Expert Review of Medical Devices

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/ierd20

Device profile of the percept PC deep brain

stimulation system for the treatment of
Parkinson’s disease and related disorders

Joohi Jimenez-Shahed

To cite this article: Joohi Jimenez-Shahed (2021) Device profile of the percept PC deep brain
stimulation system for the treatment of Parkinson’s disease and related disorders, Expert
Review of Medical Devices, 18:4, 319-332, DOI: 10.1080/17434440.2021.1909471

To link to this article: https://doi.org/10.1080/17434440.2021.1909471

© 2021 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group.

Published online: 05 Apr 2021.

Submit your article to this journal

Article views: 12032

View related articles

View Crossmark data

Citing articles: 48 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ierd20
EXPERT REVIEW OF MEDICAL DEVICES
2021, VOL. 18, NO. 4, 319–332
https://doi.org/10.1080/17434440.2021.1909471

DEVICE PROFILE

Device profile of the percept PC deep brain stimulation system for the treatment of
Parkinson’s disease and related disorders
Joohi Jimenez-Shahed
Movement Disorders Neuromodulation & Brain Circuit Therapeutics, Neurology and Neurosurgery, Icahn School of Medicine at Mount Sinai,
New York, USA

ABSTRACT ARTICLE HISTORY

Introduction: Several software and hardware advances in the field of deep brain stimulation (DBS) have Received 4 January 2021
been realized in recent years and devices from three manufacturers are available. The Percept™ PC Accepted 24 March 2021
platform (Medtronic, Inc.) enables brain sensing, the latest innovation. Clinicians should be familiar with KEYWORDS
the differences in devices, and with the latest technologies to deliver optimized patient care. Brain sensing; deep brain
Areas covered: In this device profile, the sensing capabilities of the Percept™ PC platform are stimulation; local field
described, and the system capabilities are differentiated from other available platforms. The develop­ potentials; movement
ment of the preceding Activa™ PC+S research platform, an investigational device to simultaneously disorders; parkinson’s
sense brain signals and provide therapeutic stimulation, is provided to place Percept™ PC in the disease
appropriate context.
Expert opinion: Percept™ PC offers unique sensing and diary functions as a means to refine ther­
apeutic stimulation, track symptoms and correlate them to neurophysiologic characteristics. Additional
features enhance the patient experience with DBS, including 3 T magnetic resonance imaging compat­
ibility, wireless telemetry, a smaller and thinner battery profile, and increased battery longevity. Future
work will be needed to illustrate the clinical utility and added value of using sensing to optimize DBS
therapy. Patients implanted with Percept™ PC will have ready access to future technology
developments.

1. Introduction over the age of 40 years [9], and up to 30–50% of patients

do not respond to the most common medications used to
Deep brain stimulation (DBS) is the surgical standard of care
treat it [10]. Dystonia is a condition whose main feature is
for patients with movement disorders such as Parkinson’s
sustained muscle contractions producing an abnormal pos­
disease (PD), essential tremor and dystonia whose symptoms
ture, which can be quite heterogeneous in manifestations
are inadequately controlled with usual therapies [1], with the
and etiology [11]. DBS is typically considered when dystonia
most common use being in the management of advanced PD.
is the main source of disability and affects quality of life, and
PD is characterized by motor symptoms such as tremor,
has not responded to oral treatments or botulinum toxin
bradykinesia and rigidity [2]. The mainstay of medication man­
injections [6].
agement is with levodopa, which is highly effective in control­
Although DBS has established efficacy for managing motor
ling motor symptoms, but is associated with development of
symptoms in these disease states when applied to the right
motor complications [3]. Such motor complications, including
patient at the right time, there remain several limitations to
wearing off effects and levodopa-induced dyskinesias, increase
standard DBS delivery paradigms. First, DBS is applied con­
in frequency and severity with advancing disease [4], and can
tinuously, which may contribute to stimulation-induced side
become more problematic than the primary motor features
effects [12] or adverse consequences of medication reduction
themselves. Additionally, patients frequently experience gait
[13], or to complex phenomena such as habituation of symp­
disturbances as the disease progresses along with increasing
tom control [14]. Second, the process of DBS programming
impairment in quality of life. DBS can control bothersome
can be lengthy as it is largely performed on a trial and error
motor complications and refractory tremor, and has proven
basis [12,15], though there is increasing recognition of the
efficacy in improving quality of life [5,6].
role of physiology and imaging to guide programming para­
Essential tremor (ET) is more common than PD, but is less
meter selection, including more efficient identification of the
often treated with DBS despite multiple robust lines of evi­
optimal therapeutic contact [16,17]. Third, the energy
dence favoring long-term sustained efficacy [7]. It is typically
demands of continuous stimulation and the inclusion of new
considered when patients have tried and failed multiple med­
technical capabilities are increasingly difficult to balance with
ication management options for tremor, which otherwise
the energy supplied in a reasonably sized and efficient pulse
interferes with activities of daily living and quality of life
generator without requiring recharging, and have led to
[6,8]. ET is estimated to affect over 4% of the population

CONTACT Joohi Jimenez-Shahed [email protected] Movement Disorders Neuromodulation & Brain Circuit Therapeutics, Neurology
and Neurosurgery, Icahn School of Medicine at Mount Sinai, Mount Sinai West, 1000 10th Avenue, Suite 10C, New York NY 10019, USA
© 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/),
which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
320 J. JIMENEZ-SHAHED

