Motion response of a polycrystalline diamond adaptive axis of rotation cervical total disc arthroplasty

Motion response of a polycrystalline diamond adaptive axis of rotation cervical total disc arthroplasty

Accepted Manuscript Motion response of a polycrystalline diamond adaptive axis of rotation cervical total disc arthroplasty Robert M. Havey, Saeed Kh...

3MB Sizes 0 Downloads 1 Views

Accepted Manuscript Motion response of a polycrystalline diamond adaptive axis of rotation cervical total disc arthroplasty

Robert M. Havey, Saeed Khayatzadeh, Leonard I. Voronov, Kenneth R. Blank, Gerard Carandang, David P. Harding, Avinash G. Patwardhan PII: DOI: Reference:

S0268-0033(18)30703-4 https://doi.org/10.1016/j.clinbiomech.2018.12.023 JCLB 4673

To appear in:

Clinical Biomechanics

Received date: Accepted date:

17 August 2018 20 December 2018

Please cite this article as: Robert M. Havey, Saeed Khayatzadeh, Leonard I. Voronov, Kenneth R. Blank, Gerard Carandang, David P. Harding, Avinash G. Patwardhan , Motion response of a polycrystalline diamond adaptive axis of rotation cervical total disc arthroplasty. Jclb (2018), https://doi.org/10.1016/j.clinbiomech.2018.12.023

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Motion Response of a Polycrystalline Diamond Adaptive Axis of Rotation Cervical Total Disc Arthroplasty

Robert M. Havey, MS1,2, Saeed Khayatzadeh, PhD1, Leonard I. Voronov MD, PhD1,2, Kenneth

T

R. Blank, MS, MHA1, Gerard Carandang, MS1, David P. Harding, PhD3, Avinash G.

CR

Edward Hines Jr. VA Hospital, Hines, IL, USA; 2 Loyola University Chicago, Maywood, IL,

US

1

IP

Patwardhan, PhD1,2

AN

USA; 3Dymicron, Inc., Orem, UT, USA

M

Correspondence: Robert Havey, MS

ED

PO Box 5000 (151)

PT

Hines IL, 60141, USA

AC

CE

Email: [email protected]

Word Counts:

Abstract – 249 words Narrative – 4,094 words

ACCEPTED MANUSCRIPT Abstract Background Cervical fusion is associated with adjacent segment degeneration. Cervical disc arthroplasty is considered an alternative to reduce risk of adjacent segment disease. Kinematics after

T

arthroplasty should closely replicate healthy invivo kinematics to reduce adjacent segment

IP

stresses. The purpose of this study was to assess the kinematics of a polycrystalline diamond

CR

cervical disc prosthesis.

US

Methods

AN

Nine cadaveric C3-T1 spines were tested intact and after one (C5-C6) and two level (C5-C7) arthroplasty (Triadyme-C, Dymicron, Orem, UT, USA). Kinematics were evaluated in flexion-

ED

M

extension, lateral bending, and axial rotation.

Findings

PT

Prosthesis placement at C5-C6 and C6-C7 was 0.5mm anterior and 0.6mm posterior to midline

CE

respectively. C5-C6 flexion-extension motion was 12.8 degrees intact and 10.5 degrees after arthroplasty. C6-C7 flexion-extension motion was 10.0 and 11.4 degrees after arthroplasty.

AC

C5-C6 lateral bending reduced from 8.5 to 3.7 degrees after arthroplasty and at C6-C7 from 7.5 to 5.1 degrees. C5-C6 axial rotation decreased from 10.4 to 6.2 degrees after arthroplasty and at C6-C7 from 7.8 to 5.3 degrees. Segmental lordosis increased by 4.2 degrees, and middle disc height by 1.4mm after arthroplasty. Change in center of rotation from intact to arthroplasty averaged 0.9mm posteriorly and 0.1mm caudally at C5-C6, and 1.4mm posteriorly and 0.3mm cranially at C6-C7. Page 2 of 30

ACCEPTED MANUSCRIPT

Interpretation The cervical disc arthroplasty evaluated restored flexion-extension motion to intact levels and moderately increased segmental stiffness. Disc height increased by up to 1.5mm and segmental

T

lordosis by 4.2 degrees. The unique prosthesis design allowed the axis of rotation after

AC

CE

PT

ED

M

AN

US

CR

IP

arthroplasty to closely mimic the native location.

Page 3 of 30

ACCEPTED MANUSCRIPT Introduction Anterior cervical discectomy and fusion (ACDF) is considered to be the standard surgical procedure for the treatment of disc herniation and symptoms of spondylosis and myelopathy. Degeneration of spinal segments adjacent to a previous fusion, termed adjacent segment disease

T

(ASD), has been attributed in part to the initial fusion [1, 2]. The clinical incidence of

IP

symptomatic ASD is estimated to be about 2.9% per year for the first 10 years after fusion; with

CR

two-thirds of these patients requiring re-operation [2]. Cervical total disc arthroplasty (TDA) is becoming more widely accepted as an alternative to fusion in select patients with the intention of

US

reducing adjacent level stresses and preventing or delaying onset of ASD [3, 4]. Clinical

AN

evidence and recent meta-analyses support these claims of cervical TDA showing outcomes superior to ACDF with lower ASD and reoperation rates [5, 6, 7, 8, 9, 4, 10, 11, 12].

