Journal of Robotic Surgery

https://doi.org/10.1007/s11701-020-01147-7

REVIEW ARTICLE

Robotic‑assisted cortical bone trajectory (CBT) screws using the Mazor

X Stealth Edition (MXSE) system: workflow and technical tips for safe
and efficient use
John A. Buza III1 · Christopher R. Good2 · Ronald A. Lehman Jr.3 · John Pollina4 · Richard V. Chua5 ·
Avery L. Buchholz6 · Jeffrey L. Gum1 

Received: 29 April 2020 / Accepted: 22 September 2020

© Springer-Verlag London Ltd., part of Springer Nature 2020

Abstract
Robotic-assisted spine surgery has a number of potential advantages, including more precise pre-operative planning, a high
degree of accuracy in screw placement, and significantly reduced radiation exposure to the surgical team. While the current
primary goal of these systems is to improve the safety of spine surgery by increasing screw accuracy, there are a number of
technical errors that may increase the risk of screw malposition. Given the learning curve associated with this technology,
it is important for the surgeon to have a thorough understanding of all required steps. In this article, we will demonstrate the
setup and workflow of a combined navigation and robotic spine surgery platform using the Mazor X Stealth Edition (MXSE)
system to place cortical-based trajectory (CBT) screws, including a review of all technical tips and pearls to efficiently
perform this procedure with minimal risk of screw malposition. In this article, we will review surgical planning, operating
room setup, robotic arm mounting, registration, and CBT screw placement using the MXSE system.

Keywords  Robotic surgery · Spine · Workflow · Technique · Mazor X

Introduction [2, 3]. This robot-assist platform was small, spine-mounted

which had an attached arm with six degrees of freedom,
In 2001, Mazor Robotics began work on a small spine- which provided the trajectory for pedicle screw instrumen-
mounted robot which could assist with pedicle screw instru- tation. The surgeon could then perform all instrumentation
mentation [1]. In 2004, the first generation of this technol- through this guided trajectory. Mazor Robotics released the
ogy, called Spine Assist, was FDA-approved for clinical use second generation, termed the Renaissance, in 2011, and
followed with the release of the third generation Mazor X
John A. Buza III and Jeffrey L. Gum have contributed equally to in 2016 [4, 5]. The most recent version was first utilized in
this work. January of 2019 and it combines navigation and robotics
into a single platform called the Mazor X Stealth Edition
* Jeffrey L. Gum
(MXSE) [6]. Mazor robotic systems are the most widely
[email protected]
used spinal robotics in the United States and globally. As
1
Department of Spine Surgery, Norton Leatherman Spine of mid-2018, these robotic systems were in more than 250
Center, 210 E. Gray St. Suite 900, Louisville, KY 40202, centers and 40 countries worldwide [5]. In the last several
USA
years, two competitors have hit the market. The ROSA
2
Virginia Spine Institute, Reston, VA, USA robot (Zimmer-Biomet) is a floor mounted robot which has
3
Daniel and Jane Och Spine Hospital, New York Presbyterian, combined robotic and navigation functions [7]. The Globus
Columbia University Medical Center, New York, NY, USA Medical Excelsius GPS Robot was FDA-approved in 2017,
4
Department of Neurosurgery, University at Buffalo, Buffalo, is also floor mounted, and has combined robotic and naviga-
NY, USA tion functions [8].
5
Northwest Neurospecialists, Tuscon, AZ, USA Given the widespread use of the Mazor spine robot, most
6
Department of Neurosurgery, University of Virginia Health studies in the literature evaluating the accuracy of robotic-
System, Charlottesville, VA, USA assisted screw placement have utilized one of the Mazor