system. Safety, battery longevity, and future considerations

Article highlights will also be discussed.
● ™
Percept PC is the first commercially available device capable of
sensing brain signals (local field potentials, LFPs) while simulta­
neously delivering stimulation. 2. Body of the review
● ™
BrainSense technology contained within the Percept PC DBS ™
platform can be configured by clinicians to track LFP frequency 2.1. Overview of the market
bands of interest up to 100Hz and provides the opportunity to
correlate these brain signals with symptoms reported by patients. DBS devices and platforms from multiple manufacturers are
● Brain sensing is a tool that is not currently associated with any
therapeutic claims, but evidence suggests the potential for clinical
currently available for commercial use, with specific indica­
utility, including future closed loop DBS, in the treatment of patients tions that may vary by geographic region (Table 1). While
with movement disorders such as Parkinson’s disease. each device has specific attributes and some unique features,
● ™
Additional system capabilities included in the Percept PC platform
the basic principles of stimulation remain identical. The goal is
enhance the patient experience with DBS, including 3T magnetic
resonance imaging compatibility, wireless telemetry, a smaller and to apply a discreet electrical field within specific nodes of the
thinner battery profile, and increased battery longevity. Software pathologic network responsible for disease manifestations, in
upgrades can be used to unlock new or expanded capabilities as
they become available.
a manner that regularizes neuronal patterns and prevents
transmission of pathologic bursting or oscillatory signals [34].
In movement disorders, the most common DBS targets are the
subthalamic nucleus, the globus pallidus interna, and the
ventral intermediate nucleus of the thalamus.
increased frequency of battery replacement procedures The process of DBS programming entails shaping the field
[18–20]. of stimulation within that target nucleus in order to engage
There has been a marked increase in the past two decade the neuronal elements that will lead to symptom reduction,
in our understanding of the neurophysiologic markers under­ while avoiding spread of current to those in surrounding
lying symptoms of movement disorders that are treated with regions that may cause side effects [35]. The tools available
DBS, accompanied by the development of DBS technologies for field optimization within each manufacturer’s DBS platform
that can be used to both record and modulate these markers are what differentiates them, representing not only a broad
[21]. Local field potentials (LFPs) represent the aggregate array of features including electrode design, range of stimula­
activity of a population of neuronal elements [22] and can tion parameters available, and advanced field shaping tools,
be recorded using both the implanted micro- and macro- but also innovative capabilities such as LFP sensing and lead
electrodes used during DBS surgery and therapy [23,24]. LFP visualization software (Table 2). Additional characteristics such
analysis has provided important insights into the pathophy­ as magnetic resonance imaging (MRI) compatibility, and the
siology of movement disorders and our understanding of size and recharge ability of the implantable pulse generators
medication and DBS benefits. In PD, bradykinesia and rigidity (IPG) are further distinguishing features, though they do not
correlate with beta band (8–30 Hz) activity, and the degree of directly affect programming.
improvement in these symptoms following levodopa or sti­ Potential advantages of lead visualization are the opportu­
mulation in the STN correlates with the magnitude of beta nity to use an anatomic approach to complex field shaping
band suppression [25]. Gamma band (31–200 Hz) and high and for the individualization of a patient’s therapeutic stimula­
frequency oscillation (>300 Hz) power recorded in either brain tion parameters, as well as improved efficiency of program­
target in PD patients increases following dopaminergic treat­ ming sessions [42]. On the other hand, LFP recordings have
ment [26,27], and an increase in gamma power may be seen the potential to inform electrode placement during surgery,
during levodopa-induced dyskinesia [28]. In essential tremor, identify the likely therapeutic contact, detect motor fluctua­
beta frequency oscillations (8–27 Hz) in the ViM are correlated tions and other clinical states, and serve as a control variable
with tremor frequency measured peripherally by surface EMG for closed loop stimulation [43]. The availability of co-existing
[29]. Dystonia is characterized by higher power in the 3–12 Hz features such as new lead and IPG designs, the widened range
frequency range in the GPi, and correlation between LFP of stimulation parameters, and other advanced field shaping
power and dystonic muscle activity has also been demon­ tools enhance the ability to effectively apply these innovative
strated [30]. These disease states are increasingly recognized features in clinical practice [44].
as network disorders with abnormal synchronization through­ While these technological advances provide greater pro­
out the cortical-basal ganglia-thalamocortical circuit. For gramming capabilities for clinicians and the possibility of data-
example in PD, phase amplitude coupling (PAC) between the driven and individualized approaches to programming opti­
beta band oscillations and broad band gamma activity has mization, the individual systems or features have not been
been shown within the motor cortex, within the STN, and compared from an efficacy, efficiency or cost-effectiveness
between the STN and motor cortex [31–33]. Such coupling is standpoint. There are no evidence-based recommendations
diminished by levodopa and DBS. for use of individual devices in specific patient populations.
This device profile will review the specifications and func­ Decisions related to device selection are determined by
tionality of the Percept™ PC, the first commercially available a combination of factors including access to the devices within
DBS platform capable of in vivo sensing. A market overview the health system, clinician preference or experience, and
will provide context for the enhancements included in this patient preference [45].
Table 1. Commercially available deep brain stimulation platforms and their indications by geographic region.
MEDTRONIC ABBOTT BOSTON SCIENTIFIC PINS MEDICAL
US
Activa & Percept
™ ™ Infinity™ Vercise PC
™ Vercise Gevia
™ Vercise Genusd
™
● PD – bilateral GPI or STN ● PD – bilateral STN or GPi ● PD – bilateral STN ● PD – bilateral STN ● PD – bilateral STN or GPi
● Tremor (ET or PD) – unilateral ViM ● Tremor: unilateral or bilateral ViM or GPi or GPi
a
● Dystonia – unilateral or bilateral
GPi or STN
EU
Activa & Percept
™ ™ (see below, OUS) Vercise PC, Gevia and Genusd
™
● PD – bilateral GPi or STN ● PD – unilateral or bilateral STN or GPi
● Tremor (ET or PD) – unilateral or ● Tremor – thalamic stimulation for ET or PD
bilateral ViM ● Dystonia – unilateral or bilateral STN or GPi
● Dystonia – unilateral or bilateral GPi
or STN
OTHER Japan, Australia, Brazil OUS ROW Japan only China only
Activa & Percept
™ ™ Infinity™ Vercise PC & Gevia
™ Vercise Genus
™ ● Clinical applications include PD,
d
● PD – GPi or STN ● PD – unilateral or bilateral STN or GPi or ● PD – unilateral or bilateral STN or GPi ● PD – unilateral or bilateral STN or tremor, dystonia
e
● Tremor (ET or PD) – ViM thalamus ● Tremor – thalamic stimulation for ET or PD GPi
f
● Dystonia – GPi or STN ● Disabling tremor – unilateral or bilateral ● Dystonia – unilateral or bilateral STN or GPi ● Tremor – thalamic stimulation for
ViM ET or PD
● Dystoniac – unilateral or bilateral STN ● Dystonia – unilateral or bilateral
or GPi STN or GPi
NOTES
PD = Parkinson’s disease; ET = essential tremor; STN = subthalamic nucleus; GPi = globus pallidus interna; ViM = ventral intermediate nucleus of the thalamus; US = United States; EU = European Union; OUS = outside United
States; ROW = rest of world
a
approved under a humanitarian device exemption; chronic, intractable (drug refractory) primary dystonia, including generalized and/or segmental dystonia, hemidystonia, and cervical dystonia (torticollis), in patients seven
years of age or above
b
intractable, chronic dystonia, including primary and secondary dystonia for patients who are at least 7 years old
c
includes Neural Navigator software
d
Japan does not specify laterality; Australia – bilateral; Brazil – unilateral or bilateral
e
Japan does not specify laterality; Australia – unilateral; Brazil – unilateral or bilateral
f
Japan does not specify laterality; Australia – unilateral or bilateral; Brazil – bilateral
EXPERT REVIEW OF MEDICAL DEVICES
321
322 J. JIMENEZ-SHAHED

Table 2. Tools and design features for stimulation field shaping and their implications.
Tool/Design Description and Implications Reference(s) Platforms
feature
Lead and/or IPG Segmented contacts Allows axial current steering away from regions causing side effects, or toward fiber [36] Infinity ™
design tracts that are relevant to clinical improvement; single segment activation widens the Vercise ™
therapeutic window compared to omnidirectional stimulation
Independent current
control to each
Caps the total amount of current and distributes portions of the current independently
through 2 or more contacts, independent of changes in impedance; allows for axial
[37] Vercise ®
contact and longitudinal current steering
Independent frequency Allows individual frequencies to be programmed on each lead or contact that is ™
Infinity
control connected to the same IPG Vercise™
Range of Pulse Width <60µsec Pulse width affects the intensity of the field of stimulation; lower pulse width widens the [38] Percept ™
stimulation therapeutic window ™
Infinity
parameters Vercise™
PINS
Programming Decision support tool* Visual representations of stimulation responses, occurrences of stimulation-induced [36] ™
Infinity
software symptom relief and side effects; facilitates clinician review and selection of optimal Vercise™
stimulation parameters
Advanced field Interleaving, or multi- A current fractionalization approach that rapidly alternates multiple stimulation sets at [37] Activa™
shaping tools stim set a shared frequency, but can have other different stimulation parameters (PW and Percept™
Amp); allows complex field shaping Infinity™
PINS
Anodic stimulation Hyperpolarizes membranes immediately below the electrode, but elicits action [39] Vercise™
potentials at a distance from the electrode via the return current, potentially
mitigating stimulation side effects
LFP Sensing BrainSense ™ A specific frequency range of LFP signals can be recorded simultaneously while [46] Percept ™
delivering stimulation, can be monitored in relation to stimulation adjustment, and PINS
can be correlated with patient reported events, such as medication intake or clinical
symptoms
Visualization SureTune ™ VTA models represented in relation to the lead and relative to an anatomic atlas, with [36] Activa™
the possibility of manual adjustment of anatomic regions
GUIDE ™ VTA models represented in relation to the lead and relative to an anatomic atlas [42] Vercise™
™
GUIDE XT,
†
VTA models and lead location in relation to patient-specific anatomy [40,41] Vercise™
STIMVIEW ™ XT
IPG = implantable pulse generator; LFP = local field potential; PW = pulse width; Amp = amplitude
™ ™ ™
*Infinity software = Informity ; Vercise software = Neural Navigator
†
™ ™
GUIDE XT and STIMVIEW XT are not currently available in the US market

