ofadequate (for both dosage and duration) therapeutic courses of multiple classes oftreatment; a comprehensive evaluation that concludes that the patient’s conditionis severe, chronic, disabling, and intractable; an assessment of the patient’s socialsituation, its impact on illness severity and vice versa, and the potential formeaningful recovery.’’ (Rabins et al. 2009)1

25.3.4 Obligations to Research Subjects: Responsibilitiesto Promote the Welfare of Research Participants

The different pieces of guidance relate to a diverse set of issues within the theme ofobligations to research subjects in psychiatric DBS trials. To address the need fortransparency with regard to potential conflicts of interest, Nuttin et al. (2003)

1 Parts of this quote appear as a list of bullet points in the original article.

25 Ethical Guidance for the Use of Deep Brain Stimulation 281

suggest that investigators disclose potential conflicts of interest to regulatorybodies, ethics committees, and ‘‘potential enrolees’’ during the informed consentprocess. Investigators also have an ‘‘obligation to collect prospective short-termand long-term follow-up data, including both therapeutic and adverse effects. Allof these data must be made publically available’’ (Rabins et al. 2009). Also,research subjects should be free to withdraw from the study (Nuttin et al. 2003), orhalt their participation in the study, without ‘‘financial barriers or burdens’’ (Ra-bins et al. 2009). At the same time, some of the guidelines focused on the com-mitment and obligation of the team to the long-term follow-up, and the provisionof care for participants. For instance, Carter et al. (2011) suggest that a minimalethical requirement to trialing DBS in addiction would be the ‘‘commitment of theresearch group to subsequent maintenance of the device’’ and the ‘‘provision ofpsychosocial support post-DBS.’’ Kuhn et al. (2009) also suggest that teams recruita ‘‘near-by person’’ to provide support to and to monitor the patient.

25.3.5 Conditions for Informed Consent: Issues of Importanceand Possible Barriers to Informed Consent for DBSin Psychiatric Disorders

One of the most basic obligations of the investigative team is to follow a process offreeand autonomous informed consent with the research subject who participates in thestudy. The guidance on psychiatric DBS trials reaffirms the importance of informedconsent in conducting these studies, and of ensuring that participants have the capacityto consent and are fully aware of all risks and benefits. Early on, Nuttin et al. (2003)suggested that ‘‘the use of DBS should be limited solely to those patients with decision-making capacity who are able to provide their own informed consent.’’ Other aspects tobe highlighted in the informed consent process have been put forward. Carter et al.(2011) recommend that subjects be made aware of ‘‘post-operative requirements (e.g.,programming, battery replacement)’’ and Rabins et al. (2009) specifically state that‘‘the consent process should state explicitly that, even with positive outcomes, DBS fordisorders of MBT is unlikely by itself to improve all aspects of the individual’s mood,function and interpersonal relationships: DBS is only one aspect of a comprehensivetreatment program.’’ One aspect of consent for DBS in psychiatric disorders which hasbeen given more attention of late is the issue of decisional capacity. Both Rabins et al.(2009) and Dunn et al. (2011) caution against the assumption that candidate patientswith psychiatric disorders do not have decisional capacity to consent to DBS. AlthoughDunn et al. (2011) suggest that the ‘‘severity and intractability of disease may rea-sonably alter an individual’s valuation of risks and potential benefits when consideringan intervention,’’ they maintain that this does not make treatment-resistant depressedpatients different from other refractory and/or desperate patients consenting toinvestigational therapies. However, consensus achieved by Rabins et al. (2009) cau-tions investigators to work diligently to identify and protect research subjects from

282 E. Bell and E. Racine

therapeutic misconceptions that may arise during the consent process, ‘‘because of thedramatic nature of the intervention and the risk of unrealistic expectations, specialattention must be given throughout the informed consent process to the identifica-tion—through conversation and direct questioning of potential subjects’ understand-ing of a protocol and motivations for participation—and correction of false beliefs andtherapeutic misconceptions.’’

25.3.6 Social Responsibilities: Social Considerations and Impactsof DBS in Psychiatric Disorders Which Extend the Purviewof Research Ethics

Some unique issues are raised in the guidance with regard to social issues that gobeyond standard research ethics and take into account the social impacts and influ-ences of research and the reporting of research results. With respect to the issue ofpublication bias, Carter et al. (2011) recommend ‘‘balanced publishing of researchresults, including negative results’’ and Rabins et al. (2009) identify that ‘‘care mustbe taken to avoid a positive bias, if those with poor outcomes are lost to follow up.’’Strong recommendations are also built into the guidance to prevent potential non-therapeutic uses of DBS and to guide investigators to ethically acceptable conditionsfor its use. For instance, ‘‘[t]he surgery should be performed only to restore normalfunction and relieve patient’s distress and suffering’’ (Nuttin et al. 2003) and‘‘motivation of the medical team [should be] to treat a medical illness [addiction] andnot as a form of extrajudicial punishment’’ (Carter et al. 2011).

25.3.7 Scientific Practices: Maintaining Scientific Rigorin Investigational Trials of DBS for the Treatmentof Psychiatric Disorders

The last theme that emerges throughout the guidance relates to the ethical andscientific responsibilities of investigators conducting a research study. Good sci-entific evidence is necessary for the selection of targets and in developing DBS forthe treatment of psychiatric disorders. Consensus achieved by Rabins et al. (2009)identified a need ‘‘for more basic research to support site selection for DBS ofMBT’’ and Lipsman et al. (2010) propose a ‘‘data-driven, evidence-based rationalefor disease & target selection’’ which ‘‘surpasses a consensus-derived threshold ofinformation for surgical intervention.’’ Kuhn et al. (2009) also propose ‘‘scientificpreclarification’’ in the definition of the target area and stimulation parameters.Other scientific recommendations made by Rabins et al. (2009) are to study andcompare the efficacy of DBS for the treatment of psychiatric disorders with currentneurosurgical approaches and ablative therapies, and to determine long-term safety

25 Ethical Guidance for the Use of Deep Brain Stimulation 283

over 10–15 years. Last, to bolster the scientific data over the long term, Rabinset al. (2009) propose the creation of a registry of de-identified data where the datafrom all individuals undergoing DBS for the treatment of psychiatric disorderswould be available.