13
Vol.:(0123456789)
 Journal of Robotic Surgery

platforms. Most recent reports of the Mazor system have (CBT) screws, including surgical planning, operating room
focused on traditional pedicle screw trajectory, with reported setup, robotic arm mounting, registration, pedicle cannula-
accuracy in the range of 97–99% using the Gertzbein–Rob- tion, and CBT screw insertion. Each stage will include tips to
bins grading system [4, 7, 9–19]. Overall, although there avoid errors which may ultimately lead to screw malposition.
are limitations to the current literature, most studies have We will include the lessons which we have learned from the
found a fairly high and reproducible accuracy for pedicle early adoption of this technology at our institution, which has
screw placement using robotic-assist systems. Additionally, allowed robotic-assisted spine surgery to be performed in a
with the adoption of any new technology, it is important safe, efficient, and reproducible manner.
to consider the learning curve associated with its use. It
can be very beneficial for new adopters to understand the Surgical planning
learning process and have reasonable expectations to avoid
early abandonment. While there are a limited number of One of the main purported advantages of robotic-assisted
studies, it appears that the learning curve is in the range of spine surgery is improved pre-operative planning using soft-
approximately 15–30 cases [4, 20, 21]. In a study of 102 ware for pre-operative planning. The MXSE system is CT
patients undergoing robotic-assisted pedicle screw place- based and requires that the patient undergo a CT scan which
ment, Hu et al. found that the rate of successful placement can be done either preoperatively (CT to Fluoro) or intraop-
increased after the first 30 cases [4]. In a matched cohort eratively (Scan and Plan). For the CT to Fluoro workflow,
comparison of 95 patients undergoing robot-assisted versus the CT scan is performed using 1 mm-thick slices and then
fluoroscopic-guided pedicle screw placement in the lumbar loaded to the MXSE and planned prior to incision. This is
spine, Schatlo et al. found that the highest rate of conversion our preferred workflow as it avoids intraoperative time for
from robot-assisted to manual placement occurred in the planning. When utilizing the scan and plan workflow, an
first 5–25 patients [20]. It is likely that for most spine sur- intraoperative CT such as O-arm is utilized and the data
geons who have experience performing fluoroscopic-assisted are transferred to the MXSE and then planned, while the
pedicle screw placement, the use of robotic-assisted pedicle O-arm is being removed. Both modalities allow the surgeon
screw insertion can be adopted fairly quickly. to plan CBT screw trajectory, diameter, and length. There
While most studies on robotic-assisted screw placement are several advantages for CBT while planning. The starting
have evaluated traditional pedicle screw trajectory, recent points may be placed more medially, or in a more divergent
reports have evaluated cortical bone trajectory (CBT) screws pattern (in the cranial-caudal plane) to reduce the size of
using robotic assistance [22, 23]. CBT screws were first the skin incision and dissection in an effort to be more mini-
described in 2009 by Santoni et al. as a minimally invasive mally invasive. This reduction of tissue dissection is con-
alternative to traditional pedicle trajectory screws during sistent with the benefits of CBT. When planning, we align
posterior lumbar interbody fusion procedures [24]. CBT the tulips in the coronal plane for a straight rod, and in the
screws have a number a potential advantages. The corti- sagittal plane, we diverge the screws, so that their insertion
cal trajectory has an inferomedial entry point with a more pathway merges on the skin truly minimizing the incision
superolateral trajectory, which requires a smaller incision to size (Fig. 1a–b). We have modified our sagittal screw tra-
place instrumentation. In addition, lateral soft-tissue dissec- jectory over time. The UIV or most cranial screws need to
tion is limited to the region of the pars interarticularis, with start on the pars just distal to the proximal facet and must
less retraction required. In addition, CBT screws have been avoid facet violation at all costs to reduce adjacent segment
shown to have higher insertional torque and pullout strength breakdown. However, the remaining screws can be altered
compared with traditional pedicle screw trajectory [25–27]. to start within the facet (trans-facet), since those levels are
Clinical outcome studies in patients with degenerative lum- being fused (Fig. 2a–c). It is important to prep the surface to
bar disorders have demonstrated similar pain, functional, be relatively flat to avoid skive into the joint or off trajectory.
and disability scores as traditional pedicle screws but with Additionally, since the screw size has already been planned,
a lower complication profile [28–30]. While recent reports the surgical team can already have prepared (improve effi-
have described the use of robotic-assisted surgery to place ciency) and potentially sterile packed individually to, in
CBT screws, the surgical technique of this procedure has not theory, reduce waste.
been well described in the literature [22, 23].
Considering the learning curve associated with robotic-
assisted spine surgery, it is important that surgeons incorpo- Operating room setup
rating this technology into their practice be familiar with the
technical pitfalls that may occur during all stages of robotic- The patient may be placed prone or lateral on a radiolu-
assisted spine surgery. This article will review the technical cent table. The hands should be placed above the head, so
aspects of utilizing the MXSE for cortical bone trajectory that the arms will not interfere when acquiring images. It is

