targeting in dbs - semantic scholar · 2017-03-21 · a large therapeutic window between beneficial...
TRANSCRIPT
Yurui Gao, BME, Vanderbilt U
Targeting in DBS
Yurui Gao
Department of Biomedical Engineering, Vanderbilt University
CONTENT
WHAT IS DBS? ........................................................................................................................................... 2
History of DBS ....................................................................................................................................... 2
Hardware of DBS ................................................................................................................................... 3
Surgical Procedure of DBS ..................................................................................................................... 4
Mechanisms of DBS............................................................................................................................... 5
DBS vs. ablation ..................................................................................................................................... 7
WHAT ARE THE TARGETS IN DBS? ............................................................................................................ 8
Ventral Thalamic Nuclei ........................................................................................................................ 9
Subthalamic Nucleus ............................................................................................................................. 9
Globus Pallidus pars interna ............................................................................................................... 10
HOW TO TARGETING IN DBS? ................................................................................................................. 11
By brain atlas ....................................................................................................................................... 12
By visible surrounding anatomical landmarks .................................................................................... 14
By direct locating (SWI/7T MRI/DTI) ................................................................................................... 15
By microelectrode recordings ............................................................................................................. 16
By macro stimulation tests ................................................................................................................. 18
References .................................................................................................................................................. 20
Abbreviations
DBS: deep brain stimulation MRI: Magnetic Resonance Imaging CNS: central nerve system PD: Parkinson disease Vim: ventralis intermedius nucleus ET: essential tremor STN: subthalamic nucleus FDA: Food and Drug Administration GPi: globus pallidus pars interna TS: Tourette syndrome GPe: globus Pallidus pars externa OCD: obsessive compulsive disorder SNr: substantia nigra reticulata IPG: internal pulse generator
Yurui Gao, BME, Vanderbilt U
WHAT IS DBS?
DBS (deep brain stimulation) is a surgical treatment involving the implantation of brain
pacemaker, which sends electrical impulses to target structure deep in the brain for controlling
specific neural activity. The FDA approved DBS as a treatment for ET, PD and dystonia. DBS is
also commonly used to treat chronic pain as well as depression.
History of DBS
The modulation of brain activity by direct electrical stimulation of the brain can be traced to
1870[ (1)]. It was shown that electrical stimulation of the motor cortex in dogs can elicit limb
movement. Four years later, electrical stimulation was introduced as a tool for improving
neurosurgical procedures in humans [ (2)]. The emergence and development of human
stereotaxic devices provided neurosurgeons a chance to investigate the effects of stimulating
deeper cortical structures [ (3)] as well as to identify these deeper structures by observing the
effects of stimulation. Although stimulation had intraoperative use before 1960, its goal
focused on delineating the ‘best’ target for a subsequence lesion.
In the early 1960s, high-frequency (100Hz) stimulation of the ventrolateral thalamus could
diminish tremor [ (4)]. In the early 1970s, it is reported that chronic stimulation was used to
therapeutically treat pain [ (5)], movement disorders, or epilepsy [ (6)]. It took decades of years
to implement combining the implantable pacemaker technology with chronically implanted
deep brain electrode for long-term chronic DBS [ (7)]. Since then, DBS has become increasingly
used for treating a variety of disorders. The first widespread use of DBS in the US and Europe
was for the treatment of ET or the tremor of PD. Also DBS has been reported for treating
Yurui Gao, BME, Vanderbilt U
Fig.1. DBS hardware in patient http://www.riversideonline.com/source/images/image_popup/r7_deepbrain.jpght
tp://professional.medtronic.com/downloads/deep-brain-stimulation
/128_3389LeadExtension2.jpg
Lead
IPG
Extension
Lead
Extension
dystonia involved the thalamus [ (8)] and the GPi [ (9)]. There have been a few recent reposts of
DBS for TS [ (10)] and OCD [ (11)]. Furthermore, the advancements in neuroimaging, especially
MRI and intraoperative electrophysiological recordings have increased stereotaxic surgical
targeting accuracy.
