Deep Brain Stimulation How Does It Work

8/13/2019 Deep Brain Stimulation How Does It Work

1/7

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 75 SUPPLEMENT 2 MARCH 2008 S59

JERROLD L. VITEK, MD, PhDDirector, Neuromodulation Research Center,

Department of Neurosciences, Lerner Research Institute,

Cleveland Clinic, Cleveland, OH

Deep brain stimulation: How does it work?

ABSTRACT

Deep brain stimulation has significantly improved themotor symptoms in patients with Parkinsons disease(PD) and other movement disorders. The mechanismsresponsible for these improvements continue to beexplored. Inhibition at the site of stimulation hasbeen the prevailing explanation for the symptom

improvement observed with deep brain stimulation.Research using microelectrode recording during deepbrain stimulation in the MPTP monkey model of PDhas helped clarify how electrical stimulation of struc-tures within the basal gangliathalamocortical circuitimproves motor symptoms, and suggests that activa-tion of output and the resultant change in pattern ofneuronal activity that permeates throughout thebasal ganglia motor circuit is the mechanism respon-sible for symptom improvement.

Whether deep brain stimulation can dramat-

ically help patients with Parkinsons dis-ease (PD) and other movement disordersis no longer questioned. Rather, how it

works is not well understood: how do patients withseemingly diverse conditions show improvement withthe same intervention?

Patients with advanced PD often freeze when try-ing to walk and have tremor, rigidity, bradykinesia,and gait and balance problems. With deep brain stim-ulation, a patient typically experiences a markedimprovement in these motor symptoms.

Similarly, patients with hypokinetic disorders suchas generalized dystonia who have extensive involun-tary movements involving multiple body parts mayexperience a significant reduction in these movementsand regain function during deep brain stimulation. Inmy experience, it is not unusual for patients who werenot ambulatory as a result of their dystonic move-ments to regain function to the point where they can

walk unassisted and, in some cases, participate in phys-ical activities such as racquetball or jogging on a tread-mill. One of my patients with generalized dystoniacould walk no farther than several meters before deepbrain stimulation but afterward was able to run on atreadmill. This patient did not gain this type of func-tion immediately after stimulation, but after sustainedefforts at programming his stimulation device over the

course of 1 year he was able to travel to Europe, hikein the mountains, and jog on a treadmill.

In addition to treating movement disorders, deepbrain stimulation is being used experimentally to treatpatients with behavioral disorders such as depressionand obsessive-compulsive disorder that are refractiveto standard therapy. Broadening our understanding ofthe mechanisms responsible for success with deepbrain stimulation is important since it may help toimprove current applications and develop new ones.This article discusses our research in deep brain stim-ulation using microelectrode recording of structureswithin the basal gangliathalamocortical circuit inthe MPTP monkey model of PD.

INSIGHTS INTO MECHANISMS OF STIMULATIONPROMISE TECHNOLOGICAL REFINEMENTS

One rationale for attempting to better understandhow deep brain stimulation works is that such knowl-edge may enable us to improve the technology to bet-ter apply the technique.

Electrode design is one important area of potentialimprovement. Diseases that may one day be treatedwith deep brain stimulation will likely require elec-trodes of different shapes than those used currently, toaccommodate other targets in the brain. At present, asingle lead shape is used to stimulate the subthalamicnucleus (STN) and the globus pallidus internus (GPi)for treating PD. Possible future targets include theglobus pallidus externus (GPe), various subnuclei of thethalamus, portions of the striatum, and other subcorti-cal and cortical structures that have different geometricconfigurations and physiologic characteristics. Sincethese structures and regions of the brain differ from oneanother in size and shape, it is highly likely that newelectrode designs will be needed to take advantage of

Dr. Vitek reported that he serves as a consultant and board member forAdvanced Neuromodulation Systems, Inc., and serves as a consultant forMedtronic, Inc., from which he has also received fees for teaching/speaking.


