Biomechanics of anterior plating failure in treating distractive flexion injury in the caudal subaxial cervical spine

Biomechanics of anterior plating failure in treating distractive flexion injury in the caudal subaxial cervical spine

Accepted Manuscript Biomechanics of anterior plating failure in treating distractive flexion injury in the caudal subaxial cervical spine Zhong Wang,...

1MB Sizes 9 Downloads 66 Views

Accepted Manuscript Biomechanics of anterior plating failure in treating distractive flexion injury in the caudal subaxial cervical spine

Zhong Wang, Hui Zhao, Ji-ming Liu, Rui Chao, Tai-bang Chen, Li-wen Tan, Feng Zhu, Jian-hua Zhao, Peng Liu PII: DOI: Reference:

S0268-0033(17)30270-X doi:10.1016/j.clinbiomech.2017.10.017 JCLB 4410

To appear in:

Clinical Biomechanics

Received date: Accepted date:

24 March 2017 18 October 2017

Please cite this article as: Zhong Wang, Hui Zhao, Ji-ming Liu, Rui Chao, Tai-bang Chen, Li-wen Tan, Feng Zhu, Jian-hua Zhao, Peng Liu , Biomechanics of anterior plating failure in treating distractive flexion injury in the caudal subaxial cervical spine. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Jclb(2017), doi:10.1016/j.clinbiomech.2017.10.017

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 Segment-specific stability of cervical spine. 1

Biomechanics of anterior plating failure in treating distractive flexion injury in the caudal subaxial cervical spine

RI PT

Zhong Wanga, MD, Hui Zhaob, PhD, Ji-ming Liuc, ME, Rui Chaoad, PhD, MD, Tai-bang Chene, MD,

Department of Spine Surgery, Daping Hospital, Institute of Surgery Research, Daping Hospital, Third

Military Medical University, Chongqing, China

MA

b

NU

a

SC

Li-wen Tanf, PhD, Feng Zhug, PhD, Jian-hua Zhaoa*, PhD, MD, Peng Liua*, MD

Chongqing Key Laboratory of Vehicle Crash/Bio-Impact and Traffic Safety, Department 4, Institute of

c

ED

Surgery Research, Daping Hospital, Third Military Medical University, Chongqing, China China Automotive Engineering Research Institute, No. 9 Jinyu Avenue, New North Zone, Chongqing,

CE PT

China d

Department of Orthopaedic Surgery, Chongqing Emergency Medical Center, the Fourth People’s

e

f

AC

Hospital of Chongqing, Chongqing, China Department of Orthopaedic surgery, Kunming General Hospital, Yunnan, China

Institute of Digital Medicine, Third Military Medical University, Chongqing, China

g

Department of Mechanical Engineering, Embry-Riddle Aeronautical University, Daytona Beach,

Florida, USA *

Co-corresponding authors

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 2

Email: Zhong Wang: [email protected] Hui Zhao: [email protected]

RI PT

Ji-ming Liu: [email protected] Rui Chao: [email protected] Tai-bang Chen: [email protected]

SC

Li-wen Tan: [email protected]

MA

Jian-hua Zhao: [email protected]

NU

Feng Zhu: [email protected]

ED

Peng Liu: [email protected]

Respondence and reprint requests should be addressed to Jian-hua Zhao, Department of Spine Surgery,

CE PT

Daping Hospital, Institute of Surgery Research, 10 Branch Road, Changjiang Street, Yuzhong District, Chongqing 400042, China; E-mail: [email protected],

AC

Or:

Peng Liu, Department of Spine Surgery, Daping Hospital, Institute of Surgery Research, 10 Branch Road, Changjiang Street, Yuzhong District, Chongqing 400042, China; E-mail: [email protected]

Abstract: 246 words, Main text: 4388 words, Figures: 7, Table number: 1

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 3

Abstract Background Operative level is a potential biomechanical risk factor for construct failure during anterior

RI PT

fixation for distractive flexion injury. No biomechanical study of this concept has been reported, although it is important in clinical management. Methods

SC

To explore the mechanism of this concept, a previously validated three-dimensional C2-T1 finite

NU

element model was modified to simulate surgical procedure via the anterior approach for treating

MA

single-level distractive flexion injury, from C2-C3 to C7-T1. Four loading conditions were used including no-compression, follower load, axial load, and combined load. Construct stability at the

ED

operative level was assessed. Findings

CE PT

Under these loading conditions with the head’s weight simulated, segmental stability decreases when the operative level shifts cephalocaudally, especially at C6-C7 and C7-T1, the stress of

AC

screw-bone interface increases cephalocaudally, and in the same operative level, the caudal screws always carries more load than the cephalad ones. All these predicted results are consistent with failure patterns observed in clinical reports. In the contrast, under other loading conditions without the weight of head, no obvious segmental divergence was predicted. Interpretation This study supports that the biomechanical mechanism of this phenomenon includes eccentric load from head weight during sagittal movements and difference of moment arms. Our study suggests

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 4

that anterior fixation is not recommended for treating distractive flexion injury at the caudal segments of the subaxial cervical spine, especially at C6-C7 and C7-T1, because of the intrinsic instability in these segments. Combined posterior rigid fixation with anterior fixation should be considered for these

RI PT

segments.

AC

CE PT

ED

MA

NU

SC

Key Words: Anterior plate, cervical spine trauma, biomechanics, finite element analysis

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 5

1.