25.4 Reflections on Guidance Available for Psychiatric DBS

Our review of available guidance shows that several ethical issues have beencross-identified as needing dedicated attention. To conclude this chapter, we makecomments and observations on three questions summoned by our review: (1) isspecific ethical guidance on DBS needed; (2) what issues are not well attended bycurrent guidance; (3) how should current ethical guidance be made available andtranslated for use by researchers and clinicians?

25.4.1 First Question: Is Specific ethical guidance on DBSNeeded? Is It All That Is Needed?

We have witnessed efforts to develop specific ethical guidance for psychiatric DBS(Table 25.2). A skeptic’s response to this emergence of guidance could be to pointout that many of the issues discussed in current guidance have been identified anddiscussed in the general medical ethics literature. A common assumption of the-oretical bioethicists is that bioethics inquiry should pursue novel ethical challengesand try not to reinvent ‘‘the bioethics wheel’’ (Parens and Johnston 2007). Indeed,we must be careful not to miss out on previous general ethical guidance andperhaps more importantly, not exaggerate the specificity of techniques and inter-ventions such as DBS to make them appear like exceptions. This is a risk ofspecialized ethical guidance. The reinvention of a bioethics wheel would miss outon important comparisons and parallels to be drawn while potentially inflating theuniqueness of the risks and benefits of an intervention such as DBS. However, anargument can be made for the importance of understanding the ethical contextwithin which DBS is surfacing and therefore to engage interdisciplinary groups to‘‘appropriate’’ previous scholarship. In this sense, the recommendations of work-ing groups and collaborative teams represent a process through which generalethical guidance is appropriated and contextualized, and responses to publicconcerns are considered and put forward. There are good practical and ethicalreasons why such a process constitutes an obligatory passage for context-sensitivemedical ethics. The effort of contextualization and specification made by inter-disciplinary groups can render explicit important aspects, which are not apparentin general ethical guidance. For example, recommendations for oversight throughinstitutional review boards or research ethics boards and of humanitarian uses stem

284 E. Bell and E. Racine

from a context of debatable DBS practices. There are also compelling historicaland practical reasons that underlie the ethical importance of multidisciplinaryteams in patient evaluations and follow-up for DBS. Further, collaborative effortsto generate guidance allow different individuals and disciplines to meet and dis-play transparent goals and objectives. They also make explicit the rationale forspecific actions and interventions, allowing the expression, communication, anddiscussion of such rationales easier. All in all, we must think carefully about howspecific guidance is interpreted in the context of existing research and clinicalguidance to understand correctly its function and purpose. Guidance developedspecifically for DBS needs to connect to general research, clinical ethics practice,and ethics scholarship while at the same time recognizing the role of specializedguidance for clinicians and patients.

25.4.2 Second Question: What Issues Are Not Well Attendedby Current Guidance? Should We Be Attending to OtherIssues?

In spite of common efforts to identify and address the ethical challenges of psy-chiatric DBS, some issues and problems do not seem to be as well captured bycurrent explicit guidance, in spite of empirical evidence or scholarly analysis oftheir existence. Some of this inattention may be explained by the specificity ofgoals and the limited focus of working groups generating explicit guidance but, ifthey are beyond the scope of issues that can be tackled by working groups, theycould be addressed in the future through other processes. One set of issuesreceiving limited attention is that of fair access to DBS and the allocation ofresources, which are certainly issues that surface in countries such as Canadawhere health resources and technologies are under fiscal and governmental

Fig. 25.1 Identification,development,implementation, andevaluation of evidence-basedethics requires an iterativeknowledge cycle of researchand experimentation

25 Ethical Guidance for the Use of Deep Brain Stimulation 285

pressure (Bell et al. 2011). Other concerns revealed in the patient’s experience ofDBS, including challenges related to relationships and disruptions of personalnarratives, have not been considered extensively in current guidance, althoughsome issues have been studied empirically (and have demonstrated patient-reported challenges) in DBS for the treatment of Parkinson’s disease (Haahr et al.2010; Agid et al. 2006; Schupbach et al. 2006). What is still unclear is if possiblepersonality disruptions are serious enough to warrant more guidance in psychiatricapplications of DBS. Another area lacking guidance relates to how and what kindof preclinical animal research is required before a clinical study in humans ismedically and ethically justified. Moreover, guidance related to the ethicallyappropriate time at which to translate innovative and experimental procedures intostandard care are lacking and call for principled and rationalized justification.Some topics may also overshadow other important ones. For example, informedconsent plays a huge role in contemporary ethics. Yet, sometimes consent may begiven an overly important role in ethical analysis. Since respect for autonomy mustbe factored into a larger ethical equation where beneficence and social justicecome into play, matters such as resource allocation and scientific validity haveimportant implications for balancing ethical principles. Informed consent cannot,and should not, necessarily offset the need to examine concerns of justice andscientific validity; the ability to obtain consent from participants in a sociallyunfair or poorly designed study would not achieve a satisfactory ethical resolution.

25.4.3 Third Question: How Should Current ethical guidanceBe Made Available and Translated for Use by Researchersand Clinicians? Can Evidence-Based Ethics in DBS SpurUptake of ethical guidance?

