Biomechanical analysis of lateral interbody fusion strategies for adjacent segment degeneration in the lumbar spine

Accepted Manuscript Title: Biomechanical analysis of lateral interbody fusion strategies for adjacent segment degeneration in the lumbar spine Author: Melodie F. Metzger, Samuel T. Robinson, Ruben C. Maldonado, Jeremy Rawlinson, John Liu, Frank L. Acosta PII: DOI: Reference:

S1529-9430(17)30092-X http://dx.doi.org/doi: 10.1016/j.spinee.2017.03.005 SPINEE 57259

To appear in:

The Spine Journal

Received date: Revised date: Accepted date:

5-10-2016 21-2-2017 15-3-2017

Please cite this article as: Melodie F. Metzger, Samuel T. Robinson, Ruben C. Maldonado, Jeremy Rawlinson, John Liu, Frank L. Acosta, Biomechanical analysis of lateral interbody fusion strategies for adjacent segment degeneration in the lumbar spine, The Spine Journal (2017), http://dx.doi.org/doi: 10.1016/j.spinee.2017.03.005. 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.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

Biomechanical Analysis of Lateral Interbody Fusion Strategies for Adjacent Segment Degeneration in the Lumbar Spine Melodie F. Metzger, PhD1; Samuel T. Robinson, BS1; Ruben C. Maldonado1; Jeremy Rawlinson, PhD2; John Liu, MD3; Frank L. Acosta, MD3 1. Orthopedic Biomechanics Laboratory, Department of Orthopedic Surgery, Cedars-Sinai Medical Center, Los Angeles, CA 2. Spinal Applied Research, Medtronic, Memphis, TN 3. Department of Neurological Surgery, Keck School of Medicine, University of Southern California, Los Angeles, CA

Corresponding Author Contact Information: Melodie Metzger, Ph.D. Orthopedic Biomechanics Laboratory Cedars-Sinai Medical Center 8700 Beverly Blvd., DAV6006 Los Angeles, CA90048 Email: [email protected] Phone: 310-423-7765; Fax: 310-248-8016

Abstract

29

Background context. Surgical treatment of symptomatic adjacent segment disease (ASD)

30

typically involves extension of previous instrumentation to include the newly-affected level(s).

31

Disruption of the incision-site can present challenges and increases the risk of complication.

32

Lateral-based interbody fusion techniques may provide a viable surgical alternative that avoids

33

these risks. This study is the first to analyze the biomechanical effect of adding a lateral-based

34

construct to an existing fusion.

1 Page 1 of 17

1

Purpose.

To determine whether a minimally-invasive lateral interbody device, with and

2

without supplemental instrumentation, can effectively stabilize the rostral segment adjacent to

3

a two-level fusion when compared to a traditional posterior revision approach.

4

Study Design/Setting. A cadaveric biomechanical study of lateral-based interbody strategies as

5

add-on techniques to an existing fusion for the treatment of ASD.

6

Methods. Twelve lumbosacral specimens were non-destructively loaded in flexion, extension,

7

lateral bending, and torsion. Sequentially, the tested conditions were: intact, two-level TLIF

8

(L3-L5), followed by LLIF (lateral lumbar interbody fusion) procedures at L2-L3 including:

9

interbody alone, a supplemental lateral plate, a supplemental spinous process plate, and then

10

either cortical screw or pedicle screw fixation. A three-level TLIF was the final instrumented

11

condition. In all conditions, three-dimensional kinematics were tracked and range of motion

12

(ROM) calculated for comparisons. Institutional funds (<$50K) in support of this work were

13

provided by Medtronic Spine.

14

Results.

15

significantly reduced motion in flexion, extension, and lateral bending (p<0.05). Supplementing

16

with a lateral plate further reduced ROM during lateral bending and torsion, whereas a spinous

17

process plate further reduced ROM during flexion and extension. The addition of posterior

18

cortical screws provided the most stable LLIF construct, demonstrating ROM comparable to a

19

traditional 3-level TLIF.