2/7

this geometric and physiologic variability. Future elec-trodes may vary in size and shape from those used cur-rently, incorporate three-dimensional designs, andrequire a current source that allows the pattern of stim-ulation to be varied based on the physiologic changesthat characterize each neurologic disorder.

Directionality may be another important feature ofelectrode design. With presently used electrodes,electric current spreads in all directions. To spread thecurrent or increase the volume of tissue affected bystimulation, one must increase the voltage beingpassed through the lead. This results in a larger regionof tissue being affected by stimulation, but the currentdensity varies based on distance from the stimulationsite, with neural tissue close to the site being affecteddifferently from tissue that is farther away. Moreover,the current cannot be directed or aimed in one direc-tion or the other. A split-band design could spread

current in opposing directions, and a three-dimen-sional directional design involving several contactscould affect a volume of tissue more homogeneously.

PROGRESS IN DEFINING PD PATHOPHYSIOLOGY

As with any disease, defining the problem and under-standing the underlying pathophysiology are essentialfirst steps to finding an effective treatment for PD. Inthe 1930s and 1940s, numerous attempts were made totreat PD with surgical therapies. Surgical targets werechosen throughout the length of the neuraxis, includ-

ing the cortex, the internal capsule, the basal ganglia,the thalamus, the cerebral peduncle, and the spinalcord itself. The underlying pathophysiology was notwell understood, however, so the rationale for surgerywas weak at best. For example, lesioning the corteximproved parkinsonian tremor, but it also caused paral-ysis and was associated with considerable morbidity.

Evidence of a common circuitOver time, a number of anatomic and physiologic stud-ies provided evidence that there may be a commonanatomy or circuit that malfunctions between thediverse disorders that are now improved with deepbrain stimulation. It is now recognized that PD anddystoniadisorders that involve a paucity of move-ment and excessive movement, respectivelybothresult from disorders of the basal ganglia. Similarly, thebasal gangliathalamocortical circuit appears to playan integral role in behavioral disorders such as depres-

sion, schizophrenia, autism, and obsessive-compulsivedisorder. This basal gangliathalamocortical circuitincludes connections from the cortex, through thebasal ganglia, and back to the cortex through the thal-amus (Figure 1). Different regions within nodal points(striatum, GPe, GPi, STN, thalamus) of the circuitaffect movement, cognition, and behavior, so that mal-function in different regions of each nodal point in thecircuit may result in different neurologic disorders.

In PD, degeneration of dopamine-producing neurons

DEEP BRAIN STIMULATION

Cortex

CM VA/VL

GPi/SNr

Striatum

GPe

STN

SNc

PPN

FIGURE 1. Schematic diagramof the basal gangliathalamo-cortical circuitry under normaland parkinsonian conditions.Inhibitory connections are shownas black arrows, excitatoryconnections as gray arrows.Parkinsonism leads to differen-tial changes in the two striato-pallidal projections, which areindicated by the thickness ofthe connecting arrows. Basalganglia output to the thalamusis increased.

Reprinted, with permission,from Wichmann T, et al. Basal ganglia:anatomy and physiology. In: Factor SA,

Weiner WJ, eds. Parkinsons Disease:Diagnosis and Clinical Management,2nd ed.

New York, NY: Demos; 2008:255.

Brainstem/spinal cord

Cortex

CM VA/VL

GPi/SNr

Striatum

GPe

STN

SNc

PPNBrainstem/spinal cord

GPe = external segment of the globus pallidusGPi = internal segment of the globus pallidusSTN = subthalamic nucleus

SNr = substantia nigra pars reticulataSNc = substantia nigra pars compactaCM = centromedian nucleus of thalamus

Normal Parkinsonism

VA/VL = ventroanterior and ventrolateralnuclei of thalamus

PPN = pedunculopontine nucleus

S60 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 75 SUPPLEMENT 2 MARCH 2008


