Nucleus Arthroplasty Volume II

Nucleus
Arthroplasty
Technology
in Spinal Care
Volume II: Biomechanics &

Development
Table of Contents
2
Introduction
Deputy Editorial Board
CHAPTER 7
Biomechanics of the Degenerated Disc
CHAPTER 8
11
Principles and Mechanical Requirements of Nucleus Implants
CHAPTER 9
17
Kinematic Demands of Nucleus Arthroplasty Technology
CHAPTER 10
24
Device Stiffness vs. Load-Sharing with Nucleus Arthroplasty Devices
CHAPTER 11
34
Endplate Mechanics
CHAPTER 12
41
Repair of the Anulus Fibrosus of the Lumbar Disc
CHAPTER 13
49
Integration of Nucleus Arthroplasty Technology into the Continuum of Care
IBC
Conclusion
his monograph series is a groundbreaking project in the

rapidly emerging field of non-fusion spinal surgery. The
full range of nucleus replacement technologies is examined
with discussion on biomechanical and physiological properties of the disc, detailed information on each cutting-edge
device technology, indications, and patient selection criteria.
Nucleus Arthroplasty Technology in Spinal Care is
published for the medical profession by Raymedica, LLC,
Minneapolis, MN 55431.
The views expressed in this series are those of the authors
and do not necessarily represent those of Raymedica, LLC.
ACKNOWLEDGEMENT
We, Raymedica, LLC, and the authors of this volume, wish
to acknowledge our debt of gratitude for the important
contribution of Steven J. Seme, Developmental Editor. His
guidance has added a great deal to the teaching value of
this volume.
Copyright 2006 and 2007 Raymedica, LLC. All rights
reserved. Printed in the U.S.A.
www.nucleusarthroplasty.com
Introduction
Reginald J. Davis, MD, FACS
Frank P. Cammisa, Jr., MD, FACS
CHIEF OF NEUROSURGERY
ASSOCIATE PROFESSOR OF CLINICAL SURGERY
Baltimore Neurosurgical Associates, PA

Baltimore, MD 21204
Hospital for Special Surgery

New York, NY 10021
Federico P. Girardi, MD
William C. Hutton, DSc
ASSISTANT PROFESSOR
PROFESSOR AND DIRECTOR OF
OF ORTHOPEDIC SURGERY
ORTHOPEDIC RESEARCH

New York, NY 10021
Emory University
Atlanta, GA 30322
he first documented works describing the diagnosis and

treatment of the spine, spinal disorders, and spinal instability
date back to 1900-2500 B.C. Interestingly, the documents recommended against the treatment of spinal cord injury. The development of therapeutic treatments has a long history starting with
the cane, the first load-sharing device. Today, our efforts to
improve therapies to treat spine disease persist. We continue to
recognize problems, identify issues, and define variables in an
effort to better understand spinal degeneration and to develop
innovative solutions that utilize a wide array of materials and
technologies. Our field has had a rich history of advancements,
accomplishments, and inventiveness. We owe a great debt to the
pioneers who, armed with little more than a detailed knowledge
of anatomy, heralded in the era of spinal surgery. Their trials,
errors, innovations, and teachings have guided our efforts to
ultimately improve clinical outcomes.
Early on, it was recognized that the disc played a vital role in overall
spine health. With great effort and ingenuity, the unique anatomical,
biomechanical, and physiological properties of the disc were elucidated and incorporated into elegant treatment algorithms. We now
have access to an almost overwhelming flow of information about
lumbar disc arthroplasty from countless sources. Central to the evolution of therapies is a better appreciation of the complexities of the
lumbar disc. By combining knowledge gleaned from anatomical
dissection, biochemical processes, and resultant physiology with a
disciplined foundation in biomechanics, we have created a fabric
of understanding never before enjoyed. Spine arthroplasty is now
an important and evolving area within the treatment of spinal
disorders. This sub-discipline represents the coalescence of many

areas of study focused on the development of new and exciting
solutions to address clinical problems.
These significant advances in our understanding of the spine represent a culmination of efforts occurring across many fronts. Our
increased understanding of the biological factors at work in disc
disease has been a driving force in the development and emergence of new materials and delivery methods. The critical role
that advanced biocompatible alloys, polymers, and viscoelastic
hydrogels play in the innovation of disc arthroplasty technologies
cannot be over emphasized.
Technological advancements have played a vital role in supporting
and expanding our knowledge of motion preserving disc technologies. The latest imaging technologies allow a much more detailed
appreciation of pathological processes, such as disc degeneration,
and provide the ability to monitor the results of an intervention.
Computerized finite element analysis offers a risk-free environment
in which to test hypotheses and predict clinical impact. Biochemical
advancements yield an intimate understanding of the chemical environment including chemical mediators and potential intervention
portals. This wealth of knowledge can be used to great advantage
when developing disc arthroplasty technologies.
Not to be overlooked, the socioeconomic challenges involved in the
development of new technologies, such as the Nucleus Arthroplasty
motion preservation system, have also become more apparent.
The all important variable of proper patient selection continues to
require constant reassessment and vigilance. Increasingly, third-party
payers control access to care and treatment choice to an alarming
degree. Such considerations can no longer be ignored in the quest
for ideal patient management methods.
This publication has been constructed to provide an overview of
the current biomechanical developments in Nucleus Arthroplasty
technology. Key elements include the Principles and Mechanical
requirements of Nucleus Devices, Kinematic Demands, Endplate
Mechanics, Device Stiffness vs. Load Sharing, and Anular Closure
Techniques. In addition, Volume II of this series will provide
insight into the potential market and the current players working
in the forefront of Nucleus Arthroplasty technology development