20

Conclusion. The data presented suggests a lateral-based interbody fusion supplemented with

21

additional minimally-invasive instrumentation may provide comparable stability to a traditional

22

posterior revision approach without removal of the existing two-level rod in an ASD revision

23

scenario.

The addition of a lateral interbody device super-adjacent to a 2-level fusion

24 25 26 27

Keywords: minimally-invasive, lateral interbody fusion, LLIF, adjacent segment disease, lumbar, spine, degeneration, biomechanics, stability, range of motion, supplemental fixation

28 29 30 2 Page 2 of 17

1

INTRODUCTION

2 3

As pedicle screw-rod instrumentation becomes more widespread, spine surgeons are

4

inevitably faced with a growing number of patients presenting with symptomatic adjacent

5

segment degeneration (ASD) [1-6]. The surgical treatment of ASD, in most cases, requires

6

extension of the lumbar fusion to include the newly-degenerated segment. Traditionally,

7

techniques to accomplish this extension include a posterior approach utilizing pedicle screws

8

with complete replacement of the existing posterior rod, with or without supplemental

9

interbody fusion, and usually with a transforaminal lumbar interbody fusion (TLIF) procedure [1,

10

7]. The minimally-invasive lateral lumbar approach, on the other hand, may be particularly well

11

suited for treatment of ASD with less morbidity than either transforaminal or anterior lumbar

12

interbody approaches [8, 9].

13 14

Few biomechanical studies exist that examine the stability of lateral constructs with

15

supplemental fixation [10-12]. To date, no study has analyzed the biomechanical effect of

16

lateral-based constructs on levels adjacent to an existing fusion, as would be the case when

17

treating symptomatic ASD. Therefore, the purpose of the following study was to determine the

18

mechanical parameters of several lateral-based constructs for adjacent segment degeneration

19

at L2-L3 after a two-level posterior lumbar fusion from L3-L5 and compare them to a traditional

20

three-level rod extension. It was hypothesized that a stand-alone lateral lumbar interbody

21

fusion (LLIF) graft alone would not provide adequate stability when proximally added to an

22

existing 2-level fusion, but when supplemented with additional minimally-invasive 3 Page 3 of 17

1

instrumentation it would have comparable stability to a three-level TLIF with pedicle screw-rod

2

fixation.

3 4

MATERIAL AND METHODS

5 6

Specimen Preparation

7 8

Fresh-frozen cadaveric lumbar spines (L1-S1) were procured from an approved tissue bank and

9

stored at -30oC. Specimens were screened radiographically for any visual flaws and to ensure

10

adequate disc height. Afterwards, bone mineral density (BMD) was quantified using dual-

11

energy x-ray absorptiometry (DXA) scans. BMD and t-scores were averaged over the L1-L4

12

vertebrae and a minimum cut-off value of t=-2.5 was used to rule out any severely osteoporotic

13

spines.

14 15

Prior to testing, each specimen was thoroughly cleaned of nonstructural soft tissue while

16

preserving the disc, facet joint capsules, and ligamentous structures. The cranial (L1) and

17

caudal (S1) vertebrae were potted to approximately half-axial height in cylindrical metal cups

18

using two-part epoxy resin (BJB Enterprises, Tustin, CA), taking care to maintain the natural

19

curvature and sacral inclination of each spine. Wood screws provided additional anchorage of

20

the vertebral bodies in the potting material. Specimens were kept moist using saline-soaked

21

gauze throughout testing.

22 4 Page 4 of 17

1

Biomechanical Testing

2 3

Biomechanical testing was conducted on a MTS Bionix Testing System augmented with a biaxial

4

Spine Simulator to impose bending rotations (MTS Bionix 370.02, MTS Systems Corp, Eden

5

Prairie, MN). Forces and moments were measured using an ATI Mini45 load cell (ATI Industrial

6

Automation, Apex, NC) located directly below the caudal end of the specimen. After specimens

7

were rigidly secured onto the testing machine, preliminary forces and torques were offset

8

manually by adjusting the rotation angles or translational position of the caudal vertebrae,

9

allowing the test to begin from a neutral posture. All tests were performed on the same day to

10

minimize tissue degradation.