The development of ethical guidance for DBS has opened the door for a translationof ethical deliberation and scholarship to practice and real-world experimentation.In fact, collaborative teams (such as those we found generating guidance onpsychiatric DBS) are clearly important stakeholders not only in the development ofguidance for practice but also in the implementation of guidance. At the sametime, research has to generate evidence to support a process of ongoing monitoringfor the identification of issues as well as to validate ethical concerns alreadyaddressed in guidance. Furthermore, research can enable better characterization ofthe issues as well as the development, implementation, and evaluation of enactedsolutions (see Fig. 25.1). Such a process situates ethics best practices in DBS andother fields within a research and knowledge cycle to ensure that recommendationsare informed by evidence, supported by research, and subsequently evaluated(Kim 2004; Racine et al. 2011). For a more extensive discussion of the role ofpragmatic and evidence-based neuroethics, see Racine (2010).

286 E. Bell and E. Racine

25.5 Conclusion

This chapter has reviewed how the emergence of DBS in psychiatric conditionsand for other innovative uses has sparked or reinitiated concerns about the ethicaluse of DBS. The community of DBS practitioners and scholars has promptlyresponded by developing specific ethical guidance, which addresses some con-cerns that although not necessarily unique to DBS need attention and discussionwithin interdisciplinary forums. Important convergences in the ethical guidance,among others reviewed in this chapter, can be found in the areas of ethicaloversight, interdisciplinary collaboration, and subject selection. The efforts madeby clinicians and others to craft this guidance raised general questions about theutility of such specific guidance and the means to introduce it to allow practices toevolve dynamically in response to collaborative ethical deliberation. In this sense,the work of groups and scholars involved in this area not only helps to addressimportant issues needing immediate attention but also contributes to shaping thedevelopment of a flexible, responsive, transparent, and evidence-based ethics.

References

Agid Y, Schupbach M, Gargiulo M, Mallet L, Houeto JL, Behar C, Maltete D, Mesnage V,Welter ML (2006) Neurosurgery in Parkinson’s disease: the doctor is happy, the patient lessso? J Neural Transm Suppl 70:409–414

Bell E (under review) Ethical issues in psychiatric applications of deep brain stimulation:learning from Canadian healthcare providers

Bell E, Mathieu G, Racine E (2009) Preparing the ethical future of deep brain stimulation. SurgNeurol 72:577–586

Bell E, Maxwell B, McAndrews MP, Sadikot A, Racine E (2011) Deep brain stimulation andethics: perspectives from a multi-site qualitative study of Canadian neurosurgical centers.World Neurosurg 76:537–547

Carter A, Bell E, Racine E, Hall W (2011) Ethical issues raised by proposals to treat addictionusing deep brain stimulation. Neuroethics 4:129–142

Dunn LB, Holtzheimer PE, Hoop JG, Mayberg HS, Roberts LW, Appelbaum PS (2011) Ethicalissues in deep brain stimulation research for treatment-resistant depression: focus on risk andconsent. Am J Bioeth Neurosci 2:29–36

Fins JJ, Mayberg HS, Nuttin B, Kubu CS, Galert T, Sturm V, Stoppenbrink K, Merkel R,Schlaepfer TE (2011) Misuse of the FDA’s humanitarian device exemption in deep brainstimulation for obsessive-compulsive disorder. Health Aff (Millwood) 30:302–311

Glannon W (2008) Deep-brain stimulation for depression. HEC Forum 20:325–335Haahr A, Kirkevold M, Hall EO, Ostergaard K (2010) From miracle to reconciliation: a

hermeneutic phenomenological study exploring the experience of living with Parkinson’sdisease following deep brain stimulation. Int J Nurs Stud 47:1228–1236

Hildt E (2006) Electrodes in the brain: some anthropological and ethical aspects of deep brainstimulation. IRIE 5:33–39

Kim SY (2004) Evidence-based ethics for neurology and psychiatry research. NeuroRx 1:372–377Kuhn J, Gaebel W, Klosterkoetter J, Woopen C (2009) Deep brain stimulation as a new

therapeutic approach in therapy-resistant mental disorders: ethical aspects of investigationaltreatment. Eur Arch Psychiatry Clin Neurosci 259(Suppl 2):S135–S141

25 Ethical Guidance for the Use of Deep Brain Stimulation 287

Lipsman N, Bernstein M, Lozano AM (2010) Criteria for the ethical conduct of psychiatricneurosurgery clinical trials. Neurosurg Focus 29:E9

Lipsman N, Mendelsohn D, Taira T, Bernstein M (2011) The contemporary practice ofpsychiatric surgery: results from a survey of North American functional neurosurgeons.Stereotact Funct Neurosurg 89:103–110

Mendelsohn D, Lipsman N, Bernstein M (2010) Neurosurgeons’ perspectives on psychosurgeryand neuroenhancement: a qualitative study at one center. J Neurosurg 113:1212–1218

Mian MK, Campos M, Sheth SA, Eskandar EN (2010) Deep brain stimulation for obsessive-compulsive disorder: past, present, and future. Neurosurg Focus 29:E10

National Commission for the Protection of Human Subjects of Biomedical and BehavioralResearch (1977) Report and recommendations: Psychosurgery. DHEW publication no.(OS)77-0001, Washington

Nuttin B, Gybels J, Cosyns P, Gabriels L, Meyerson B, Andreewitch S, Rasmussen S, GreenbergB, Friehs G, Rezai A, Montgomery E, Malone D, Fins JJ (2002) Deep brain stimulation forpsychiatric disorders. Neurosurgery 51:519

Nuttin B, Gybels J, Cosyns P, Gabriels L, Meyerson B, Andreewitch S, Rasmussen SA,Greenberg B, Friehs G, Rezai AR, Montgomery E, Malone D and Fins JJ (2003) Deep brainstimulation for psychiatric disorders. Neurosurg Clin N Am 14:xv–xvi

Parens E, Johnston J (2007) Does it make sense to speak of neuroethics? Three problems withkeying ethics to hot new science and technology. EMBO Rep 8 Spec No S61-4