Introduction The principles for treating unstable subaxial cervical spine (SCS) injuries have been largely

established. However, the optimal treatment for distractive flexion injury (DFI), one of the three

RI PT

common SCS injuries (Allen, et al., 1982; Urrutia, et al., 2016), remains unclear. Literatures demonstrate that DFI treatment, which include anterior, posterior, and combined approaches, have differing advantages and disadvantages, complicating decision-making for clinicians regarding

SC

treatment choice (Yonenobu, et al., 1985; Ripa, et al., 1991; Saunders, et al., 1991; Emery, et al.,1998;

NU

Razack, et al.,2000; Wang, et al.,2003; Johnson, et al.,2004; Koller, et al.,2009). Anterior cervical

MA

decompression, fusion and plating (ACDFP) is the one of the most popular operative modalities (Goffin, et al., 1989; Aebi, et al., 1991; Reindl, et al., 2006; Villavicencio, et al., 2007; Koller, et al.,

ED

2009; Maharaj, et al., 2015). Compared with posterior surgery, ACDFP causes less iatrogenic injury to muscles, solid interbody strut grafting, and direct spinal canal decompression via discectomy (Eismont,

CE PT

et al., 1991; Randle, et al., 1991; Garvey, et al., 1992; Robertson and Ryan, 1992). Several clinical reports of satisfactory outcomes suggest that ACDFP may be beneficial for

AC

cervical DFI; reported general rates of pseudarthrosis and mechanical failure are low (Yonenobu, et al., 1985; Aebi, et al., 1991; Randle, et al., 1991; Ripa, et al., 1991; Saunders, et al., 1991; Garvey, et al., 1992; Emery, et al., 1998; Lifeso and Colucci, 2000; Razack, et al., 2000; Wang, et al., 2003; Johnson, et al., 2004; Koller, et al., 2009). However, the mechanical failure rate in the caudal segments is higher than in the cephalad segments. Reviewing the literature and our clinical practice, most ACDFP construct failures occur at the caudal SCS segments, mostly frequently at C6-C7, C7-T1, and occasionally at C5-C6. A study investigated the radiographic failure of a single-segment ACDFP in 87

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 6

patients with DFI. No failure occurred at segments of C3-C4 and C4-C5, 6% failure occurred at C5-6 (2/33), 24% at C6-C7 (8/33), and 33% at C7-T1 (1/3) (Johnson et al., 2004). Koller et al. (2009) observed the mid/long-term clinical outcomes of 29 cases of SCS injuries treated with ACDFP and

RI PT

found that 3 non-fusion cases all occurred at C6-C7 segment. These studies indicated that the chance of construct failure in surgeries was significantly higher at the C6-C7 level than at the more cephalad levels. In a series of old DFI, which means DFI without treatment for more than 3 weeks, recurrence of

SC

dislocation after operations was only observed at C6-C7 level, even with combined fixation of a

NU

posterior wire and ACDFP (Liu et al., 2008). This implies that DFI at C6-C7 level needs more rigid

MA

implants to maintain segmental stability. Wang et al. (2003) observed in a fusion that ended at C7 that the corpectomy involved appeared to have an increased probability of graft migration or displacement.

ED

The above reports imply that operative level affects the stability provided by the ACDFP in effectively treating DFI. However, the number of failures (mechanical or functional) in any of these

CE PT

studies is too low to apply the desired logistic regression modelling appropriate to control for interaction effect or confounding of the independent variables on the binary outcome of ACDFP alone

AC

in treating DFI (Wang, et al., 2003; Johnson, et al., 2004; Koller, et al., 2009). To our knowledge, this level-dependent stability of a certain implant in a certain region of the spine has never been examined longitudinally and related mechanisms remain unclear. We hypothesized that biomechanical factors contribute to this increased anterior plating failure in caudal segments, in the case of posterior ligamentous complex (PLC) dysfunction. Intrinsic stability of the caudal segments decreases more markedly after DFI compared to the cephalad segments. ACDFP alone cannot provide enough stability in these segments. The caudal implants carry more load than the

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 7

cephalad ones. Therefore assessing the biomechanical association between operative level and implants load and segmental stability is essential. Consequently, we addressed the following research questions in a series of experiments: (1) Does operative level affect segmental stability? (2) What is the

RI PT

mechanism behind this phenomenon? (3) In consideration of immediate postoperative stability, which

2.

Development of the finite element model (FEM)

NU

2.1.

Methods

SC

segment of SCS is ACDFP not suggested for treating DFI?

MA

A previously developed and validated three-dimensional FEM of the C2-T1 ligamentous subaxial cervical spine was used in this study (Wang, et al., 2016). This model was created using a series of

ED

cryosections of a Chinese female donor (22-year-old, weight 52 kg, and height 162 cm at the time of death). A total of 399 slices (No. 1584 to No. 1792) were used. Based on Cusick’s scale (1996), these

CE PT

segments were graded as “normal”, ensuring that the SCS of CVH2 was degeneration-free. The original color photographs were segmented manually by several spine surgeons and an

AC

experienced anatomist using commercial software Adobe Photoshop (Adobe Systems, California, USA). Quality of segmentation and surface smoothing were checked using commercial software Amira (5.3.3, Visage Imaging, Carlsbad, CA). Meshing and refinement was processed using HyperMesh (V12.0, Altair, Michigan, America) based on the acquired surface files. Cortical and cancellous bone was modelled using isotropic elastic four-node tetrahedral elements. Lordosis of C2-C7 in current model was 23.7°, measured by Cobb’s method (Fig. 1A), which was in the reported range of the SCS (Linder, 2000). The facet joints of C1-C2 were adjusted to be parallel to

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 8

the horizontal plane. Facet cartilage was modelled using hexahedral elements with an isotropic-elastic material model. The uncovertebral joint is defined as a joint between the uncinate process and the adjacent anvil of the superior vertebrae. The cartilage of the uncovertebral joints was modelled as an

RI PT

elongation of the cartilaginous end-plate. Surface to surface contact algorithm is used in defining joint interaction and friction coefficient was assumed to be 0.1 (Womack, et al., 2008). The volume and inclination of uncovertebral joints were defined with reconstructed coronal slices using an improved

SC

method based on a previously published method (Kumaresan, et al., 1997; Yoganandan, et al., 2001).

NU

The major ligaments modeled in current model included anterior longitudinal ligament (ALL),

MA

posterior longitudinal ligament (PLL), capsular ligament (CL), ligamentum flavum (FL), and interspinous ligaments (ISL). These ligaments were modeled using 2-node, tension-only, nonlinear

ED

spring elements, defined with force-displacement curves from previous literatures as shown supplemental material (Yoganandan, et al., 2000; Wheeldon, et al., 2008).