3/7

in the substantia nigra pars compacta reduces dopaminelevels in the striatum. In MPTP monkey models of PDthere is also a loss of dopamine-producing cells in thesubstantia nigra pars compacta. These animals develop

the cardinal motor symptoms of PD and are considereda good model of the human disorder. By recording fromthe basal gangliathalamocortical circuit in this model,we and others have observed excessive activity in theSTN and GPi.14 In addition, cells in these regions inthe monkey model were more likely to discharge inbursts compared with cells from healthy monkeys, andthey showed a higher degree of synchronized oscillatoryactivity among neighboring neurons.5,6

Ultimate goal: The ability to individualize therapyUnderstanding how such changes relate to parkinson-ian symptoms will enable us to develop stimulation

strategies that are focused on ameliorating the particu-lar physiologic changes in PD. Since PD can lead todistinctly different clinical pictures, it would be ideal tobe able to individualize therapy based on the particularmotor symptoms each patient experiences. This mayrequire stimulation strategies that affect either a partic-ular region of the targeted structure or a particularphysiologic change that occurs in the disease state.

THE RATE HYPOTHESIS:ALTERED CELLULAR DISCHARGE RATESCAUSE PARKINSONIAN MOTOR SYMPTOMS

A good model for PD was lacking prior to the 1980s.As a result, there was little understanding of thepathophysiologic basis for this disorder. A break-through in the mid-1980s revolutionized research inthis field. A group of young people developed parkin-sonian symptoms, and it was discovered that they hadall used recreational designer drugs containing animpurity: the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). Now given to primatesto simulate PD, MPTP causes all of the classic symp-toms of PD except tremor (this may vary from speciesto species), including freezing, slowness, stiffness, andgait and balance problems. Like humans with PD, pri-

mates with MPTP-induced PD even develop dyskine-sia after prolonged treatment with levodopa.

Experimentation with MPTP monkeys in the late1980s led to the rate hypothesis, which basicallystates that when dopamine production is reducedfrom the substantia nigra compacta (as in PD),changes in striatal activity lead to suppression of GPeactivity and a reduction in inhibitory output from theGPe to the STN. This decrease in inhibitory outputallows the STN to be overactive, which, in conjunc-

tion with a reduction of direct striatal inhibition ofthe GPi, causes excessive GPi activity and a suppres-sion of thalamic activity to the cortex (Figure 1).

When recording electrodes were placed in these

structures in the monkey brain, rate changes werereported to occur in each of these structures in theparkinsonian state.14,7 Action potentials recordedfrom the GPi in MPTP-treated monkeys occurred ata much faster rate than those in healthy monkeys.

Pallidotomy revisited:Dramatic symptom improvement is possibleOn the basis of the above and other studies in theMPTP monkey model of PD, investigators in the1990s reasoned that reduced dopamine in PD led toexcessive activity in portions of this circuit. While Iwould like to say that this led to the rationale forlesioning the STN and GPi for the treatment of PD,this approach had already been taken in the early1930s and 1940s and continued into the 1960s; it waslargely stopped with the introduction of levodopa andwas restarted again after the realization that chroniclevodopa therapy was associated with a variety of sideeffects, including the development of excessive invol-untary movement and motor fluctuations.

Pallidotomy (lesioning of the pallidum), althoughtried as a treatment for PD in the 1930s and 1940s, hadbeen abandoned as a result of its inconsistent benefitand lack of effect on parkinsonian tremor. It underwent

a resurgence in the 1990s through the work of a groupin New York8 that revived Lars Leksells pallidotomyapproach of the 1960s9 at a time when basic sciencestudies provided the rationale for surgical therapy tocreate lesions in the GPi. These basic science studiesalso provided critical new information about the opti-mal site for lesioning, which led to improved and moreconsistent outcomes.1013 In the early years, lesions werecreated in the anterior (nonmotor) portion of the pal-lidum but led to inconsistent results. In the 1990s, witha better understanding of the portion of the palliduminvolved in motor control, destroying brain tissue bycreating a lesion in the posterolateral motor region of

the pallidum resulted in such dramatic improvement inmotor signs that waiting lists of up to 4 years were com-mon for patients who wanted the procedure.