M

For maximum benefit, spine biomechanics after TDA must closely replicate healthy in

ED

vivo biomechanics to reduce stresses imposed on adjacent segments by fusion. Biomechanical studies have demonstrated that some TDA designs can approximate physiologic quantity and

PT

quality of motion at the index level and allow normal kinematics at adjacent levels [13, 14, 15,

CE

16]. There are many designs of cervical TDA ranging from ball and socket fixed axis of rotation designs with three degrees of freedom to non-articulating polymer core designs [3]. Non-

AC

articulating designs are intended to better accommodate the variable axes of rotation in flexionextension, lateral bending and axial rotation [15]. The purpose of this biomechanical study was to evaluate the kinematics of an articulating unconstrained, unrestrained, incompressible TDA with five degrees of freedom [3]. We tested the hypothesis that cervical disc replacement would allow physiologic range of motion (RoM) and provide quality of motion (stiffness, center of rotation (CoR), hysteresis, and neutral zone) Page 4 of 30

ACCEPTED MANUSCRIPT similar to that of a healthy motion segment. Methods Triadyme-C™ Cervical TDA

The hypothesis was tested using the Triadyme-C™ (Dymicron, Orem, UT, USA),

IP

T

cervical total disc arthroplasty. This TDA has an articulating design with a three lobe articulation

CR

in which the mating surfaces are non-congruent (Fig. 1). This design stipulates that the radii of the lobes are smaller than the corresponding pockets. As a consequence of this design, load

US

transfer from the superior lobes to the inferior pockets occurs over a small surface area resulting in very high contact stresses. These high-stress point loads would cause high amounts of wear

AN

debris and lead to device failure with traditional biocompatible materials such as cobalt chrome,

M

ultra high molecular weight polyethylene (UHMWPE), and ceramics. Industrial polycrystalline

ED

diamond, specially formulated and processed for biocompatibility, was determined to be a more suitable material to resist the high contact stresses. This TDA design allows for three primary

PT

rotations, coupled with anteroposterior and medial-lateral translation. The device is noncompressible. Each component is made of roughly equal parts polycrystalline diamond

CE

and titanium/titanium-carbide (Ti/TiC), sintered together under high temperature and

AC

pressure. The layers are fully fused and chemically bonded. Primary fixation is achieved with the dual-keel design, and secondary fixation is accommodated by a porous titanium plasma spay coating on the endplates.

Specimens and Experimental Set-Up Page 5 of 30

ACCEPTED MANUSCRIPT Nine (9) fresh-frozen human cervical cadaveric spines (age: 38.3 (SD 5.8) years, 5 Male, 4 Female) with no previous spinal surgery were used. Specimens were screened radiographically to exclude those with evidence of disc ossification and bridging osteophytes. The spines were cleaned and stripped of extraneous tissue, leaving the discs, facets, and intrinsic ligaments intact.

T

All tests were performed at room temperature and the specimens were kept moist during testing

CR

IP

with saline soaked towels.

The motion of the C3, C4, C5, C6 and C7 vertebrae relative to T1 were measured

US

optoelectronically (Optotrak® Certus, Northern Digital, Waterloo, Ontario). A six-component

AN

load cell (Model MC3A-6-250, AMTI Inc., Newton, MA) placed under the specimen measured applied compressive preload and moments. Fluoroscopic imaging (GE OEC 9800 Plus, GE

M

Healthcare, Chicago, IL) was used to measure intact disc height for propert TDA sizing and to

ED

document device placement as per standard clinical technique. The follower load technique was used to apply compressive preload to the cervical spine

CE

PT

during the range of motion experiments in flexion and extension [17, 15, 18].

3-D CT Based Specimen Specific Analysis

AC

Three dimensional (3-D) Computed Tomography (CT) based specimen specific analysis was used during testing to acquire all data necessary to accurately evaluate segmental range of motion, disc height, changes in segmental lordosis and CoR for each motion segment [19]. The benefit of this technique is that the 3-D motion of any anatomical feature or relationship between anatomical features of different vertebrae can be measured throughout each specimen’s collected motion data set. The methodology for specimens specific analysis is described in detail in Havey Page 6 of 30

ACCEPTED MANUSCRIPT et al [19].

Experimental Protocol Specimens were tested under the following conditions (Fig. 2); i) intact, ii) C5-C6 TDA

T

(n=9), iii) C6-C7 TDA (n=7). In each condition, specimens were subjected to RoM testing in

IP

flexion-extension with compressive preloads of 0 N and 150 N, and lateral bending and axial

CR

rotation without compressive preload. In all testing, specimens were subjected to +/-1.5 Nm moments [17, 20, 15, 15]. Loading was performed at a quasistatic rate (0.2 Nm/second) to allow

US

sufficient time for viscoelastic relaxation. Load-displacement data was collected for a minimum

AN

of three loading cycles and until two continuous and reproducible cycles were obtained.

M

Surgical Technique

ED

After testing the intact spine (Fig. 2A), a discectomy was performed at C5-C6 using standard instruments. The posterior longitudinal ligament (PLL) was resected in all

PT

implantations. The endplates were preserved but scraped clean and medial uncinatectomies were

CE

performed when necessary for arthroplasty placement leaving at least the lateral two-thirds intact. A single level cervical total disc arthroplasty (Triadyme-C™, Dymicron, Orem, UT) was

AC

then implanted at C5-C6 using the technique and instruments provided by the manufacturer (Fig. 2B). Trial sizes were used to to determine the correct TDA footprint. Proper placement was confirmed by fluoroscopy. The specimens were again tested as described above. Following the single level TDA evaluation, a second TDA implantation was performed at C6-C7 in seven of the nine specimens using the techniques described above (Fig. 2C).

Page 7 of 30

ACCEPTED MANUSCRIPT Data Analysis The load versus displacement curves were analyzed to obtain RoM at the implanted and adjacent segments for each protocol step. Quality of motion was assessed through measure of the neutral zone (NZ), hysteresis, and segmental high flexibility zone (HFZ) stiffness at the

T

implanted level for all test conditions under each loading mode.

IP

The 3-D specimen specific models were used to evaluate kinematic measures that are

CR

difficult or impossible to accurately make using traditional techniques. Change in segmental lordosis, CoR, and disc height were measured from the specimen specific models.

US

Activities of daily living generally take place within the high flexibility zone of the spine.

AN

This is the region in the segmental RoM in which anterior and posterior ligament tension as well as facet loading are at a minimum. The segmental CoR (projection of the flexion-extension axis

M

of rotation on the sagittal plane) for the implanted motion segments (C5-C6, C6-C7) was

ED

measured between the start of the HFZ in extension, to its end point in flexion. The measurement was performed in this manner to eliminate the effects of tensioning soft tissues and facet loading

PT

on the position of the CoR.