Hardware of DBS
Figure 1 displays the major components of DBS hardware: the implanted pulse generator (IPG),
extension and lead (also called electrode). The IPG is a battery-power neurostimulator encased
in a titanium housing, which sends electrical pulses to the brain interfere with neural activity at
the target sites. DBS leads are 4 thin coiled wires insulated in polyurethane. Each wire ends in a
1.5mm platinum iridium electrode, resulting in 4 electrodes ending at the tip of the lead. These
quadripolar leads is implanted in
the brain with the electrodes
positioning at the target cerebral
sites and thus they deliver
stimulation using either one
electrode or a combination of
electrodes. IPG and lead are
connected by extension, an insulated wire that runs from the head, down the side of the neck,
behind the ear to the IPG, which is placed subcutaneously below the clavicle or abdomen. The
stretch coil extension is 15% extensible, which accommodates the natural movement of the
head, neck and shoulders.
Yurui Gao, BME, Vanderbilt U
All the hardware of DBS is surgically implanted inside the body. Under local anesthesia, a
hole about 14mm in diameter is drilled in the skull and the electrode is inserted, with feedback
from the patient for optimal placement. The placement of the IPG and lead then occurs under
general anesthesia.
Surgical Procedure of DBS
The surgical procedure of DBS is similar to the ablative stereotactic neurosurgery, except the
final stage of electrode implantation.
First, a stereotactic head frame is fixed to the skull to establish a 3D coordinate system to
allow correlation of the preoperative imaging with each point in the brain and to prevent
intraoperative head movement. Holloway and colleagues [ (12)] proposed frameless stereotaxy
using bone fiducial markers to improve patient comfort. Thereafter CT and/or MRI scanning are
performed to identify the anterior and posterior commissures (also ventriculography). The
target sites of DBS may be or may not be directly indentified in the CT and/ or MRI image due to
the limit of resolution, therefore, direct or indirect targeting methods depending on imaging
process help to locate the target structure in patient’s brain.
Microelectrode recording (MER) and mircostimulation and/or macrostimulation is
performed to determine the best possible location for the DBS electrodes: optimal benefit and
a large therapeutic window between beneficial and adverse effects. After the optimal location
for implantation of the electrode has been decided, the DBS lead (macroelectrode) is secured
to the skull with a plastic cap, metal plate, or cement. The IPG can be implanted the same day.
Yurui Gao, BME, Vanderbilt U
Fig. 2. Surgical procedure of DBS in MRI [ (36)].
Stimulation settings are adjusted by means of a programmer placed on the skin overlying
the IPG. The adjustable stimulation parameters include frequency (0-185Hz), pulse width (60-
450sµ), voltage (0-10.5V), and stimulating contacts (monopolar or bipolar stimulation).
Mechanisms of DBS
Although the application of DBS has a long history, the therapeutic mechanism of DBS is
currently incompletely understood. In addition, no single mechanism has emerged to account
for the effect of DBS in different brain regions and in different diseases. The effects of DBS have
been well documented to be frequency dependent, with the greatest relief of symptoms at
>100 Hz and no therapeutic relief at <50 Hz. Stimulation pulse width (duration) determines
which neural elements are preferentially affected; longer pulse widths (1-10ms) influence the
cell soma, whereas shorter pulse widths (30-200µs) mainly affect axons. Furthermore, DBS
Yurui Gao, BME, Vanderbilt U
results in highly localized stimulation because relatively low currents are used which result in
small current spread (about 2-3 mm with an intensity of 2mA) and the intensity of stimulation
decreases proportionally to the square of the distance.
The general therapeutic stimulation parameters of DBS have been derived primarily by
trial and error, from the almost immediate effects on, such as tremor, rigidity, bradykinesia,
paresthesia and chronic pain in patients during surgery. The exact DBS parameters, including
the stimulus amplitude, duration and frequency band, vary with the treatment and with the
targeted brain region. At present, with most commercially available stimulators, these
parameters can be externally changed and fine-tuned over time. However, DBS allows only
open-loop continuous stimulation, which does not take into account the continuous neural
feedback from the individual patient.
It is suggested that the mechanism is reversible inhibition of the target site since the results
and effects are comparable to ablation. Several studies proposed that high-frequency
stimulation increases the excitatory response from the implanted site [ (13)]. GPi stimulation
may result in activation of GPe GABAergic axons [ (14)] that inhibit GPi neurons. A similar
phenomenon may exist with STN DBS. Two additional hypotheses have been proposed as the
mechanisms of DBS: (1) depolarization
blockade of myelinated axons; (2)
‘neuronal jamming’ whereby activation
of fibers transfers non-physiological
and incomprehensible messages to
Yurui Gao, BME, Vanderbilt U
Fig. 3. Complications of DBS in MRI. Left: Hemorrhagic cortical event. Right: Ischemia
downstream target nuclei, which are then disregarded. It is likely that the neuroanatomical of
various effects of DBS are complex incorporating a combination of various mechanisms (See the
above table: How does DBS work?[ (15)]).