Although unilateral pallidotomy led to markedimprovement in motor symptoms on the contralateralside, attempts at bilateral lesions to improve bothsides of the body, as well as axial symptoms, were asso-ciated with marked hypophonia and, in some reports,cognitive decline. This led physicians and scientiststo search for a procedure that could be performed


VITEK


4/7

bilaterally without the high incidence of side effectsassociated with lesioning proceduresand thus to thebirth of deep brain stimulation.

Deep brain stimulation as lesion simulationDuring the early experience with pallidotomy, the areato be lesioned would first be stimulated with the lesion-ing probe to observe its effects and thereby determinethe precise area in which to create a lesion. At thetime, no mechanism existed to leave the stimulator inplace rather than create a lesion. But after the develop-ment of implantable stimulation devices, chronic stim-ulation could be delivered bilaterally to the pallidumand STN, resulting in a markedly improved treatment.Since side effects associated with stimulation arereversible, the ability to perform such procedures onboth sides of the brain and to adjust stimulation param-

eters in order to optimize benefits while minimizingside effects made deep brain stimulation the procedureof choice for patients with advanced PD and led to itsexploration for treatment of other neurologic disorders.

Because stimulation produced the same or similarbenefit as a lesion, most physicians thought that stimu-lation must work in a similar manner, ie, by decreasingoutput from the stimulated structure. The rationale forthis hypothesis received support from the rate modelof PD, which postulated that PD motor symptoms

occur as a result of overactivity in the STN and GPi. Itwas postulated that deep brain stimulation improvedclinical symptoms by suppressing output from the stim-ulated structurein other words, deep brain stimula-

tion effectively caused a physiologic ablation.14,15

FURTHER RESEARCH GIVES RISE TOTHE PATTERN HYPOTHESIS

Deep brain stimulation in the monkey modelTo test the effects of deep brain stimulation, we haveperformed it in primates with MPTP-induced parkin-sonism. Custom-made leads sized to fit a monkey brainare implanted in the same deep brain structures thatare targeted when treating PD in humans. Each animallead has four contacts 0.5 mm in size. We implant apulse generator, connect the pulse generator to thelead, and set stimulation parameters to improve motorsymptoms to mimic a human therapeutic setting asclosely as possible. We then record from the basal gan-glia structures before, during, and after stimulation thatimproves the monkeys motor symptoms. This allows usto determine which changes in neuronal activity in thebasal ganglia circuit during stimulation are associatedwith an improvement in motor symptoms.

Chamber placement and orientation as well as leadplacement are determined with the help of a softwareprogram and information from magnetic resonanceimaging and computed tomography, similar to theprocess for neurosurgery in humans.16 The software

also allows for mapping the location of every cell fromwhich recordings are taken (Figure 2).In earlier studies examining the mechanism under-

lying deep brain stimulation, neural activity wasrecorded only after stimulation, so that activity thatoccurred during stimulation had to be inferred fromthat which occurred immediately after stimulation wasstopped. We developed a method to subtract artifactproduced from stimulation without losing data. Thismethod has been validated, is now used in a number oflaboratories, and has revolutionized our ability to studythe effect of stimulation on neuronal activity.17

A paradoxical finding

Based on the rate hypothesis, we expected thatincreased output from the GPi would cause parkinson-ian symptoms and predicted that stimulation of theSTN should suppress its output, which would suppressexcitatory activity to the GPi from the STN and therebyreduce its output. Reduction of the inhibitory outputfrom the GPi to the thalamus would, in turn, lead to arestoration of thalamocortical function and a reductionin the motor signs associated with PD. However, stimu-lating the STN was found to increase GPi activity.18


FIGURE 2. Image generated by software designed to assist inlead placement in a monkey model of Parkinsons disease.16 Thevarious subcortical structures are represented in different colors.In this example, the thalamus is yellow, the subthalamic nucleus isgray, the globus pallidus externus (GPe) is purple, and the globuspallidus internus (GPi) is red/pink. The lead is passing through theGPe and GPi with the contacts denoted by the purple bands. Eachcell from which recordings are made is denoted by a white symbol.