Rabins P, Appleby BS, Brandt J, DeLong MR, Dunn LB, Gabriels L, Greenberg BD, Haber SN,Holtzheimer PE 3rd, Mari Z, Mayberg HS, McCann E, Mink SP, Rasmussen S, Schlaepfer TE,Vawter DE, Vitek JL, Walkup J, Mathews DJ (2009) Scientific and ethical issues related to deepbrain stimulation for disorders of mood, behavior, and thought. Arch Gen Psychiatry 66:931–937

Racine E (2010) Pragmatic neuroethics: improving understanding and treatment of the mind-brain. MIT Press, Cambridge

Racine E, Bell E, Di Pietro NC, Wade L, Illes J (2011) Evidence-based neuroethics forneurodevelopmental disorders. Semin Pediatr Neurol 18:21–25

Schechtman M (2010) Philosophical reflections on narrative and deep brain stimulation. J ClinEthics 21:133–139

Schlaepfer TE, Fins JJ (2010) Deep brain stimulation and the neuroethics of responsiblepublishing: when one is not enough. JAMA 303:775–776

Schupbach M, Gargiulo M, Welter ML, Mallet L, Behar C, Houeto JL, Maltete D, Mesnage V,Agid Y (2006) Neurosurgery in Parkinson disease: a distressed mind in a repaired body?Neurology 66:1811–1816

Synofzik M, Schlaepfer TE (2008) Stimulating personality: ethical criteria for deep brainstimulation in psychiatric patients and for enhancement purposes. Biotechnol J 3:1511–1520

Synofzik M, Schlaepfer TE (2011) Electrodes in the brain–ethical criteria for research andtreatment with deep brain stimulation for neuropsychiatric disorders. Brain Stimul 4:7–16

288 E. Bell and E. Racine

Chapter 26History of ‘‘Psychiatric’’ Deep BrainStimulation: A Critical Appraisal

Marwan I. Hariz

26.1 Introduction

Deep brain stimulation (DBS) is an established method for surgical treatments ofmovement disorders. Today, most of the investigational applications of DBS are inthe field of neuropsychiatry, especially obsessive–compulsive disorder (OCD),Gilles de la Tourette syndrome, and major depressive disorder. It is a common beliefthat chronologically DBS in psychiatry follows DBS in movement disorders; forexample, in 2004 Kopell et al. (2004) wrote: ‘‘Over the last decade, deep brainstimulation (DBS) has revolutionized the practice of neurosurgery, particularly in therealm of movement disorders. It is no surprise that DBS is now being studied in thetreatment of refractory psychiatric diseases.’’ Also, Selten et al. (2008) wrote: ‘‘TheDBS procedure was originally introduced for the treatment of movement disorders,but nowadays it is being studied as a possible treatment option for intractable states ofneuropsychiatric conditions.’’ It is also common belief that DBS in psychiatrystemmed from the observation of psychiatric and behavioral side effects of DBS inthe subthalamic nucleus (STN) in Parkinson’s disease patients. Schläpfer et al.(Schläpfer and Bewernick 2009) wrote: ‘‘The observation of induced psychiatric sideeffects (e.g., changes in mood, hypomania, reduction of anxiety) gave the impulse totry DBS also for psychiatric disorders.’’ Finally, it is assumed that the old stereotacticsurgery for psychiatry illness was not multidisciplinary enough, with neurosurgeonsacting alone, many times without consulting psychiatrists. As an example of thisstatement, in 2006 Fins et al. (2006) wrote the following:

M. I. Hariz (&)Professor of Functional Neurosurgery, UCL Institute of Neurology,Box 146 Queen Square, London WC1N 3BG, UKe-mail: m.hariz@ucl.ac.uk

M. I. HarizDepartment of Clinical Neuroscience, Umeå University, Umeå, Sweden

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5_26, � Springer-Verlag Berlin Heidelberg 2012

289

It is ethically untenable for this work to proceed by neurosurgeons in isolation withoutpsychiatrists determining the diagnosis and suitability of patients for treatment. The merefact that electrodes can be placed is not a moral warrant for their insertion…. Such errantbehavior is especially inappropriate because it represents a recapitulation of the excessesassociated with psychosurgery…. If this generation of neuroscientists and practitionershope to avoid the abuses of that earlier era, and avoid conflation of neuromodulation withpsychosurgery, it is critical that neuromodulation be performed in an interdisciplinary andethically sound fashion.

The aim of this chapter is to scrutinize these statements that are representativeof leading opinions in contemporary literature, in the light of available historicalliterature on the subject.

26.2 Materials and Methods

The author attempted to trace, through a literature search in scientific journals, aswell as in published books and proceedings from scientific meetings, the origins ofchronic DBS to find out what its first applications in humans were, and who wasinvolved in the practice of early DBS.

26.3 Results

26.3.1 Origins of DBS

Stereotactic functional neurosurgery started with a cooperation between ErnstSpiegel, a neurologist, and Henry Wycis, a neurosurgeon (Spiegel et al. 1947).They introduced the stereotactic technique in humans with the explicit aim toavoid the side effects of lobotomy by making a very focal lesion in pertinentpathways and nuclei in psychiatric patients. Indeed, in their seminal article from1947 (Spiegel et al. 1947) describing the first human stereotactic apparatus, theywrote: ‘‘This apparatus is being used for psychosurgery…. Lesions have beenplaced in the region of the medial nucleus of the thalamus (medial thalamotomy).’’Soon after in 1952, neurophysiologist and neuropsychiatrist José Delgadodescribed a technique for implantation of electrodes for long-term recording andchronic stimulation to evaluate its value in psychotic patients (Delgado et al.1952). In 1953, in an article about depth stimulation of the brain, Bickford et al.(1953) wrote the following: ‘‘An observation that may have some practical sig-nificance was that several of our psychotic patients seem to improve and becomemore accessible in the course of stimulation studies lasting several days.’’ Theythought that a likely explanation for this phenomenon ‘‘was that the local stimu-lation was having a therapeutic effect comparable to that of electroshock.’’ Theywrote further: ‘‘This aspect of localized stimulation studies requires further