2.2 Introduction to the Percept™ PC DBS platform enables a geometrically symmetric sensing configuration around
the stimulating contact, thereby improving common mode rejec­
The Medtronic Percept™ PC DBS device [46] is the first commer­
tion of the stimulus artifact [44]. The Percept™ PC IPG is
cially available platform for patients with movement disorders
embedded with patented software for real-time LFP signal pro­
that is capable of in vivo brain sensing. All other DBS devices
cessing and analysis, which are in turn stored on the IPG for later
deliver electrical stimulation without recording, while Percept™
download by the clinician. The sensing noise floor is <300 nano­
PC does both. The system, like all other DBS systems, is comprised
volts per root Hertz (nV/rtHz) and the input sensing range is
of a DBS electrode, extension wire and IPG (Figure 1).
0.55–400 microvolts root mean square (µVrms). With implanted
Accompanying non-implanted hardware includes a clinician pro­
leads in place, a survey of LFP activity can be conducted. This
gramming tablet, communicator and a patient controller (Figure
‘BrainSense survey’ is conducted with stimulation OFF and gen­
2). Electrical current is generated in the neurostimulator (IPG) and
erates a graph of the differential in LFP signal between any two
travels via the extension wire to the DBS lead and into the
contact pairs (Figure 4). It takes about 90 seconds to generate this
targeted tissue. One or two DBS electrodes can be connected to
data, which is automatically processed via fast Fourier transform
the dual channel IPG. The power source for the IPG is a hybrid
and presented as LFP magnitude (µVp, microvolts peak) vs. fre­
combined silver vanadium oxide primary cell, which has implica­
quency (Hz) from each of the contact combinations (6 per hemi­
tions for the ability to reliably predict when battery replacement
sphere). This survey allows the clinician to determine if any signal
will be needed [47]. The Percept™ PC IPG is compatible with
is detectable and from which contact pairs. With a low pass filter
Medtronic lead models 3387 and 3389 (used for movement
of 100 Hz, the Percept™ IPG is able to record signals in the delta,
disorders and epilepsy), lead model 3391 (used for psychiatric
theta, alpha, beta, and low gamma ranges [44].
disorders), and extension model 37,086. The lead models differ in
A ‘signal test’ is then performed, during which an impedance
terms of length and spacing of the contacts.
check (to exclude recording pairs where a short or open circuit is
present) and artifact check (to exclude electrocardiogram and
2.2.1. BrainSense™ technology motion artifacts) occur using the contact combinations where
Sensing functions in Percept™ PC are accomplished using simultaneous stimulation and sensing can occur (Figure 3). The
BrainSense™ technology, by measuring LFPs using the contacts system will display any LFP peaks that are present, and automati­
adjacent to the stimulating contact (Figure 3). LFPs represent the cally select the largest peak that is in the beta or gamma frequency
aggregate electrical activity of the group of neurons surrounding range, with power >1.1 (µVp). The clinician can examine the power
the recording contact [22]. A monopolar stimulation configura­ and frequency of any other LFP peaks that may be present, and
tion is required in order to record these brain signals because it ultimately identify the frequency signal of interest to be tracked
 EXPERT REVIEW OF MEDICAL DEVICES 323

™ PC implantable pulse generator has a volume of 33 cm3 and mass of 61 g. Its dimensions are 68 mm x 55 mm x 11 mm. image provided by
Figure 1. The Percept
Medtronic, Inc.

over time. Signals with lower power (≤1.1 µVp) are more difficult to Patients can be asked to participate in annotating this
track. Absence of a suitable or expected signal can be related to data using the ‘Events’ function. A 30-second snapshot of
electrode positioning, medication state, artifacts, or abnormal LFP data is taken directly after a patient marks an event
impedances. All data is sampled at 250 Hz in the time domain that is predetermined by the clinician. Examples of clinically
with two low pass filters at 100 Hz, one high pass filter at 1 Hz and relevant events include ‘took medications’, ‘dystonia epi­
another high pass filter that can be configured by the clinician to sode’, or ‘dyskinesia’, and can be identified and/or selected
be at 1 Hz or 10 Hz. The signal is transformed to the frequency at the discretion of the clinician and patient (Figure 2). Up to
domain and the power is detected for display. four types of events can be identified for annotation, and up
LFP signals can be recorded in two scenarios – out-of-clinic, to 400 snapshots can be stored (200 per hemisphere). Events
and in-clinic. Out-of-clinic recordings are captured between can also be marked and time stamped without LFP snapshots
square wave pulses and displayed as an average power of (up to 900 events) for later review by the clinician upon
10 minute epochs in the ‘Timeline’ view of the programming telemetry connection in clinic. The events will be displayed
tablet (Figure 5). The recorded frequency band of LFP activity in a timeline or summary format along with corresponding
is approximately 5 Hz wide surrounding the band selected by LFP recordings.
the physician (for example, with a selected frequency of 20 Hz, Another feature of the BrainSense™ technology is the abil­
the system may record the LFP signal in the 18–23 Hz range). ity to evaluate the dynamics of the LFP band of interest over
Data from out-of-clinic recordings can only be viewed when time, referred to as setting ‘LFP thresholds’. Once the patient is
the patient returns to clinic and the clinician programming at a stable therapeutic setting, and LFP bands of interest are
tablet is connected via telemetry to the patient’s IPG. Up to identified to fluctuate in response to stimulation levels, LFP
60 days of data can be stored, after which the oldest data is thresholds can be set (Figure 6). This allows the clinician to set
overwritten. reference points of LFP magnitude to track them over time. In
324 J. JIMENEZ-SHAHED

™
Figure 2. The Percept PC patient controller. the first panel shows the therapy status. the second panel shows the different stimulation groups programmed in to
the device, which the patient can select as needed. The third panel shows the customized events selected for the patient to mark through the day as needed. Image
provided by Medtronic, Inc.

™
Figure 3. Sensing configurations in the Percept PC platform are shown in (3a) stimulation can be provided at contact 1 with sensing at contacts 0–2, at contact 2
with sensing at contacts 1–3, or at contacts 1 and 2 with sensing at contacts 0–3. in this signal test, the system has identified peak beta band activity at 22.46 Hz on
contact 1 (3b), 13.67 Hz on contact (3c), and 15.63 Hz on contacts 1 and 2 (3d). the strongest beta power is in contact 1 (3b) which is selected as active therapy to
be clinically programmed. The 5 Hz band of LFP activity around 22.46 Hz (3b) will be passively sensed and the LFP magnitude of 3.36 indicates a strong signal for
passive sensing. image provided by Medtronic, Inc.

this scenario, stimulation amplitude is used as an actuator to threshold recordings are enabled, passive sensing out-of-clinic
influence the LFP for measurement purposes [48]. A low will yield data that can be visualized in graphical format as
amplitude is identified where the signal of interest is of percentage of time spent where the LFP power is above,
greater magnitude, and the corresponding LFP power is cap­ below or between these thresholds. In the case of beta band
tured as the ‘upper LFP threshold’. A higher stimulation ampli­ tracking in PD, these data may indicate the proportion of time
tude (that does not elicit side effects) is identified where the a patient spends in the ‘symptomatic’ (beta power is stronger,
signal of interest has lower magnitude, and the corresponding above the upper threshold) or ‘treated’ (beta power is weaker,
LFP power is captured as the ‘lower LFP threshold’. Once these below the lower threshold) states.
 EXPERT REVIEW OF MEDICAL DEVICES 325

™ ™
Figure 4. The BrainSense survey performed on the Percept PC platform in a patient with Parkinson’s disease treated with left STN DBS, captured when OFF
medications. This function shows the LFP frequency range measured at six different contact combinations (listed on the left side). The highlighted sensing channel, 0
to 3 (white line), shows a peak in the beta frequency range at 20.51 Hz and with a magnitude of 1.15. Beta peaks at the same frequency are seen in other sensing
channels but with a lower magnitude.

™
Figure 5. The ‘timeline’ view of the Percept PC platform in an unmedicated patient with Parkinson’s disease treated with left STN DBS. Figure (5a) shows the
results of passive LFP sensing in the band of interest during a 24-hour period on 13 November 2020. The orange line at the bottom indicates that stimulation was
OFF during this period. (5b) shows the LFP magnitude in the band of interest during a 24-hour period on 28 November 2020. the button in the top right corner
indicates that stimulation was ON, and the orange line at the bottom shows that stimulation was titrated by the patient just before 12:00pm without resulting
change in LFP magnitude. (5c) shows that on 19 November 2020, the patient marked an episode of toe curling at 9:13am. the inset shows the LFP snapshot across
a range of frequencies for the 30-second time period after the event was marked.

In-clinic LFP activity can be visualized using the ‘Streaming’ clinician (Figure 7). This data, along with all out-of-clinic LFP
view. Here the clinician can view the tracked LFP power in real sensing and diary information retrieved during an in-clinic
time from both hemispheres (if there are two electrodes in session can be exported to a JavaScript Object Notation
place and configured for sensing) and can track any changes (JSON) file for offline analysis.
in the LFP that may result from stimulation adjustment or DBS programming using the Percept™ PC DBS device is
during examination tasks or other activities requested by the accomplished in a similar fashion to programming using
326 J. JIMENEZ-SHAHED

™
Figure 6. Capturing LFP Thresholds and the LFP chart. LFP thresholds are set in the BrainSense setup activity and LFP charts are visualized in the ‘events’ tab of
the clinician programmer. (6a) shows the amplitudes used to set LFP thresholds for chronic passive sensing in the left STN. The upper threshold is set using an
amplitude of 1.2 mA (solid teal line), which produced incomplete clinical benefit for the patient. The LFP magnitude at that amplitude is set as the upper threshold
(dotted teal line). The lower threshold is set using an amplitude of 2.8 mA (solid yellow line), which was just below the level at which side effects were experienced.
The LFP magnitude at that amplitude is set as the lower threshold (dotted yellow line). (6b) shows the LFP chart based on this dual threshold sensing over 4 days of
recordings. The proportion of time in a 24-hour period spent with actual LFP magnitude above the upper threshold, below the lower threshold, or between
thresholds is indicated by the colored blocks.