activities. This is an incredibly exciting field as technologies
focused on the repair and replacement of the diseased disc
nucleus will catapult us far beyond the treatment options we
have available today.
In conclusion, we can say that the spine arthroplasty specialist
of today is well prepared to deliver the most advanced solutions
to the clinical puzzle of disc disease with technologies based on
a rich tradition of innovation and compassion coupled with a
tremendous wealth of physiological knowledge and assessment
tools. As spine surgery evolves from mechanical solutions to
therapeutic solutions both surgeons and patients will benefit.
We hope you will find this series on Nucleus Arthroplasty
technology to be a valuable asset.
Deputy Editorial Board
aymedica has selected Drs. Reginald J. Davis, MD, FACS, Federico P. Girardi, MD, Frank P. Cammisa, Jr., MD, FACS,
and William C. Hutton, DSc to edit this series of monographs on Nucleus Arthroplasty technology, because of their
special interest in this dynamic area of medicine. They are well respected for their clinical work and travel widely to speak
and educate physicians. Drs. Davis, Girardi, and Cammisa are noted for their expertise in spine surgery and advanced
training in minimally invasive surgical techniques.
Dr. Davis is founder of Baltimore Neurosurgical Associates, chief

of Neurosurgery at the Greater Baltimore Medical Center, and a
faculty member at the Johns Hopkins School of Medicine and
the University of Maryland. He is a Fellow of the American
College of Surgeons and a Diplomate of the American Board of
Surgery. Dr. Davis received his medical degree from Johns
Hopkins University School of Medicine, Baltimore, Maryland.
Dr. Girardi is assistant professor of orthopedic surgery, Weill

Medical College of Cornell University and is attending orthopedic
surgeon at the Hospital for Special Surgery, New York, New York.
He specializes in the treatment of spinal disorders including degenerative disc disease (DDD), spinal deformities, metabolic fractures,
and spinal tumors. Dr. Girardi received his medical degree from
the Universidad Nacional de Rosario, Rosario, Argentina.
He has broad experience in advanced procedures such as spinal

stabilization, intradiscal electrothermal therapy, and microendoscopic discectomy and has conducted physician training programs on these procedures. His professional affiliations include
the AANS-CNS Section on Disorders of the Spine, the American
Association of Neurological Surgeons, the Congress of
Neurological Surgeons, and the North American Spine Society.
He has performed extensive clinical research in the areas of minimally invasive surgery, clinical outcomes, and spinal imaging. He
is also interested in basic research on bone, disc, and nerve tissue
regeneration and in the investigation of alternatives to spinal
fusion for the treatment of DDD. His professional affiliations
include the North American Spine Society, Scoliosis Research
Society, the European Spine Society, the International Society for
the Study of the Lumbar Spine, and the Spine Arthroplasty Society.
Dr. Cammisa is associate professor of clinical surgery, Weill

Medical College of Cornell University and is the Chief of Spinal
Surgical Service at The Hospital for Special Surgery in New York,
New York, where he also serves as an associate scientist in the
research division. Dr. Cammisa received his medical degree from
the College of Physicians and Surgeons at Columbia University,
New York, New York.
Dr. Hutton is professor and director of orthopedic research,

Emory University in Atlanta, Georgia. He also attended
Universities in Glasgow, Birmingham, and London. Before coming
to Atlanta, he worked at educational institutes in London and
Adelaide, Australia. In Adelaide, he was professor of biomechanics
and chairman of the Department of Mechanical Engineering.
His clinical interests include non-fusion and motion preservation

technologies, minimally invasive, laparoscopic, and computer
assisted spinal surgery; microsurgery and athletic spinal injuries.
He is an active member of many spine societies, academic committees and editorial review boards. He has lectured widely and
published in numerous peer-reviewed journals and books.
His major area of interest is biomechanics with a particular

focus on the spine. Dr. Hutton has published over 180 papers
in peer review journals. He has won many prizes for his work,
most recently (2004) the Russell S. Hibbs Award from the
Scoliosis Research Society. At present, he has a Research Career
Science Award from the Department of Veterans Affairs. He is
a member of the International Society for the Study of the
Lumbar Spine.
Chapter 7
Biomechanics of
the Degenerated Disc
Andrew A. Sama, MD
ASSISTANT PROFESSOR OF ORTHOPEDIC SURGERY
Weill Medical College of Cornell University