290 M. I. Hariz

investigation since it may lead to a most specific, less damaging, and more ther-apeutically effective electrostimulation technic than can be achieved by the rela-tively crude extracranial stimulation methods in use at present’’ (Bickford et al.1953). Meanwhile, Delgado continued to investigate the use of DBS and devised atechnique of ‘‘radio communication with the brain’’ through chronically implantedelectrodes attached to a receiver subcutaneously implanted in the scalp, which hecalled ‘‘stimoceiver,’’ specifically for use in psychosurgical patients (Delgado et al.1968, 1973). In parallel, a group at Tulane University in New Orleans led bypsychiatrist Robert Heath was heavily involved for three decades, starting in theearly 1950s, in studies of chronic depth stimulation in patients with schizophreniaand in the search for the brain’s ‘‘pleasure center’’ (Baumeister 2000). Some ofHeath’s work at Tulane University included studies of ‘‘rewarding’’ and ‘‘aver-sive’’ subcortical structures (Heath 1963), and dealt with surgical control ofbehavior and initiation of heterosexual behavior in a homosexual male (Moan andHeath 1972), and other aspects of modulation of behavior and emotion usingchronic DBS (Heath 1977). The Tulane University experience in this field wasanalyzed in 2000 by psychologist Alan Baumeister and was published under thetitle ‘‘The Tulane Electrical Brain Stimulation Program. A historical case study inmedical ethics’’ (Baumeister 2000). Baumeister wrote: ‘‘The central conclusion ofthe present review is that the Tulane electrical brain stimulation experiments hadneither a scientific nor a clinical justification… The conclusion is that theseexperiments were dubious and precarious by yesterday’s standards.’’ Long beforeBaumeister’s verdict, in 1977 neurosurgeon Lauri Laitinen (1977) commented, inhis article entitled ‘‘Ethical aspects of psychiatric surgery’’ on one of Heath’sarticles published in 1972 (Heath 1972): ‘‘There is no doubt that in this study allstandards of ethics had been ignored. The ethical responsibility of the editors whoaccept reports of this kind for publication should also be discussed.’’

In view of the above, it is difficult to give any credit to the claim of Fins et al.(2006) that ‘‘it is ethically untenable for this work to proceed by neurosurgeons inisolation’’ when history shows that those conducting such work ‘‘in isolation,’’ anddisclosing such an ‘‘inappropriate… errant behavior,’’ were not neurosurgeons. It isinteresting in this context to note that a wide-held belief which was repeatedlypublished is that ‘‘one of the most notable surgeons was the American neurosurgeonWalter Freeman… Freeman began to apply his relatively untested procedure, theprefrontal lobotomy, in which he transorbitally inserted an ice pick into the frontalcortex’’ (Malone and Pandya 2006). In fact, Freeman was a neuropsychiatrist, and thetruth is that he was actually abandoned by his neurosurgeon James Watts, followingFreeman’s increasingly uncritical and erratic attitude to lobotomy (El-Hai 2005).When neurosurgeons are today made scapegoats by some, who would remember thatthe Norwegian psychiatrist Ornulf Odegård (1953), who was the director of Nor-way’s main psychiatrist hospital for more than 30 years, wrote the following in 1953:‘‘Psychosurgery can be easily performed by the psychiatrist himself with the tool hemight have in his pocket, and strangely enough it may be harmless and effective.’’

Coming back to the DBS in the early days, this method continued to be rarelytested primarily for behavioral disorders well into the 1970s (Escobedo et al.

26 History of ‘‘Psychiatric’’ Deep Brain Stimulation 291

1973). Meanwhile, starting in the 1960s, Bechtereva et al. (1977), from the formerLeningrad, pioneered chronic stimulation of the thalamus and basal ganglia intreatment of Parkinson’s disease.

26.3.2 ‘‘Modern’’ Applications of DBS in Psychiatry

The first use of modern-era DBS in psychiatric disorders had nothing to do withthe observation of psychiatric and behavioral side effects of STN DBS, as claimedby some (Schläpfer and Bewernick 2009). When Vandewalle et al. (1999) pio-neered DBS for the treatment of Gilles de la Tourette syndrome, and Nuttin et al.(1999) pioneered DBS for the treatment of obsessive–compulsive disorder, both in1999, they were simply targeting the very same brain structures that were ste-reotactically lesioned in the past for the same disorders.

26.4 Discussion

A review of the old scientific literature on DBS bears witness to the inaccuracy ofseveral contemporary statements, in which DBS is portrayed as a novel treatmentmodality that has only recently been introduced in psychiatric disease on the basis ofobservations of its effect during application in movement disorders, and in whichneurosurgeons are erroneously blamed for (mal)practices of the past, and criticizedfor neglecting multidisciplinarity and ethical rules. How ironic these accusations areis evidenced by a recent publication in the entitled ‘‘Scientific and ethical issuesrelated to deep brain stimulation for disorders of mood, behavior, and thoughts’’(Rabins et al. 2009). This article summarized the results of a 2-day consensus con-ference held to examine scientific and ethical issues in the application of DBS inpsychiatry in order to ‘‘establish consensus among participants about the design offuture clinical trials of deep brain stimulation for disorders of mood, behavior, andthought’’ and to ‘‘develop standards for the protection of human subjects partici-pating in such studies’’ (Rabins et al. 2009). There was no neurosurgeon among the30 participants at the meeting, 19 of whom were authors of the article.

26.5 Conclusions

a. DBS was not originally introduced for the treatment of movement disorders.From the very beginning DBS was a tool to study and eventually treat psy-chiatric illness, and to modify behavior.

b. The first application of modern DBS in psychiatric illness tried to mimic le-sional surgery by implanting electrodes in the same targets that were lesionedbefore for the same conditions.