Figure 7. In this ‘Streaming’ view, the top panel shows the LFP power in the selected band in real time while the bottom panel shows the trends in LFP power since
the beginning of the streaming session. In addition, both panels show that LFP thresholds have been set (as described in Figure 6) – the green dotted line for the
upper threshold and yellow dotted line for the lower threshold. In this snapshot, stimulation at 3.1 mA, 60 µsec and 130 Hz (right side of the screen) is associated
with LFP magnitude that usually falls below the lower threshold, indicating suppression of the band of interest. No stimulation changes were made during this
streaming session, as indicated by the level line adjacent to the label ‘mA’ in both panels. however, stimulation can be adjusted by the clinician to observe for any
effects on LFP magnitude. Image provided by Medtronic, Inc.

Medtronic’s previous Activa™ platform with some exceptions is wider with Percept™ PC (20–450µsec that can be titrated in
related to differences in the range of options that are avail­ 10µsec steps, vs. 60–450µsec for Activa™ devices) as is the
able. For example, Activa™ devices can be programmed in range of frequencies (2–250 Hz vs. 30–250 Hz for Activa™
either constant current or constant voltage mode, whereas devices in constant current mode). Stimulation can be imple­
Percept™ PC devices are restricted to current mode (Table mented with or without sensing. Telemetry sessions lasting 1
3). The step size in terms of amplitude titration is more refined h (including streaming) are estimated to reduce IPG longevity
with Percept™ PC devices, where increments of 0.05 mA can by one day [48]. Continuous passive sensing in the out-of-
be used for current strength titration from 0 to 12.5 mA, after clinic setting requires much less energy than telemetry. In an
which a 0.1 mA step size is available, through 25.5 mA of IPG with a 5-year estimated longevity (based on stimulation
current. The range of pulse widths available for programming parameters and other factors), for each month that continuous
 EXPERT REVIEW OF MEDICAL DEVICES 327

Table 3. IPG specifications and range of programming parameters for different deep brain stimulation platforms.
Pulse width Frequency Rechargeable
Amplitude (mA or V) (µsec) (Hz) IPG? Contact Connections
Percept ™ 0–25.5 mA 20–450 2–250 No 2 sets of 8
PC
Activa ™ 0–25.5 mA (current mode) or 0–10.5 V (voltage 60–450 2–250 Yes 2 sets of 8 (PC, RC) or
mode) 1 set of 8 (SC)
Infinity™ 0–12.75 mA 20–500 2–240 No 2 sets of 8
Vercise™ 0.1–20 mA 10–450 2–255 Yes 1 set of 8 or 2 sets of 8
PINS 0–25.0 mA (current mode) or 0–10.0 V (voltage 30–450 2–250 Yes 2 sets of 4 (dual channel) or 1 set of 4 (single
mode) channel)
Neuropacea 0–12.0 mA 40–1000 1–333 Yes 1 set of 4
IPG = implantable pulse generator; PC = primary cell; RC = rechargeable; SC = single channel
a
stimulation is delivered in bursts of duration 10 msec to 5 seconds

passive sensing is turned on, the corresponding reduction in 2.3. Clinical profile and post-marketing findings
longevity is estimated to be 5.4 days [49].
There are no clinical studies on Percept™ PC devices to date.
Medtronic’s research program with the Activa™ PC+S (primary
2.2.2. Additional specifications cell plus sensing) platform was used as the supporting evi­
Additional specifications in the Percept™ PC represent improve­ dence for approval of Percept™ PC devices. The Activa™ PC+S
ments upon the Activa™ platform. The system is conditionally is a chronically implanted, fully internalized but investigational
compatible with 3 T MRI scanners. Clinicians can review system LFP sensing system designed for use in humans in the context
components within the clinician programmer to determine MRI of clinical research protocols. The prototype of the Activa™ PC
eligibility and then configure the ‘MRI mode’ for the patient. In this +S was a bi-directional brain machine interface that was
mode, the stimulation is switched to a bipolar therapy group that intended to simultaneously sense meaningful brain signals in
allows it to remain active during scanning, and once enabled, the presence of therapeutic stimulation, and was first devel­
patients can switch into this mode using their own controller, oped utilizing an existing neurostimulator [53]. It was estab­
and back to therapeutic stimulation after the scan is complete. lished on the premise that a variety of brain diseases will have
During programming sessions, wireless telemetry allows more a detectable biomarker encoded in LFP activity that can be
freedom of patient movement to assess symptom changes in used as a control signal for closed loop stimulation. This
response to adjustments. The IPG itself is 20% smaller and 20% prototype was further developed and studied in an in vivo
thinner than the Activa™ PC IPG and has more rounded corners to model to modulate the circuit of Papez [54]. In this study,
enhance patient comfort (Figure 1). The IPG itself, as previously a biomarker was identified and mapped into classifier and
mentioned, is manufactured to allow real-time prediction of control-policy algorithms that were stored on the device, in
remaining battery life, and can realize a 15% increase in longevity order to continuously titrate stimulation amplitude to achieve
over the Activa™ PC IPG without BrainSense™ technology usage. the desired network effect. Further effort was then undertaken
With median energy use in a typical patient with PD and up to to eliminate stimulation artifact from the recorded signals in
2 months of BrainSense™ usage, the IPG is predicted to last over order to optimize sensing and algorithm performance, such
5 years before requiring replacement. Lastly, software upgrades that the system could chronically detect seizure activity using
can be used to release new or expanded capabilities once they a classification algorithm and adjust stimulation on the basis
become available, without need for a new system to be implanted. of its output [44], thereby providing the initial proof of con­
Future additions to the Percept™ platform are anticipated to cept of closed loop DBS.
include a directional lead and a rechargeable IPG. The Activa™ PC+S was first tested in non-human primates
to detect movement-related changes in cortex over physio­
logically relevant frequency bands with a detectable signal
2.2.3. Safety, contraindications and cost-effectiveness for 24 months, suggesting utility in the evaluation of neuro­
There are no new safety concerns regarding the Percept™ PC stimulation effects in the long term [55]. When implanted in
DBS device, which carries the same contraindications, warn­ five non-human primates with experimentally-induced par­
ings and precautions as other DBS devices manufactured by kinsonism, the Activa™ PC+S demonstrated the ability to
Medtronic, Inc., and which are detailed in the ‘Information for record dynamic changes in LFPs related to different clinical
Prescribers’ manual [50]. Data security is ensured by both states, such as at rest, and during passive joint manipulation
application-level and tablet-based encryption [46]. LFP trans­ and reaching behavior, and concurrent to deep brain stimu­
mission uses the same encrypted Medtronic propriety teleme­ lation [56]. This device was developed as an investigational
try as all other telemetry with the implanted device [51]. yet translational research platform to be used in order to
Further protections for the therapy applications on the clin­ enable the process of biomarker discovery and control algo­
ician programmer are provided by anti-tampering and anti- rithm development and protoyping, with the eventual goal
reverse engineering capabilities [52]. There are no contraindi­ of creating a fully developed closed loop neuromodulation
cations to sensing but the use of sensing to inform clinical system [54].
decision making remains unestablished. There are no cost- Various LFP characteristics are well-recognized to relate to
effectiveness data for the Percept™ PC DBS device. symptoms of PD [22] and can be used to differentiate medication
328 J. JIMENEZ-SHAHED