DBS vs. ablation
There are two surgical-based techniques to benefit the affective disorder diseases we
mentioned above: ablation (thalamotomy and pallidotomy) techniques and DBS. DBS appears
to have the same effect as ablation-blocking neuronal transmission in the target nucleus to
improve of tremor and Parkinsonism. However, DBS electrode implantation causes fewer
permanent neurologic adverse effects compared with ablation because much less brain tissue is
destroyed [ (16)]. Radiofrequency ablations may be inadvertently expanded or be misplaced to
involve important structures close to the intended target, which potentially cause permanent
damage of tissue and loss of function. With DBS, stimulation parameters can be finely adjusted
to maximize beneficial results and minimize adverse effects caused by mild current spread to
adjacent structures. DBS electrodes can also be surgically repositioned if they are placed
suboptimally at the first time. Compare to the controllability of DBS, the effects of ablation are
irreversible and unchangeable without additional surgery.
Ablation is usually performed with the patient
awake; however, DBS requires an additional urgery
performed under general anesthesia to implant the IPG
and extension wire. Also, IPG needs to be replaced by
surgery every 2 to 7 years due to the battery out of
Yurui Gao, BME, Vanderbilt U
Fig.4. Targets in DBS. Left: describes the inhibitory and/or excitatory relationship among the targets in DBS. Right: shows the anatomical location of targets in hand-drawn picture [ (35)].
power. Approximately 25% to 30% of patients with DBS experience hardware complication [
(17)], typically mechanical hardware breakage or skin erosion over the hardware, which usually
require replacement of damaged or infected hardware and treatment with antibiotics. In
patients with severe postural instability and frequent falls, ablative surgery may be preferred
since falls may damage or displace DBS hardware. In addition, DBS is much more expensive
procedure compared than ablation; thus in many regions of the world, ablation is the only
affordable surgical option when medication is not effective to the progress of the disease.
WHAT ARE THE TARGETS IN DBS?
The most popular target sites for DBS are the vertralis intermedius nucleus of the thalamus
(Vim), the subthalamic nucleus (STN), and the globus pallidus pars interna (GPi).
Yurui Gao, BME, Vanderbilt U
Ventral Thalamic Nuclei
The cerebellar afferent receiving zone of thalamus (human VIM nucleus) has been the primary
target for the treatment of tremor. These nuclei receive excitatory glutamatergic afferents from
the deep cerebellar nuclei, excitatory glutamatergic afferents from the cerebral cortex, and
inhibitory GABAergic inputs from the reticular nucleus of the thalamus. The output from these
nuclei primarily targets motor areas of cerebral cortex but has also been shown to project to
striatum. Thus, although Vim is commonly viewed as a simple relay for information from the
cerebellum to cerebral cortex, the synaptic connections are complex and DBS likely influences
multiple elements.
Anatomically, Vim is a complexly connected structure locating in the ventrobasal thalamus.
Dorsally, Vim is bordered by the dorsal region of the ventral tier nuclei. Posterior to the Vim lies
the principal sensory nucleus ventral caudalis (Vc) or ventral posterior lateral nucleus (VPL).
Anterior to the cerebellar receiving area lies the region that receives afferents from the basal
ganglia (GPi, SNr), incorporating ventro-oralis anterior (Voa), ventral anterior nucleus (VA). The
medial border of Vim is with Vim.i, which is anatomically similar to Vim but receives
proprioceptive afferents with receptive fields in the face region. Laterally, Vim is bordered by
the posterior limb of the internal capsule. Organization within Vim is strictly somatotopic, with
face followed by hand followed by leg from medial to lateral.
Subthalamic Nucleus
The STN has become the most commonly used target for DBS in the treatment o f PD. The STN
is an important node in basal ganglia circuits, serving as major target for cortical afferents and
Yurui Gao, BME, Vanderbilt U
also receiving multiple inputs from other basal ganglia components. The STN receives
glutamatergic excitatory afferents from the frontal lobe of the cerebral cortex, GABAergic
inhibitory afferents from the GPe, and excitatory afferents from the parafascicular nucleus of
the thalamus. The output from the STN is glutamatergic and excitatory to both segments of the
globus, to the substantia nigra pars reticulata (SNr), and to the pedunculopontine area. Thus
the DBS in the STN has the potential to influence a variety of afferent and efferent targets and
may have different effects on different neurons.