13
Journal of Robotic Surgery

Fig. 1  Planning views from

an L3-S1 CBT construct after
having an ALIF at L5-S1. a
Coronal view showing the tulips
lining up to keep the rod as
straight as possible in this plane.
b Sagittal view showing that the
trajectory should be modified to
align and intersect on the skin to
minimize the incision size

Fig. 2  Comparison of CBT constructs. a A classic CBT construct straight in as a trans-facet orientation. c The axial view of the plan-
with both UIV and LIV screws starting on the pars with an up/out ning process of the modified LIV screws
direction. b A modified CBT construct with the LIV screws more

important to secure the patient to the operating room table Robotic arm mounting
prior to beginning the procedure. If the table does not have
a frame, a bed frame must be attached to the operating room After the patient is placed on the operating room table
table before transferring the patient onto the operating room and prior to skin preparation, the next step is to mount the
table. The robotic base should be placed at the foot of the MXSE system onto the bed frame. Given the weight of the
operating room table, as this will allow room for the robotic robotic arm, there is an on-board automated lifting mecha-
arm mount and room for fluoroscopy. C-arm fluoroscopy nism stored within the MXSE workstation which allows
may approach the patient from either side, but should be for mounting onto the bed frame. We prefer to mount the
on the side opposite the sterile field. The MXSE Robotics robotic arm at the foot of the bed on the patient’s right
image adaptor must be attached to the C-arm image intensi- side (when prone), although it can be placed on either side
fier prior to surgery (Fig. 3).

13
 Journal of Robotic Surgery

Fig. 3  The Mazor Robotics image adaptor attached to the C-arm Fig. 4  The Mazor robotic arm mounted to the operating room table
image intensifier

all soft tissues from the planned trajectory and allow the
per surgeon preference (Fig. 4). After the robotic system surgeon to visualize the cannula dock directly onto the bone
is mounted to the bed, it can be draped. It is important surface. This direct visualization is very advantageous when
to drape the arm after the patient has been placed on the early in adoption as it allows the surgeon to trust the system.
operating room table. Additionally, we prep any obvious uneven surfaces such as
an overgrown facet to minimize the skive potential by having
the flattest surface possible. All of this should be done prior
Incision and exposure to mounting to, as well, minimize shift between the patient
and the system once registered.
After mounting the robotic arm, prior to draping the arm
or the patient we place a spinal needle at the approxi-
mate center of the incision and obtain a lateral radiograph Mounting the patient to the robotic arm
(Fig. 5a). After adjusting as necessary, this minimizes our
incision. We then perform our surgical approach exposing For additional stability, the MXSE system requires that the
to the lateral edge of the facets with another intraoperative patient is directly mounted to the robotic arm. There are
lateral radiograph to confirm level(s). For a one-level fusion, multiple different options available for affixing a coupler
we typically make a 3 cm incision centered over the level to the patient’s bony anatomy, including a spinous pro-
to be fused (Fig. 5b). We prefer to use a McCulloch retrac- cess clamp, dual spinous process clamp, link bridge to the
tor for 1–2-level fusions, utilizing the wide blades with the spinous process, MIST (MIS Thoracic) bridge to the spinous
retractor rack placed on the side opposite (cranially) the process, and a Schanz screw/pin into the pelvis. The Schanz
robotic arm. It is important to perform a meticulous sub- pin is intended for open procedures from L4 to S2, or mini-
periosteal dissection to remove all muscle and soft tissue mally invasive procedures from L1 to S2. The Schanz pin
off of the lamina and projected screw starting points, with is our most commonly performed and preferred technique
preservation of the superior facet capsule. This will remove for bone mounting in CBT cases. To place the Schanz pin,

13
Journal of Robotic Surgery

Fig. 5  a Spinal needle place-

ment used for localization of
incision. b Typical 3–4 cm
incision length for a one or two
level spinal fusion