‘ON’, medication ‘OFF’, tremor, dyskinesia, and sleep states. 5. Conclusion

Although the majority of work has been done in the STN, some
As presented in this device profile, Percept™ PC builds on the
data regarding LFP features in the GPi and ViM of PD patients is
existing Activa™ clinical and research platforms to provide
also described [57]. Less information is available about the LFP
both innovative technologies and general device improve­
features in essential tremor and dystonia. Some investigators have
ments to DBS. The hallmark feature is its BrainSense™ tech­
studied the utility of LFPs in informing the care of patients treated
nology, which allows clinicians to select and track brain signals
with DBS for management of movement disorders including
(LFPs) of interest and relate them to clinical symptoms
intraoperative targeting [58,59] and selection of stimulation para­
reported by the patient. Additional device features such as
meters such as the optimal therapeutic stimulation contact [16].
3 T MRI compatibility, wireless telemetry, IPG design and
Studies reporting on LFP data collected from the Activa™ PC+S
increased longevity all have the potential to improve the
system have yielded important information regarding LFP
patient experience with DBS. The platform itself holds no
dynamics in relation to technical difficulties [60,61], to different
new therapeutic or efficacy claims, and as such does not
actions or tasks [62,63], to time and improvement of PD motor
pose new safety concerns.
symptoms [63–65], to inter-individual variability and disease sever­
ity [63,66], and to varied stimulation parameters and medications
[65,67], all of which will have implications for understanding the
6. Expert opinion
feasibility and long-term utility of sensing, especially as they relate
to the future possibility of closed loop DBS. The discovery that LFP recordings from the deep brain nuclei
that are targeted for DBS therapy in patients with movement
disorders provide informational content related to the disease
3. Alternative devices
state has opened up an exciting world of research aimed at
There are no commercially available alternative sensing understanding the neurophysiologic correlates of disease.
devices for DBS in movement disorders. Tables 1 and Tables Along with this, the possibility of using these unique brain
2 compare the indications and capabilities of Percept™ to signals to further individualize and personalize DBS therapy
those of competing DBS platforms. A research device with through closed loop stimulation of pathologic network activity
similar LFP sensing capabilities is available with has garnered increasing interest and dedicated investigation
a rechargeable IPG (Model G106R, Pins Medical Co., Ltd, over the last decade. Important steps in this process are to
China) [68,69]. Published reports describe its chronic record­ understand the type and nature of the neurophysiologic bio­
ing capabilities to differentiate sleep from wakefulness [70] markers that are present in diseased networks and how best
and its ability to suppress STN beta band activity in PD to modulate them with stimulation to optimize therapeutic
patients [71]. Commercially available devices from the same efficacy.
company are available in China (Table 3) without sensing The unique development pathway for the sensing technol­
capabilities. Two other systems that can record LFP signals ogy contained in the Percept™ PC involved first creating
from externalized DBS macroelectrodes and deliver adaptive a translational research platform (the Activa™ PC+S) that was
DBS have also been described in research use [72,73]. Lastly, intended to enable biomarker discovery and control algorithm
the RNS® System (NeuroPace, Mountain View, CA, USA) is development for future closed loop stimulation. As a result of
a responsive stimulation system indicated for adjunctive ther­ experience gained through the Activa™ PC+S and its proto­
apy in reducing partial onset seizure frequency in adults that types, the Percept™ PC now represents the first opportunity
are refractory to two or more antiepileptic medications [74]. It to observe and capture brain signals of interest in the broader
is a closed loop system that delivers programmable bursts of population of patients receiving DBS while also delivering
stimulation upon seizure detection, with a published report in stimulation, all outside the research environment. Within the
a series of five subjects with medically refractory Tourette BrainSense™ platform, the immediate benefits of brain sen­
syndrome (TS) in which stimulation was delivered in the sing and diary functions include implications for programming
centromedian-parafascicular complex of the thalamus [75]. and correlating symptoms with neurophysiologic features.
In this report, changes in tic symptoms were shown to corre­ Using the Percept™ PC, clinicians will gain experience with
late with modulation of gamma oscillations. In another report passive sensing in patients with movement disorders and gain
of a single TS patient treated in the same target with the same familiarity with how LFP signals relate to factors such as
device [76], the control signal used for adaptive stimulation electrode placement, medication states, stimulation, disease
was a spectral feature in the 5–15 Hz band, which was asso­ characteristics, and clinical symptoms. Well-designed post-
ciated with effective tic reduction. market clinical trials will be needed to illustrate the clinical
utility and added value of using sensing to optimize DBS
therapy, and in which populations. One such study, the
4. Regulatory status
ADAPT-PD trial (NCT04547712) has recently launched with
Percept™ PC is approved for use in the United States, has CE the goal of investigating an adaptive DBS algorithm for perso­
mark approval for the European Union, and is also approved in nalized therapy in PD. In this single-blind study, 100 partici­
Japan, Brazil and Australia with indications for the treatment pants implanted with the Percept™ PC device will be enrolled
of Parkinson’s disease, tremor and dystonia (Table 1). The and randomized to a crossover sequence of adaptive DBS
United States Food and Drug Administration approved using either a single or dual threshold mode. The primary
Percept™ PC as a Class III device on June 25, 2020. outcome measure will be the change in duration of ON time
 EXPERT REVIEW OF MEDICAL DEVICES 329

without troublesome dyskinesia between baseline and 1 and fully realized. Alternately, external factors such as patient
2 months post-randomization. and clinician preferences, and institutional factors may influ­
In addition, the Percept™ PC harbors added benefits to the ence decisions to implant Percept™ PC devices. Renewed
care of DBS patients. Design features may improve patient com­ interest is likely to be seen as new hardware (such as
fort, reduce the frequency of IPG exchanges required, and may a directional lead or rechargeable IPG) or software upgrades
remove obstacles to the non-DBS care of patients with move­ with demonstrated utility become available. When a fully
ment disorders (ie, the patient-enabled MRI mode with 3 T con­ closed-loop system is developed within the Percept™ plat­
ditional safety). A new controller allows patients a more intuitive form, it will require that currently implanted patients transi­
option to engage with their therapy and partner with their tion to a rechargeable IPG. Regardless, the promise of an
treating clinician in recognizing and understanding symptoms. evolving menu of features available on the Percept™ PC
Patients with existing Medtronic DBS leads can readily transition platform is likely to provide multiple opportunities to opti­
to a Percept™ PC at the time of IPG replacement and potentially mize DBS care delivery and the patient experience now and
capture its benefits, including sensing. into the future.
While the many new features in Percept™ PC represent
a major step forward in terms of tools available to clinicians
in order to optimize the care of movement disorders patients 6.1. Five-year view
with DBS, it is important to recognize a few of its limitations. In
Percept™ PC and its predecessor, the Activa™ PC+S, were
newly implanted patients, capitalizing on BrainSense™ tech­
developed with the goal in mind of identifying disease bio­
nology during post-operative management still depends
markers that can be used in closed loop neuromodulation
greatly on accurate electrode placement. To date, the
systems. Closed loop DBS for movement disorders such as
Percept™ PC is only able to record LFPs post-implantation,
Parkinson’s disease holds promise to reduce side effects and
so alternate methods need to be used to verify that
improve efficiency while maintaining or even enhancing motor
a potentially useful signal is present intra-operatively.
symptom control. Several pilot studies of closed loop DBS have
Second, the device only records from specific stimulation
already shown evidence of short-term success in small num­
configurations, the use of which again depends on electrode
bers of patients at reducing PD or tremor symptoms at lower
placement, but also patient-specific disease features. For
therapeutic current [83,84] but longer term studies in larger
example, when stimulation of the dorsal contact is required
patient groups will be needed. Additional work is needed to
in the STN for management of dyskinesia in PD,
identify whether closed loop DBS is better deployed as an on-
a corresponding sensing configuration may not be possible
demand therapy or with continuous modulation based on
to achieve. Third, the Percept™ PC will only record a specified
a changing control signal, and whether that control signal
5 Hz-wide band of LFP activity, or can portray the frequency
should be based on intrinsic signals such as LFPs or peripheral
band of interest as being above, below or between specified
ones such as wearable sensors [85]. A combination of
thresholds of LFP magnitude. Interpretation of the nature of
approaches will likely need to be available in order to address
patient-marked events may therefore depend on accurate
the unique needs of individual patients. Continued innovation
identification of relevant LFP bands, or when peak beta fre­
across these areas is likely to result in further refinements in
quency is being tracked, may be limited to understanding
DBS delivery methods and improved outcomes for patients
whether they occur during medication ‘ON’ or ‘OFF’ states.
with movement disorders, can be applied to other disease
Fourth, although Percept™ PC is available for use in tremor
states or to non-motor features, and will hopefully be readily
and dystonia patients, the LFP patterns in these disorders are
accessible to patients with movement disorders through sim­
not as well described as in PD and further research is required
ple software upgrades.
to understand how sensing can be used to optimize care in
these populations [77–79]. Lastly, there is increasing evidence
regarding the informational content within the high-frequency Funding
oscillations of deep brain nuclei (i.e., LFPs occurring in the
This paper was not funded.
200–450 Hz range) which can be linked to lower frequency
oscillations via phase amplitude coupling [80–82], but these
are outside the range of detection by the Percept™ PC.
Reviewer disclosures
As with any new technology, initial high rates of adoption
are likely to be seen as clinicians explore the potential for One peer reviewer is a shareholder of Newronika SPA. Peer reviewers on
value added to their practices and care of their patients. this manuscript have no other relevant financial relationships or otherwise
to disclose.
However, there is an anticipated learning curve that will be
required for physicians to familiarize themselves with
a neurophysiologic approach to understanding disease and
Declaration of interest
DBS care which may be more time consuming. This com­
bined with the limitations of Percept™ PC in its current form J Jimenez-Shahed has received consulting fees from Abbott/St. Jude
may therefore temper initial interest for some practitioners. Medical for service on a study advisory board. The author has no other
relevant affiliations or financial involvement with any organization or
However, continued research efforts will progressively eluci­ entity with a financial interest in or financial conflict with the subject
date strategies for the application of brain sensing, to which matter or materials discussed in the manuscript apart from those
patients implanted now will have access when they become disclosed.
330 J. JIMENEZ-SHAHED