CE

Measurement of implant placement was performed by superimposing the mid sagittal plane slice of the motion segment’s caudal most vertebral body obtained from the CT scan

AC

reconstruction, over the sagittal plane x-ray obtained during testing. Scaling of the combined images was related back to the CT scan which was in millimeters. The offset between midline of the TDA and that of the superior endplate of the inferior vertebral body was recorded. A repeated-measures analysis of variance (ANOVA) with Bonferroni correction for multiple comparisons was used to assess the effects of the cervical disc arthroplasty on motion segment kinematics in each loading mode. Comparisons were made at C5-C6 and C6-C7 Page 8 of 30

ACCEPTED MANUSCRIPT between the intact and implanted segment to determine to what extent the disc arthroplasty restored intact kinematics.

T

Results

IP

Implant Size and Placement

CR

Nine cadaveric cervical spine specimens were tested. All specimens were operated at the C5-C6 level and implanted with TDA following intact testing. In all specimens the C5-C6 TDA

US

height was 6mm. In seven of the nine specimens, a second TDA was implanted at C6-C7

AN

resulting in a two-level construct. Five of these TDA at C6-C7 were 6 mm height and two were 7 mm.

M

TDA implantation was performed by a surgeon experienced in cervical arthroplasty.

ED

Placement of the TDA at C5-C6 was 0.5 (SD 0.8) mm anterior to the midline and at C6-C7, 0.6

PT

(SD 1.0) mm posterior to midline.

CE

Range of motion

Applied moment versus angular displacement graphs depict the behavior of the motion

AC

segment in the intact condition, and after TDA implantation without and with 150N of compressive follower preload at both implanted levels (Fig. 3A,B). Flexion-extension curves after TDA and without compressive preload tended to be smooth, exhibiting good quantity and quality of motion. Addition of compressive preload caused seven of the 16 implanted segments to exhibit monotonic motion in flexion or extension suggesting that the motion segment could have produced more motion if higher moments were delivered to the specimen. Regardless of the Page 9 of 30

ACCEPTED MANUSCRIPT preload magnitude, eight of the 16 implanted segments exhibited a flat response in extension suggesting all available extension motion was exhausted before reaching 1.5Nm (Fig. 3B).

T

Flexion-Extension RoM

IP

In the absence of a compressive preload (0N), the intact C5-C6 flexion-extension angular

CR

RoM was 12.6 (SD 2.6) degrees and decreased slightly to 11.7 (SD 1.8) degrees after TDA placement (P=0.32). Under compressive load of 150 N, the intact C5-C6 RoM was 12.8 (SD 2.5)

US

degrees which decreased to 10.5 (SD 2.1) degrees with TDA (P=0.03) (Fig. 4, Tables 1, 2).

AN

At C6-C7 without compressive preload, the angular motion of the intact spine was 11.5 (SD 3.4) degrees in flexion-extension and increased to 12.4 (SD 3.3) degrees with TDA

M

(P=0.07). Under 150 N of compressive load, intact flexion-extension RoM at C6-C7 was 10.0

ED

(SD 3.4) degrees which increased to 11.4 (SD 3.0) degrees (P=0.15) after TDA (Fig. 4, Tables 1,

CE

Lateral Bending RoM

PT

2).

Under moments of +/-1.5 Nm in lateral bending, the angular motion at C5-C6 was

AC

significantly reduced from 8.5 (SD 2.8) degrees intact to 3.7 (SD 1.0) degrees after TDA (P<0.01). C6-C7 lateral bending changed from 7.5 (SD 2.8) degrees intact to 5.1 (SD 2.3) degrees after TDA placement (P=0.07) (Fig. 4, Tables 1, 2).

Axial Rotation RoM C5-C6 intact axial rotation RoM was 10.4 (SD 1.1) degrees, decreasing to 6.2 (SD 1.9) Page 10 of 30

ACCEPTED MANUSCRIPT degrees after TDA (P<0.01). C6-C7 intact axial rotation motion was 7.8 (SD 1.7) degrees, which decreased to 5.3 (SD 0.9) degrees after TDA (P=0.02) (Fig. 4, Tables 1, 2).

T

Flexion-Extension Neutral Zone

IP

TDA had no significant effect on flexion-extension NZ compared to the intact condition.

CR

Under 0N preload, the C5-C6 NZ of the intact spine was 2.0 (SD 1.0) degrees and 1.5 (SD 0.4) degrees after TDA placement (P=0.15). Under 150 N of compressive load the intact C5-C6 NZ

US

measured 2.6 (SD 1.7) degrees and 1.6 (SD 0.7) degrees with TDA (P=0.08) (Tables 1, 2).

AN

At C6-C7 without preload, the intact flexion-extension NZ was 1.6 (SD 0.7) degrees and 1.1 (SD 0.4) degrees after TDA (P=0.14). With 150N of compressive load the intact NZ was 2.1

ED

M

(SD 1.3) degrees and after TDA 2.0 (SD 0.8) degrees (P=0.84) (Tables 1, 2).

HFZ Stiffness

PT

Placement of the TDA at C5-C6 significantly increased flexion and extension HFZ

CE

stiffness both with and without compressive preload compared to the intact condition (P<0.05). This was not true at C6-C7 however, as TDA placement did not significantly affect flexion or

AC

extension stiffness compared to intact (P>0.42) (Tables 1, 2).

Hysteresis Under 0N preload flexion-extension hysteresis increased significantly at both C5-C6 and C6-C7. However, with compressive preload, hysteresis was not different than the intact condition at either implanted level (Tables 1, 2). Page 11 of 30

ACCEPTED MANUSCRIPT

Segmental Lordosis Segmental lordosis was measured for the intact motion segments and after TDA placement. Rather than taking a static measure with the specimen in its neutral posture,

T

measurement of segmental lordosis was performed throughout the segmental extension to flexion

IP

RoM. Measurement of segmental lordosis was then taken as the specimen passed neutral posture

CR

at 0N applied moment. At C5-C6 the segmental lordosis increased after TDA on average by 3.3 (SD 2.6) degrees and at C6-C7 by 5.4 (SD 1.8) degrees. The reduction in extension motion or

US

flat response in extension previously mentioned, tended to occur in segments that had large

AN

increases in segmental lordosis and occurred in six of the seven C6-C7 and two of the nine C5-

M

C6 segments.