a stab incision is made directly over the posterior superior “no fly zone scan” (Fig. 7). It is important to make sure that
iliac spine (PSIS). The Schanz pin is then advanced directly before initiating this scan, that the sterile sleeve slack is held
into the PSIS (Fig. 6a). For patients that have an easily pal- tight against the surgical arm, so that it does not obstruct the
pable PSIS, we tend to use the 80 mm Schanz pin. For larger camera, and to place blue towels over the surgical wound
patients with increased soft tissue over the PSIS, we tend to with lights removed. We then perform a “snapshot” that tells
use the 120 mm Schanz pin. The Schanz pin is advanced into the navigation system where the robotic arm is and merges
the PSIS until it is stable. We prefer to toggle the Schanz the two together.
pin periodically after advancement; when the entire patient Following this scan, the surgeon defines the region of
moves in coordination with the Schanz pin, it is deemed sta- interest using the navigation probe at the cranial and cau-
ble (Fig. 6b). This step is critical to minimize shift between dal boundaries which is typically the top and bottom of the
the patient and MXSE system. If the Schanz pin is noted to surgical wound for short segment fusions. The next step
be loose, or toggles independently of the patient, it is rec- includes obtaining AP and oblique fluoroscopy images for
ommended be repositioned until stable fixation is achieved. registration (Fig. 8). These images must be obtained with
After placement of the Schanz pin, the Schanz screw ball the surgical arm in the respective (AP or oblique) position,
adaptor is connected to the Schanz pin (Fig. 6c). This is then as this allows the system to calibrate the position of the arm
attached to the bone mount bridge (Fig. 6d). If preferred, one in space. For the AP image, the 3D marker is attached to
may connect directly the Schanz pin with the appropriate the target extender, which is directly after obtaining the AP
connection (either a Schanz arm or Dynamic Schanz con- image, the surgeon must ensure that all levels to be operated
nector), but we prefer use the Schanz screw ball adaptor. We on are included in the image, and that the entire 3D marker
then toggle the system again to confer that the patient, bed, pattern is visible in the image. The 3D marker should ideally
and MXSE system all move as a single unit. At this point, be located in the center of the AP fluoro shot. Once an AP
the robotic arm is linked to the patient’s anatomy and the image is obtained that meets these criteria, the AP image is
surgeon may begin the process of registration. acquired by the software and the same steps may be repeated
for the oblique image. After registration, the surgeon must
perform the verification process before activating the robotic
Registration arm. Visual fluoroscopic registration is performed for all
registered vertebrae as shown.
The MXSE system has two available workflows as described
in surgical planning. We prefer the most common workflow
(CT to fluoro), which is described hereafter. After initiating Cortical cannulation and screw placement
the robot using the MXSE computer screen, the Mazor X eye
camera should be moved into a position where it recognizes The surgeon activates the surgical arm by selecting the
the reference marker, and will perform a 3D scan to define planned trajectory at the desired level to be executed. The
the working volume for the surgeon. This is referred to as the surgical arm will move into position during this time and

13
 Journal of Robotic Surgery

Fig. 6  a The Schanz pin is

advanced directly into the poste-
rior iliac spine. b Schanz pin is
toggled to ensure that adequate
fixation has been achieved. c
The Schanz screw ball adaptor
is connected to the Schanz
screw. d The Schanz screw ball
adaptor is directly mounted to
the bone mount bridge

will make an audible noise until the arm comes to a com- the teeth to engage without forcefully pushing the arm off
plete stop. The trajectory of the planned pedicle screw trajectory. The teeth are very sharp and one should utilize
should be assessed at each stage. If there is any suspicion this to help minimize skive. Then, follow the light taps with
for inaccuracy, the surgeon should verify system accuracy by a few more aggressive taps. The surgeon can confirm that the
either placing a navigated instrument on known anatomy and drill guide is securely positioned by applying gentle torque
confirming or placing the stylus probe into the arm guide. If to the head to the drill guide to ensure that it does not easily
there is any doubt, we recommend to redo the snapshot step rotate. To detect skive throughout these steps, we encourage
and potentially re-register. Once the robotic arm is in posi- surgeons to watch the black end-effector (EE) very close. If
tion, the surgeon may insert the appropriate length working it deviates even a small amount, it has a high potential to
cannula. Once the cannula is in position, the drill guide can skive and go off-axis. Additionally, there should be no soft-
be gently inserted into the cannula by allowing it to fall into tissue pressure on the system.
place with its own momentum (Fig. 9a). Once the drill guide Following successful docking of the drill guide, the
is in place, it should be lightly malleted into place to allow appropriate drill tip (short or long) can be attached to the