CE PT

The intervertebral disc was modeled as a central nucleus with a surrounding ring-shaped annulus, coved with two cartilaginous end-plates. Cartilaginous end-plate was modeled with 8-node

AC

isotropic-elastic solid elements. Thickness of cartilaginous end-plates was 1.2 mm at the perimeter, and gradually decreased to 0.6mm in the center (Leahy and Puttlitz, 2012). The nucleus comprised 25% of the cross-sectional area of the total cervical disc based on literature data and our MRI pilot test (Womack, 2009). The nucleus was modeled using hyper-elastic, incompressible, two-parameter Mooney-Rivlin formulation (Schmidt, et al., 2006). Annulus was modelled as a two-component structure: ground substance (GS) with reinforced concentric lamina of fiber. GS was modeled using hyper-elastic hexahedral elements (C3D8H), assumed to be equal for all discs. The annulus fiber was

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 9

arranged in 8 concentric rings, attached to the cartilaginous end-plate. Volume of the fiber comprised approximately 20% of the surrounding annulus. At the midsagittal position, the external fiber was arranged in a 30° pattern to the horizontal in the outmost layer, and gradually increased to 45° in the

RI PT

innermost layer (Holzapfel, et al., 2005). The annulus fiber was modeled with 2-noded, tension-only, nonlinear spring elements and its nonlinear behavior was described with a function based on experimental data of Shirazi-Adl et al. (1986). The fibers of different lamellae were adjusted according

SC

to the literature (innermost layers 7–8: 0.65, layers 3–4: 0.9, layers 5–6: 0.75, outermost layers 1–2: 1.0)

NU

(Schmidt, et al., 2006).

MA

The initial material properties were based on previous studies, and parts of them were adjusted, as shown in Table 1. In total, 54,667 spring elements, 777,789 solid elements were used to construct the

ED

initial model. Material properties of the model structure and instrumentations were adopted from the literature (Shirazi-Adl, et al., 1986; Yoganandan, et al., 2000, 2001; Holzapfel, et al., 2005; Schmidt, et

CE PT

al., 2006; Wheeldon, et al., 2008; Hussain, et al., 2009; Panzer and Cronin, 2009) as shown in Table 1. Mesh quality of initial model was shown in supplemental material. Model validation

AC

2.2.

The current model was validated against a number of in vitro studies of human cervical spine including pure-bending moment and the ones with pre-compression (Miura, et al.,2002; Wheeldon, et al., 2006; Nightingale, et al., 2007; Yoganandan, et al., 2007, 2008). Range of motion (RoM) of each segment was the major indicator for validation. The predicted RoMs (flexion-extension, lateral bending and axial rotation) under 2 Nm pure bending moment were compared with the experimental data obtained from young, healthy normal cadaver cervical spine (Wheeldon, et al., 2006; Yoganandan, et

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 10

al., 2007, 2008). Follower load technique was used for applying a 100 N pre-compression and predicted RoMs of flexion-extension were compared (Miura, et al., 2002). Furthermore, C4-C5 segment was extracted for further comparison including low-level loading (predominantly disc

RI PT

response) and higher-level loading (including the facet joints and ligaments) (Nightingale, et al., 2007). All the chosen experimental studies were load controlled and all the simulations had the same boundary and loading conditions with the experimental studies. For comparison purpose the predicted results

Simulation of single-level DFI treated by ACDFP

MA

2.3.

NU

RI). The RoM for validation included the neutral zone.

SC

have been presented in a same manner. Simulations were conducted using Abaqus (Simulia, Providence,

Six additional FEMs were developed based on the intact model to simulate single-level DFI

ED

treated by ACDFP surgery (C2-C3, C3-C4, C4-C5, C5-C6, C6-C7, and C7-T1) (Fig. 2A). Interspinous ligaments, ligamentum flavum, capsular ligament, and the posterior longitudinal ligament were

CE PT

removed at the injured level to simulate a stage 3 or 4 DFI. To simulate single-level ACDFP, the anterior longitudinal ligament, disc and cartilaginous endplate were removed. Furthermore, a

AC

cylindrical strut bone graft was placed centrally between the intervertebral space occupying 50% of the opposing endplate areas (Hussain, et al., 2009). The bone grafts were not rigidly fused to the relative endplate and allowed compression but not tension transmission. The surface between the bone graft and adjacent endplate was described using a contact algorithm and the friction coefficient was set as 0.4. The anterior plate-screw fixation system was rigidly fixed at the operative level. The screw-plate and screw-bone interface were fully constrained for all 6 degrees of freedom. The screws were inserted parallel to the superior endplate and the medial inclination angle to the sagittal plane was

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 11

approximately 13°. Screws at different levels were assumed to have the same dimensions, drilling depth, and drilling angle (Fig. 1B). 2.4.

Boundary and Loading conditions

RI PT

To simulate in vivo segmental response in single-level ACDFP, four types of loading conditions were used in this study: no compression, follower load, axial load, and combined load (NC, FL, AL, and CL) (Fig. 2B). Intact and six single-level operative C2-T1 models were tested under these four

No compression (NC)

NU

2.4.1.

SC

loading conditions. The inferior surface of T1 was fully constrained in all loading conditions.

MA

A 2-Nm quasi-static pure bending moment (flexion-extension) was applied to the superior surface of C2. Follower load (FL)

ED

2.4.2.

A 100-N preload was applied to simulate physiologic compressive loads from paravertebral

CE PT

muscles and was simulated at each motion segment by a pair of 2-node spring elements. After applying the 100-N FL, a 2-Nm pure sagittal bending moment was applied to the superior surface of C2. The

2015). 2.4.3.

AC

RoM was normalized after FL was applied (Patwardhan, et al., 2000; Miura, et al., 2002; Bell, et al.,

Axial load (AL)

Here, the loads were applied in the order: NC and AL. First, a 2-Nm pure bending moment was applied to the superior surface of C2. Second, a 50-N AL was applied to the superior surface of C2 in the vertical direction to simulate the head’s weight (Bell, et al., 2015). 2.4.4.

Combined Load (CL)

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 12

A combined load was applied in this order: FL, NC, and AL; a 100-N pre-compressive FL, followed by a 2-Nm sagittal moment and a 50-N AL.

3.1.

Results

RI PT

3.

Model validation

In general, 50 test points were compared. RoM of 44 test points was within the experimental

SC

corridor (±1SD) (Fig. 3) (Miura, et al., 2002; Wheeldon, et al., 2006; Nightingale, et al.,2007;

NU

Yoganandan, et al.,2007, 2008). The minimum differences in segmental motions between the FE model

MA

and experimental corridor were pure bending moment under 2Nm (flexion C5-C6 0.9°, C6-C7 1.2°, axial rotation C4-C5 0.3°, C5-C6 0.7°, C6-C7 0.5°) and Pre-compression under 2Nm (flexion C6-C7

ED

0.5°). To authors’ opinion, such small differences in segmental motions of the initial model from in vitro experimental corridor can be justified, as such motion differences always exist between and

CE PT

among various in vitro studies because of differences in spinal anatomy and tissue behaviour. Current model matches the level 1 requirement of validation of Viceconti et al. (2005). Kinematics changes

AC

3.2.