11 12

Specimens were non-destructively loaded in extension, flexion, left and right lateral bending,

13

and left and right torsion. After three preconditioning cycles at 80%, a maximum bending

14

moment of 7.5 N-m was applied at a rate of 0.2Nm/s and held for 20 seconds to allow for creep

15

deformation. Data was collected during this 20 second- hold and averaged.

16 17

Three-dimensional vertebral kinematic response was measured via an optical infrared camera

18

system (Optotrak Certus, Northern Digital Inc., Waterloo Canada). Infrared emission triads

19

were mounted on all vertebral bodies and relative motion was tracked to a predefined

20

anatomical set of axes with an accuracy of 0.1mm.

21

5 Page 5 of 17

1

Each specimen was first tested in the intact, un-instrumented condition. Afterwards, all

2

instrumentation was performed by a trained spinal surgeon (FLA) according to accepted

3

surgical techniques. A 2-level fusion was instrumented from L3 to L5 using TLIF interbody cages

4

(CAPSTONE® PEEK Spinal System, Medtronic Spinal and Biologics, Memphis, TN) and bilateral

5

pedicle screw fixation. Specimens were then retested with this 2-level ‘base’ fusion.

6

Subsequent instrumentation and testing was conducted with a LLIF procedure and 'add-on'

7

constructs at the supra-adjacent (L2-L3) level, Figure 1. The interbody used with the LLIF

8

approach was defined as the LLIF cage in this study. These constructs included (1) stand-alone

9

LLIF (CLYDESDALE® PEEK Spinal System) cage, (2) LLIF cage supplemented with a spinous

10

process plate (CD HORIZON SPIRE™ Spinal System), (3) LLIF cage with a lateral plate (CD

11

HORIZON ENGAGE™ Direct Lateral Plate System), (4) LLIF cage with a short bilateral cortical

12

screw-rod construct (CD HORIZON® SOLERA™ Spinal System), and (5) LLIF cage with a short

13

bilateral pedicle screw-rod construct (CD HORIZON® SOLERA™ Spinal System). For the last two

14

LLIF constructs, specimens were separated into two group (A) cortical screws or (B) pedicle

15

screws, since these two subgroups could not be tested sequentially. The lateral access (cage

16

insertion and plate) was on the left for all specimens. For the final test, specimens were

17

instrumented from L2 to L5 with a 3-Level TLIF with bilateral pedicle screw fixation to replicate

18

the standard ASD revision by extension of the lumbar fusion to all three levels.

19 20

Data Analysis

21

6 Page 6 of 17

1

Relative vertebral motion between all vertebrae was converted into 3-2-1 Euler angles and

2

subsequently translated to range of motion (ROM) data in the sagittal, coronal, and transverse

3

planes relative to the initial neutral position using MATLAB software (vR2009b, The MathWorks

4

Inc., Natick, MA). ROM at L2-L3 was normalized to the intact condition (% intact) to control for

5

differences in flexibility among specimens. The percent of total ROM was also calculated (%

6

total) for each segment to determine any gross alterations in mobility beyond what is natural

7

for each segment (L2-L3 ROM/total specimen ROM). Average ROM and percent total values

8

were compared to the corresponding intact ROM and TLIF ROM values using repeated

9

measures analysis of variance (ANOVA) with a Bonferroni correction for multiple comparisons

10

(SAS 9.2, Cary, NC). The percent intact was also compared to the TLIF data using ANOVA.

11

Tukey-Kramer was used for post-test pairwise comparisons. Repeated measures ANOVA was

12

applied separately for the two screw-rod construct subgroups (A and B). The level for statistical

13

significance was set at p<0.05.