New York, NY 10021
ASSISTANT PROFESSOR

New York, NY 10021
INTRODUCTION
s people age, the lumbar discs undergo a progressive

alteration of chemical composition and biomechanical
properties. This combination of chemical and biomechanical
changes can lead to back pain. Nucleus Arthroplasty interventions are emerging that intend to replace the nucleus of the
diseased lumbar disc for treatment of patients with discogenic
pain. Understanding the natural progression of disc degeneration is paramount to understanding the function of such
devices. This chapter summaNUCLEUS ARTHROPLASTY INTERVENTIONS ARE
rizes what is generally underEMERGING THAT INTEND TO REPLACE THE NUCLEUS
stood about the degenerative
OF THE DISEASED LUMBAR DISC FOR TREATMENT OF
cascade and the resulting
PATIENTS WITH DISCOGENIC PAIN.
biomechanical changes.
Nucleus
Anulus
Apophyseal
Joint
Intervertebral
Disc
Figure 1
Major components of spinal complex.
anulus. The hydrophilic proteoglycans in the nucleus control the

diurnal influx and efflux of water and nutrients that keep the discs
healthy and give them their normal function. The osmotic and
hydrostatic properties of the disc are not static.3, 4 As people age,
the discs can begin to desiccate and become depleted of nutrients.
Disc waste products accumulate, decreasing the concentration of
viable cells, resulting in changes in the biomechanical properties
of the disc.5
THERE IS A COMPLEX ARRANGEMENT OF BONES AND

CARTILAGE THAT MAKE UP EACH SPINAL MOTION SEGMENT. THE INTERVERTEBRAL DISC ALONG WITH THE
APOPHYSEAL JOINTS ALLOW FOR FLEXIBILITY WHILE ALSO
HELPING TO MAINTAIN THE STABILITY OF THE SPINE.
AGING AND LUMBAR DISC DEGENERATION

There is a complex arrangement of bones and cartilage that make
up each spinal motion segment. The intervertebral disc along with
the apophyseal joints allow for flexibility while also helping to
maintain the stability of the spine (Figure 1).1 The disc is sometimes seen as a cushion between vertebral bodies to help absorb
and distribute applied forces. Each disc is made up of an anulus,
a nucleus pulposus, and the cartilage endplates. The intricate
lamellar arrangement between the fibers of the anulus fibrosus
exteriorly, and the nucleus pulposus within, serve to absorb the
forces applied to the spine. In healthy discs, there is a harmonious
relationship between the components, such that compressive axial
forces produce an increase in hydrostatic pressure in the nucleus
pulposus, and this pressure is transmitted to the anulus where it
is absorbed as tensile stress.2 Thus, the anulus sees tensile stress as
well as compressive stress. The need to support this complex stress
explains, to some extent, the orientation of the fibers of the
THE DEGENERATIVE CASCADE

As explained in Book IFundamentals, Chapter 1 of this series,
the result of the disc aging and degenerating is a net decrease in
the amount of aggregated proteoglycan and an increase in the
non-aggregated proteoglycans, which leads to lower osmotic
water binding capacity and loss of compressive resistance in the
lumbar disc. As forces are applied to these biologically altered
discs, there is a greater likelihood that damage will occur to the
anulus fibrosus creating a vicious cycle of degeneration.6, 7
Kirkaldy-Willis et al8 described lumbar disc degeneration as a
cascade that impacts the three joint complex. The first stage of
the cascade is known as the Dysfunctional Stage and corresponds
to the early onset of disc degeneration. The net result of the
Dysfunctional Stage is a tearing of the outer anulus fibrosus secondary to repetitive micro trauma. This is clinically manifested
THE RESULT OF THE DISC AGING AND DEGENERATING IS A NET DECREASE IN THE
AMOUNT OF AGGREGATED PROTEOGLYCAN AND AN INCREASE IN THE NON-AGGREGATED
PROTEOGLYCANS, WHICH LEADS TO LOWER OSMOTIC WATER BINDING CAPACITY AND
LOSS OF COMPRESSIVE RESISTANCE IN THE LUMBAR DISC.
a Dysfunction
b Instability
c Stabilization
Figure 2
MRI images of degenerative cascade. (a) Dysfunction StageHigh Intensity Zone lesion of the posterior anulus fibrosus with decrease
signal intensity of the disc on T2-weighted image. (b) Instability StageDisc collapse with continued decrease in signal intensity.
(c) Stabilization StageContinued disc collapse, osteophyte formation, stenosis of neural structures.
THE KIRKALDY-WILLIS STAGES DESCRIBED
MRI CHARACTERISTICS OF
LUMBAR DISC DEGENERATION
as a mechanical low back pain that is episodic. During this stage

there is also a dehydration of the nucleus pulposus.8
The Kirkaldy-Willis stages described above can be radiographically characterized by multiple imaging techniques, but magnetic
resonance imaging (MRI) evaluation is the gold standard. Stage I
(Dysfunctional) is usually manifested on MRI imaging techniques
by the presence of a High Intensity Zone lesion of the posterior
anulus fibrosus, and an overall decrease signal intensity of the disc
on T2-weighted sequences in the sagittal plane (Figure 2a).9
The second stage is called the Instability Stage. This is representative of more significant damage to the disc secondary to a
delamination of the layers of the anulus fibrosus. Vertebral segmental instability can occur resulting in further damage and loss
of proteoglycan composition in the nucleus pulposus.8
Stage II (Instability) is manifested on MRI with further loss

of disc space height and progressive desiccation resulting in a
darker disc (Figure 2b). Disc herniations can present in Stage II
because a loss of anular integrity may result in herniation of the
nucleus pulposus.9
The third stage is the Stabilization Stage. This occurs when there
is resorption of the nucleus pulposus and worsening of intervertebral disc space collapse. This is the stage where osteophytes
form secondary to anular traction on vertebral endplates and
spinal stenosis may result.8
Stage III (Stabilization) is manifested on MRI by worsening of

disc space collapse and by the presence of osteophyte formation
coupled with soft tissue redundancy or hypertrophy, resulting
in central canal, subarticular lateral recess, or foraminal stenosis
(Figure 2c).9
CAN BE RADIOGRAPHICALLY CHARACTERIZED