In all loading conditions, RoM at the operative level was higher in flexion than in extension. In NC and FL, RoMs at different operative levels were similar and in the range of 2°. In AL and CL, RoM at the operative level notably increased at C6-C7 and C7-T1, compared with the cephalad segments. Rate of motion change (RMC), which means percent change of segmental motion, was used to analyse segmental stability provided by ACDFP. Cephalocaudal increase of RMC was observed in NC, AL, and CL loading conditions (Fig. 4).

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 13

Rate of motion change = 3.3.

𝑅𝑜𝑀 𝑜𝑓 𝑜𝑝𝑒𝑟𝑎𝑡𝑖𝑣𝑒 𝑙𝑒𝑣𝑒𝑙 × 100% 𝑅𝑜𝑀 𝑜𝑓 𝑖𝑑𝑒𝑛𝑡𝑖𝑐𝑎𝑙 𝑙𝑒𝑣𝑒𝑙 𝑖𝑛 𝑡ℎ𝑒 𝑖𝑛𝑡𝑎𝑐𝑡 𝑚𝑜𝑑𝑒𝑙

Stress distribution and maximal stress of implants Construct maximal stress (MS) was observed in the conjunction area between screw and cortical

RI PT

bone, and implant MS was higher in flexion than in extension in all loading conditions. Cephalad screws MS was lower than that of caudal screws at the same operative level. In flexion, evident

6). Force and moment transmitted by screws

NU

3.4.

SC

cephalocaudal increase of implants MS was observed in AL and CL, rather than in NC and FL (Fig. 5,

MA

A cross-section plane just under the cap of screws, which did not contact the bone or plate, was set to measure the force and moment transmitted through these screws. Generally, the moment transmitted

ED

through the screws was higher in flexion than in extension. Caudal screws transmitted more moment than cephalad screws. In flexion, a noticeable cephalocaudal increase of moment transmitted through

CE PT

cephalad and caudal screws was observed in AL and CL. The force transmitted through the screws was higher in extension than in flexion. In extension, the force transmitted showed a cephalocaudal increase

4.

AC

at different operative levels, except at C7-T1 (Fig. 7). Discussion

In previous in vitro biomechanical studies on cervical spine instrumentation, mostly single or two-level functional spine units (FSU) were used (Panjabi, 1988; Pitzen, et al., 2002; Snyder, et al., 2007; Horn, et al., 2009). It is assumed that mechanical behaviour of the FSU is equivalent to other FSUs in the same anatomical region of the spine. However, this assumption is flawed. A study showed that RoM and laxity zone parameters are significantly different between FSU and intact spine

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 14

specimens and suggested that flexion studies should be performed on multi-segment specimens to appropriately represent the anatomical boundary condition (Dickey and Kerr, 2003). RoM and laxity zone parameters are significantly different between FSU and intact spine specimens. To our knowledge,

RI PT

the current study is the first biomechanical study to compare ACDFP at different SCS levels using a full-length ligamentous FEM, and the level-specific response of SCS was also studied.

With current DFI FEMs, we found a notable cephalocaudal decrease of segmental stability

SC

provided by ACDFP, especially at C6-C7 and C7-T1. Segmental RoM and RMC are commonly used to

NU

evaluate the immediate postoperative stability provided by instruments. RoM and RMC increased

MA

significantly in C6-C7 and C7-T1 segments in AL and CL (Fig. 4), inferring that the stability provided by the ACDFP markedly decreases at these two levels. Moreover, the load transmitted through the

ED

anterior implants showed a similar cephalocaudal increase in AL and CL (Fig. 5, 6). These results agree with the outcomes of clinical studies where the incidence of ACDFP construct failure in C6-C7 and the

CE PT

level below was significantly higher than in the cephalad segments (Johnson, et al., 2004; Liu, et al., 2008; Koller, et al., 2009). These findings confirmed our hypothesis. Beyond the related operative

AC

techniques, the intrinsic mechanical factor is a major potential cause of implants failure in caudal segments, regarding PLC dysfunction. PLC acts as tension band in flexion motion and contributes to segmental stability. In PLC failure, the load during flexion motion is sustained completely by the anterior cervical spine locking plate. Intrinsic stability of caudal segments decreases more noticeably after DFI compared to the cephalad segments and the implants in the caudal segments also carry more load. Our predicted results, combined with previous clinical observation that the implant failure rate increases dramatically from C6 and the segment below, suggested that more rigid fixation, such as

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 15

combined posterior and anterior instrumentation, may be safer in preventing mechanical failure of implants used in DFI. Mechanisms of this phenomenon were earlier explored. Johnson et al. (2004) hypothesized that

RI PT

surgical exposure and ability to obtain intraoperative radiographs were limited at this lower level, particularly in large-sized patients. Wang et al. (2003) hypothesized that association of a fusion extending down to C7 with graft displacement is possibly related to cervical lordosis in conjunction

SC

with kyphotic angulation associated with the sagittal inclination at the cervicothoracic junction. This

NU

sharp angular change is probably associated with increased stress at the graft endplate interface, resulting in a higher probability of graft extrusion. The SCS caudal end has also been included in the

MA

idea of the cervicothoracic junction in some studies and viewed as a “structurally precarious” region

ED

with complex biomechanics (Ulrich, et al., 2001; Kreshak, et al., 2002; Post, et al., 2006; Ramieri, et al., 2011). However, the concept of the cervicothoracic junction itself is vague and difficult to be applied to

CE PT

clinical practice. Traditionally, the C7 to T3 segments are considered part of this junction. However, in this study, the mechanical characteristics notably changed from C6 downwards.