14 15

RESULTS

16 17

A total of 12 specimens, 6 males and 6 females, were included in the study with a mean age of

18

58 years (range, 34-69). The average BMD for all 12 specimens was 1.10 g/cm2 (range, 0.92–

19

1.57) and the average t‐score was ‐0.83 (range, -2.43–2.98) as determined by DXA. Male and

20

female spines were evenly distributed between the cortical and pedicle screw subgroups and

21

there were no significant differences in bone quality parameters (BMD or t-score) between the

22

two subgroups. 7 Page 7 of 17

1 2

Average intact total (L1-S1) specimen range of motion was significantly reduced in all bending

3

modes after the insertion of a 2-level base fusion from L3 to L5, Figure 3. Compared to the

4

conventional 3-level TLIF, total specimen ROM was greater during lateral bending when a

5

stand-alone LLIF was added to the rostral-adjacent segment (L2-L3) of the base fusion and

6

when supplemented with a lateral plate, spinous process plate, or cortical screws (only right

7

bending was significant), p<0.05. When the LLIF interbody was supplemented with pedicle

8

screws-rods overall specimen ROM in extension and right torsional bending was significantly

9

lower than ROM with the 3-level TLIF.

10 11

Average ROM at the add-on level (L2-L3) for all LLIF treatment options were compared to both

12

the intact L2-L3 disc and the L2-L3 segment when included in the standard 3-level TLIF

13

procedure, Figure 4. During extension, the stand-alone LLIF reduced ROM at L2-L3 from an

14

average intact value of 2.5 to 1.6 degrees (p=0.06). The addition of supplemental

15

instrumentation further reduced extension ROM to 1.3o with the lateral plate, 0.9o with the

16

spinous plate, 0.5o with pedicle screws-rods, and 0.4o with cortical screws-rods. These last three

17

add-on options had comparable rigidity in extension to the 3-level TLIF (0.6o). The LLIF alone

18

significantly reduced flexion ROM from the intact average of 5.6o to 2.0o (p<0.001). When the

19

lateral plate was added, flexion ROM remained similar to the LLIF cage alone, but was further

20

reduced with the spinous plate (1.6o, p<0.001), pedicle screws-rods (0.6o, p<0.05), and bilateral

21

cortical screws-rods (1.3o, p<0.001). During lateral bending, the TLIF and all of the LLIF add-on

22

constructs demonstrated a significant reduction in motion at L2-L3 compared to intact values 8 Page 8 of 17

1

(p<0.05). Intact left and right axial rotation at L2-L3 was less than 2o and was significantly

2

reduced when the LLIF was supplemented with either a lateral plate or pedicle screws-rods,

3

both of which were also significantly less than the ROM at L2-L3 when included in the standard

4

3-level TLIF construct. Torsional range of motion at L2-L3 with the LLIF and added cortical

5

screws-rods was also significantly less than intact ROM at L2-L3, but was not significantly

6

different than ROM at L2-L3 with the 3-level TLIF.

7 8

Range of motion data at L2-L3 was normalized to the intact state and the LLIF treatment

9

options were then compared to the 3-level TLIF, Figure 5. In flexion and extension, TLIF ROM

10

was approximately 20% of intact ROM at L2-L3, which was significantly lower than the 2-level

11

base fusion, LLIF cage alone, and LLIF with lateral plate, p<0.05. During lateral bending L2-L3

12

percent of intact ROM was significantly reduced with the 3-level TLIF, which was significantly

13

lower than that determined for the 2-level fusion, LLIF cage alone, LLIF with lateral plate, and

14

LLIF with spinous plate, p<0.05. During right and left axial rotation the TLIF was about 60% of

15

intact ROM which was statistically lower than the percent of intact for the 2-level base fusion

16

and statistically higher than the normalized ROM at L2-L3 with the lateral plate (left torsion

17

only).

18 19

Percent of total ROM at L2-L3 was significantly increased from intact values after

20

instrumentation of L3-L5 with a 2-level base fusion in all bending planes (p<0.05), Figure 6. The

21

LLIF alone was not statistically different from intact during extension and torsion, but

22

demonstrated a significantly lower percent of total during flexion and lateral bending, p<0.05. 9 Page 9 of 17

1

The lateral plate reduced percent of total from that of the intact segment during flexion, left

2

bending, and left torsional rotation (p<0.05). The addition of the spinous plate only had a

3

significant reduction in percent of total ROM compared to intact during flexion and extension.