BY MULTIPLE IMAGING TECHNIQUES BUT MRI
EVALUATION IS THE GOLD STANDARD.
THE PAIN GENERATOR

The radiographic findings and histomorphology of disc degeneration are objective signs of the presence of a disease state that can
be measured, but these findings may not always correlate with the
patients symptoms.10 The pain generators in degenerative disc
disease (DDD) are multifactorial and can be chemically mediated
or mechanically induced. The loss of disc structure and biomechanical integrity results in alterations of the load-sharing properties of the lumbar disc and the vertebral bodies.3 The nerve
endings in the facet joints, spinal ligaments, and para-spinous
musculature are then stimulated and result in pain. The chemical
mediation of back pain occurs from the release of cytokines and
free radicals along with cellular debris from matrix degeneration.
This results in nocioceptive stimulation of nerve endings and
resultant back pain. These conclusions are based on Buckwalters
articles.5 As discs become progressively more degenerated, they
demonstrate an ingrowth of blood vessels and nerve fibers
beyond the outer anulus which may also contribute to the onset
of back pain. Normal lumbar discs do not typically have blood
vessels or nerve fibers inside of the outer anulus.
BIOMECHANICAL CHANGES
According to Horst et al, non-degenerated discs exhibit fluid-like
properties, whereas degenerated discs have properties that are
more like those of a solid. The normal nucleus pulposus behaves
similar to a viscous fluid: with degeneration, it shows an increase
in shear modulus and becomes stiffer and more elastic.11 As the
tissue transitions from fluid-like to more solid-like properties, there
is an associated decrease in hydrostatic pressurization. Hydrostatic
pressurization of the nucleus pulposus allows the intervertbral disc
to support large loads which may be several times total body
weight.16 The majority of these loads are typically carried through
the anterior column of the spine. However, as the nucleus pulposus
degenerates, a large proportion of load transmission is shifted to
the posterior elements and facet joints.13, 17 This results in increased
facet loading and degeneration as well as an increase in back pain.
Studies have shown that there also are changes in the kinematics
of the discs as a result of varying degrees of degeneration. The
relationships between the soft tissue laxity and changes in kinematics of the lumbar spine were studied extensively by Mimura,
Fujiwara, Krismer, and Frei.12-15 Despite the fact that there were
mixed results from the studies, an assessment of the general
trends suggests that segmental motion is increased and the
motion segment becomes more unstable in the earlier phases of
degeneration when the anulus becomes slack. As degeneration
continues to more advanced stages, a re-stabilization seems to
occur as disc space height is lost, the nucleus becomes more
fibrotic, and there is a net decrease in the flexibility.
Nucleus Arthroplasty technologies are emerging as an alternative
early surgical treatment for patients with degenerative disc disease. Removing the diseased disc nucleus and replacing it with a
nuclear replacement device could improve and possibly restore
the load bearing and kinematic properties of the degenerated
segment to a more physiologic level.19 Providing resistance to
compressive loads while reducing segment instability may help
ensure transmission of loads applied to the spine will be predominantly maintained in the anterior column, thereby unloading the facet joints and decreasing pain.2 Also, by maintaining
segment height and anular stability, the stimulation of the free
nerve endings in the outer anulus may be minimized and therefore decrease discogenic back pain. Biomechanical studies evaluating the kinematic restoration, axial load sharing, endplate
mechanics and anular repair with Nucleus Arthroplasty repair
are discussed in more detail in the following chapters.
PROVIDING RESISTANCE TO COMPRESSIVE LOADS

WHILE REDUCING SEGMENT INSTABILITY MAY HELP
ENSURE TRANSMISSION OF LOADS APPLIED TO THE
SPINE WILL BE PREDOMINANTLY MAINTAINED IN THE
ANTERIOR COLUMN, THEREBY UNLOADING THE
FACET JOINTS AND DECREASING PAIN.
According to Buckwalter it appears that the degenerative process

affects the nucleus pulposus and cartilage endplate more significantly than the anulus fibrosus with respect to changes in
material properties.5, 17
CONSIDERING THE IMPORTANT BIOMECHANICAL ROLE OF THE NUCLEUS