AC

Besides these hypotheses, the present study provided another possible mechanism: the in vivo biomechanical factor. Various methods of modelling physiologic loading have been reported in studies (Foley, et al., 1999; Miura, et al., 2002; DiAngelo and Foley, 2004; Wheeldon, et al., 2006). To identify specific factors contributing to this phenomenon, four loading conditions representing different scenarios were employed in the present study. NC loading (pure bending moment) represents the standard in vitro experimental protocol for the spine (Wilke, et al., 1998). FL is a condition adding a longitudinal preload to NC to mimic the physiologic compressive load generated by paravertebral

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 16

muscles, which will increase stability to the system without markedly affecting the segmental motion pattern (Miura, et al., 2002). However, applying a pure bending moment with or without follower load resulted in an approximately equal segmental contribution to the overall extension to flexion-extension

RI PT

motion path (Fig. 6, 7) (Bell, et al., 2015). Both NC and FL loading conditions are unable to replicate the in vivo segmental motion pattern. The most important difference of NC, FL with AL, and CL is the weight of the head. Taking the head’s weight into consideration in the loading procedures, AL and CL

SC

become more similar in in vivo biomechanical environment compared to NC and FL. It is reported that

NU

applying an eccentric axial compressive load as the weight of head replicated the segmental motion

MA

pattern of average in vivo segmental kinematics (DiAngelo and Foley, 2004). Recently, an in vitro study suggested that the combination of eccentric axial compressive load and follower load were

ED

supposed to have a synergistic effect, producing more physiologic kinematics (Bell, et al., 2015). In the present study, the AL and CL loading conditions show higher biofidelity in predicting SCS construct

CE PT

failure. The tendency of cephalocaudal increase was similar in both loading conditions. Compared with AL, CL achieved a higher degree of agreement with in vivo physiologic biomechanical environment of

AC

the cervical spine than AL, with a preload to stabilize the spine, suggesting that the eccentric axial compressive load produced by the head’s weight and paravertebral muscle force in flexion motion is the primary contributor to the phenomenon of level-dependent stability provided by ACDFP in treating DFI. The moment transmitted through the screws was higher in flexion than in extension and demonstrated a regular cephalocaudal increase under AL and CL loading conditions, consistent with the stress distribution pattern (Fig. 7). This implies that the lever arm is another contributor to this phenomenon. The lever arm of the lower segment is much longer than that of the upper segments. In

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 17

flexion, a cephalocaudal increase of lever arms at different segments of SCS generated different flexion moments produced by eccentric axial compressive load at different segments, which was reflected by the cephalocaudal increase of implants stress at injured segments.

RI PT

An interesting finding of present study is that maximal stress of the caudal screws was always higher than that of the cephalad segments in AL and CL, consistent with the construct failure pattern of ACDFP in previous clinical studies, where most constructs failed with loosening of the screws in the

SC

caudal vertebra and progressive anterior subluxation and kyphosis (Johnson, et al., 2004). This suggests

NU

that biomechanical factors majorly contribute to the failure of the screw-bone interface in such

MA

situations.

In current study, stress of screws was compared among different levels. In general, the magnitude

ED

of screws stress was higher in AL and CL, and showed an obvious cephalocaudal increase. The maximal stress peaked at the caudal screw of C7-T1 in flexion in AL, which is about 600 MPa.

CE PT

According to our previous experiments, the failure strength of titanium alloy (TC 4) is about 800-950 MPa. The implant is likely to fail instantly under 600 MPa stress. However, the stress is really high and

AC

most of our experiment supports that a stress over 500 MPa may dramatically decrease the use cycle and low-cycle fatigue cracks are more likely to happen. It must be cautious in interpreting these results because these predicted results were driven by the assumptions and necessary simplifications. For example, the screw-plate and screw-bone interface were fully constrained, the material of bone and implant was elastic and no failure point was set, and the loading used was also highly idealized and simplified. However, the results of FE study represent trends rather than precise values and we think the current FE results represent the cephalocaudal increase load of implants in AL and CL.

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 18

For the model validation, range of motion was the primary indicator in current study. Some FE model of spine validated the intradiscal pressure (IDP) as well. However, IDP was not validated in current model for the following reasons. First, there is not a well-accepted method for simulating

RI PT

nucleus. Second, most experimental data of IDP is based on lumbar spine, and there are little studies about the IDP of cervical spine (Oxland, 2016), most of which lack important data and are not suitable for model validation. Finally, in this study, we focused on the load of the implants and stability

SC

provided by the implants. All the intervertebral soft tissue was removed including nucleus. So, we think

NU

that the IDP may have very little influence on our predicted results. For the above reasons, we did not

MA

validate the model using IDP.

The current study has certain limitations. First, the spectrum of DFI varied because of the variable

ED

extent of injury-related soft tissue disruption and individual neck morphology. To focus on the effect of operative level on the segmental stability provided by ACDFP, we simulated stage 3 or 4 DFI by PLC

CE PT

dysfunction. Second, the loading conditions were highly idealized and could not precisely represent daily SCS motion. In vivo loading of the spine is complicated and varies from one individual to another

AC

with a strong dependence on the level of activity and particular movements. The loading condition design disregarded postoperative orthosis that could help restrict the motion of involved segments of the spine. None of the loadings in current study directly simulated muscle forces and thus were able to simulate the realistic loading conditions. A recent approach is to use firstly a musculoskeletal model to predict muscle forces and then apply them to passive FE model of lumbar spine to simulate realistic in vivo loading, which makes it easier to simulate the muscle forces (Azari, et al., 2017). Third, the material properties of bone and implant were assumed to be elastic and no failure point was set. Other

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 19

factors that could result in implant failure were neglected, such as stress shield, yield of bone, and interface of bone-screw. Fourth, the material properties of disc component used were based on the experiments of lumbar spine. We also tried to assign material properties of disc component based on

RI PT

cervical spine, but most of the experimental data of disc is based on lumbar. Maybe in the future, the material properties of cervical disc will be acquired and this problem should be solved. Thus, the present results should be driven by all the simplifications and assumptions mentioned above. However,

SC

the results of this FEM study represent trends rather than precise values because of necessary

NU

simplifications and assumptions concerning material properties, geometry of different tissues, applied

MA

loads, and contact behaviour. Therefore, we think current FE results represent the biomechanical influence of operative level on the corresponding immediate postoperative segmental stability and load

5.

Conclusions

ED

of implants.