5/7

Despite increased rates, the incidence and intensity ofsymptoms were reduced. Further complicating the pic-ture, we were contemporaneously exploring the effect ofcreating lesions in other parts of the basal ganglia thatalso led to increased rates of GPi activity, but in thiscase we observed that the increased rates were associatedwith a worsening of motor symptoms. In short, we hadtwo laboratories working in parallel that had apparentlyobtained opposite results: increased GPi activity wasassociated with improved symptoms in one laboratoryand with worse symptoms in the other.18,19

Patterns of activity are more important than rateThis seeming paradox may be explained by evaluatingthe data with a post-stimulus time histogram (Figure 3).

Simple recordings of activity show seemingly randomaction potentials over time; however, if activity isrecorded repeatedly during stimulation and the overalldata are averaged, action potentials are observed tooccur in a definite pattern, with action potentials in GPineurons occurring mainly at 3 ms and 6 ms after a stim-ulation pulse in the STN. The number of cells showinga particular pattern of response could be changed byvarying the stimulation parameters. This shift in thepopulation of neurons that showed such a stereotyped

pattern of response under stimulation parameters thatimproved motor symptoms may offer part of the expla-nation for our apparent paradox: stimulation thatimproved motor symptoms regularized that spike train,while the lesions we produced in the GPe that increasedthe rate did not change the irregularity in the spiketrain. These observations provided compelling data tosupport the hypothesis that motor symptoms associatedwith PD, and possibly other movement and nonmove-ment disorders, may occur as a result of changes in thepattern of neuronal activity rather than changes in rate.

Knowledge that stimulation activated output fromthe stimulated region and changed the pattern of neu-ronal activity led us to ponder whether other targets,

or even other ways to deliver stimulation, might workbetter to improve parkinsonian symptoms.

A focus on GPe stimulationAs a result of these observations, we reasoned thatsince GPe activity is also altered in PD and its rates arereduced, driving the output from this region that isinhibitory to the STN and GPi may help to reduce andregularize that activity at a point in the circuit thatcould provide even greater improvement in the motorsymptoms associated with PD. Based on this hypothe-


VITEK

A Globus pallidus internus (GPi) B Globus pallidus externus (GPe)

FIGURE 3. Examples of neuronal responses occurring during subthalamic nucleus stimulation in (A) a GPi cell and (B) a GPe cell.Top: Analog signal overlays of 100 sweeps made by triggering at 10-ms intervals in the prestimulation period (before start of stimulation) and bytriggering on the stimulation pulse in the on-stimulation period. Arrows indicate residual stimulation artifacts after artifact template subtraction.Middle: Peristimulus time histograms (PSTHs) reconstructed from successive 7.0-ms time periods in the prestimulation period and from the inter-stimulus periods (7.3 ms) in the on-stimulation period. The first PSTH bin is omitted in the on-stimulation period because of signal saturation andresidual stimulation artifacts.Asterisks denote a significant increase at P< .01, and daggers denote a significant decrease at P< .01 (Wilcoxonsigned rank test). Bottom: Mean firing rate calculated every 1 sec on the basis of the PSTH, illustrating the time course of the firing rate.Reprinted,with permission, from Hashimoto T, et al. Stimulation of the subthalamic nucleus changes the firing pattern of pallidal neurons. J Neurosci 2003; 23:19161923.Copyright 2003 by Society for Neuroscience.


6/7

sis, we performed direct stimulation of the GPe in theMPTP monkey model of PD and evaluated its effect onmotor behavior and neuronal activity in the circuit.