292 M. I. Hariz

c. Although ‘‘it is ethically untenable for this work to proceed by neurosurgeons inisolation without psychiatrists determining the diagnosis and suitability ofpatients for treatment’’ (Fins et al. 2006), it was indeed psychiatrists, neurol-ogists, and neurophysiologists who were in the past working ‘‘in isolation.’’

Conflicts of Interest The author has occasionally received reimbursement for travel expensesand honoraria for speaking at meetings from Medtronic.

References

Baumeister AA (2000) The Tulane electrical brain stimulation program. A historical case study inmedical ethics. J Hist Neurosci 9:262–278

Bechtereva NP, Kambarova DK, Smirnov VM, Shandurina AN (1977) Using the brain’s latentabilities for therapy: chronic intracerebral electrical stimulation. In: Sweet BW, Obrador S,Martín-Rodrígez JG (eds) Neurosurgical treatment in psychiatry, pain and epilepsy.University Park Press, Baltimore, pp 581–613

Bickford RG, Petersen MC, Dodge HW Jr, Sem-Jacobsen CW (1953) Observations on depthstimulation of the human brain through implanted electrographic leads. Mayo Clin Proc 28:181–187

Delgado JM, Hamlin H, Chapman WP (1952) Technique of intracranial electrode implacementfor recording and stimulation and its possible therapeutic value in psychotic patients. ConfinNeurol 12:315–319

Delgado JM, Mark V, Sweet W, Ervin F, Weiss G, Bach-Y-Rita G, Hagiwara R (1968)Intracerebral radio stimulation and recording in completely free patients. J Nerv Ment Dis147:329–340

Delgado JMR, Obrador S, Martín-Rodriguez JG (1973) Two-way radio communication with thebrain in psychosurgical patients. In: Laitinen LV, Livingstone KE (eds) Surgical approachesin psychiatry. Medical and Technical Publishing, Lancaster, pp 215–223

El-Hai J (2005) The lobotomist. Wiley, HobokenEscobedo F, Fernández-Guardiola A, Solís G (1973) Chronic stimulation of the cingulum in

humans with behaviour disorders. In: Laitinen LV, Livingstone KE (eds) Surgical approachesin psychiatry. Medical and Technical Publishing, Lancaster, pp 65–68

Fins JJ, Rezai AR, Greenberg BD (2006) Psychosurgery: avoiding an ethical redux whileadvancing a therapeutic future. Neurosurgery 59:713–716

Heath RG (1963) Electrical self-stimulation of the brain in Man. Am J Psychiatry 120:571–577Heath RG (1972) Pleasure and brain activity in man: deep and surface electroencephalograms

during orgasm. J Nerv Ment Dis 154:3–18Heath RG (1977) Modulation of emotion with a brain pacemaker. Treatment for intractable

psychiatric illness. J Nerv Ment Dis 165:300–317Kopell BH, Greenberg B, Rezai AR (2004) Deep brain stimulation for psychiatric disorders.

J Clin Neurophysiol 21:51–67Laitinen LV (1977) Ethical aspects of psychiatric surgery. In: Sweet WH, Obrador S, Martín-

Rodríguez JG (eds) Neurosurgical treatment in psychiatry, pain and epilepsy. University ParkPress, Baltimore, pp 483–488

Malone DA Jr, Pandya MM (2006) Behavioral neurosurgery. Adv Neurol 99:241–247Moan CE, Heath RG (1972) Septal stimulation for the initiation of heterosexual behavior in a

homosexual male. J Behav Ther Exp Psychiatry 3:23–30Nuttin B, Cosyns P, Demeulemeester H, Gybels J, Meyerson B (1999) Electrical stimulation in

anterior limbs of internal capsules in patients with obsessive compulsive disorder. Lancet354:1526

26 History of ‘‘Psychiatric’’ Deep Brain Stimulation 293

Odegård O (1953) Nye framsteg i psychiatrien. Tidskrift for den Norske Laegeforening123:411–414

Rabins P, Appleby BS, Brandt J, DeLong MR, Dunn LB, Gabriëls L, Greenberg BD, Haber SN,Holtzheimer PE 3rd, Mari Z, Mayberg HS, McCann E, Mink SP, Rasmussen S, SchlaepferTE, Vawter DE, Vitek JL, Walkup J, Mathews DJ (2009) Scientific and ethical issues relatedto deep brain stimulation for disorders of mood, behavior, and thought. Arch Gen Psychiatry66:931–937

Schläpfer TE, Bewernick BH (2009) Deep brain stimulation for psychiatric disorders–state of theart. Adv Tech Stand Neurosurg 34:37–57

Spiegel EA, Wycis HT, Marks M, Lee AS (1947) Stereotaxic apparatus for operations on thehuman brain. Science 106:349–350

Stelten BM, Noblesse LH, Ackermans L, Temel Y, Visser-Vandewalle V (2008) Theneurosurgical treatment of addiction. Neurosurg Focus 25(1):E5

Vandewalle V, van der Linden C, Groenewegen HJ, Caemaert J (1999) Stereotactic treatment ofGilles de la Tourette syndrome by high frequency stimulation of thalamus. Lancet 353:724

294 M. I. Hariz

Index

AAblative

procedures, 114surgery, 12, 193

Abnormal activity, 25, 26Abstinence, 144Activation, 26, 27, 29, 206, 208, 209, 211, 212Activation threshold, 4, 7Activity modulation, 26Addiction, 90, 232, 276Addictive behavior, 46, 166, 233Adenosine, 196–197, 209–210Afferent inputs, 24Amygdala, 22, 37, 78, 87, 147, 221, 229–232

basolateral nucleus, 37, 147, 243central nucleus, 37, 243

Anger, 165Anhedonia, 81, 82, 88, 89, 160, 230Animal

models, 57, 62–68, 106, 142, 176, 218models of addiction, 150studies, 218

Anteriorcapsulotomy, 35, 36, 40, 96cingulate cortex, 11, 81cingulate gyrus, 55commissure, 12, 36, 97limbs of the internal capsule (ALIC), 36,