CE PT

Generally, in the case of posterior ligamentous SCS dysfunction, operative level is an independent risk factor for construct failure, if ACDFP alone is used to reconstruct segmental stability. The stability

AC

provided by ACDFP decreases when the operative level shifts from the cephalad to caudal segments, especially at C6-C7 and C7-T1. The predicted results combined with previous clinical observation suggested that it may be safer to use more rigid fixation such as combined operation rather than ACDFP alone for these caudal segments. Based on the comparison of results acquired from four loading conditions, the biomechanical mechanism behind this phenomenon includes eccentric load from the combination of the head’s gravity and contraction force of the paravertebral muscles during the sagittal movements of the cervical spine, and difference of lever arms in cephalad and caudal

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 20

segments. 6.

Acknowledgement The authors gratefully acknowledge the support of Wu Chun-hui, Ph.D., and Elsevier Webshop

7.

RI PT

for language help. Funding Sources

This project was supported by the National Natural Science Foundation of China (Grant No.

NU

SC

81171718).

MA

References

Aebi M, Zuber K, Marchesi D. 1991. Treatment of cervical spine injuries with anterior plating. Indications,

ED

techniques, and results. Spine (Phila Pa 1976). 16(3 Suppl): S38-45. Allen BL, Ferguson RL, Lehmann TR, O'Brien RP. 1982. A mechanistic classification of closed, indirect fractures and dislocations of the lower cervical spine. Spine (Phila Pa 1976). 7(1): 1-27.

CE PT

Azari, F., Arjmand, N., Shirazi-Adl, A., Rahimi-Moghaddam, T. 2017. A combined passive and active musculoskeletal model study to estimate L4-L5 load sharing. J Biomech Bell KM, Yan Y, Debski RE, Sowa GA, Kang JD, Tashman S. 2015. Influence of varying compressive loading methods on physiologic motion patterns in the cervical spine. J Biomech. . Cusick, J.F., Yoganandan, N., Pintar, F., Gardon, M. 1996. Cervical spine injuries from high-velocity forces: a pathoanatomic and radiologic study. J Spinal Disord 9, 1-7.

AC

DiAngelo DJ, Foley KT. 2004. An improved biomechanical testing protocol for evaluating spinal arthroplasty and motion preservation devices in a multilevel human cadaveric cervical model. Neurosurg Focus. 17(3): E4. Dickey JP, Kerr DJ. Effect of specimen length: are the mechanics of individual motion segments comparable in functional spinal units and multisegment specimens. Med Eng Phys. 2003. 25(3): 221-7. Eismont FJ, Arena MJ, Green BA. 1991. Extrusion of an intervertebral disc associated with traumatic subluxation or dislocation of cervical facets. Case report. J Bone Joint Surg Am. 73(10): 1555-60. Emery SE, Bohlman HH, Bolesta MJ, Jones PK. 1998. Anterior cervical decompression and arthrodesis for the treatment of cervical spondylotic myelopathy. Two to seventeen-year follow-up. J Bone Joint Surg Am. 80(7): 941-51. Foley KT, DiAngelo DJ, Rampersaud YR, Vossel KA, Jansen TH. 1999. The in vitro effects of instrumentation on multilevel cervical strut-graft mechanics. Spine (Phila Pa 1976). 24(22): 2366-76.

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 21

Garvey TA, Eismont FJ, Roberti LJ. 1992. Anterior decompression, structural bone grafting, and Caspar plate stabilization for unstable cervical spine fractures and/or dislocations. Spine (Phila Pa 1976). 17(10 Suppl): S431-5. Goffin J, Plets C, Van den Bergh R. 1989. Anterior cervical fusion and osteosynthetic stabilization according to Caspar: a prospective study of 41 patients with fractures and/or dislocations of the cervical spine. Neurosurgery. 25(6): 865-71. Holzapfel GA, Schulze-Bauer CA, Feigl G, Regitnig P. 2005. Single lamellar mechanics of the human lumbar

RI PT

anulus fibrosus. Biomech Model Mechanobiol. 3(3): 125-40. Horn EM, Reyes PM, Baek S, et al. Biomechanics of C-7 transfacet screw fixation. J Neurosurg Spine. 2009. 11(3): 338-43.

Hussain M, Natarajan RN, Fayyazi AH, Braaksma BR, Andersson GB, An HS. 2009. Screw angulation affects bone-screw stresses and bone graft load sharing in anterior cervical corpectomy fusion with a rigid

SC

screw-plate construct: a finite element model study. Spine J. 9(12): 1016-23.

Johnson MG, Fisher CG, Boyd M, Pitzen T, Oxland TR, Dvorak MF. The radiographic failure of single segment

29(24): 2815-20.

NU

anterior cervical plate fixation in traumatic cervical flexion distraction injuries. Spine (Phila Pa 1976). 2004.

Koller H, Reynolds J, Zenner J, et al. 2009. Mid- to long-term outcome of instrumented anterior cervical fusion for subaxial injuries. Eur Spine J. 18(5): 630-53.

MA

Kreshak JL, Kim DH, Lindsey DP, Kam AC, Panjabi MM, Yerby SA. 2002. Posterior stabilization at the cervicothoracic junction: a biomechanical study. Spine (Phila Pa 1976). 27(24): 2763-70. Kumaresan, S., Yoganandan, N., Pintar, F.A. 1997. Methodology to quantify human cervical spine uncovertebral

ED

joint anatomy. Journalof Musculoskeletal Research 1, 131-139. Leahy, P.D., Puttlitz, C.M. 2012. The effects of ligamentous injury in the human lower cervical spine. J Biomech 45, 2668-2672.

CE PT

Lifeso RM, Colucci MA. 2000. Anterior fusion for rotationally unstable cervical spine fractures. Spine (Phila Pa 1976). 25(16): 2028-34.

Linder, A. 2000. A new mathematical neck model for a low-velocity rear-end impact dummy: evaluation of components influencing head kinematics. Accid Anal Prev 32, 261-269. Liu P, Zhao J, Liu F, Liu M, Fan W. 2008. A novel operative approach for the treatment of old distractive flexion injuries of subaxial cervical spine. Spine (Phila Pa 1976). 33(13): 1459-64.