As an interesting sidelight, it should be noted that

long before we developed this hypothesis, we hadobservations from a 1994 experiment (only recentlypublished20) in which bradykinesia was improved uponacute stimulation in the GPe prior to making a lesionin the GPi. With sustained stimulation in this patient,we observed development of dyskinetic movements.Since we reasoned that lesions in this region wouldworsen parkinsonian symptomsa rationale recentlysupported by a publication from our laboratory in200619and since we had no means by which to stim-ulate this region chronically at the time, this observa-tion was filed away and we continued with lesioningthe GPi for the treatment of these patients.

However, with the advent of chronic deep brainstimulation, we opted to reexplore this series of experi-ments in MPTP-treated monkeys. A lead was placedsuch that three of its contacts were in the GPe and onewas in the GPi. Bradykinesia was assessed by determin-ing the time it took for the monkey to retrieve raisinsfrom a Klver board. By inducing symptoms on one sideonly, we were able to use the healthy side as a control.We observed that before stimulation, retrieval tookmore than twice as long on the affected side.Stimulation of only 2 V had no effect, but increasing thevoltage to 5.5 V significantly improved retrieval time.21

Plotting the data using post-stimulus time his-tograms showed that stimulation of the GPe inhibitedthe STN, confirming our hypothesis that stimulationactivated the output from the stimulated structure(the GPe sends inhibitory projections to the STN).The responses observed were dramatic, with themajority of cells in the STN showing almost completesuppression of activity (Vitek et al, unpublished data).

In light of this observation, we expected that the rateof activity in the GPi would be reduced. Interestingly,although the rate was changed in most cells comparedwith control, what was most striking was the relativelystereotyped pattern of inhibition and excitation that

occurred following each pulse of GPe stimulation.Although shifted in absolute frequency, the pattern thatoccurred was similar to that observed during STN stim-ulation, with alternating periods of excitation and inhi-bition evident in the post-stimulus time histogram.

Further evaluation of the data revealed a change inburst and oscillatory activity in the STN. Analysis ofthe data showed a shift in the distribution of power fromlow to high frequencies. Stimulation reduced activity inthe low-frequency range and increased power in higher

frequencies, similar to that in normal movement.Further analysis of the spike trains revealed that

entropy (a reflection of noise in the spike signal) wasreduced under stimulation parameters that resulted in

a reduction in symptoms. In contrast, stimulationparameters that resulted in worsening symptomsincreased measures of entropy (Dorval, data submit-ted for publication).

PATTERN CHANGES AFFECT INFORMATIONPROCESSING ACROSS THE BASAL GANGLIATHALAMOCORTICAL NETWORK

There is a lack of consensus about the precise physio-logic effect of deep brain stimulation for improvingsymptoms in movement disorders. Many researcherscontinue to believe that deep brain stimulation worksthrough inhibition. An alternate explanation is thatat effective stimulation parameters, the net effect isactivation of output from the stimulated structure.Various modalities, including modeling,22,23 microdial-ysis,24 functional magnetic resonance imaging,25 andpositron emission tomography,26,27 provide additionalevidence that activation occurs during stimulation.

While one cannot discount a role for rate changes inmediating the effects of deep brain stimulation, there isnow increasing evidence suggesting that patternchanges induced in the network as a result of stimula-tion-induced activation of output from the stimulatedstructure play an integral role in this process.

Research often leads to unpredictable outcomes.The prevailing hypothesis a decade ago concerning thepathophysiologic basis of PD (and still believed inmany centers) was that rate is the controlling factor.But we have seen in our animal models that symptomsimprove with increased rate in the GPi during stimula-tion in the STN. Similarly, GPi rates are abnormallylow in patients with dystonia and in PD patients duringdyskinesia, yet lesioning in the GPi that further reducesits output leads to improvement in these conditions.Based on these observations, it would appear that rateis unlikely to be the critical factor; we now must takeinto account other factors, such as pattern, oscillation,

and synchronization, as well as changes in the networkdynamics. Deep brain stimulation is changing theinformational content of the neural network, andthese changes are occurring across populations of neu-rons through the whole basal ganglia circuit. Knowinghow these changes result in improvement in the neu-rologic disorder being treated will be critical to ourunderstanding of not only how deep brain stimulationworks, but how to make it work better and how toapply it effectively to other neurologic disorders.