Anatomically, the STN is a complex, biconvex lens-shaped, triply oblique structure. Its
maximal length is 13.2 mm rostrocaudally and 8 to 9 mm mediolaterally. STN is bordered on its
anerior and lateral sides by the corticobulbospinal tract, while posteromedially lies the
prelemniscal radiation and the red nucleus. The dosal border of STN is with the lenticular
fasciculus anteriorly, the thalamic fasciculus posteriorly, with the thalamus dorsal to the latter.
The ventral border is formed by substantia nigra reticulate (SNr).. STN is organized into motor,
limbic, and associative regions. In the axial plane, the STN slopes form anteromedial to
posterolateral, parallel to the cerebral peduncle below. Whereas the limbic region lies in the
anteromedial portion, the sensormotor region of STN lies in the posterolateral portion, and
within this area it lies dorsally. Some mapping strategies are directed toward locating the motor
region by confirming the presence of movement evoked neural responses.
Globus Pallidus pars interna
The GPi is another commonly used DBS target for the treatment of PD and meanwhile is
increasingly targeted for DBS treatment of dystonia. The GPi labeled in figure 4 is one of the
Yurui Gao, BME, Vanderbilt U
primary output nuclei of the basal ganglia and the main output representation of limb
movements [Mink 1996]. The GPi receives excitatory glutamatergic afferents from the STN,
inhibitory GABAergic afferents form striatum, inhibitory inputs from the GPe, and nigral
dopamine afferents. Due to the size and geometry, the effect of stimulation in the GPi is more
likely to restrict to the nucleus, but the potential remains for the possible spread to adjacent
structures and pathways, especially the GPe.
Anatomically, GPi is the internal segment of globus pallidus (GP) and the motor region of
the GPi is the target for DBS in PD and dystonial. The GPe forms the lateral border of GPi, and
medially, the dorsal border. The posterior limb of internal capsule forms the posteromedial
border. The ansa lenticularis, the main outflow of GPi, lies below GPi coursing medially and
anteriorly. Below this white matter lies the optic tract coursing posterolaterally toward the
lateral geniculate nucleus. GPi is separated from GPe by the internal medullary lamina.
HOW TO TARGETING IN DBS?
Given the small size of DBS targets, targeting is a critical step in the DBS surgical procedure.
Accurate preoperative targeting influences directly and significantly the operating time and
most importantly, the outcome of the surgery. There are directly or indirectly methods
performed preoperatively or intraoperatively.
Yurui Gao, BME, Vanderbilt U
Fig.5. AC-PC coordinate localization. The anterior and posterior commissures can be identified on this T1-weighted images [ (30)]
Fig.6. AC-PC coordinate localization. The anterior and posterior commissures can be identified on this T1-weighted images [ (30)]
By brain atlas
The targeting based on a standardized brain atlas is also
called ‘indirect targeting’, a common procedure used for
DBS targeting purposes mainly when the targets of DBS is
not clearly visible in MR T2-weighted images. While atlas-
based coordinates are biased by the inherent inaccuracies
of atlases, coordinates can be derived from the experience
of teams and represent the average values of the
coordinates of these best contacts, gathered in a rather
large number of cases. Registering these coordinates to internal landmarks such as AC-PC
length and height of the thalamus above the AC-PC plane provides normalized coordinates less
dependent on individual variations. The registration uses the midcommissural point (MCP) or
the PC as reference, which is calculated after selecting the AC and PC coordinates (Fig.5).