AC

Maharaj, M.M., Mobbs, R.J., Hogan, J., Zhao, D.F., Rao, P.J., Phan, K. 2015. Anterior cervical disc arthroplasty (ACDA) versus anterior cervical discectomy and fusion (ACDF): a systematic review and meta-analysis. Journal of spine surgery (Hong Kong) 1, 72-85. Miura T, Panjabi MM, Cripton PA. 2002. A method to simulate in vivo cervical spine kinematics using in vitro compressive preload. Spine (Phila Pa 1976). 27(1): 43-8. Nightingale RW, Carol CV, Ottaviano D, et al. 2007. Flexion and extension structural properties and strengths for male cervical spine segments. J Biomech. 40(3): 535-42. Oxland, T.R. 2016. Fundamental biomechanics of the spine-What we have learned in the past 25 years and future directions. J Biomech 49, 817-832. Panjabi MM. 1988. Biomechanical evaluation of spinal fixation devices: I. A conceptual framework. Spine (Phila Pa 1976). 13(10): 1129-34.

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 22

Panzer MB, Cronin DS. 2009. C4-C5 segment finite element model development, validation, and load-sharing investigation. J Biomech. 42(4): 480-90. Patwardhan AG, Havey RM, Ghanayem AJ, et al. 2000. Load-carrying capacity of the human cervical spine in compression is increased under a follower load. Spine (Phila Pa 1976). 25(12): 1548-54. Pitzen TR, Matthis D, Barbier DD, Steudel WI. 2002. Initial stability of cervical spine fixation: predictive value of a finite element model. Technical note. J Neurosurg. 97(1 Suppl): 128-34. Post NH, Cooper PR, Frempong-Boadu AK, Costa ME. 2006. Unique features of herniated discs at the

RI PT

cervicothoracic junction: clinical presentation, imaging, operative management, and outcome after anterior decompressive operation in 10 patients. Neurosurgery. 58(3): 497-501; discussion 497-501. Ramieri A, Domenicucci M, Ciappetta P, Cellocco P, Raco A, Costanzo G. 2011. Spine surgery in neurological lesions of the cervicothoracic junction: multicentric experience on 33 consecutive cases. Eur Spine J. 20 Suppl 1: S13-9.

SC

Randle MJ, Wolf A, Levi L, et al. 1991. The use of anterior Caspar plate fixation in acute cervical spine injury. Surg Neurol. 36(3): 181-9.

Razack N, Green BA, Levi AD. 2000. The management of traumatic cervical bilateral facet fracture-dislocations

NU

with unicortical anterior plates. J Spinal Disord. 13(5): 374-81.

Reindl, R., Ouellet, J., Harvey, E.J., Berry, G., Arlet, V. 2006. Anterior reduction for cervical spine dislocation. Spine 31, 648-652.

MA

Ripa DR, Kowall MG, Meyer PR, Rusin JJ. 1991. Series of ninety-two traumatic cervical spine injuries stabilized with anterior ASIF plate fusion technique. Spine (Phila Pa 1976). 16(3 Suppl): S46-55. Robertson PA, Ryan MD. 1992. Neurological deterioration after reduction of cervical subluxation. Mechanical

ED

compression by disc tissue. J Bone Joint Surg Br. 74(2): 224-7. Saunders RL, Bernini PM, Shirreffs TG, Reeves AG. 1991. Central corpectomy for cervical spondylotic myelopathy: a consecutive series with long-term follow-up evaluation. J Neurosurg. 74(2): 163-70.

CE PT

Schmidt H, Heuer F, Simon U, et al. 2006. Application of a new calibration method for a three-dimensional finite element model of a human lumbar annulus fibrosus. Clin Biomech (Bristol, Avon). 21(4): 337-44. Shirazi-Adl A, Ahmed AM, Shrivastava SC. 1986. Mechanical response of a lumbar motion segment in axial torque alone and combined with compression. Spine (Phila Pa 1976). 11(9): 914-27. Snyder JT, Tzermiadianos MN, Ghanayem AJ, et al. 2007. Effect of uncovertebral joint excision on the motion response of the cervical spine after total disc replacement. Spine (Phila Pa 1976). 32(26): 2965-9.

AC

Ulrich C, Arand M, Nothwang J. 2001. Internal fixation on the lower cervical spine--biomechanics and clinical practice of procedures and implants. Eur Spine J. 10(2): 88-100. Urrutia J, Zamora T, Campos M, et al. 2016. A comparative agreement evaluation of two subaxial cervical spine injury classification systems: the AOSpine and the Allen and Ferguson schemes. Eur Spine J. 25(7): 2185-92. Viceconti, M., Olsen, S., Nolte, L.P., Burton, K. 2005. Extracting clinically relevant data from finite element simulations. Clin Biomech (Bristol, Avon) 20, 451-454. Villavicencio, A.T., Pushchak, E., Burneikiene, S., Thramann, J.J. 2007. The safety of instrumented outpatient anterior cervical discectomy and fusion. Spine J 7, 148-153. Wang JC, Hart RA, Emery SE, Bohlman HH. 2003. Graft migration or displacement after multilevel cervical corpectomy and strut grafting. Spine (Phila Pa 1976). 28(10): 1016-21; discussion 1021-2.

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 23

Wang Z, Zhao H, Liu JM, Tan LW, Liu P, Zhao JH. 2016. Resection or degeneration of uncovertebral joints altered the segmental kinematics and load-sharing pattern of subaxial cervical spine: A biomechanical investigation using a C2-T1 finite element model. J Biomech. 49(13): 2854-2862. Wheeldon, J.A., Stemper, B.D., Yoganandan, N., Pintar, F.A. 2008. Validation of a finite element model of the young normal lower cervical spine. Ann Biomed Eng 36, 1458-1469. Wheeldon JA, Pintar FA, Knowles S, Yoganandan N. 2006. Experimental flexion/extension data corridors for validation of finite element models of the young, normal cervical spine. J Biomech. 39(2): 375-80.

RI PT

Wilke HJ, Wenger K, Claes L. 1998. Testing criteria for spinal implants: recommendations for the standardization of in vitro stability testing of spinal implants. Eur Spine J. 7(2): 148-54.

Womack, W., Woldtvedt, D., Puttlitz, C.M. 2008. Lower cervical spine facet cartilage thickness mapping. Osteoarthr. Cartil. 16, 1018-1023.

Womack, W.J. Computational modeling of the lower cervical spine: Facet cartilage distribution and disc

SC

replacement ,2009.