ED

Disc Height

Height of each motion segment was measured in the neutral posture using the same

PT

methods as for segmental lordosis (Tables 1, 2). Middle height was used to minimize any effect

CE

change in segmental lordosis has on disc height measurement. At C5-C6 the average increase in disc height was 1.4 (SD 0.8) mm without preload and 1.5 (SD 0.9) mm with preload. At C6-C7

AC

the average increase in disc height was 1.2 (SD 0.5) mm and 1.3 (SD 0.4) mm.

Center of Rotation COR measurements were performed in the HFZ to eliminate the influence of tensioning soft tissues and apophyseal facet joint loading. The kinematic data from -0.1Nm extension to 0.65Nm flexion were used to calculate the CoR location. Page 12 of 30

ACCEPTED MANUSCRIPT At C5-C6 the average HFZ-COR location of the intact segment was 1.3 (SD 1.2) mm posterior to midline and 3.1 (SD 2.1) mm caudal to the C6 superior endplate. The change in location after TDA averaged 0.9 (SD 1.0) mm posteriorly (n=8, P<0.05) and 0.1 (SD 2.2) mm caudally (n=8, P=0.9) (Fig. 6A). One specimen was excluded from CoR analysis due to an error

T

during the TDA data collection. At C6-C7 the average intact HFZ-COR location was 2.1 (SD

IP

0.6) mm posterior to midline and 0.7 (SD 0.9)mm caudal to the C6 superior endplate. CoR

CR

location after C6-C7 TDA averaged 1.4 (SD 0.8) mm posteriorly (n=7, P<0.01) and 0.3 (SD 2.0)

US

mm cranially (n=7, P=0.7) (Fig. 6B).

AN

Discussion

Eliminating pain and mimicking healthy kinematics without introducing reactive

M

wear debris are the ultimate goals of disc arthroplasty. Native discs have mobile CoR’s

ED

which act to minimize forces on the uncovertebral and facet joints and to provide negative feedback to the motion segment that controls the quantity and quality of

PT

motion. Uncontrolled CoR mobility due to arthroplasty designs with highly mobile

CE

articulating components, may be perceptible by the patient as instability. The biologic response may be stabilization using muscle activation patterns or bone growth. Altered

AC

muscle activation patterns may result in fatigue and pain and higher muscle compressive forces will in turn likely increase wear debris generation and continued or accelerated degeneration at the index and adjacent segments. The cervical disc arthroplasty evaluated in this study permited a physiologically normal range of motion at both C5-C6 and C6-C7 in flexion-extension. RoM in lateral bending and axial rotation decreased compared to the intact condition. Some quality of motion measures were Page 13 of 30

ACCEPTED MANUSCRIPT affected by arthroplasty depending on the operative level and presence of preload. There was no change in the measured neutral zone at either level, but both flexion and extension HFZ stiffness increased significantly at C5-C6 after TDA while remaining unchanged at C6-C7. Hysteresis significantly increased at both levels without preload but was not statistically different than the

T

intact condition with compressive preload.

IP

This research study was performed on relatively young specimens without fluoroscopic

CR

evidence of disc degeneration, facet degeneration or loss of disc height. An advantage of testing relatively healthy specimens is the ability to compare postoperative motions to the native disc.

US

This provides a specimen specific and motion segment specific control for data comparison. The

AN

biomechanical testing performed in this study was intended to mimic the immediate postoperative condition. As such, soft tissue changes such as annular scar tissue formation, and

ED

to the quazistatic rate of loading [21].

M

bony remodeling are not incorporated, although soft tissue relaxation may be accounted for due

The kinematic behavior of the intact spine and all arthroplasty devices is affected by

PT

segmental compression. Consequently, compressive follower preload of 150 N was applied

CE

during flexion-extension testing. Although application of the follower load technique provides a key component of the in vivo environment [17], the complicated musculature of the neck creates

AC

loading conditions nearly impossible to reproduce completely on a cadaveric spine. Never the less, compressive loading is necessary in order to more completeley understand the dynamic response of motion preservation implants and how they coexist with the remaining segmental constraints such as remaining soft tissues, uncovertebral and facet joints. If proper followerload path optimization is not performed, the preload cables will impart segmental shear forces and moments which will alter the segmental posture and resulting kinematic signature [17, 15]. Page 14 of 30

ACCEPTED MANUSCRIPT After TDA placement, flexion-extension RoM decreased on average to 11.5 (SD 2.6) degrees under compressive follower preload, or by just over one degree, remaining within what is considered to be normal in vitro (range: 8.4–13.9 deg) [15, 14, 22, 20, 16, 23, 24] and in vivo RoM [25, 26, 27, 28]. This slight decrease in RoM may be due to frictional loss between the

T

articulating surfaces of the TDA, mismatch in CoR between the native and implanted segments,

IP

or preload artifact. The preload cable path was not re-optimized or adjusted after TDA

CR

implantation in order to understand the immediate postoperative kinematic response. As a result,

moment and limiting the RoM in extension [29].

US

the increase in segmental lordosis may have altered the optimal cable path resulting in artifact

AN

Lateral bending and axial rotation RoM decreased after TDA implantation by an average of 44% and 36%, respectively. Similar reduction in AR and LB RoM after TDA is well

M

documented in the literature [15, 30, 14, 22, 20, 30]. These reductions in motion are somewhat

ED

variable between studies and can be attributed to TDA design, implanted level and degenerative condition, surgical technique and tissue resection, and change in segmental lordosis and segment

PT

height affecting the remaining soft tissue tension. In this study the PLL was resected but the

CE

lateral uncinates were left intact and an annular window only wide enough for implant insertion was made. A wider annulotomy or uncinatectomy for the purpose of performing a

[20].

AC

decompression would likely allow more coupled motion and result in greater RoM in LB and AR

Specimen specific 3-D technology allowed measurement of both disc height and segmental lordosis before and after TDA implantation. Significant increases in both middle disc height and segmental lordosis were found after TDA, regardless of the preload applied. When this study was performed, the smallest TDA height available was 6mm. A 5mm implant height Page 15 of 30

ACCEPTED MANUSCRIPT may have mitigated the increases in disc height and segmental lordosis. In vivo, degenerative segments have some loss of disc height resulting in a reduction of segmental lordosis [31]. Restoration of segmental lordosis and overall sagittal balance should be taken into account during selection of the TDA height.