Typical initial anatomical coordinates for the ventral and sensorimotor STN are 11 to 13 mm
lateral to the midline, 4 to 5 mm ventral to the intercommissural plane, and 3 to 4 mm
posterior to the MCP. Coordinates for the
sensorimotor GPi are selected as 19 to 21
mm lateral to the midline, 2 to 3 mm
anterior to the MCP, and 4 to 5 mm
ventral to intercommissural plane. The
Vim is targeted 11 to 12 mm lateral to the
Yurui Gao, BME, Vanderbilt U
Fig.7. 3D atlas-based targeting. Left:.3D polygonal surface of left thalamus and STN in brain atlas; Right: 3D rendering of the atlas nuclei. [ (33)]
wall of the III ventricle, at the level of the intercommissural plane. Most commonly, the goal is
to target the topography of the upper extremity, which carries the greatest impact for quality
of life. In terms of anterior-posterior position, the Vim is typically located between 2/12 and
3/12 of the AC-PC distance rostral to the posterior commissure. These coordinates may not represent
the ideal location for implantation but are used as initial anatomical targets, which are verified by
physiological recordings. Figure 6 demonstrates a typical digitized version of the Schaltenbrand and
Wahren atlas registered to the MRI space to conform to the patient’s anatomical structures. These atlas-
and-registration-based targets are than cross-correlated with the direct visualization target.
Since the scanning performed in a 3D space, when oblique slices produced by scanner, 2D
registration is not qualified to register atlas to MRI slice by slice. 3D atlas-based targeting may
overcome this difficulty. Briefly, 3D surfaces (fig. 7 left) representing visible subcortical nuclei
were segmented from a high-resolution MRI. These surfaces were then aligned with 2D
Schaltenbrand and Wahren
slices [8] to define outlines
of nonvisible nuclei such as
STN and lofted to create the
additional 3D surfaces.
Atlas-based targeting
has been the traditional method in stereotactic neurosurgery. The limitation of this method is
that it does not take into account the discrete anatomical variations among individuals. Not all
brains correspond to the topographical distribution depicted in a given atlas. Even the
deformable atlases available in computerized planning stations may not conform adequately to
Yurui Gao, BME, Vanderbilt U
Fig.8. Direct STN targeting in T2-weighted MRI image. (white arrow-> STN anterior; black arrow->lateral to the red nucleus) [ (30)]
the patient’s anatomy. In this sense, direct targeting seems to have obvious benefits over
indirect.
Regarding the registration methods, it is roughly divided into rigid and non-rigid algorithm
according to the characteristic of deformation from source image to reference image. Rigid
registration is referred as primarily affine transformation of source image involving translation,
rotation as well as scaling. Rigid registration is also used as a pre-alignment step for non-rigid
transformation. Non-rigid registration includes intensity-based algorithm (demons/ B-spline/
thin-plate-splin/ adaptive bases algorithm [ (18)]) and segmentation-based registration. Mutual
information [ (19)] is the most commonly used standard for automatically measure the
goodness of the registration result. More details will be skipped in this paper.
By visible surrounding anatomical landmarks
A target within STN can be chosen at the ventral part of the
somatosensory STN, which is the lateral part of the nucleus.
The axial and coronal T2 images are particularly important
for adequate visualization of the STN as a sharp contrast can
often be seen between the nucleus and the surrounding
white matter. The red nucleus can also be clearly seen and
the STN lies anterior and lateral to the red nucleus. The
anterior border of the read nucleus can be used as
landmark for the anteroposterior localization of the STN target [ (20)]. Figure 8 shows the
visualization result in the axial T2 images.
Yurui Gao, BME, Vanderbilt U
Fig.9.Direct GPi targeting. A: GPi (arrow) is seen on an inversion recovery axial image. B: The optic tract (tip of the trajectory line) can be identified ventral to the GPi and used as a landmark for direct targeting. [11]
Fig.10.Phase image from SWI sequence acquisition in 3T MRI. The STN, SN, and RN are sharply delineated on phase image. [8]
The GPi and the adjacent optic
tract and internal capsule can be
visualized via inversion recovery and
T2 images, Proper windowing of the
images can also provide
visualization of the internal capsule
together with the GPi and GPe. For
GPi DBS, implantation target choice
may be different from that used for pallidotomies, which are typically located at the
posteroventral portion of the nucleus. Implantation at that position may not be optimal for DBS
because the DBS stimulation field may require higher amplitudes and potential spread to the
internal capsule. Thus for DBS, a more anterior and lateral target is preferable, which will allow
the stimulation amplitude to be increased without significant involvement of the descending
white matter fibers. The optic tract courses ventrally to the globus pallidus and can be used as
landmark for the approach (Fig.9).