References with repeated replacements: just a matter of energy delivered?

Brain Stimul. 2019;12(4):845–850. PMID: 30876884. Epub 2019/
Papers of special note have been highlighted as either of interest (•) or of 03/17.
considerable interest (••) to readers. 19. Niemann M, Schneider GH, Kuhn A, et al. Longevity of implantable
1. Krack P, Volkmann J, Tinkhauser G, et al. Deep brain stimulation in pulse generators in bilateral deep brain stimulation for movement
movement disorders: from experimental surgery to evidence-based disorders. Neuromodulation. 2018;21(6):597–603. Epub 2017/ 12/
therapy. Mov Disord. 2019;34(12):1795–1810. PMID: 31580535. 22. PMID: 29265529.
Epub 2019/ 10/04. 20. Sette AL, Seigneuret E, Reymond F, et al. Battery longevity of neurosti­
•• presents the current clinical state of the field of DBS for mulators in Parkinson disease: a historic cohort study. Brain Stimul.
movement disorders. 2019;12(4):851–857. PMID: 30842036. Epub 2019/ 03/08.
2. Kalia LV, Lang AE. Parkinson’s disease. Lancet. 2015;386 21. Parastarfeizabadi M, Kouzani AZ. Advances in closed-loop deep
(9996):896–912. Epub 2015/ 04/24. PMID: 25904081. brain stimulation devices. J Neuroeng Rehabil. 2017;14(1):79.
3. Jimenez-Shahed J. A review of current and novel levodopa formu­ Epub 2017/ 08/13. PMID: 28800738; PMCID: PMCPMC5553781.
lations for the treatment of Parkinson’s disease. Ther Deliv. 2016;7 22. Thompson JA, Lanctin D, Ince NF, et al. Clinical implications of local field
(3):179–191. Epub 2016/ 02/20. PMID: 26893250. potentials for understanding and treating movement disorders.
4. Kim HJ, Mason S, Foltynie T, et al. Motor complications in Stereotact Funct Neurosurg. 2014;92(4):251–263. PMID: 25170784.
Parkinson’s disease: 13-year follow-up of the CamPaIGN cohort. Epub 2014/ 08/30.
Mov Disord. 2020;35(1):185–190. Epub 2020/ 01/23. PMID: •• review of LFPs in movement disorders.
31965629; PMCID: PMCPMC7063985. 23. Telkes I, Ince NF, Onaran I, et al. Localization of subthalamic
5. Schuepbach WM, Rau J, Knudsen K, et al. Neurostimulation for nucleus borders using macroelectrode local field potential
Parkinson’s disease with early motor complications. N Engl J Med. recordings. PMID: 25570528. Annu Int Conf IEEE Eng Med Biol
2013;368(7):610–622. PMID: 23406026. Epub 2013/ 02/15. Soc. 2014;2014:2621–2624. DOI:10.1109/EMBC.2014.6944160.
6. Paschen S, Deuschl G. Patient evaluation and selection for movement PMID: 25570528.
disorders surgery: the changing spectrum of indications. Prog Neurol 24. Telkes I, Ince NF, Onaran I, et al. Spatio-spectral characterization of
Surg Epub 2018/ 01/15. 2018;33:80–93. PMID: 29332075 local field potentials in the subthalamic nucleus via multitrack
• describes issues related to patient selection for DBS in movement microelectrode recordings. PMID: 26737552. [Annu Int Conf IEEE
disorders. Eng Med Biol Soc. 2015;2015:5561–5564. Epub 2016/ 01/07].
7. Deuschl G, Raethjen J, Hellriegel H, et al. Treatment of patients with 25. Ray NJ, Jenkinson N, Wang S, et al. Local field potential beta
essential tremor. Lancet Neurol. 2011;10(2):148–161. Epub 2011/ activity in the subthalamic nucleus of patients with Parkinson’s
01/25. PMID: 21256454. disease is associated with improvements in bradykinesia after
8. Huss DS, Dallapiazza RF, Shah BB, et al. Functional assessment and dopamine and deep brain stimulation. Exp Neurol. 2008;213
quality of life in essential tremor with bilateral or unilateral DBS and (1):108–113. PMID: 18619592. Epub 2008/ 07/16.
focused ultrasound thalamotomy. Mov Disord. 2015;30(14):1937–1943. 26. Brown P, Oliviero A, Mazzone P, et al. Dopamine dependency of
Epub 2016/ 01/16. PMID: 26769606. oscillations between subthalamic nucleus and pallidum in
9. Clark LN, Louis ED. Essential tremor. Handb Clin Neurol. Parkinson’s disease. J Neurosci. 2001;21(3):1033–1038. PMID:
2018;147:229–239. Epub 2018/ 01/13. PMID: 29325613; PMCID: 11157088; PMCID: PMCPMC6762327. Epub 2001/ 02/07.
PMCPMC5898616. 27. Foffani G, Priori A, Egidi M, et al. 300-Hz subthalamic oscillations in
10. Zesiewicz TA, Elble RJ, Louis ED, et al. Evidence-based guideline Parkinson’s disease. Brain. 2003;126(Pt10):2153–2163. PMID:
update: treatment of essential tremor: report of the quality stan­ 12937087. Epub 2003/ 08/26.
dards subcommittee of the american academy of Neurology. 28. Alonso-Frech F, Zamarbide I, Alegre M, et al. Slow oscillatory activ­
Neurology. 2011;77(19):1752–1755. PMID: 22013182; PMCID: ity and levodopa-induced dyskinesias in Parkinson’s disease. Brain.
PMCPMC3208950. Epub 2011/ 10/21. 2006;129(Pt7):1748–1757. PMID: 16684788. Epub 2006/ 05/11.
11. Albanese A, Di Giovanni M, Lalli S. Dystonia: diagnosis and 29. Kane A, Hutchison WD, Hodaie M, et al. Enhanced synchronization
management. Eur J Neurol. 2019;26(1):5–17. Epub 2018/ 07/24. PMID: of thalamic theta band local field potentials in patients with essen­
30035844. tial tremor. Exp Neurol. 2009;217(1):171–176. Epub 2009/ 02/24.
12. Koeglsperger T, Palleis C, Hell F, et al. Deep brain stimulation PMID: 19233174.
programming for movement disorders: current concepts and 30. Chen CC, Kuhn AA, Trottenberg T, et al. Neuronal activity in globus
evidence-based strategies. Front Neurol Epub 2019/ 06/25. pallidus interna can be synchronized to local field potential activity
2019;10:410. PMID: 31231293; PMCID: PMCPMC6558426 over 3–12 Hz in patients with dystonia. Exp Neurol. 2006;202
• describes strategies for programming DBS devices without (2):480–486. Epub 2006/ 08/26. PMID: 16930593.
sensing. 31. De Hemptinne C, Ryapolova-Webb ES, Air EL, et al. Exaggerated
13. Schüpbach WMM. Impulsivity, impulse control disorders, and sub­ phase-amplitude coupling in the primary motor cortex in Parkinson
thalamic stimulation in Parkinson’s disease. Basal Ganglia. 2012;2 disease. Proc Natl Acad Sci U S A. 2013;110(12):4780–4785. PMID:
(4):205–209. 23471992; PMCID: PMCPMC3606991. Epub 2013/ 03/09.
14. Kronenbuerger M, Fromm C, Block F, et al. On-demand deep brain 32. De Hemptinne C, Swann NC, Ostrem JL, et al. Therapeutic deep
stimulation for essential tremor: a report on four cases. Mov Disord. brain stimulation reduces cortical phase-amplitude coupling in
2006;21(3):401–405. PMID: 16211619. Epub 2005/ 10/08. Parkinson’s disease. Nat Neurosci. 2015;18(5):779–786. PMID:
15. Rebelo P, Green AL, Aziz TZ, et al. Thalamic directional deep brain 25867121; PMCID: PMCPMC4414895. Epub 2015/ 04/14.
stimulation for tremor: spend less, get more. Brain Stimul. 2018;11 33. Van Wijk BC, Beudel M, Jha A, et al. Subthalamic nucleus phase-
(3):600–606. PMID: 29373260. Epub 2018/ 01/27. amplitude coupling correlates with motor impairment in
16. Ince NF, Gupte A, Wichmann T, et al. Selection of optimal program­ Parkinson’s disease. Clin Neurophysiol. 2016;127(4):2010–2019.
ming contacts based on local field potential recordings from sub­ PMID: 26971483; PMCID: PMCPMC4803022. Epub 2016/ 03/15.
thalamic nucleus in patients with Parkinson’s disease. 34. Miocinovic S, Somayajula S, Chitnis S, et al. History, applications,
Neurosurgery. 2010;67(2):390–397. PMID: 20644424; PMCID: and mechanisms of deep brain stimulation. JAMA Neurol. 2013;70
PMCPMC4319368. Epub 2010/ 07/21. (2):163–171. Epub 2013/ 02/15. PMID: 23407652.
17. Duffley G, S A, Lutz B, et al. Mobile application for Parkinson’s 35. Kringelbach ML, Jenkinson N, Owen SL, et al. Translational princi­
disease deep brain stimulation (MAP DBS): an open-label, multisite, ples of deep brain stimulation. Nat Rev Neurosci. 2007;8
randomized, controlled clinical trial. (submitted). 2020. (8):623–635. Epub 2007/ 07/20. PMID: 17637800.
18. Israeli-Korn SD, Fay-Karmon T, Tessler S, et al. Decreasing battery 36. Merola A, Romagnolo A, Krishna V, et al. current directions in deep
life in subthalamic deep brain stimulation for Parkinson’s disease brain stimulation for parkinson’s disease-directing current to
 EXPERT REVIEW OF MEDICAL DEVICES 331