PULPOSUS, AN ATTEMPT TO RESTORE OR RECREATE THE FUNCTION OF A
HEALTHY NUCLEUS APPEARS TO BE A TARGET POINT FOR INTERVENTION.
CONCLUSION
Taking into consideration the varying degrees of disc degeneration
and the biomechanical changes that occur, one of the challenges of
treatment becomes defining a point for surgical intervention.
Considering the important biomechanical role of the nucleus pulposus, an attempt to restore or recreate the function of a healthy
nucleus appears to be a target point for intervention.18 Patients who
are found to have early or mid-stage degenerative disc disease, and
who show signs of progression of degeneration, are potential candidates for nucleus replacement. Removing the damaged nucleus
pulposus and replacing it with a Nucleus Arthroplasty device to
restore the biomechanical properties of the given motion segment
may be beneficial in breaking the degenerative cascade.
REFERENCES
1. White AA, Panjabi MM. Clinical Biomechanics of the Spine. Philadelphia:
Lippincott Williams & Wilkins; 1990.
2. Goins ML, Wimberley DW, Yuan PS, Fitzhenry LN, Vaccaro AR. Nucleus
pulposus replacement: an emerging technology. Spine J 2005 Nov-Dec;
5(6 Suppl):317S-24S.
3. Setton LA, Chen J. Mechanobiology of the intervertebral disc and relevance to
disc degeneration. J Bone Joint Surg Am 2006 Apr;88 Suppl 2:52-7.
4. Buckwalter JA, Mow VC, Bowden SD, Eyre DR, Weidenbaum M. Intervertebral
Disk Structure, Compostion, and Mechanical Function. In: Buckwalter JA,
Ainhorn TA, Simon SR, editors. Orthopaedic Basic Science Biology and
Biomechanics for the Musculoskeletal System. 2nd ed. Rosemont, IL:
American Academy of Orthopaedic Surgeons; 2000. p.548.
5. Buckwalter JA, Boden SD, Eyre DR, Mow VC, Weidenbaum M. Intervertebral
Disk Aging, Degeneration, and Herniation. In: Buckwalter JA, Ainhorn TA,
Simon SR, editors. Orthopaedic Basic Science Biology and Biomechanics for
the Musculoskeletal System. 2nd ed. Rosemont, IL: American Academy of
Orthopaedic Surgeons; 2000. p.558.
6. Kurowski P, Kubo A. The relationship of degeneration of the intervertebral
disc to mechanical loading conditions on lumbar vertebrae. Spine 1986
Sep;11(7):726-31.
10
7. Rohlmann A, Zander T, Schmidt H, Wilke HJ, Bergmann G. Analysis of the

influence of disc degeneration on the mechanical behaviour of a lumbar
motion segment using the finite element method. J Biomech
2006;39(13):2484-90.
8. Kirkaldy-Willis WH, Wedge JH, Yong-Hing K, Reilly J. Pathology and pathogenesis of lumbar spondylosis and stenosis. Spine 1978 Dec;3(4):319-28.
9. Benneker LM, Heini PF, Anderson SE, Alini M, Ito K. Correlation of radiographic and MRI parameters to morphological and biochemical assessment
of intervertebral disc degeneration. Eur Spine J 2005 Feb;14(1):27-35.
10. Kjaer P, Albert H, Jensen TS, Leboeuf-Yde C, Bendix T, Wedderkopp N, et al.
Back pain, radiology and end plate changes by means of Modic. Ugeskr
Laeger 2006 Apr 24;168(17):1668,9; author reply 1669.
11. Horst M, Brinckmann P. 1980 Volvo award in biomechanics. Measurement of
the distribution of axial stress on the end plate of the vertebral body. Spine
1981 May-Jun;6(3):217-32.
12. Mimura M, Panjabi MM, Oxland TR, Crisco JJ, Yamamoto I, Vasavada A.
Disc degeneration affects the multidirectional flexibility of the lumbar spine.
Spine 1994 Jun 15;19(12):1371-80.
13. Fujiwara A, Lim TH, An HS, Tanaka N, Jeon CH, Andersson GB, et al. The
effect of disc degeneration and facet joint osteoarthritis on the segmental
flexibility of the lumbar spine. Spine 2000 Dec 1;25(23):3036-44.
14. Krismer M, Haid C, Behensky H, Kapfinger P, Landauer F, Rachbauer F.
Motion in lumbar functional spine units during side bending and axial
rotation moments depending on the degree of degeneration. Spine 2000
Aug 15;25(16):2020-7.
15. Frei H, Oxland TR, Rathonyi GC, Nolte LP. The effect of nucleotomy on
lumbar spine mechanics in compression and shear loading. Spine 2001
Oct 1;26(19):2080-9.
16. Johannessen W, Elliott DM. Effects of degeneration on the biphasic material
properties of human nucleus pulposus in confined compression. Spine 2005
Dec 15;30(24):E724-9.
17. Niosi CA, Oxland TR. Degenerative mechanics of the lumbar spine. Spine J
2004 Nov-Dec;4(6 Suppl):202S-8S.
18. Di Martino A, Vaccaro AR, Lee JY, Denaro V, Lim MR. Nucleus pulposus
replacement: basic science and indications for clinical use. Spine 2005 Aug
15;30(16 Suppl):S16-22.
19. Le Huec JC, Aunoble S, Basso Y, Tournier C, Yamada K. Biomechanical
Considerations for Total Lumbar Disk Replacement. In: Kim DH, Cammisa
FP, Fessler RG, editors. Dynamic Reconstruction of the Spine. Thieme
Medical Publishers; 2006. p.149.
Chapter 8
Principles and Mechanical