T

Neutral zone as described by Panjabi is a measure of the angular deformation at 0 N,

IP

between the loading and unloading range of motion curves in flexion-extension [32]. NZ is a

CR

composite measure, and a change in its dimension can be caused by several factors including viscoelastic relaxation of the specimen’s soft tissues or TDA materials, change in segmental

US

stiffness, change in RoM, shift in the axis of rotation and the rate of loading. Discrete measures

AN

of hysteresis and segmental stiffness in the HFZ are specific measures of quality of motion and can be used to compare TDA technologies. However, NZ has been included in the results for

M

comparison to previously published studies.

ED

The sigmoidal shape of an intact motion segments load displacement curve (Fig. 3) is determined by the soft tissue stiffness, and primarily dependent on the disc health and hydration.

PT

A healthy disc has a gradual stiffness change between the neutral posture or HFZ and the low

CE

flexibility or stiff region. As the motion segment becomes degenerative, disc height decreases and the laxed soft tissues provide less resistance to motion causing the HFZ stiffness to decrease

AC

and segmental RoM to increase. Under compressive preload, the TDA caused the flexion response to be more monotonic than the intact motion. In other words, the typical bi-linear response in flexion became more linear. One of the goals of TDA is to restore the HFZ stiffness to encourage simultaneous recruitment of all motion segments. Ideally, stiffness of the reconstructed motion segment should be as close as possible to the native adjacent discs. An arthroplasty that is too stiff will require Page 16 of 30

ACCEPTED MANUSCRIPT additional motion from the adjacent native discs. An arthroplasty that reduces stiffness below the adjacent discs risks excessive wear, increased muscle activity to stabilize the unstable segment and risks stress shielding the adjacent native discs, which may be detrimental to their long term viability.

T

After TDA, the HFZ stiffness increased at C5-C6 while remaining similar to intact at C6-

IP

C7. This increase in stiffness can potentially be attributed to two factors. First, the disc health

CR

of the two segments may have played a role. The HFZ standard deviations suggests that the C5-C6 levels were very consistent and of similar health, while the C6-C7 levels were

US

more variable. This increased variability can be due to degenerative changes in some of

AN

the specimens. At C5-C6, all segments but one (8 of 9) increased stiffness after TDA by an average of 0.06Nm/deg under no preload. At C6-C7 the response was different with

M

three specimens increasing stiffness by an average 0.03Nm/deg and four specimens

ED

decreasing stiffness by an average -0.03Nm/deg. Under preload, the response was similar. The significant increase in stiffness at C5-C6 is due to the homogenous response

PT

of the intact motion segments, which was lacking at C6-C7. Secondly, an increase in

CE

segmental height after TDA disproportionately increased stiffness at C5-C6. The intact disc height at C5-C6 was significantly smaller than at C6-C7 (P<0.01). This in combination with the

stiffness.

AC

smallest TDR size of 6mm, likely increased soft tissue tension resulting in increased HFZ

Hysteresis after TDA implantation was not different than the intact condition under compressive follower preload but was different without compressive load. Biomechanical evaluation of cervical arthroplasty, anecdotally tends to show increased segmental hysteresis compared to the intact motion segment [15, 33, 34, 35]. This change in hysteresis is common Page 17 of 30

ACCEPTED MANUSCRIPT with arthroplasty, and likely contributors are: friction between the articulating surfaces, altered segmental mechanics due to soft tissue dissection and soft tissue tensioning. A direct comparison of changes in hysteresis between TDA devices is not possible since this measure is not commonly reported in the literature and how much of the observed change can be attributed to

T

TDA mechanics vs altered tissue mechanics is difficult to determine. Similarly, the long-term

IP

health costs are not known, but likely have consequences to the sequence of motion of individual

CR

segments within the global cervical motion. The effect of a change in hysteresis can only be

RoM, not just using flexion-extension radiographs.

US

studied with long-term follow up of clinical studies investigating motion quality throughout the

AN

The facet joints and ligamentous structures provide stability to a motion segment and influence the CoR location. If not closely aligned with the native anatomy, a fixed CoR design

M

may cause abnormal quantity and quality of motion resulting in altered facet loading and

ED

degenerative changes such as heterotrophic ossification [36]. Activities of daily living occur primarily in the HFZ making traditional measures of CoR from full extension to full flexion

PT

potentially inaccurate. A new CoR measure has been presented to measure CoR in the high

CE

flexibility zone where it is less affected by the facet joints and tensioned soft tissues. The overall results of this study show TDA maintained the flexion-extension CoR in the

AC

anteroposterior direction within 1.1 (SD 0.9) mm and in the cranial-caudal direction within 0.2 (SD 2.1) mm of the intact CoR location.

Conclusion The Triadyme-C™ cervical disc arthroplasty evaluated in this study restored RoM in flexion-extension to intact levels. Overall the TDA maintained good quality of motion at both Page 18 of 30

ACCEPTED MANUSCRIPT implanted levels. The TDA moderately increased motion segment stiffness through the neutral posture and had minimal effect on neutral zone and hysteresis. The TDA produced disc height distraction of 1.0 to 1.5mm depending on the implanted level and provided an increase in segmental lordosis by 4.2 degrees on average under physiologic compressive loads. One of the

T

design constraints of this TDA was that it have an adaptive CoR to accommodate the native CoR

IP

of each motion segment. CoR location measured in this study closely mimiced the native CoR

AC

CE

PT

ED

M

AN

US

CR

location within each implanted motion segment.

Page 19 of 30

ACCEPTED MANUSCRIPT References

[1]

M. Alhashash, M. Shousha and H. Boehm, "Adjacent Segment Disease After Cervical Spine Fusion: Evaluation of a 70 Patient Long Term Follow-up," Spine, vol. 43, no. 9, pp.

A. S. Hilibrand, G. D. Carlson, M. A. Palumbo, P. K. Jones and H. H. Bohlman,

IP

[2]

T

605-609, May 2018.