4

The LLIF with supplemental pedicle screws-rods and cortical screws-rods, as well as the

5

traditional 3-level TLIF, all had a significantly reduced percent of total ROM compared to intact

6

in all bending modes except torsion, p<0.05.

7 8

DISCUSSION

9 10

This study represents the first mechanical analysis of lateral-based interbody strategies as an

11

add-on technique to existing fusions for the treatment of adjacent segment disease. While

12

these results are only capable of describing immediate postoperative stability, they provide

13

preliminary evidence that less-invasive LLIF procedures may effectively stabilize the level supra-

14

adjacent to a two-level fusion when used in conjunction with supplemental instrumentation.

15 16

When implanted proximal to an existing fusion, the LLIF interbody alone increased stability,

17

particularly in the anterior column, demonstrated by a significant reduction in motion at L2-L3

18

compared to the un-instrumented disc during flexion. Adding a lateral plate effectively reduced

19

motion in the coronal plane, whereas the addition of a spinous process plate helped minimize

20

sagittal plane motion, as expected based on the anatomical placement and design of each

21

respective plate. Adding a short cortical segment to the LLIF at the proximal add-on level

10 Page 10 of 17

1

appeared to be the most effective minimally-invasive add-on technique included in this study

2

for all bending planes with the exception of torsional motion.

3 4

Previous biomechanical studies investigating lateral cages further strengthen our results [10,

5

12-14]. Laws et al. compared Anterior Lumbar Interbody Fusion (ALIF) and LLIF cages both with

6

and without supplemental instrumentation and, similar to our results at L2-L3, reported a

7

significant decrease in motion with stand-alone LLIF cages in flexion, extension and lateral

8

bending when compared to the intact disc [12]. They reported no significant reduction in

9

motion in any plane with a stand-alone ALIF. Similarly, when they added a lateral plate to the

10

LLIF it helped reduced motion, primarily in torsion and lateral bending, while adding bilateral

11

pedicle screws resulted in the greatest level of stabilization. Fogel et al. also showed significant

12

reductions from intact with a stand-alone lateral cage that was further reduced when

13

supplemented with a lateral plate, a spinous process plate, a combination of both a lateral and

14

a spinous process plate, ipsilateral pedicle screw, and bilateral pedicle screws [10].

15

Interestingly, when both plates were added to the lateral cage it provided comparable rigidity

16

to a cage with bilateral pedicle screws. While this two-plate configuration was not tested in the

17

present study, it provides another potential less-invasive add-on technique.

18 19

Clinically, posterior extension surgery remains the gold standard treatment for ASD despite the

20

increased risk of complication associated with disrupting previously formed scar tissue, most

21

likely because few alternatives exist. Recently, two case studies have surfaced suggesting the

22

minimally-invasive lateral approach may be a viable alternative that avoids these risks by using 11 Page 11 of 17

1

a different route to access the spine [8, 9]. Both of these studies used stand-alone lateral

2

interbodies with no additional instrumentation, reasoning the larger surface area of the cage

3

combined with preservation of the anterior and posterior ligaments provide adequate stability.

4

While their results demonstrated successful rates of arthrodesis, further clinical studies with

5

sufficient sample sizes and longer follow-up periods are warranted.

6 7

One point that arises when discussing fusion instrumentation is the definition of ‘adequate

8

stability’. While the FDA defines successful fusion as radiographic ROM of less than 5o, studies

9

indicate that ROM greater than 2o is associated with nonunion[15, 16]. Clinically, adequate

10

stability is that which allows the formation of a solid bony union. Since biomechanical studies

11

are unable to achieve this outcome, we chose to compare ROM to the clinical gold standard of

12

fusion. LLIF techniques that provided stability similar to that of a TLIF with bilateral pedicle

13

screw-rod fixation were considered adequately stable. It is possible that other techniques

14

investigated in this study may provide enough load sharing to enable bony union, but further

15

clinical studies are warranted.