Requirements of Nucleus Implants
Prof. Dr. Hans-Joachim Wilke

PROFESSSOR
Institute of Orthopaedic Research and Biomechanics

University of Ulm
Ulm, Germany 89081
BASIC BIOMECHANICAL CONSIDERATIONS
on-fusion technologies in spinal surgery gain more and

more popularity. Constantly, new ideas are created and
turned into new products. Each idea has its own philosophy with
the principle goal to maintain the motion in the treated segment.
In contrast to total disc implants, nucleus implants are designed
to preserve as many spinal structures as possible. These implants
restore and maintain disc height and/or original mobility.
In many cases, disc degeneration or disc prolapse is treated just by

nucleotomy and decompression; this can produce a good outcome due to decompression of the nerve
roots. However, in some
THE AVERAGE AMOUNT OF REMOVED NUCLEUS MATERIAL
patients this treatment is
OF 3 G LEADS TO A HEIGHT LOSS OF 2.4 MM, WHICH CAN
insufficient and back
BE ASSOCIATED WITH AN INCREASE IN THE SEGMENTAL
pain recurs after a while.
RANGE OF MOTION OF ABOUT 20-30%, AND AN INCREASE
This may be explained
IN THE NEUTRAL ZONE OF UP TO 100%.
by the nucleotomy,
which causes a reduction in disc height proportional to 0.8 mm/g removed material.
Thus the average amount of removed nucleus material of 3 g leads
to a height loss of 2.4 mm, which can be associated with an
11
Nucleus Implants
Polymer
Mechanical
Tissue Engineered
Steelball, Fernstrm
Pre-Formed
Regain, Biomet Inc.
In situ
Formed
Natural
Scaffolds
Synthetic
Scaffolds
NUBAC, Pioneer Surgical

CL-Disc, Interpore Cross
Hydrogel
IPD, Dynamic Spine
Polyurethane
Collagen Gels
Hyaluronic Acid
PDR, TranS1
knitted titanium, Buck
Fleece of PLA
Textiles
Newcleus, Zimmer Spine
with jacket
w/o jacket
Thermo
Responsive
Polymer
Dual Disc cylinders,

Ray and Corbin
Silicone
Hydrogel
Other
Material
Aquarelle, Stryker
NuCore, Spine Wave DiscCell, Gentis
BioDisc, CryoLife
PDN, Raymedica, LLC
NeuDisc,
Replication Medical, Inc.
SaluDisc, Spine Medica
with balloon
w/o balloon
DASCOR, Disc
Dynamics Inc.
Gelifex SP/ IP,

SYNTHES Inc.
SINUX ANR, DePuy
PDN-SOLO, Raymedica, LLC

HydroFlex, Raymedica, LLC
PNR, TranS1
increase in the segmental range of motion of about 20-30%, and

an increase in the neutral zone of up to 100%. This instability
can lead to more stress in the remaining anulus and facet joints
resulting in further degeneration. This height loss may also be
associated with bulging of the anular ring leading to an unphysiological strain pattern in the anulus; this may affect the cells in
these structures. Bulging of the disc is about 1mm per 1 kN up
to 2.5 kN axial preload.1 It seems that this disc bulging is slightly
higher anteriorly than posteriorly.2 The data suggests that this
bulging may even increase further when the spine is bent. This
effect sometimes is compared with a rather flat car tire (this
problem is sometimes called flat-tire syndrome). Removing the
disc from one side creates a slight asymmetry due to the hole.
Finite element analysis suggests that the regions of the disc that
show the largest bulging also show the highest strain. This seems
to be the posterolateral region of the disc, where the strain is
exaggerated after removal of the nucleus.
12
PMMA, Hamby and

Glaser
Figure 1
Classification chart of the different nucleus replacement devices.
VARIOUS TYPES OF NUCLEUS DEVICES

The advantage of a nucleus prosthesis as compared to a total
disc prosthesis is that the nucleus device generally allows the
preservation of the existing anatomical structures including
the anulus, vertebral endplates, and ligaments.
In theory, an optimum nucleus replacement should restore the
mobility and re-establish the intact disc height, thereby restoring
the nominal stresses and strains of the collagen fibres in the anulus. This seems to be a superior situation as compared to leaving
the disc alone after a discectomy.
The origin of the idea to replace the nucleus goes back to the
1950s. The first idea was to fill the nuclear cavity with polymethylmethacrylate (PMMA) or silicone.3,4 Since that time a variety of
solutions to replace the nucleus have been developed.5 This present article tries to classify the different ideas into the three categories: mechanical nucleus devices, polymer implants, and tissue
engineered nucleus implants.
MECHANICAL IMPLANTS
Preformed
Fernstrm implanted the first mechanical nucleus replacement