CR

"Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior

[3]

US

cervical arthrodesis," J Bone Joint Surg Am, vol. 81, pp. 519-528, Apr 1999. W. R. Sears, P. F. McCombe and R. C. Sasso, "Kinematics of Cervical and Lumbar Total

R. C. Sasso, J. D. Smucker, R. J. Hacker and J. G. Heller, "Artificial disc versus fusion: a

M

[4]

AN

Disc Replacement," Semin Spine Surg, vol. 18, pp. 117-129, 2006.

2933-2940, Dec 2007.

P. V. Mummaneni, J. K. Burkus, R. W. Haid, V. C. Traynelis and T. A. Zdeblick,

PT

[5]

ED

prospective, randomized study with 2-year follow-up on 99 patients," Spine, vol. 32, pp.

"Clinical and radiographic analysis of cervical disc arthroplasty compared with allograft

[6]

AC

2007.

CE

fusion: a randomized controlled clinical trial," J Neurosurg Spine, vol. 6, pp. 198-209, Mar

M. F. Gornet, J. K. Burkus, M. E. Shaffrey, H. Nian and F. E. Harrell, "Cervical Disc Arthroplasty with Prestige LP Disc Versus Anterior Cervical Discectomy and Fusion: Seven-Year Outcomes," Int J Spine Surg, vol. 10, p. 24, 2016.

[7]

R. J. Davis, K. D. Kim, M. S. Hisey, G. A. Hoffman, H. W. Bae, S. E. Gaede, R. F. Rashbaum, P. D. Nunley, D. L. Peterson and J. K. Stokes, "Cervical total disc replacement

Page 20 of 30

ACCEPTED MANUSCRIPT with the Mobi-C cervical artificial disc compared with anterior discectomy and fusion for treatment of 2-level symptomatic degenerative disc disease: a prospective, randomized, controlled multicenter clinical trial: clinical article," J Neurosurg Spine, vol. 19, pp. 532545, Nov 2013. M. S. Hisey, J. E. Zigler, R. Jackson, P. D. Nunley, H. W. Bae, K. D. Kim and D. D.

T

[8]

IP

Ohnmeiss, "Prospective, Randomized Comparison of One-level Mobi-C Cervical Total

CR

Disc Replacement vs. Anterior Cervical Discectomy and Fusion: Results at 5-year Follow-

[9]

US

up," Int J Spine Surg, vol. 10, p. 10, 2016.

D. Coric, P. D. Nunley, R. D. Guyer, D. Musante, C. N. Carmody, C. R. Gordon, C.

AN

Lauryssen, D. D. Ohnmeiss and M. O. Boltes, "Prospective, randomized, multicenter study

M

of cervical arthroplasty: 269 patients from the Kineflex|C artificial disc investigational device exemption study with a minimum 2-year follow-up: clinical article," J Neurosurg

J. E. Zigler, R. Delamarter, D. Murrey, J. Spivak and M. Janssen, "ProDisc-C and anterior

PT

[10]

ED

Spine, vol. 15, pp. 348-358, Oct 2011.

cervical discectomy and fusion as surgical treatment for single-level cervical symptomatic

CE

degenerative disc disease: five-year results of a Food and Drug Administration study,"

[11]

AC

Spine, vol. 38, pp. 203-209, Feb 2013. P. C. McAfee, C. Reah, K. Gilder, L. Eisermann and B. Cunningham, "A Meta-Analysis of Comparative Outcomes Following Cervical Arthroplasty or Anterior Cervical Fusion," Spine, vol. 37, no. 11, pp. 943-952, 2012. [12]

L. Dong, D. Wang, X. Chen, T. Liu, Z. Xu, M. Tan and D. Hao, "A Comprehensive MetaAnalysis of the Adjacent Segment Parameters in Cervical Disk Arthroplasty Versus

Page 21 of 30

ACCEPTED MANUSCRIPT Anterior Cervical Discectomy and Fusion," Clin Spine Surg, vol. 31, no. 4, pp. 162-173, 2018. [13]

D. J. DiAngelo, K. T. Foley, B. R. Morrow, J. S. Schwab, J. Song, J. W. German and E. Blair, "In vitro biomechanics of cervical disc arthroplasty with the ProDisc-C total disc

C. M. Puttlitz, M. A. Rousseau, Z. Xu, S. Hu, B. K. Tay and J. C. Lotz, "Intervertebral

IP

[14]

T

implant," Neurosurg Focus, vol. 17, p. E7, Sep 2004.

CR

disc replacement maintains cervical spine kinetics," Spine, vol. 29, pp. 2809-2814, Dec

[15]

US

2004.

A. G. Patwardhan, M. N. Tzermiadianos, P. P. Tsitsopoulos, L. I. Voronov, S. M. Renner,

AN

M. L. Reo, G. Carandang, K. Ritter-Lang and R. M. Havey, "Primary and coupled motions

M

after cervical total disc replacement using a compressible six-degree-of-freedom

[16]

ED

prosthesis," Eur Spine J, vol. 21 Suppl 5, pp. S618--629, Jun 2012. N. R. Crawford, S. Baek, A. G. Sawa, S. Safavi-Abbasi, V. K. Sonntag and N. Duggal,

PT

"Biomechanics of a Fixed-Center of Rotation Cervical Intervertebral Disc Prosthesis," Int J

[17]

CE

Spine Surg, vol. 6, pp. 34-42, Dec 2012. A. G. Patwardhan, R. M. Havey, A. J. Ghanayem, H. Diener, K. P. Meade, B. Dunlap and

AC

S. D. Hodges, "Load-carrying capacity of the human cervical spine in compression is increased under a follower load," Spine, vol. 25, pp. 1548-1554, Jun 2000. [18]

H.-J. Wilke, K. Wenger and L. Claes, "Testing criteria for spinal implants: recommendations for the standardization of in vitro stability testing of spinal implants," European Spine Journal, vol. 7, pp. 148-154, 1998.