16 17

By investigating multiple add-on instrumentation techniques, this study provides important

18

comparative data across various treatment options for supra-adjacent ASD. Nonetheless, there

19

are several limitations that should be noted. First, our data only represents the immediate

20

postoperative response and cannot account for any long-term effects on stability such as device

21

settling, fatigue, and bony in-growth. Repeated testing methods were implemented, increasing

22

the potential for accumulative wear and tear. Although, quasi-static testing techniques, 12 Page 12 of 17

1

physiologic non-destructive loading, and careful observation for any increase in ROM caused by

2

tissue fatigue likely minimized this effect. Another limitation was the reduction in sample size

3

that occurred when specimens were divided into two subgroups (A/B, n=6 each) for the LLIF

4

with supplemental cortical and pedicle screw add-on techniques. This reduction resulted in

5

lower statistical power between subgroups, although clear and verifiable trends were apparent

6

and the subgroup sample size remained comparable to other cadaveric biomechanical studies

7

[17, 18]. Lastly, our study only investigated the biomechanics of supra-adjacent segment

8

disease and cannot be confidently translated to levels below an existing fusion. We recognize

9

ASD also occurs inferiorly and recommend further research studies into alternative treatment

10

options for ASD below an existing fusion.

11 12

In conclusion, LLIF supplemented with either a lateral plate or cortical screws-rods posteriorly

13

may provide comparable stability without removal of the existing two-level rod in a revision

14

scenario. Further clinical research is necessary to determine the efficacy and safety of this

15

procedure, and to assess other potential advantages such as reduced surgical time, blood loss,

16

and recovery time.

17 18

Acknowledgements: The authors would like to thank Lea Kanim for her assistance with the

19

statistical analysis.

20 21

13 Page 13 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

REFERENCES 1. Celestre PC, Montgomery SR, Kupperman AI, Aghdasi B, Inoue H, Wang JC. Lumbar clinical adjacent segment pathology: predilection for proximal levels. Spine (Phila Pa 1976). 2014;39(2):172-6. 2. Sears WR, Sergides IG, Kazemi N, Smith M, White GJ, Osburg B. Incidence and prevalence of surgery at segments adjacent to a previous posterior lumbar arthrodesis. Spine J. 2011;11(1):11-20. 3. Campbell PG, Yadla S, Malone J, et al. Complications related to instrumentation in spine surgery: a prospective analysis. Neurosurg Focus. 2011;31(4):E10. 4. Lee CS, Hwang CJ, Lee SW, et al. Risk factors for adjacent segment disease after lumbar fusion. Eur Spine J. 2009;18(11):1637-43. 5. Imagama S, Kawakami N, Kanemura T, et al. Radiographic Adjacent Segment Degeneration at Five Years After L4/5 Posterior Lumbar Interbody Fusion With Pedicle Screw Instrumentation: Evaluation by Computed Tomography and Annual Screening With Magnetic Resonance Imaging. J Spinal Disord Tech. 2013. 6. Xia XP, Chen HL, Cheng HB. Prevalence of adjacent segment degeneration after spine surgery: a systematic review and meta-analysis. Spine (Phila Pa 1976). 2013;38(7):597-608. 7. Choi KC, Kim JS, Shim HK, Ahn Y, Lee SH. Changes in the adjacent segment 10 years after anterior lumbar interbody fusion for low-grade isthmic spondylolisthesis. Clin Orthop Relat Res. 2014;472(6):1845-54. 8. Wang MY, Vasudevan R, Mindea SA. Minimally invasive lateral interbody fusion for the treatment of rostral adjacent-segment lumbar degenerative stenosis without supplemental pedicle screw fixation. J Neurosurg Spine. 2014;21(6):861-6. 9. Palejwala SK, Sheen WA, Walter CM, Dunn JH, Baaj AA. Minimally invasive lateral transpsoas interbody fusion using a stand-alone construct for the treatment of adjacent segment disease of the lumbar spine: review of the literature and report of three cases. Clin Neurol Neurosurg. 2014;124:90-6. 10. Fogel GR, Parikh RD, Ryu SI, Turner AW. Biomechanics of lateral lumbar interbody fusion constructs with lateral and posterior plate fixation: laboratory investigation. J Neurosurg Spine. 2014;20(3):291-7. 11. Mantell M, Cyriac M, Haines CM, Gudipally M, O'Brien JR. Biomechanical analysis of an expandable lateral cage and a static transforaminal lumbar interbody fusion cage with posterior instrumentation in an in vitro spondylolisthesis model. J Neurosurg Spine. 2015:1-7. 12. Laws CJ, Coughlin DG, Lotz JC, Serhan HA, Hu SS. Direct lateral approach to lumbar fusion is a biomechanically equivalent alternative to the anterior approach: an in vitro study. Spine (Phila Pa 1976). 2012;37(10):819-25. 13. Cappuccino A, Cornwall GB, Turner AW, et al. Biomechanical analysis and review of lateral lumbar fusion constructs. Spine (Phila Pa 1976). 2010;35(26 Suppl):S361-7. 14. Perez-Orribo L, Kalb S, Reyes PM, Chang SW, Crawford NR. Biomechanics of lumbar cortical screw-rod fixation versus pedicle screw-rod fixation with and without interbody support. Spine (Phila Pa 1976). 2013;38(8):635-41. 15. FDA. Guidance Document for the Preparation of IDEs for Spinal Systems. Office of Device Evaluation: Orthopedic Devices Branch Division of General and Restorative Devices; 2000; Available from:

14 Page 14 of 17

1 2 3 4 5 6 7 8 9 10

http://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocu ments/ucm073772.pdf. 16. Santos ER, Goss DG, Morcom RK, Fraser RD. Radiologic assessment of interbody fusion using carbon fiber cages. Spine (Phila Pa 1976). 2003;28(10):997-1001. 17. Wheeler DJ, Freeman AL, Ellingson AM, et al. Inter-laboratory variability in in vitro spinal segment flexibility testing. J Biomech. 2011;44(13):2383-7. 18. Wilke HJ, Rohlmann A, Neller S, et al. Is it possible to simulate physiologic loading conditions by applying pure moments? A comparison of in vivo and in vitro load components in an internal fixator. Spine (Phila Pa 1976). 2001;26(6):636-42.

11

15 Page 15 of 17

1

Figure 1. Chart outlining the order in which testing was performed. Specimens were

2

repeatedly tested seven times. For the sixth test, specimens were evenly divided into

3

two groups to investigate the use of a LLIF interbody cage with either a short bilateral

4

cortical or pedicle screw-rod construct.

5 6

Figure 2. Graphic representation of each instrumented condition from the base 2-level TLIF

7

with LLIF alone to the lateral construct and supplemental fixation. The initial intact condition is

8

not shown and the final construct is the TLIF with full rod extension to the adjacent segment.

9 10

Figure 3. Average total specimen ROM (L1-S1). Intact specimens had significantly greater

11

range of motion compared to all instrumented cases (p<0.05). * Indicates statistical differences

12

between standard 3-level TLIF and other constructs (p<0.05).

13 14

Figure 4. Graphs showing range of motion at L2-L3 for each bending mode as mean 

15

standard deviation. *indicates statistical difference from TLIF, ** indicates statistical

16

difference from intact (p<0.05)

17 18 19 20 21 22 23

Base = LLIF = L.P. = S.P.P. = C.S. = TLIF =

2-level TLIF from L3-L5 Base + LLIF interbody alone at L2-L3 Base + LLIF interbody and lateral plate at L2-L3 Base + LLIF interbody and spinous process plate at L2-L3 P.S.= Base + LLIF interbody and pedicle screws at L2-L3 Base + LLIF interbody and cortical screws at L2-L3 3-level TLIF from L2-L5.

24

16 Page 16 of 17

1

Figure 5. ROM, as a percent of intact, at L2-L3. * indicates a statistically significant

2

difference when compared to TLIF ROM (p<0.05).

3 4

Figure 6. Motion at L2-L3 as the percent of total motion for all spines tested. *indicates

5

statistically significant differences when compared to 3-Level TLIF ROM (p<0.05).

6 7

17 Page 17 of 17