7/7

FUTURE DIRECTIONS

Future research should focus on multiunit recordingsimultaneously across nodal points in the basal gangliathalamocortical circuit to assess population and net-work dynamics. This approach would provide infor-mation on the real-time effects of stimulation in thenetwork. Until now, most studies have collectedrecordings from one cell at a time. This is a very labor-intensive process and limits our ability to relate whathappens at one point in the circuit to what happens atanother point. Multiunit recording across multiplenodes within the basal gangliathalamocortical circuitwill help us address this question and tell us what hap-pens across populations of neurons at multiple sites inthe motor circuit and how this is changed during stim-ulation. Such an approach will help us to better under-

stand the pathophysiologic basis for the developmentof neurologic disorders and how stimulation works toimprove these disorders. This information is a criticalstep toward the ability to knowingly change networkactivity in a way that is predictable and more compat-ible with the normal state, as well as toward the appli-cation of this technology to other disorders.

The potential for clinical applications of deepbrain stimulation is dramatic, but we must proceedwith caution. Indications should be based on soundscientific rationale, and outcomes must be accuratelyand systematically documented. Move forward wemust, but with cautionmost certainly.

AcknowledgmentsThe author thanks Dr. Jianyu Zhang for his work in preparing Figure 2 andDrs. Svjetlana Miocinovic and Cameron McIntyre for their work in devel-oping the software program Cicerone that was used to prepare this figure.The author also thanks Drs.Takao Hashimoto, Jianyu Zhang, and WeidongXu for their vital contributions to our deep brain stimulation researchprogram, without which none of this work would have been possible.

REFERENCES1. DeLong MR. Primate models of movement disorders of basal gan-

glia origin. Trends Neurosci 1990; 13:281285.2. Filion M, Tremblay L. Abnormal spontaneous activity of globus

pallidus neurons in monkeys with MPTP-induced parkinsonism.Brain Res 1991; 547:142151.

3. Miller WC, DeLong MR. Altered tonic activity of neurons in the

globus pallidus and subthalamic nucleus in the primate MPTP modelof parkinsonism. In: Carpenter MB, Jayaraman A, eds. The BasalGanglia II. Structure and Function: Current Concepts. New York,

NY: Plenum; 1987:415427.4. Bergman H, Wichmann T, Karmon B, DeLong MR. The primate

subthalamic nucleus. II. Neuronal activity in the MPTP model ofparkinsonism. J Neurophysiol 1994; 72:507520.

5. Nini A, Feingold A, Slovin H, Bergman H.Neurons in the globuspallidus do not show correlated activity in the normal monkey, butphase-locked oscillations appear in the MPTP model of parkinson-ism. J Neurophysiol 1995; 74:18001805.

6. Raz A, Vaadia E, Bergman H. Firing patterns and correlations ofspontaneous discharge of pallidal neurons in the normal and the

tremulous 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine vervetmodel of parkinsonism. J Neurosci 2000; 20:85598571.

7. Elder C, Vitek J. The motor thalamus: alteration of neuronal activ-ity in the parkinsonian state. In: Kultas-Ilinsky K, Ilinsky IA, eds.Basal Ganglia and Thalamus in Health and Movement Disorders.

New York, NY: Kluwer Academic Plenum Publishers; 2001:257265.8. Fazzini E, Dogali M, Sterio D, Eidelberg D, Beric A. Stereotactic

pallidotomy for Parkinsons disease: a long-term follow-up of unilat-eral pallidotomy. Neurology 1997; 48:12731277.

9. Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of parkin-sonism by stereotactic thermolesions in the pallidal region. A clini-cal evaluation of 81 cases. Acta Psychiatr Scand 1960; 35:358377.

10. Vitek JL, Bakay RA, Freeman A, et al. Randomized trial of palli-dotomy versus medical therapy for Parkinsons disease. Ann Neurol2003; 53:558569.