37, 40, 81, 96, 225Antidepressant effect, 72, 83, 106Anxiety, 165, 220, 242, 243, 245, 246

disorder, 35, 39, 43, 87symptoms, 36, 75

Anxiolytic effect, 86, 164, 220Apathy, 36, 164

Astrocytes, 205Attention, 220Aversive, 39, 222, 247Axon, 24–30Axonal, 22, 28

activation, 25, 27, 195fibers passing, 24pathways, 22, 28, 29response, 24trajectories, 22

BBasal ganglia, 45, 54, 195, 226Basal ganglia–thalamocortical circuits, 12, 54,

63, 64, 65, 67, 162Battery life, 99, 196Bed nucleus of the stria terminalis (BNST),

36–40, 65, 97, 221Behavioral

addiction, 46disorders, 115, 123, 124inhibition, 39therapy, 123

Binge eating, 166Biophysical markers, 30Biphasic stimulation, 2Bipolar disorder, 74, 89Bipolar stimulation, 5Brain

activity, 21stimulation reward, 221

Brain-derived neurotrophic factor (BDNF),109

D. Denys et al. (eds.), Deep Brain Stimulation,DOI: 10.1007/978-3-642-30991-5, � Springer-Verlag Berlin Heidelberg 2012

295

Brainstem, 12–18, 37, 38, 226, 227,229–231, 233

CCarbon

fiber microelectrode, 264, 270nanofiber, 264, 267, 270nanotube, 264

Caudate nucleus, 28, 36, 44, 74, 82, 197, 221,228, 230, 234

Charge, 2, 3, 6density, 2, 105

Chronaxie, 4Chronic unpredictable stress (CUS), 106Clinical benefit, 187Cognitive effects, 49, 86Compulsive

behavior, 63–66, 120, 169checking, 62shopping, 166

Computational models, 27, 187, 255Conditioned fear, 38, 220, 246Conditioned place preference (CPP), 144Connectivity, 22, 226, 229Construct validity, 62, 148, 218Continuous reinforcement schedule, 142Corticostriatal–thalamo-

cortical (CSTC), 22–27, 45, 226circuits, 25, 36networks, 22, 24

DDecisional threshold, 58Deep brain stimulation (DBS), 1, 8, 21, 24–28,

43, 45, 46, 53, 61, 64, 71, 77, 81–83,95, 113, 135, 138, 160, 208–212, 225,227, 228, 230, 233–236, 263–265, 267,269–271, 289–292

electrodes, 5, 6, 27, 72, 73, 184–189, 209,219, 254–260, 263

recordings, 185Depressed, 25, 26Depression, 11, 12, 18, 22–27, 78, 100, 103,

163model, 245

Depressive-like behavior, 109Diffusion tensor imaging (DTI), 22, 30, 225Dopamine, 26, 44, 62, 82, 123, 132, 177, 222,

243–245, 248, 264, 267, 269, 270D2/3-receptor binding, 235

dysregulation syndrome, 150, 166, 200release, 196, 222

Dopaminergic drugs, 161, 162, 163, 166Dorsal raphe nucleus (DRN), 176Dorsal striatum, 147Dose–response curves, 143Drug

addiction, 131–139, 144, 221self-administration, 141, 142treatment, 114

EElectric field, 24, 25, 27, 28Electrode

array, 257, 259, 260design, 255encapsulation, 2geometries, 256impedance, 2technology, 187

Electroencephalogram (EEG), 187, 245, 246Electrophysiological, 76, 178

activity, 184, 194recordings, 59, 218

Electrophysiology, 184, 245, 246Endophenotypes, 220Ethical

aspects, 291considerations, 189guidance, 273, 275–277, 284, 286, 287

Ethics, 89, 273–275, 277, 278, 280, 282–287,291

Extended amygdala, 37Extinction, 144Extinction and reinstatement in the CPPmodel,

145Extracellular potential, 24, 25Extreme capsule, 13

FFace validity, 62, 148Fast scan cyclic voltammetry (FSCV), 197,

198, 267Fear, 37–40, 165, 220, 221, 243, 245–247

signals, 87Field steering, 254, 260Fixed-ratio schedule of reinforcement, 142Forced swim test (FST), 106, 179Functional MRI (fMRI), 40, 194, 225Future developments, 186, 187

296 Index

GGABA (GABAergic), 24, 66, 82, 179, 206,

245, 247Gamma oscillations, 185Glia, 205–210Globus pallidus (GPi), 64, 161, 234Globus pallidus externa, 114, 120Globus pallidus interna, 114, 120Glutamate, 23, 83, 206, 243–245

release, 197

HHigh-frequency stimulation, 26, 27, 53, 54History, 291Human enhancement, 275Hyperdirect prefronto-subthalamic pathway,

57Hypersexuality, 166Hypervigilance, 39Hypodopaminergic syndrome, 160, 162, 164,

165Hypomania, 49, 55, 75, 88, 98, 121, 231, 289Hypothalamic-pituitary-adrenal axis, 38Hypothalamus, 72

IImpulse control disorders, 166Impulsivity, 49, 55, 98, 162, 169, 232, 249Informed consent, 274, 276, 278, 282, 283,

286Infralimbic cortex, 104Insula, 40, 72, 147, 154, 231, 232Insular cortex, 147Internal capsule, 14, 47, 104, 219–221Internal capsule/nucleus accumbens, 119, 121Intracranial

EEG, 89self-stimulation, 221

Intra-operative recordings, 184

LLateral

habenula, 105, 153, 179hypothalamus, 153

Learned helplessness (LH), 106, 179Levodopa, 161, 178Lobotomy, 114, 274, 290, 291Local field potentials, 154, 184Local neurons, 24Low-frequency stimulation, 105, 153, 208, 227