maximize clinical benefit. Neurol Ther. 2020;9(1):25–41. Epub 2020/ 52. Medtronic. DBS security reference guide. Minneapolis, MN:
03/12. PMID: 32157562; PMCID: PMCPMC7229063. Medtronic, Inc.; 2020. Available from: http://manuals.medtronic.
•• reviews current technologies included in various DBS com/content/dam/emanuals/neuro/NDHF1550-189563.pdf
platforms. 53. Stanslaski S, Cong P, Carlson D, et al. An implantable bi-directional
37. Zhang S, Silburn P, Pouratian N, et al. Comparing current steering brain-machine interface system for chronic neuroprosthesis
technologies for directional deep brain stimulation using research. Annu Int Conf IEEE Eng Med Biol Soc.
a computational model that incorporates heterogeneous tissue 2009;2009:5494–5497. Epub 2009/ 12/08. PMID: 19965049.
properties. Neuromodulation. 2020;23(4):469–477. PMID: 54. Afshar P, Khambhati A, Stanslaski S, et al. A translational platform
31423642; PMCID: PMCPMC7318189. Epub 2019/ 08/20. for prototyping closed-loop neuromodulation systems. Front
38. Steigerwald F, Timmermann L, Kuhn A, et al. Pulse duration set­ Neural Circuits. 2012;6:117. Epub 2013/ 01/25. PMID: 23346048;
tings in subthalamic stimulation for Parkinson’s disease. Mov PMCID: PMCPMC3551193.
Disord. 2018;33(1):165–169. PMID: 29165837; PMCID: 55. Ryapolova-Webb E, Afshar P, Stanslaski S, et al. Chronic cortical and
PMCPMC5813170. Epub 2017/ 11/23. electromyographic recordings from a fully implantable device: pre­
39. Kirsch AD, Hassin-Baer S, Matthies C, et al. Anodic versus cathodic clinical experience in a nonhuman primate. J Neural Eng. 2014;11
neurostimulation of the subthalamic nucleus: a (1):016009. PMID: 24445430. Epub 2014/ 01/22.
randomized-controlled study of acute clinical effects. Parkinsonism 56. Connolly AT, Muralidharan A, Hendrix C, et al. Local field potential
Relat Disord. 2018;55:61–67. Epub 2018/ 05/23. PMID: 29784559. recordings in a non-human primate model of Parkinsons disease
40. Boston Scientific Corporation. Guide(TM) XT software user guide - using the Activa PC + S neurostimulator. J Neural Eng. 2015;12
version 2.0.2. Valencia, CA: Boston Scientific Neuromodulation (6):066012. PMID: 26469737; PMCID: PMCPMC5130227. Epub 2015/
Corporation; 2018. Available from: https://www.bostonscientific. 10/16.
com/content/dam/Manuals/au/current-rev-en/92093042-02_GUIDE 57. Halje P, Brys I, Mariman JJ, et al. Oscillations in cortico-basal ganglia
%E2%84%A2_XT_Software_User_Guide_s.pdf circuits: implications for Parkinson’s disease and other neurologic
41. Boston Scientific Corporation. Vercise (TM) neural navigator 3 pro­ and psychiatric conditions. J Neurophysiol. 2019;122(1):203–231.
gramming manual. Valencia, CA: Boston Scientific Epub 2019/ 05/03. PMID: 31042442.
Neuromodulation Corporation. Available from: https://www.boston 58. Telkes I, Jimenez-Shahed J, Viswanathan A, et al. Prediction of STN-
scientific.com/content/dam/Manuals/us/current-rev-en/92376212- DBS electrode implantation track in parkinson’s disease by using
02_Vercise_Neural_Navigator_3_(Windows%2010)_Programming_ local field potentials. Front Neurosci. 2016;10:198. Epub 2016/ 06/
Manual_en-US_s.pdf 01. PMID: 27242404; PMCID: PMCPMC4860394.
42. Carl B, Bopp M, Sass B, et al. Visualization of volume of tissue 59. Ozturk M, Telkes I, Jimenez-Shahed J, et al. Randomized,
activated modeling in a clinical planning system for deep brain double-blind assessment of LFP versus SUA guidance in STN-DBS
stimulation. J Neurosurg Sci. 2020 PMID: 32031356. 10.23736/ lead implantation: a pilot study. Front Neurosci. 2020;14:611. Epub
S0390-5616.19.04827-6 Epub 2020/ 02/08. 2020/ 07/14. PMID: 32655356; PMCID: PMCPMC7325925.
43. Thompson JA, Tekriwal A, Felsen G, et al. Sleep patterns in 60. Neumann WJ, Staub F, Horn A, et al. Deep brain recordings using
Parkinson’s disease: direct recordings from the subthalamic an implanted pulse generator in Parkinson’s disease.
nucleus. J Neurol Neurosurg Psychiatry. 2018;89(1):95–104. PMID: Neuromodulation. 2016;19(1):20–24. PMID: 26387795; PMCID:
28866626. Epub 2017/ 09/04. PMCPMC4881811. Epub 2015/ 09/22.
44. Stanslaski S, Afshar P, Cong P, et al. Design and validation of a fully 61. Steiner LA, Neumann WJ, Staub-Bartelt F, et al. Subthalamic beta
implantable, chronic, closed-loop neuromodulation device with dynamics mirror Parkinsonian bradykinesia months after neurosti­
concurrent sensing and stimulation. IEEE Trans Neural Syst mulator implantation. Mov Disord. 2017;32(8):1183–1190. PMID:
Rehabil Eng. 2012;20(4):410–421. PMID: 22275720. Epub 2012/ 01/ 28639263; PMCID: PMCPMC5575541. Epub 2017/ 06/24.
26. 62. Quinn EJ, Blumenfeld Z, Velisar A, et al. Beta oscillations in freely
45. Okun MS. Tips for Choosing a Deep Brain Stimulation Device. JAMA moving Parkinson’s subjects are attenuated during deep brain
Neurol. 2019;76(7):749–750. Epub 2019/ 04/30. PMID: 31034008. stimulation. Mov Disord. 2015;30(13):1750–1758. PMID: 26360123.
46. Medtronic. Clinician programmer application for deep brain stimu­ Epub 2015/ 09/12.
™
lation – clinican programming guide – percept pc model b35200 63. Hanrahan SJ, Nedrud JJ, Davidson BS, et al. Long-term task- and
neurostimulator – application version 2.0. Minneapolis, MN: dopamine-dependent dynamics of subthalamic local field poten­
Medtronic, Inc.; 2020. Available from: http://manuals.medtronic. tials in Parkinson’s disease. Brain Sci. 2016;6(4):57. PMID: 27916831;
com/content/dam/emanuals/neuro/M988717A_a_015_view_color. PMCID: PMCPMC5187571. Epub 2016/ 12/06.
pdf 64. Trager MH, Koop MM, Velisar A, et al. Subthalamic beta oscillations
•• manufacturer-published manual summarizing the specifica­ are attenuated after withdrawal of chronic high frequency neuro­
™
tions of the Percept PC. stimulation in Parkinson’s disease. Neurobiol Dis. 2016;96:22–30.
47. Bock DC, Marschilok AC, Takeuchi KJ, et al. Batteries used to power Epub 2016/ 10/19. PMID: 27553876.
implantable biomedical devices. Electrochim Acta. 65. Neumann WJ, Staub-Bartelt F, Horn A, et al. Long term correlation
2012;84:155–164. of subthalamic beta band activity with motor impairment in
48. Medtronic. Percept(TM) PC neurostimulator with brainsense(TM) patients with Parkinson’s disease. Clin Neurophysiol. 2017;128
Technology - Supplemental Information for Sensing Only (11):2286–2291. PMID: 29031219; PMCID: PMCPMC5779610. Epub
Configuration. Minneapolis, MN: Medtronic, Inc.; 2020. 2017/ 10/17.
49. Medtronic. System eligibility - battery longevity - deep brain sti­ 66. Canessa A, Pozzi NG, Arnulfo G, et al. Striatal dopaminergic inner­
mulation systems. Minneapolis, MN: Medtronic, Inc.; 2019. Available vation regulates subthalamic beta-oscillations and cortical-
from: http://manuals.medtronic.com/content/dam/emanuals/ subcortical coupling during movements: preliminary evidence in
neuro/M929534A111A_view_d1581623080025.pdf subjects with Parkinson’s disease. Front Hum Neurosci.
50. Medtronic. Medtronic deep brain stimulation therapy - 2016;10:611. Epub 2016/ 12/22. PMID: 27999534; PMCID:
implanted neurostimulators - information for prescribers. PMCPMC5138226.
Minneapolis, MN: Medtronic, Inc.; 2020. Available from: http:// 67. Blumenfeld Z, Koop MM, Prieto TE, et al. Sixty-hertz stimulation
manuals.medtronic.com/content/dam/emanuals/neuro/ improves bradykinesia and amplifies subthalamic low-frequency
M927893A_a_101_view.pdf oscillations. Mov Disord. 2017;32(1):80–88. PMID: 27859579. Epub
51. Medtronic. HCP Brochure. Minneapolis, MN: Medtronic, Inc.; 2020. 2016/ 11/20.
Available from: https://www.medtronic.com/content/dam/medtro 68. Qian X, Hao HW, Ma BZ, et al. Implanted rechargeable electroen­
nic-com/products/neurological/dbs/documents/dbs-percept-pc- cephalography (EEG) device. Electron Lett. 2014;50(20):1419–1420.
hcp-brochure.pdf PMID: WOS: 000343237000010.
332 J. JIMENEZ-SHAHED