device in 1966. This implant consisted of a stainless steel ball
that was placed in the disc space.6 Clinically, it proved unsuccessful because the ball subsided into the endplates and served
as a fusion device. Later, other mechanical devices were developed that were made of less stiff materials and incorporated a
larger contact area with the endplates. As an example, Regain
(Biomet, Inc.) is made of one piece of highly-polished, pyrolytic
carbon with a Youngs modulus similar to cortical bone.
NUBAC (Pioneer Surgical Technology) is a ball and socket
nucleus replacement device made of PEEK with tantalum markers. Other mechanical devices include the CL-Disc (Biomet,
Inc.) which is composed of solid, zirconia ceramic, incorporating a porous titantium keel; and the IPD (Dynamic Spine) that
consists of metallic springs and is fixed to the vertebrae.
Hydrogel
The TranS1 PDR (Percutaneous Disc Reconstruction) can also

be considered as a mechanical device; this device allows a supported nucleus or mechanical nucleus replacement. It has all the
benefits of the PNR with an in situ formed silicone absorber (as
described later), and a metal-on-metal central pivot that helps
share the load. The device is implanted using a TranS1 pre-sacral
axial approach to the lumbar spine. With the pre-sacral approach,
the integrity of the anulus and ligaments is maintained, as are any
future surgical options.
A new idea describes an implant made of knitted titanium filaments Buck (Germany), which provides the physiological stiffness
and motion of a spinal segment and seems to have no dislocation
tendency. All these mechanical implants require suturing the anulus after implantation.
THE FIRST MECHANICAL NUCLEUS IMPLANT

WAS SUGGESTED IN 1966 BY FERNSTRM WHO
IMPLANTED A STAINLESS STEEL BALL AS A
SPACER INTO THE DISC.
POLYMER IMPLANTS
As the mechanical devices are often too stiff, many researchers have
experimented with various polymers to create softer nucleus
implants. These softer nucleus implants are used as either preformed
shapes or alternatively the shape is allowed to develop in situ.
Hydrogel is the most preferred material for preformed nucleus

devices made of polymers. It has been demonstrated that PVA
(non-ionic hydrogel) has a similar swelling pressure characteristic of the natural nucleus. Due to their hydrophilic characteristic,
the hydrogels can swell. This means that they can be implanted
in the dehydrated state through smaller channels in the anulus.
After implantation the hydrogel increases in size; this can reduce
the risk of expulsion.
In 1988, Charles Dean Ray, MD, FACS had the idea for a nucleus
implant that consisted of a doppel woven spiral of flexible, hightensile-strength polymeric fibers and tissue ingrowth-promoting,
polyglycolic acid filaments. The cylinder contained a viscous
hygroscopic semifluid. This resulted in the development of the
prosthetic disc nucleus (PDN) device marketed by Raymedica,
LLC. In the first stages of development, the PDN device consisted
of two parts, an anterior and posterior hydrogel pellet comprised
of hydrolyzed poly-acrylonitrile polymer (Hypan), each enclosed
in a woven polyethylene jacket. These were both oriented in a
transverse position within the disc cavity. Once implanted, the
hydrophilic hydrogel pellet absorbed fluid and increased in volume
until restricted by the confines of the now tight, polyethylene
jacket. The original intact disc height was restored after implantation of the device. However, an effect from the hydration could
not be measured.7 Because of a high expulsion rate, this two-part
device was modified into a single, but bigger, implant called
PDN-SOLO. The current version, the HydraFlex device, is a
softer, faster hydrating, more contoured device compared to the
PDN-SOLO, and its preformed shape fits the endplate surface
geometry better with the goal to minimize the risk of subsidence.
A hydrogel without a jacket was also used. The Aquarelle (1998,
Stryker Spine) is made of a semihydrated poly-vinyl-alcohol
(PVA) hydrogel allowing 80% water content. This provides viscoelastic properties similar to the nucleus. Unfortunately, clinical
trails have demonstrated a high expulsion rate.
In 2000, Replication Medical presented the NeuDisc; a modified hydrolyzed poly-acrylonitrile polymer (Aquacryl) reinforced by a Dacron mesh. This anisotropic implant is able to
absorb up to 90% of its weight. It takes 18 hours to reach full
size in laboratory testing.
Another hydrogel implant called SaluDisc (Spine Medica)
is made of Salubria. This material contains water in similar
proportions to human tissue.
13
Polyurethane
Silicone
Polycarbonate urethane (PCU) was used for the Newcleus

(Zimmer Spine) in 2003. This consisted of a preformed spiral
which allowed implantation using a minimally invasive technique. The implant material was able to absorb water up to
35% of its own weight. Functionally, it acted as a spacer with
some shock-absorbing capabilities.
A silicone-based artificial nucleus replacement SINUX ANR

(SiniTech AG, Depuy Spine) is made of a liquid polymethylsiloxane (PMSO) polymer. This is also injected into the void of the
disc and cures in situ in approximately 15 minutes.
In situ formed
Many in situ formed nucleus prostheses have been tried out since
the first attempt by Nachemson in the early 1960s. The concept is
to inject a curable polymer into the nuclear space. The advantage
of an in situ formed nucleus replacement is that it can be injected
non-invasively using a small needle. It is hoped that this type of
implant will be at a low risk for expulsion. In contrast to the preformed implants, which are implanted into the cavity and can
increase disc height, the injectable polymers cannot be injected
with enough pressure to increase disc height.
It has yet to be proved that in situ cured polymer can provide enough
mechanical strength to support the applied load plus have the
required mechanical in vivo fatigue life. One potential disadvantage
of in situ formed devices is toxicity due to unreacted monomers.
Therefore, for in situ formed devices, the polymerization process
is critical to ensure long-term biocompatability.
MANY IN SITU FORMED NUCLEUS PROSTHESES HAVE