By direct locating (SWI/7T MRI/DTI)
Since the high iron concentration in the STN causes T2*
shortening and resonance frequency increasing,
susceptibility-weighted imaging (SWI) which exploits the
magnetic susceptibility differences of tissues, provides
additional possible contrast mechanism to visualize STN
Yurui Gao, BME, Vanderbilt U
Fig.11. In vivo image from 7T MR system with a brain-optimized SENSE coil [ (29)]. 7T MR image reveals detailed structures of the thalamus including STN et al.
without using anatomic landmarks. Using postprocessing techniques, a single SWI acquisition
can generate 3 sets of inherently co-registered images: T2*-weighted maps of signal-intensity
magnitude images phase image and a combination the two. Vertinsky [ (21)] and colleagues
assessed the visibility as well as delimitation of STN using SWI and concluded that STN were
directly visualized on SWI phase image.
Moreover, ultra-high MR system (e.g. 7T MR system) provides an increasing SNR, allowing
much higher spatial and enhanced
susceptibility contrast. Those small
structures which cannot be well delineated
at lower field have opportunity to be seen
and segmented much more exactly.
However, the clinical application of 7T MR
still delayed by many critical problems such
as the regional signal variation resulting
from the inhomogeneous B0 and B1 field.
By microelectrode recordings
A fundamental purpose of microelectrode recording (MER) is to verify the stereotactic location
of the target brain region for the electrode implantation. A neurophysiologist records neural
activity as a microelectrode is advanced through stereotactic brain space and identifies specific
nuclei on the properties of the neural signals recorded [ (22)]. While not performed at all DBS
centers, this verification process can help alleviate two limitations. First, the surgical target
Yurui Gao, BME, Vanderbilt U
Fig.12. Microelectrode mapping of the STN. A: Serial-track approach, B: Five-simultaneous-track approach. [ (34)]
cannot always be directly identified on preoperative imaging data. Second, the intracranial
pressure changes once the burr hole is drilled, resulting in a possible shift of the subcortical
structures, thereby altering the position of the target point in stereotactic space [ (23)].
MER begins at various distances from the target. Take the target STN as an example, the
distances between recording and STN range from several millimeters (Fig.12B) to up to 40 mm
(Fig. 1A), depending on whether striatal
and/or thalamic activity is recorded.
Findings depend on the obliquity of the
planned tracks. When the entry point is
more lateral, the track misses the striatum
or thalamus, residing completely within the
corona radiate and internal capsule (Fig 1B).
When recording begins more proximally
(~40mm from above target) and with more
medial entry points, the caudate is
encountered with its characteristic pattern
of phasically active units. Below this, the
internal capsule is entered, from which
occasional fibers may be recorded. More
anterior tracks will pass in front of the
thalamus, next encountering the Zona
Yurui Gao, BME, Vanderbilt U
Fig.13.Semimicroeletrode mapping of the STN.
incerta (ZI), in which small infrequent neurons may be recorded with low tonic firing rates.
More posterior tracks pass into the thalamus, approximately 7mm above the STN. The shell of
the thalamus contains the reticular nucleus with its characteristic bursting pattern discharges
(15±19Hz). The ventral basal complex is
encountered next, the cell of which fire at an
average 28 Hz. These cells are easily
distinguished from the STN itself as their firing
rate is significantly lower, and the overall
background of the thalamus is substantially
quieter. After 1 to 2 mm, the STN is entered,
characterized by an abrupt increase in
background activity. Mean firing rates have been reported in the 34-74 Hz, with standard
deviation is 25Hz [ (24)]. Below the STN, the electrode traverse into SNr whose mean firing rate
in most report is higher[ (25)] than STN but in some notes is lower [ (26)]. Rather, the pattern
of neuronal activity, motor responsiveness, and background activity offer greater reliability in
determining electrode location. These findings are mirrored in semimicroelectrode recordings
[Fig 13], where the overall background of the STN provides a sharp contrast to that of the
internal capsule and ZI above and the SNr below.
By macro stimulation tests
Following mapping via microeletrode recording we described above, macrostimulation is
performed with either the microelectrode using current in the milliamp range, the uninsulated
Yurui Gao, BME, Vanderbilt U
tip of the inner guide cannula through which the microelectrode was passed, a dedicated
macroelectrode (e.g. a radio-frequency ablation electrode), or DBS electrode.
Within the internal capsule, orofacial and/or appendicular contractions are noted, as are
contraversive eye movements. Within the thalamus, various effects have been reported (e.g.
vegetative effects) in some instances and a mild reduction in distal muscle tone ventrally [13].