[19]

R. M. Havey, J. Goodsitt, S. Khayatzadeh, M. Muriuki, T. Potluri, L. I. Voronov, L. M.

Page 22 of 30

ACCEPTED MANUSCRIPT Lomasney and A. G. Patwardhan, "Three-Dimensional Computed Tomography-Based Specimen-Specific Kinematic Model for Ex Vivo Assessment of Lumbar Neuroforaminal Space," Spine, vol. 40, pp. E814--822, Jul 2015. [20]

J. T. Snyder, M. N. Tzermiadianos, A. J. Ghanayem, L. I. Voronov, A. Rinella, A. Dooris,

T

G. Carandang, S. M. Renner, R. M. Havey and A. G. Patwardhan, "Effect of uncovertebral

IP

joint excision on the motion response of the cervical spine after total disc replacement,"

R. M. Havey, L. I. Voronov, P. P. Tsitsopoulos, G. Carandang, A. J. Ghanayem, M. A.

US

[21]

CR

Spine, vol. 32, pp. 2965-2969, Dec 2007.

Lorenz, M. R. Zindrick and A. G. Patwardhan, "Relaxation response of lumbar segments

AN

undergoing disc-space distraction: implications to the stability of anterior lumbar interbody

[22]

M

implants," Spine, vol. 37, pp. 733-740, Apr 2012. C. M. Puttlitz and D. DiAngelo, "Cervical Spine Arthroplasty Biomechanics," Neurosurg

M. M. Panjabi, J. J. Crisco, A. Vasavada, T. Oda, J. Cholewicki, K. Nibu and E. Shin,

PT

[23]

ED

Clin N Am, vol. 16, pp. 589-594, 2005.

"Mechanical Properties of the Human Cervical Spine as Shown by Three-Dimensional

A. E. Dmitriev, B. W. Cunningham, N. Hu, G. Sell, F. Vigna and P. C. McAfee,

AC

[24]

CE

Load–Displacement Curves," Spine, vol. 26, no. 24, pp. 2692-2700, 2001.

"Adjacent Level Intradiscal Pressure and Segmental Kinematics Following A Cervical Total Disc Arthroplasty: An In Vitro Human Cadaveric Model," Spine, vol. 30, no. 10, pp. 1165-1172, 2005. [25]

W. J. Anderst, W. F. Donaldson, J. Y. Lee and J. D. Kang, "Cervical Motion Segment Percent Contributions to Flexion-Extension During Continuous Functional Movement in

Page 23 of 30

ACCEPTED MANUSCRIPT Control Subjects and Arthrodesis Patients," Spine, vol. 38, no. 9, pp. E533-E539, 2013. [26]

A. A. White and M. M. Panjabi, "Kinematics of the Spine," in Clinical Biomechanics of the Spine, Philadelphia, J.B. Lippincott Company, 1978, pp. 61-90.

[27]

A. Holmes, C. Wang, Z. H. Han and G. T. Dang, "The range and nature of flexion-

S. P. Moroney, A. B. Schultz and J. A. Miller, "Analysis and measurement of neck loads,"

CR

[28]

IP

T

extension motion in the cervical spine," Spine, vol. 19, pp. 2505-2510, Nov 1994.

J. Orthop. Res., vol. 6, pp. 713-720, 1988.

M. J. Garcia, A. Ghanayem, M. Tzermiadianos, L. I. Voronov, P. Cahill, A. Rinella, G.

US

[29]

Carandang, C. Abjornson, R. M. Havey and A. G. Patwardhan, "Effect of Cervical Disc

M. A. Finn, D. S. Brodke, M. Daubs, A. Patel and K. N. Bachus, "Local and global

M

[30]

AN

Prosthesis Height on Kinematics," in NASS Annual Meeting, Seattle, WA, 2006.

ED

subaxial cervical spine biomechanics after single-level fusion or cervical arthroplasty," Eur Spine J, vol. 18, pp. 1520-1527, Oct 2009. M. Miyazaki, H. Hymanson, Y. Morishita, W. He, H. Zhang, G. Wu, M. H. Kong, H.

PT

[31]

CE

Tsumura and J. Wang, "Kinematic Analysis of the Relationship Between Sagittal Alignment and Disc Degeneration in the Cervical Spine," Spine, vol. 33, no. 23, pp. E870-

[32]

AC

E876, 2008.

M. M. Panjabi, "The stabilizing system of the spine. Part II. Neutral zone and instability hypothesis," J Spinal Disord, vol. 5, pp. 390-396, Dec 1992.

[33]

S. Martin, A. J. Ghanayem, M. N. Tzermiadianos, L. I. Voronov, R. M. Havey, S. M. Renner, G. Carandang, C. Abjornson and A. G. Patwardhan, "Kinematics of cervical total disc replacement adjacent to a two-level, straight versus lordotic fusion," Spine, vol. 36, pp. Page 24 of 30

ACCEPTED MANUSCRIPT 1359-1366, Aug 2011. [34]

M. J. Lee, M. Dumonski, F. M. Phillips, L. I. Voronov, S. M. Renner, G. Carandang, R. M. Havey and A. G. Patwardhan, "Disc replacement adjacent to cervical fusion: a biomechanical comparison of hybrid construct versus two-level fusion," Spine, vol. 36, pp.

M. Brody, A. Patel, A. Ghanayem, B. Wojewnik, G. Carandang, R. Havey, L. Voronov,

IP

[35]

T

1932-1939, Nov 2011.

CR

G. Vastardis, T. Potluri and A. Patwardhan, "The Effect of Posterior Decompressive

US

Procedures on Segmental Range of Motion After Cervical Total Disc Arthroplasty," Spine, vol. 39, no. 19, pp. 1558-1563, 2014.

H. S. Ahn and D. J. DiAngelo, "A Biomechanical Study of Artificial Cervical Discs Using

AN

[36]

ED

G. E. Pickett, J. P. Rouleau and N. Duggal, "Kinematic analysis of the cervical spine following implantation of an artificial cervical disc," Spine, vol. 30, pp. 1949-1954, Sep

CE

PT

2005.

AC

[37]

M

Computer Simulation," Spine, vol. 33, no. 8, pp. 883-892, 2008.

Page 25 of 30

ACCEPTED MANUSCRIPT

Figure 1. A) Triadyme-C, Tri-lobed Polycrystalline Diamond Disc Prosthesis, Dymicron, Orem, UT, USA. B) Diagram showing the layers of PCD, Ti/TiC composite and titanium plasma

IP

T

coating. The prosthesis is made by sintering diamond and Ti/TiC powders at geologic

CR

temperatures and pressures. The layers are fully fused and chemically bonded. After sintering the

US

keels are cut and titanium plasma coating applied.