11. Dogali M, Fazzini E, Kolodny E, et al. Stereotactic ventral pallido-tomy for Parkinsons disease. Neurology 1995; 45:753761.

12. Lozano AM, Lang AF, Galvez-Jimenez N, et al. Effect of GPi pallidotomyon motor function in Parkinsons disease. Lancet 1995; 346:13831387.

13. Baron MS, Vitek JL, Bakay RA, et al. Treatment of advancedParkinsons disease by posterior GPi pallidotomy: 1-year results of apilot study. Ann Neurol 1996; 40:355366.

14. Benazzouz A, Hallett M. Mechanism of action of deep brain stim-ulation. Neurology 2000; 55(12 Suppl 6):S13S16.

15. Dostrovsky JO, Levy R, Wu JP, Hutchison WD, Tasker RR,Lozano AM. Microstimulation-induced inhibition of neuronal fir-ing in human globus pallidus. J Neurophysiol 2000; 84:570574.

16. Miocinovic S, Zhang J, Xu W, et al. Stereotactic neurosurgicalplanning, recording, and visualization for deep brain stimulation innonhuman primates. J Neurosci Methods 2007; 162:3241.

17. Hashimoto T, Elder CM, Vitek JL. A template subtraction methodfor stimulus artifact removal in high-frequency deep brain stimula-tion. J Neurosci Methods 2002; 113:181186.

18. Hashimoto T, Elder CM, Okun MS, Patrick SK, Vitek JL.Stimulation of the subthalamic nucleus changes the firing pattern ofpallidal neurons. J Neurosci 2003; 23:19161923.

19. Zhang J, Russo GS, Mewes K, Rye DB, Vitek JL. Lesions in mon-key globus pallidus externus exacerbate parkinsonian symptoms.Exp Neurol 2006; 199:446453.

20. Vitek JL, Hashimoto T, Peoples J, DeLong MR, Bakay RA. Acutestimulation in the external segment of the globus pallidus improvesparkinsonian motor signs. Mov Disord 2004; 19:907915.

21. Zhang J, Russo GS, Chen X, Hashimoto T, Elder CM, Vitek JL.Deepbrain stimulation of monkey globus pallidus externus in experimentalparkinsonism. Abstract presented at: 33rd Annual Meeting of theSociety for Neuroscience; November 812, 2003; New Orleans, LA.

22. McIntyre CC, Grill WM. Extracellular stimulation of central neu-rons: influence of stimulus waveform and frequency on neuronal out-put. J Neurophysiol 2002; 88:15921604.

23. McIntyre CC, Grill WM, Sherman DL, Thakor NV. Cellulareffects of deep brain stimulation: model-based analysis of activationand inhibition. J Neurophysiol 2004; 91:14571469.

24. Windels F, Bruet N, Poupard A, et al. Influence of the frequencyparameter on extracellular glutamate and gamma-aminobutyric acidin substantia nigra and globus pallidus during electrical stimulationof subthalamic nucleus in rats. J Neurosci Res 2003; 72:259267.

25. Jech R, Urgosik D, Tintera J, et al. Functional magnetic resonanceimaging during deep brain stimulation: a pilot study in four patientswith Parkinsons disease. Mov Disord 2001; 16:11261132.

26. Perlmutter JS, Mink JW, Bastian AJ, et al. Blood flow responses todeep brain stimulation of thalamus. Neurology 2002; 58:13881394.

27. Hershey T, Revilla FJ, Wernle AR, et al. Cortical and subcorticalblood flow effects of subthalamic nucleus stimulation in PD.

Neurology 2003; 61:816821.

Correspondence:Jerrold L. Vitek, MD, PhD, Department of Neuro-

sciences, Cleveland Clinic, 9500 Euclid Avenue, NC30, Cleveland,OH 44195; [email protected].


VITEK

Documents

Deep Brain Stimulation How Does It Work