Stimulation within zona incerta (ZI) and the fields of Forel can produce a decrease in rigidity at
relatively low voltage thresholds ( (27)). Stimulation in the STN produces the maximal clinical
effects intraoperatively, however, and they are obtained at the lowest thresholds. Stimulation
within SNr is generally without clinical effects, although some have observed decrease in
rigidity and worsening of bradykinesia.
The distance and direction of refinement of initial targeting is related to the accuracy of
the initial imaging and target calculation, as well as the technique and accuracy of the mapping
procedure. One measure of both the inadequacy of the initial imaging, target calculation, and
stereotactic technique, and the subsequent ability of physiological mapping to ‘correct’ these
errors, is the amount of correction of the electrode position following mapping. Both
microelectrode recording techniques and macrostimulation techniques can be used to correct
the electrode position. Some groups relied solely on macroelectrode or microelectrode
mapping to position the electrode optimally while some relied on combination of these two
techniques. Does the use of mapping improve outcome? There are insufficient data available to
determine whether any mapping at all, as opposed to pure anatomical targeting without
modification based on intraoperative evaluation, is necessary for good results from STN DBS.
Yurui Gao, BME, Vanderbilt U
References 1. Fritsch, G and Hitzig, E. Uber die elektrische erregbarkeit des grosshirns. Arcb. anat. physiol. 1870,
Vol. 37, pp. 300-332.
2. Bartholow, R. Experimental investigation into the functions of the human brain. Am. J. Med.Sci. 1874,
Vol. 67, pp. 305-313.
3. Spiegel, E and Wycis, H. Thalamotomy and pallidotomy for treatment of choreic movements. Acta.
Neurocbir. 1952, Vol. 2, pp. 417-422.
4. Hassler, R, Riechart, T and Munginer, F. Phsiological observations in stereotaxic operations in
extrapyramidal motor disturbances. Brain. 1960, Vol. 83, pp. 337-350.
5. Hosobuchi, Y, Adams, J and Rutkin, B. Chronic thalamic stimulation for the control of facial
anesthesia dolorosa. Arcb. Neurol. 1973, Vol. 29, pp. 158-161.
6. Cooper, I. Effect of chronic stimulation of anterior cerebellum onneurological disease. Lancet. 1973, p.
206.
7. Benabid, A, Pollak, P and Gervason, C. Long-term suppression of tremor by chronic stimulation of the
ventral intermediate thalamic nucleus. Landet. 1991, Vol. 337, pp. 403-406.
8. Sellal, F, Hirsch, E and Barth, P. A case of symptomatic hemidystonia improved by
venropostertolateral thalamic electrostimulation. Mov Disord. Vol. 8, pp. 515-518.
9. Kumar, R, Lozano, A and Sime, E. Comparative effects of unilaternal and bilateral subthalamic nucleus
deep brain stimulation. Neurology. 1999, Vol. 53, pp. 561-566.
10. Diederich, N, Kalteis, K and Stamenkovic, M. Efficient internal pallidal stimulation in Gilles de la
Toureete syndrome. Mov Disord. Vol. 20, pp. 1496-99.
11. Nuttin, B, Demeulemeester, P and Gybels, H. Electrical stimulationin anterior limbs of internal
capsules in patients eith obsessive-compulsive disdor. Lancet. Vol. 354, p. 1526.
12. Holloway, K, Gaede, S and Starr, P. Frameless stereotaxy using vone fiducial markers for deep brain
stimulation. J Neurosurg. 2005, Vol. 103, pp. 404-413.
13. Hashimoto, T, Elder, CM and Okum, MS. Stimulation of the subthalamic nucleus changes the firing
pattern of pallidal neurons. J Neurosci. 2003, Vol. 233, pp. 1916-1923.
14. Benazzouz, A and Hallett, M. Mechanism of action of deep brain stimulation. Neurology. 2000, Vol.
55(suppl 6), pp. 13-16.
15. Lozano, A, Dostrovsky, J and Chen, R. Deep brain stimulation for Parkinson's disease: disrupting the
disrupt. Lancet. 2002, Vol. 1, pp. 225-231.
Yurui Gao, BME, Vanderbilt U
16. Kern, Drew S. and Kumar, Rajeev. Deep brain stimulation. The Neurologist. 2007, Vol. 13, 5, pp. 237-
252.