Yoganandan N, Kumaresan S, Pintar FA. 2000. Geometric and mechanical properties of human cervical spine ligaments. J Biomech Eng. 122(6): 623-9.

NU

Yoganandan N, Kumaresan S, Pintar FA. 2001. Biomechanics of the cervical spine Part 2. Cervical spine soft tissue responses and biomechanical modeling. Clin Biomech (Bristol, Avon). 16(1): 1-27. Yoganandan N, Pintar FA, Stemper BD, Wolfla CE, Shender BS, Paskoff G. 2007. Level-dependent coronal and

MA

axial moment-rotation corridors of degeneration-free Am. 89(5): 1066-74.

cervical spines in lateral flexion. J Bone Joint Surg

Yoganandan N, Stemper BD, Pintar FA, Baisden JL, Shender BS, Paskoff G. 2008. Normative segment-specific

33(5): 490-6.

ED

axial and coronal angulation corridors of subaxial cervical column in axial rotation. Spine (Phila Pa 1976).

Yonenobu K, Fuji T, Ono K, Okada K, Yamamoto T, Harada N. 1985. Choice of surgical treatment for

AC

CE PT

multisegmental cervical spondylotic myelopathy. Spine (Phila Pa 1976). 10(8): 710-6.

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 24

Figure legend Fig. 1. Details of the FE model developed earlier Lordosis of C2-C7 model was 23.7°.

B.

Plate (height 21-24 mm, width 10 mm, thickness 2 mm) and screw (diameter 3.5 mm, length

RI PT

A.

14 mm) was rigidly fixed in the operative segment.

Six additional FEMs were developed according to the characteristics of DFI and surgery of

Schematic of four loading conditions implemented in present study

Fig. 3. Model validation

Response of each level under 2-Nm pure bending moment: RoM of flexion (A1 positive),

CE PT

A.

ED

B.

MA

ACDFP, including level of C2-C3 to C7-T1.

NU

A.

SC

Fig. 2. Simulation of DFI treated by ACDFP and loading conditions

extension (A1 negative), lateral bending and axial rotation (A2) Response of C4-C5 extracted model under 3.5-Nm pure bending moment: RoM and curve of

AC

B.

flexion (positive) and extension (negative) C.

Sagittal kinematics under 100-N follower load and combined 1-Nm or 2-Nm pure bending

moment: RoM of flexion-extension

Fig. 4. Sagittal kinematics changes RoM of the intact model under four loading conditions showed no obvious segment-specific

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 25

characteristics (A). Segmental RoM (B) and RMC (C) of the operative segment showed evidential increase in C6-C7 and C7-T1 under axial load and combined load.

RI PT

Fig. 5. Oblique view of von Mises stress distribution of the screws in flexion In flexion, the stress concentrated in the superior and inferior surfaces of the conjunction area between the screw and cortical bone. Evident cephalocaudal increase of stress was observed under AL

MA

Fig. 6. Maximal stress of screws and plates

NU

SC

and CL loading condition.

The maximal stress of cephalad screws (A), caudal screw (B) showed obvious cephlocaudal

ED

increase in flexion in AL and CL loading conditions. The maximal stress of the cephalad screws was

CE PT

higher than the caudal ones in flexion.

Fig. 7. Moment and force transmitted through the screws

AC

A Moment transmitted through screws; B Force transmitted through screws

ACCEPTED MANUSCRIPT

AC

CE PT

ED

MA

NU

SC

RI PT

Segment-specific stability of cervical spine. 26

ACCEPTED MANUSCRIPT

AC

CE PT

ED

MA

NU

SC

RI PT

Segment-specific stability of cervical spine. 27

ACCEPTED MANUSCRIPT

AC

CE PT

ED

MA

NU

SC

RI PT

Segment-specific stability of cervical spine. 28

ACCEPTED MANUSCRIPT

AC

CE PT

ED

MA

NU

SC

RI PT

Segment-specific stability of cervical spine. 29

ACCEPTED MANUSCRIPT

AC

CE PT

ED

MA

NU

SC

RI PT

Segment-specific stability of cervical spine. 30

ACCEPTED MANUSCRIPT

AC

CE PT

ED

MA

NU

SC

RI PT

Segment-specific stability of cervical spine. 31

ACCEPTED MANUSCRIPT

AC

CE PT

ED

MA

NU

SC

RI PT

Segment-specific stability of cervical spine. 32

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 33

Table 1. Material properties of the model components Name

Element Type

Material Model

Material

Reference

properties Cortical bone

C3D4

ISO elastic

E=12000

MPa

μ=0.3 C3D4

ISO elastic

2008

RI PT

Cancellous bone

Wheeldon et al.

E=300 MPa μ=0.3

Wheeldon et al. 2008

Cartilaginous

C3D8

ISO elastic ISO elastic

C3D8H ground

Hyper-elastic

C3D8H

Hyper-elastic

AC

CE PT

SpringA

ED

MA

substance

Annulus fiber

Instrumentations

E=23.8

SpringA

3-D solid

Schmidt et al. 2006

MPa

Schmidt et al.

μ=0.3

2006

C10=0.12,

Schmidt et al.

C01=0.09

2006

C10 = -0.075 MPa

Panzer

NU

Nucleus

C3D8

SC

Cartilage of joint

Ligaments

MPa

μ=0.3

end-plate

Annulus

E=23.8

and

Cronin, 2009

C01 = 0.122 MPa C20 = -0.294 MPa C11 = 0.689 MPa C02 = 0.122 MPa

NON-linear

Stress-strain

Shirazi-Adl et al.

Spring

curve

1986 Holzapfel et al. 2005 Schmidt et al. 2006

NON-linear

Force-deflection

Yoganandan

Spring

curve

al. 2000,2001

ISO elastic

E=110000 μ=0.3

MPa

et

Hussain et al., 2009

ACCEPTED MANUSCRIPT Segment-specific stability of cervical spine. 34

Highlights In distractive flexion injury, operative level may cause anterior fixation failure



However, no biomechanical study investigating this possibility has been reported



A validated three-dimensional C2-T1 model was used to test the hypothesis



Segmental stability decreased cephalocaudally, especially at C6-C7 and C7-T1



Anterior fixation seems unsuitable for C6-T1 in this context

AC

CE PT

ED

MA

NU

SC

RI PT