13
Journal of Robotic Surgery

Fig. 7  3D scan to define the

working volume for the surgeon,
otherwise known as the “no fly
zone” scan

drill, and placed into the drill guide (Fig. 9b). The drill bit
should pass smoothly within the drill guide. It is critically
important not to apply any excessive force when using any
of the tools, as this can cause skive of the trajectory. Once
the drill bit is confirmed to pass smoothly through the drill
guide, the surgeon’s non-dominant hand should be used to
stabilize the drill guide, and the drill should be advanced
into the bone. The drill bit has a stop which prevents from
excessive drilling at a depth of 30 mm. Once the pilot
hole is drilled, the surgeon may choose to tap or place the
screw. To tap, the drill cannula (inner cannula) is removed,
leaving only the outer cannula in place. The tap can then
be inserted through the outer cannula (Fig. 9c), followed
by the appropriately sized pedicle screw (Fig. 9d). It is
again important to remember that at each of these stages,
the surgeon should avoid excessive force on the system or
surgical accessories to prevent translation or skive. Also,
as an internal check when disengaging the driver from
the screw and pulling back, if there is translation or skive
of the EE and it takes a lot of force to remove, the screw
likely skived or was off trajectory.
Once all screws trajectories have been drilled, the
robotic arm may be removed from the surgical field once
their placement has been confirmed. The bone mount
bridge may be removed from the bone mount platform, and
the surgical arm may be removed from the operation area.
The Schanz pin drilled into the PSIS may be removed. We
Fig. 8  For the AP image, the 3D marker is attached to the target prefer to close this small incision with a buried 2–0 vicryl
extender, which is directly attached to the robotic arm and dermabond.

13
 Journal of Robotic Surgery

Fig. 9  a Drill guide is gently

inserted into the cannula using
its own momentum. b The drill
tip is inserted into the drill
guide, and advanced into bone.
c After removing the inner
sleeve, the tap is inserted and
advanced into bone an appropri-
ate distance. d The proper sized
pedicle screw is inserted under
power

Shift and skive translational force if the surgeon is not meticulous about

maintaining a constant in-line trajectory with the guide and
There are two primary errors that occur during CBT screw cannula. While many translational errors can be detected
placement. The first error is generally referred to as ‘shift’ by watching the navigation screen (pedicle screw not trave-
and is a change in position of the MXSE system relative to ling in line with planned trajectory) and/or movement of
the patient. This may occur at many phases and can be from the EE, subtle shift may occur that is difficult to recognize.
leaning on the patient, bumping the system with fluoroscopy, Anatomically, this is most common on the steep slope of the
but is mitigated most by a robust connection between patient lateral edge of the facet which is a common starting point
and system. The second primary error that may occur dur- in traditional pedicle screws but not CBT. If the surgeon
ing pedicle screw placement is referred to as ‘skive.’ This applies a downward force on the instrument, the cannula
generally occurs when a downward force applied to the can- or guide may slide down the edge of the facet to an inap-
nula, drill, tap, or guide causes the instrument to change propriate position. This frequently introduces both a trans-
position relative to its bony landmark. This can occur when lational and angular error (skive). Again, while this may be
the surgeon introduces a translational force to the cannula detected with on-screen navigation, a subtle degree of skive
or instruments such as soft tissue. The hand drill and asso- may not be identified. For this reason, it is important not to
ciated instruments are heavy, and may act to introduce a introduce any unnecessary force to the drill or instruments