69. Qian X, Chen Y, Ma BZ, et al. Chronically monitoring the deep brain brain stimulation with wearable sensors in an essential tremor
rhythms: from stimulation to recording. Sci Bull. 2016;61 patient. J Neurosurg. 2017;127(3):580–587. Epub 2016/ 11/20.
(19):1522–1524. PMID: WOS: 000385198900008. PMID: 27858575.
70. Chen Y, Gong C, Hao HW, et al. automatic sleep stage classification 79. Pina-Fuentes D, Beudel M, Little S, et al. Toward adaptive deep
based on subthalamic local field potentials. Ieee T Neur Sys Reh. brain stimulation for dystonia. Neurosurg Focus. 2018;45(2):E3.
2019;27(2):118–128. PMID: WOS: 000458786600002. PMID: 30064317. Epub 2018/ 08/02.
71. Chen Y, Gong C, Tian Y, et al. Neuromodulation effects of deep 80. Telkes I, Viswanathan A, Jimenez-Shahed J, et al. Local field poten­
brain stimulation on beta rhythm: a longitudinal local field poten­
tials of subthalamic nucleus contain electrophysiological footprints
tial study. Brain Stimul. 2020;13(6):1784–1792. PMID: WOS: of motor subtypes of Parkinson’s disease. Proc Natl Acad Sci U S A.
000597945200044.
2018;115(36):E8567–E76. PMID: 30131429; PMCID:
72. Little S, Pogosyan A, Neal S, et al. Adaptive deep brain stimulation
PMCPMC6130371. Epub 2018/ 08/23.
in advanced Parkinson disease. Ann Neurol. 2013;74(3):449–457.
PMID: 23852650; PMCID: PMCPMC3886292. Epub 2013/ 07/16. 81. Jimenez-Shahed J, Telkes I, Viswanathan A, et al. GPi oscillatory
activity differentiates tics from the resting state, voluntary move­
73. Arlotti M, Rossi L, Rosa M, et al. An external portable device for
ments, and the unmedicated Parkinsonian state. Front Neurosci.
adaptive deep brain stimulation (aDBS) clinical research in
advanced Parkinson’s Disease. Med Eng Phys. 2016;38(5):498–505. 2016;10:436. Epub 2016/ 10/14. PMID: 27733815; PMCID:
PMCPMC5039204.
Epub 2016/ 04/01. PMID: 27029510.
74. NeuroPace. RNS(R) system physician manual - for the RNS(R) neuro­ 82. Ozturk M, Abosch A, Francis D, et al. Distinct subthalamic coupling
stimulator model RNS-320. Mountain View, CA: NeuroPace, Inc.; 2020. in the ON state describes motor performance in Parkinson’s dis­
75. Maling N, Hashemiyoon R, Foote KD, et al. Increased thalamic ease. Mov Disord. 2020;35(1):91–100. Epub 2019/ 07/28. PMID:
gamma band activity correlates with symptom relief following 31348550.
deep brain stimulation in humans with Tourette’s syndrome. PLoS 83. Meidahl AC, Tinkhauser G, Herz DM, et al. Adaptive deep brain
One. 2012;7(9):e44215. Epub 2012/ 09/13. PMID: 22970181; PMCID: stimulation for movement disorders: the long road to clinical
PMCPMC3435399. therapy. Mov Disord. 2017;32(6):810–819. PMID: 28597557;
76. Molina R, Okun MS, Shute JB, et al. Report of a patient undergoing PMCID: PMCPMC5482397. Epub 2017/ 06/10.
chronic responsive deep brain stimulation for Tourette syndrome: •• describes challenges associated with bringing closed loop
proof of concept. J Neurosurg. 2018;129(2):308–314. . PMID: DBS to clinical practice.
28960154; PMCID: PMCPMC7007215. Epub 2017/ 09/30. 84. Little S, Brown P. Debugging adaptive deep brain stimulation for
77. Opri E, Cernera S, Molina R, et al. Chronic embedded Parkinson’s disease. Mov Disord. 2020;35(4):555–561. Epub 2020/
cortico-thalamic closed-loop deep brain stimulation for the treat­ 02/11. PMID: 32039501; PMCID: PMCPMC7166127.
ment of essential tremor. Sci Transl Med. 2020;12(572):eaay7680. 85. Kuo CH, White-Dzuro GA, Ko AL. Approaches to closed-loop deep
PMID: 33268512. Epub 2020/ 12/04. brain stimulation for movement disorders. Neurosurg Focus.
78. Herron JA, Thompson MC, Brown T, et al. Chronic electrocortico­ 2018;45(2):E2. PMID: 30064321. Epub 2018/ 08/02.
graphy for sensing movement intention and closed-loop deep •• reviews existing literature on closed loop DBS.