Figure 2. Experimental Protocol. A) Intact, B) C5-C6 TDA, C) C6-C7 TDA

AN

Figure 3: Segmental flexion-extension load-displacement curves. (A) C5-C6 RoM curves for the intact segment under 0N compressive preload and after TDA placement under 0N and 150N

M

compressive preload. (B) C6-C7 RoM curves for intact (0N) and after TDA placement with 0N

ED

and 150N of compressive follower preload.

Figure 4: Segmental range of motion at C5-C6 and C6-C7, intact and after TDA placement.

PT

Results for 0N and 150N compressive follower preload are presented. (*) Statistically different

CE

than the intact condition (P<0.05).

Figure 5: Segmental disc height of the intact motion segments and after TDA placement. Height

AC

was measured in the neutral posture at the sagittal view midline. (*) Statistically different than the intact condition (P<0.05). Figure 6: Change in CoR location from the intact condition to TDA. The right side of each panel (C5-C6 and C6-C7) shows the motion segment and the change in position of the COR from intact (filled circle) to after TDA (arrowhead). The left panel shows each specimen’s intact and TDA CoR location. A) C5-C6 CoR moved 0.9(1.0) mm posteriorly and 0.1(2.2) mm caudally Page 26 of 30

ACCEPTED MANUSCRIPT after TDA (n=8). B) C6-C7 CoR moved 1.4(0.8) mm posteriorly and 0.3(2.0) mm cranially from intact to TDA (n=7).

Table 1: Segmental kinematics for each protocol step including: ROM, Neutral Zone, Flexion

IP

T

stiffness and Extension stiffness, Hysteresis, Lordosis, and Middle Disc Height. C5-C6 data

standard deviation for n=7 specimens.

US

* Significantly different than the intact condition (p<0.05).

CR

represents mean and standard deviation for n=9 specimens. C6-C7 data represents mean and

AN

Table 2: Statistical analysis of intact vs TDA placement. Statistical significance is shown by

AC

CE

PT

ED

M

values of p≤0.05.

Page 27 of 30

ACCEPTED MANUSCRIPT

Table 1: Segmental kinematics for each protocol step including: RoM, Neutral Zone, Flexion stiffness and Extension stiffness, Hysteresis, Lordosis, and Middle Disc Height. C5-C6 data represents mean and standard deviation for n=9 specimens. C6-C7 data represents mean

FE (0N)

AN

IP 11.5(3.4)

12.4(3.3)

10.5*(2.1)

10.0(3.4)

11.4(3.0)

7.5(2.8)

5.1(2.3)

AR

10.4(1.1)

6.2*(1.9)

7.8(1.7)

5.3*(0.9)

FE (0N)

2.0(1.0)

1.5(0.4)

1.6(0.7)

1.1(0.4)

FE (150N) 2.6(1.7)

1.6(0.7)

2.1(1.3)

2.0(0.8)

M

3.7*(1.0)

0.05(0.01)

0.11*(0.04) 0.09(0.05)

0.08(0.04)

FE (150N) 0.09(0.03)

0.21*(0.09) 0.13(0.06)

0.15(0.08)

HFZ Extension Stiffness (Nm/degree)

FE (0N)

0.06(0.04)

0.10*(0.03) 0.08(0.04)

0.08(0.03)

FE (150N) 0.08(0.03)

0.18*(0.07) 0.12(0.05)

0.11(0.04)

FE (0N)

1.53(0.63)

2.2*(0.79)

1.48(0.46)

1.93*(0.44)

FE (150N) 2.39(0.80)

2.96(0.77)

2.17(1.19)

3.09(1.14)

FE (0N)

8.2(3.1)

11.2*(4.3)

7.3(4.0)

13.6*(3.9)

FE (150N) 7.1(2.7)

10.4*(4.6)

7.3(4.0)

12.7*(4.0)

FE (0N)

5.3(0.8)

6.8*(1.5)

6.0(0.4)

7.0*(0.4)

FE (150N) 5.2(0.9)

6.6*(1.6)

5.8(0.4)

6.9*(0.4)

CE

PT

FE (0N)

Middle Disc height (mm)

11.7(1.8)

8.5(2.8)

HFZ Flexion Stiffness (Nm/degree)

Segmental lordosis (deg)

Intact

C6-C7 TDA Placement

LB

ED

Neutral Zone (degrees)

AC

12.6(2.6)

FE (150N) 12.8(2.5)

Range of Motion (degrees)

Flexion-Extension Hysteresis (Nm*deg)

C5-C6 TDA Placement

US

Intact

CR

(*) Significantly different than the intact condition (p<0.05).

T

(standard deviation) for n=7 specimens.

Page 28 of 30

ACCEPTED MANUSCRIPT

Table 2: Statistical analysis of intact vs TDA placement. Statistical significance is shown by values of p≤0.05. Intact vs. TDA

0.318

0.067

FE (150N) 0.028

0.148

LB

0.001

0.065

AR

0.000

0.018

FE (0N)

0.152

0.139

FE (150N) 0.084

0.844

0N

0.006

0.674

150N

0.004

0N

0.010

150N

0.003

0.427

0N

0.031

0.047

0.073

0.099

0.003

0.000

150N

0.005

0.000

0N

0.001

0.001

150N

0.001

0.000

IP

FE (0N)

T

C5-C6 C6-C7

150N

0.784

PT

AC

Disc height

0N

CE

Segmental lordosis

ED

Hysteresis

0.424

M

Extension Stiffness

US

Flexion Stiffness

AN

Neutral Zone

CR

Range of Motion

Page 29 of 30

ACCEPTED MANUSCRIPT

Highlights

The polycrystalline diamond cervical arthroplasty restored intact levels of flexion-extension



Arthroplasty moderately increased segmental stiffness



Arthroplasty had minimal effect on neutral zone and hysteresis



Arthroplasty increased disc height (1.0 to 1.5mm) and segmental lordosis (4.2 degrees) on

IP

CR

US

average

CE

PT

ED

M

AN

The Arthroplasty design allowed the center of rotation to closely mimic that of the native disc

AC



T



Page 30 of 30

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6