13
Journal of Robotic Surgery

during use. In addition, similar to the traditional freehand improved precision in screw size selection and planned
pedicle screw placement, the surgeon should palpate all screw position, ease of registration and reduction of oper-
bony pedicle screw pilot holes with a ball-tipped feeler to ating room time, reduced radiation to the operating room
ensure safe and accurate placement. Finally, intraoperative staff, and an enhancement of the surgeon’s ergonomics and
radiographs should be used to evaluate satisfactory screw dexterity for repetitive tasks in pedicle screw placement
placement and ensure that the goals of surgery have been [4, 12, 14, 18, 20, 21, 31–34].
achieved (Fig. 10). Finally, it is important for the surgeon to consider the
different types of robotic-assisted technology that are
available, which include the Mazor X Stealth Edition
Review and discussion robotic system (Medtronic), the ROSA robot (Zimmer-
Biomet), and the Globus Medical Excelsius GPS Robot
While the technique described above pertains to robotic- [5, 7]. [8] Given the widespread use of the Mazor spine
assisted CBT screw placement, various forms of advanced robot, most studies in the literature evaluating the accuracy
imaging technology (including fluoroscopy and naviga- of robotic-assisted screw placement have utilized one of
tion) have previously been described for the purpose of the Mazor platforms. Most recent reports of the Mazor
CBT screw placement. It is important for the surgeon to system have focused on the traditional pedicle screw tra-
weigh the pros and cons for each of these technologies jectory, with reported accuracy in the range of 97–99%
when considering safe placement of CBT screws. The using the Gertzbein–Robbins grading system [4, 7, 9–19].
major advantage of navigation or robotic technology over While there are fewer studies to date on the Globus Medi-
simple C-arm fluoroscopy is obtaining true images in the cal Excelsius GPS Robot, accuracy of traditional pedi-
axial plane, which can only be estimated by the surgeon cle screw placement appears to be in the same range of
when they are limited to orthogonal images. This may 97–98% [34–38]. The ROSA robot system has been evalu-
be especially important for CBT screw placement, as the ated in feasibility and cadaveric studies, but has only been
cortical tract is narrower and more difficult to cannulate evaluated in clinical accuracy studies in a small number of
compared to a traditional pedicle screw. The use of naviga- patients, making comparison to other robotic technologies
tion-based CBT screw placement has been described in the difficult [7, 39–41]. Further clinical studies are required
literature and does offer the advantage of imaging in the before a direct comparison of pedicle screw placement
axial plane. This technique has been shown to be safe and accuracy can be made between these different robotic
reproducible. The major advantages of robotic-assisted systems. Although these robotic systems have different
screw placement over navigation-based screw placement workflows, the principles of robotic-guided pedicle screw
include increased accuracy and safety in screw insertion, placement are similar irrespective of the system used.

Fig. 10  Grade 2 Spondylolisthe-
sis treated with robotic midline
lumbar interbody fusion with
partial reduction of spondylolis-
thesis and restoration of disc
height and foramen