17. Elias, Jeff, Sansur, Charles and Frysinger, Robert. Sulcal and ventricular trajectories in stereotactic
surgery. Journal of Neurosurgery. 2, 2009, Vol. 110, pp. 201-207.
18. Rohde, G, Aldroubi, A and Dawant, B. The adaptive bases algorithm for intensity-based nonrigid
image registration. IEEE transaction on Medical Imaging. 11, 2003, Vol. 22, pp. 1470-1479.
19. Maes, F, Collignon, A and Vandermeulen, D. Multimodality image registration by maximization of
mutual information. IEEE transactions on medicalimaging. 2, 1997, Vol. 16, pp. 187-198.
20. Andrade-Souza, Y, Schwalb, J and Hamani, C. Comparison of three methods of targeting the
subthalamuc nucleus for chronic stimulation in Parkinson's disease. Neurosurgery. 2005, Vol. 56, pp.
360-368.
21. Vertinsky, A, Coenen, V and Lang, D. Localization of the subthalamic nucleus: optimization with
susceptibility-weighted phase MR imaging. Am J Neuroradiol. 2009, Vol. 30, pp. 1717-1724.
22. Yelnik, J, Bardinet, E and Dormont, D. A three-dimentsional, histological and deformable atlas of the
human basal ganglia. Neuroimaging. 2007, Vol. 34, pp. 618-638.
23. Haplern, C, Danish, S and Baltuch, G. Brain shift during deep brain stimulation surgery for
Parkinson's disease. Stereotact funct neurosurg. 2008, Vol. 86, pp. 37-43.
24. Hutchison, D, Allan, J and Opitz, H. Neurophysiologicla identification of the subthalamic nucleus in
surgery for Parkinson's diseasee. Ann Neurol. 1998, Vol. 44, pp. 622-628.
25. Bejjani, B, Dormont, D and Pidoux, B. Bilateral subthalamic stimualtion for Parkinson's disease by
using three-dimensinal stereotactic magnetic resonance imaging and electrophysiological guidance. J
Neurosurg. 2000, Vol. 92, pp. 615-625.
26. Benazzouz, A, Breit, S and Koudsie, A. Intraoperative microrecordings of the subthalamic nucleus in
Parkinson's disease. Mov Disord. 2002, Vol. 17, pp. 145-149.
27. Zincone, A, Landi, A and Piolti, R. Phisologic study of the subthalamic volume. Neurol Sci. 21, 2001,
pp. 111-112.
28. Deep brain stimulation. Drew S. Kern, Rajeev Kumar. 5, 2007, The Neurologist, Vol. 13, pp. 237-252.
29. Cho, Z, Kim, Y and Min, H. New brain atlas - Mapping the human brain in vivo with 7.0 T MRI and
comparison with postmortem histology: Will these images change modern medicine? Internationl J of
Imaging System and Technology. 1, 2008, Vol. 18, pp. 2-8.
30. Machado, A, Rezai, A and Kopell, B. Deep brain stimulation for Parkinson's disease: surgical
technique and perioperative management. Movement Disorder. 14, 2006, Vol. 21, pp. 247-258.
Yurui Gao, BME, Vanderbilt U
31. Talairach, J and Tournoux, P. Co-Planar stereotaxic atlas of the human brain. s.l. : Thieme, 1998.
32. Schaltenbrand, G and Wahren, W. Atlas for stereotaxy of the human brain. New Youk : Thieme,
1977.
33. Lujan, L, Noecker, A and Butson, C. Automated 3-dimensional brain atlas fitting to microelectode
recordings from deep brain stimulation surgeries. Stereotact Funct Neurosurg. 2009, Vol. 87, pp. 229-
240.
34. Gross, R, Krack, P and Rodriguez-Oroz, M. Electophysiological mapping for the imaplantation of
deep brain stimulators for Parkinson's disease and tremor. Movement disorder. 2006, Vol. 21, pp. 259-
283.
35. Bootin, M. Deep brain stimulation: overview and update. J Clinical monitoring and computing. 2006,
Vol. 20, pp. 341-346.
36. Castro, F, Pollo, C and Meuli, R. Towards fast, automatic and accurate deep brain stimulation
targeting. s.l. : Signal Processing Institue, EPFL, Switzerland, 2008.