MMajor depressive disorder (MDD), 21, 22,

24–28, 30, 71, 77, 95, 225Mania, 47, 89, 163Mechanism of action, 217Medial forebrain bundle (MFB), 81, 85, 221,

231Medial prefrontal cortex (mPFC), 145Medial wall, 57Medication, 114, 120, 123Mediodorsal thalamic nucleus, 26, 65, 96Mesolimbic, 164Microfabrication technologies, 257, 260Microsensors, 196–199Microwires, 185Monopolar

stimulation, 5, 74, 98, 135, 227waveforms, 2

Mood elevation, 97Motivation, 82–84, 145–148, 151, 153, 154,

185, 229, 283Multidisciplinary teams, 58, 139, 274, 280,

285Multiple brain regions, 187Myelinated fibers, 4

NNanoelectrode, 263–265, 267, 269Negative results, 283Network, 22–27, 207–211

activity, 23, 24models of disease, 29

Neuralcircuitry, 22, 24response, 24, 25

Neuroimaging, 225–230, 234–236Neuroleptics, 123Neuromodulation, 1, 78, 194Neuronal

activities, 55, 194response, 27

Neurostimulation, 162Neurosurgery, 132, 133, 136, 274–277, 280,

281, 289Neurosurgical treatment, 123Neurotransmitter, 199, 208, 263, 264, 267,

269–271release, 193

Nonhuman primates, 16, 22, 30, 219, 248, 260Noradrenergic, 26, 131, 160Nucleus accumbens, 36, 43, 45, 47, 64, 81, 83,

85, 96, 104, 114, 119, 121, 131, 134,

Index 297

136, 145, 150, 185, 186, 220, 228, 230,232, 243

ablation, 138

OObligation to research subjects, 276, 278, 281Obsessive compulsive

behavior, 115, 162disorder (OCD), 11, 21–28, 30, 35–40, 43,

46, 53, 61, 89, 121, 225Optogenetics, 241–243Orbitofrontal cortex, 11, 17, 36, 96, 226,

232, 233Oscillatory, 56

PParkinson’s disease (PD), 53, 159, 175, 195,

231, 247, 286model, 177

Pathologicalgambling, 166network activity, 26

Patient selection, 78, 100, 122, 133, 273, 276,277, 281

Perseveration, 49, 65Personality, 138, 162, 164, 166, 168, 274, 286Plasticity, 29Polypyrrole, 264, 267Positron emission tomography (PET), 55,

72, 225Post-operative recordings, 184Predictive validity, 62, 106, 149Prefrontal cortex, 153, 176Procognitive effects, 86Programming, 97–98, 135, 260, 280Progressive ratio, 142Psychiatry, 225, 227, 236, 274, 280Psychosurgery, 290, 291Punding, 167

QQuality standards, 87–89Quantitative outcome metrics, 30Quinpirole, 62

RRadiofrequency lesions, 96, 108, 136Reinstatement, 144Reward, 46, 142, 186, 242–245

motivation, 97processing, 221system, 81–83, 90, 132

Rheobase current, 4

SScalp EEG, 184Schedule-induced polydipsia, 63Schizophrenia, 90, 291Scientific benefit, 187, 188Self-injurious behaviour (SIB), 115Serotonin (5-HT), 27, 97, 109, 163, 176,

181, 199release, 109, 178, 195reuptake inhibitors, 54, 63, 180transmission, 176

Side-effects, 36, 55, 83, 84, 86, 90, 98, 103,114, 116, 118, 119, 138, 139, 159, 175,183, 197, 218, 236, 247, 253, 254,258–260, 289

Signal attenuation, 63Sleep disturbances, 26, 160Social issues, 273–275, 283Steering brain stimulation, 254, 260Stereotactic, 133, 134, 137Stimulation, 21–30

amplitude, 6induced adverse events, 258parameters, 1–6, 24, 54, 67, 78, 98, 187,

283pulse width, 4–6, 98–100, 135recording, 263, 265titration, 30, 99

Subcallosal cingulate (SCC), 22–25, 83, 103Subcallosal cingulate cortex, 71–78Subcaudate white matter, 12Subgenual anterior cingulate cortex, 36, 227Subgenual cingulate gyrus white matter, 11, 12Substance abuse, 89, 137Subthalamic nucleus (STN), 53, 64, 114, 119,

121, 122, 125, 153, 161, 175, 207cognitive and emotional functions of the,

54limbic, associative and motor territories in

the, 54Suicide, 75, 86, 122, 167, 175Surgical treatment, 123

TTargeted, 121, 125Targets, 115, 125

298 Index

Thalamic stimulation, 200Thalamus, 16, 114, 115, 119, 120, 123, 125,

234, 235Therapeutic mechanisms of DBS, 30Therapy resistant depression, 81Threat, 37–39, 220

monitoring processes, 40Tics, 113–125Tourette syndrome (TS), 113, 225Tractography, 22, 27–30Translational

approaches, 247research, 217–223studies, 62, 105, 245

Treatment-resistant patients, 35Treatment-resistant depression (TRD),

71–76, 78

UUncinate fasciculus, 13–18Unilateral stimulation, 98, 108, 153, 235

VVentral anterior internal capsule, 12, 17Ventral anterior internal capsule/ventral stria-

tum (VC/VS), 22, 29, 95–100, 227Ventral capsule/ventral striatum (VC/VS),

25–27, 36, 95Ventral medial prefrontal cortex, 17, 55, 107Ventral prefrontal cortex, 13Ventral striatum, 12, 17, 82Ventral tegmental area (VTA), 83, 105, 145,

222Voltammetry, 264, 267Volume of activation (VOA), 255

WWhite matter tracts, 12–15, 27, 36Wireless instantaneous neurotransmitter

concentration sensor (WINCS),196–199

Index 299