13
 Journal of Robotic Surgery

Conclusion Mazor Robotics: Stock options, consulting fees; Medtronic: Consult-

ing, advisory board; Stryker/K2M: Consulting, Royalties, Advisory
Board; Augmedics: Advisory Board, Stock Options; National Spine
Given the widespread interest and rapid growth, robotic Health Foundation: Board Member.
technology will likely play an increasing role in the oper-
ating room over the coming decades. There are a number
of advantages that this technology will offer both surgeons References
and patients, including increased accuracy, efficiency, and
decreased radiation exposure. However, along with the 1. Shweikeh F et al (2014) Robotics and the spine: a review of cur-
advantages, there is a learning curve associated with the rent and ongoing applications. Neurosurg Focus 36(3):E10
adoption of any new technology that must be considered. 2. Pechlivanis I et al (2009) Percutaneous placement of pedicle
screws in the lumbar spine using a bone mounted miniature
Although robotic technology has been shown to increase robotic system: first experiences and accuracy of screw place-
pedicle screw placement accuracy, there are a number of ment. Spine (Phila Pa 1976) 34(4):392–398
technical errors during the procedure that may lead to mis- 3. Shoham M et al (2007) Robotic assisted spinal surgery–from con-
placed pedicle screws. We have summarized the steps which cept to clinical practice. Comput Aided Surg 12(2):105–115
4. Hu X, Ohnmeiss DD, Lieberman IH (2013) Robotic-assisted pedi-
we follow during CBT screw placement to ensure a safe and cle screw placement: lessons learned from the first 102 patients.
reproducible outcome. Additionally, we feel that CBT screw Eur Spine J 22(3):661–666
placement is a great starting point for surgeons when intro- 5. Fiani B et al (2020) Impact of robot-assisted spine surgery on
ducing robotic-assisted technology. It offers a direct visuali- health care quality and neurosurgical economics: a systemic
review. Neurosurg Rev 43(1):17–25
zation of cannula-bone interface to double check accuracy, 6. Tan LA, Lehman RA (2019) Robotic-assisted spine surgery using
the planning allows you to truly minimize skin incision, and the Mazor XTM system: 2-dimensional operative video. Oper
finally, non-robotic assisted CBT requires either navigation Neurosurg (Hagerstown) 16(4):E123
or fluoro (not freehand), so the utilization of the robotic sys- 7. Lefranc M, Peltier J (2015) Accuracy of thoracolumbar trans-
pedicular and vertebral body percutaneous screw placement: cou-
tem is substitutive and not additive to improve workflow (not pling the Rosa(R) Spine robot with intraoperative flat-panel CT
just add steps). guidance–a cadaver study. J Robot Surg 9(4):331–338
8. Zygourakis CC et al (2018) Technique: open lumbar decompres-
Acknowledgements None. sion and fusion with the excelsius GPS robot. Neurosurg Focus
45(VideoSuppl1):V6
Author contributions  Dr. JB and Dr. JG were involved in the creation, 9. Tsai TH et al (2016) A retrospective study to validate an intraop-
writing, and final editing of this publication. Drs. CG, RL, JP, AB, and erative robotic classification system for assessing the accuracy of
RC were involved in the creation of the review, revised it critically for kirschner wire (K-wire) placements with postoperative computed
intellectual content, and approved the final version to be published. All tomography classification system for assessing the accuracy of
authors involved agree to be accountable for all aspects of the work in pedicle screw placements. Med (Balt) 95(38):e4834
ensuring that questions related to the accuracy or integrity of any part 10. Kuo KL et  al (2016) Assessing the intraoperative accuracy
of the work are appropriately investigated and resolved. All authors of pedicle screw placement by using a bone-mounted minia-
of this manuscript certify that it is (1) is a unique submission, (2) has ture robot system through secondary registration. PLoS ONE
not been submitted and is not being considered for publication by any 11(4):e0153235
other source in any medium, and (3) has not been published, in part or 11. Onen MR, Simsek M, Naderi S (2014) Robotic spine surgery: a
in full, in any form. preliminary report. Turk Neurosurg 24(4):512–518
12. Keric N et  al (2017) Evaluation of robot-guided minimally
invasive implantation of 2067 pedicle screws. Neurosurg Focus
Funding  No funding was received for this article. 42(5):E11
13. Kim HJ et al (2017) A prospective, randomized, controlled trial of
Compliance with ethical standards  robot-assisted vs. freehand pedicle screw fixation in spine surgery.
Int J Med Robot 13:3
Conflicts of interest John Buza has no conflicts of interest. Jeffrey 14. van Dijk JD et al (2015) Clinical pedicle screw accuracy and devi-
Gum has the following disclosures: Acuity: IP royalties, Paid consult- ation from planning in robot-guided spine surgery: robot-guided
ant;; Cingulate Therapeutics, Stock or stock Options; DePuy, A John- pedicle screw accuracy. Spine (Phila Pa 1976) 40(17):E986–E991
son & Johnson Company, Paid presenter or speaker; K2M, Paid con- 15. Barzilay Y et  al (2014) Robot-assisted vertebral body aug-
sultant; MAZOR Surgical Technologies, Paid consultant; Medtronic, mentation: a radiation reduction tool. Spine (Phila Pa 1976)
Paid consultant; Nuvasive, IP royalties, Paid consultant; Stryker, Paid 39(2):153–157
presenter or speaker; Spine Deformity Journal—Reviewer, Editorial or 16. Bederman SS et  al (2017) Robotic guidance for S2-alar-
governing board; The Spine Journal—Reviewer, Editorial or govern- iliac screws in spinal deformity correction. Clin Spine Surg
ing board. Ronald Lehman has the following disclosures: Medtronic: 30(1):E49–E53
Consulting fees and royalties; Stryker: Royalty fees. Richard Chua has 17. Sukovich W, Brink-Danan S, Hardenbrook M (2006) Miniature
the following disclosures: Indirect interest as a consultant/speaker’s robotic guidance for pedicle screw placement in posterior spinal
bureau related to surgeon education. Avery Buchholz has the following fusion: early clinical experience with the SpineAssist. Int J Med
disclosures: Medtronic: Consulting fees. John Pollina has the follow- Robot 2(2):114–122
ing disclosures: Atec Spine: Consulting fees and royalties; Medtronic: 18. Devito DP et al (2010) Clinical acceptance and accuracy assess-
Surgeon Education, Christopher Good has the following disclosures: ment of spinal implants guided with SpineAssist surgical robot:
retrospective study. Spine (Phila Pa 1976) 35(24):2109–2115

13
Journal of Robotic Surgery

19. Schatlo B et  al (2015) Unskilled unawareness and the learn- part I-technical development and a test case result. Neurosurgery
ing curve in robotic spine surgery. Acta Neurochir (Wien) 59(3):641–650 (discussion 641-50)
157(10):1819–1823 (discussion 1823) 32. Kantelhardt SR et al (2011) Perioperative course and accuracy
20. Schatlo B et al (2014) Safety and accuracy of robot-assisted ver- of screw positioning in conventional, open robotic-guided and
sus fluoroscopy-guided pedicle screw insertion for degenerative percutaneous robotic-guided, pedicle screw placement. Eur Spine
diseases of the lumbar spine: a matched cohort comparison. J J 20(6):860–868
Neurosurg Spine 20(6):636–643 33. Roser F, Tatagiba M, Maier G (2013) Spinal robotics: current
21. Hyun SJ et al (2017) Minimally invasive robotic versus open applications and future perspectives. Neurosurgery 72(Suppl
fluoroscopic-guided spinal instrumented fusions: a randomized 1):12–18
controlled trial. Spine (Phila Pa 1976) 42(6):353–358 34. Elswick CM et al (2020) Robotic-assisted spinal surgery: current
22. Ver MLP et al (2020) Index episode-of-care propensity-matched generation instrumentation and new applications. Neurosurg Clin
comparison of transforaminal lumbar interbody fusion (TLIF) N Am 31(1):103–110
techniques: open traditional TLIF versus midline lumbar inter- 35. Huntsman KT et al (2020) Navigated robot-guided pedicle screws
body fusion (MIDLIF) versus robot-assisted MIDLIF. J Neurosurg placed successfully in single-position lateral lumbar interbody
Spine 24:1–7 fusion. J Robot Surg 14(4):643–647
23. Le XF et al (2020) Rate and risk factors of superior facet joint 36. Vardiman AB et al (2020) Does the accuracy of pedicle screw
violation during cortical bone trajectory screw placement: a com- placement differ between the attending surgeon and resident in
parison of robot-assisted approach with a conventional technique. navigated robotic-assisted minimally invasive spine surgery? J
Orthop Surg 12(1):133–140 Robot Surg 14(4):567–572
24. Santoni BG et al (2009) Cortical bone trajectory for lumbar pedi- 37. Jain D et al (2019) Initial single-institution experience with a
cle screws. Spine J 9(5):366–373 novel robotic-navigation system for thoracolumbar pedicle screw
25. Matsukawa K et al (2014) In vivo analysis of insertional torque and pelvic screw placement with 643 screws. Int J Spine Surg
during pedicle screwing using cortical bone trajectory technique. 13(5):459–463
Spine (Phila Pa 1976) 39(4):E240–E245 38. Benech CA et al (2020) Navigated robotic assistance results in
26. Perez-Orribo L et al (2013) Biomechanics of lumbar cortical improved screw accuracy and positive clinical outcomes: an eval-
screw-rod fixation versus pedicle screw-rod fixation with and uation of the first 54 cases. J Robot Surg 14(3):431–437
without interbody support. Spine (Phila Pa 1976) 38(8):635–641 39. Chenin L, Peltier J, Lefranc M (2016) Minimally invasive transfo-
27. Baluch DA et al (2014) Effect of physiological loads on corti- raminal lumbar interbody fusion with the ROSA(TM) Spine robot
cal and traditional pedicle screw fixation. Spine (Phila Pa 1976) and intraoperative flat-panel CT guidance. Acta Neurochir (Wien)
39(22):E1297–E1302 158(6):1125–1128
28. Phan K et al (2015) Cortical bone trajectory for lumbar pedi- 40. Chenin L et al (2017) Evaluation of screw placement accuracy in
cle screw placement: a review of published reports. Orthop Surg circumferential lumbar arthrodesis using robotic assistance and
7(3):213–221 intraoperative flat-panel computed tomography. World Neurosurg
29. Keorochana G et al (2017) Comparative outcomes of cortical 105:86–94
screw trajectory fixation and pedicle screw fixation in lumbar spi- 41. Lefranc M, Peltier J (2016) Evaluation of the ROSA Spine robot
nal fusion: systematic review and meta-analysis. World Neurosurg for minimally invasive surgical procedures. Expert Rev Med
102:340–349 Devices 13(10):899–906
30. Laporta-Hoyos O et al (2017) Proxy-reported quality of life in
adolescents and adults with dyskinetic cerebral palsy is associated Publisher’s Note Springer Nature remains neutral with regard to
with executive functions and cortical thickness. Qual Life Res jurisdictional claims in published maps and institutional affiliations.
26(5):1209–1222
31. Lieberman IH et al (2006) Bone-mounted miniature robotic guid-
ance for pedicle screw and translaminar facet screw placement:

13