BEEN TRIED OUT SINCE THE FIRST ATTEMPT BY
NACHEMSON IN THE EARLY 1960S. THE CONCEPT IS TO
INJECT A CURABLE POLYMER INTO THE NUCLEAR SPACE.
Thermo responsive polymer

One product that is injected as an in situ formed material is
DASCOR (Disc Dynamics). The material, a cool polyurethane
polymer (18 C), is injected under pressure into a polyurethane
balloon through an attached catheter. It takes a minimum of 12
to 15 minutes to solidify in situ.
Another thermo-responsive polymer is poly (N-isopropylacrylamide)
or PNIPAAm, although specific details are not yet available.
A polymer-based hydrogel which is liquid at room temperature
is used for the Gelifex (SYNTHES). This implant looks like a
porous rubber ball after solidification at body temperature.
14
TranS1 PNR (Percutaneous Nucleus Replacement) is an in situ

formed nucleus replacement technology. The nucleotomy and
device implantation is completed using the trans-sacral axial
approach to the lumbar spine, preserving the anulus and ligaments. A hollow screw is threaded axially providing distraction
of the disc space. Silicone is injected through the screw to fill
the created cavity and help maintain motion.
Hydrogel
Several in situ curable hydrogel implants are currently under
development. The Biodisc (CryoLife) is a protein hydrogel
device (PHD), which cures in 2 minutes after injection into the
disc space. The properties are supposedly similar to the human
nucleus and there is no exothermic reaction during the hardening process. The material is similar to epoxy glue and bonds to
the anulus, which hopefully reduces the expulsion risk.
The NuCore Nucleus Device (Spine Wave) is based on a
hydrogel composed of synthetic silk-elastin copolymer. This
material has no measurable exothermic reaction.
Other Polymers
In 1959, Hamy and Glaser suggested injecting PMMA into the
disc. PMMA, which is commonly used as bone cement, cures with
a high exothermic reaction. This is a cheap procedure and was
often used clinically, particularly by neurosurgeons in Germany.
Another in situ cured nucleus implant is the DiscCell (Gentis).
This is an in situ polymerising material that is injected into the
disc space. Little information is available at the present time.
Tissue Engineered Implants

Recently, it has been shown that seeding or reinserting cells inside
the intervertebral disc may preserve disc structures by slowing
down the degeneration processes.8, 9 It was also hypothesized that the
inserted cells might restore the anulus and nucleus tissue. These cell
injection methods could not produce a restoration of disc height.
THE IDEA OF A TISSUE-ENGINEERED NUCLEUS IMPLANT IS TO SEED CELLS

I N A T H R E E - D I M E N S I O N A L M AT R I X . T H I S M AT R I X W O U L D S E R V E A S A
SCAFFOLD TO PRODUCE A STRUCTURE OF MECHANICAL STIFFNESS AND
A D E Q U AT E M E C H A N I C A L P R O P E R T I E S .
The idea of a tissue-engineered nucleus implant is to seed cells

in a three-dimensional matrix. This matrix would serve as a scaffold to produce a structure of mechanical stiffness and adequate
mechanical properties. The seeded cells could be intervertebral
disc cells or mesenchymal progentitor cells (MPCs).
configurations. Nevertheless, the following principles are

suggested in order to compare the different nucleus devices.
Mechanical tests
Static tests
Natural Scaffolds
Nucleus scaffolds can be produced from natural materials such
as collagen or hyaluronic acid. A recent study presented a threedimensional collagen-I matrix made of rat-tail collagen (ARS
Arthro, AG, Esslingen, Germany) that might be suitable to serve
as scaffold for cell seeding and eventual nucleus replacement.10
The problems of restoring disc height and implant expulsion
using a tissue-engineered nucleus have yet to be solved.11 Scaffolds
made of hyaluronic acid could be utilized, although the implant
might degrade after regeneration of the disc.
Natural material composites are currently being developed for
use as nucleus scaffolds. In any case, it is questionable whether
a natural scaffold would be sufficiently load-bearing.
Synthetic Scaffolds
Better load-bearing characteristics may be obtained with synthetic
materials, which may be fabricated of fleece or other combinations of textiles. Because of a higher density and stiffness, they
might be introduced into the disc space and further compressed
to increase the stiffness, allowing more volume to be inserted.
Such approaches are currently under study at universities in
Dresden, Heidelberg and Ulm.
Prior to any implantation, mechanical tests on the implant must

be carried out to ensure that the implant can sustain certain loads
and has sufficient fatigue life. For example, static loads are used to
determine the stiffness and the yield point of the implant.
Dynamic tests
Dynamic loads at a rate of 4 Hz or less are necessary to prove
that the implants are able to withstand at least 10,000,000 cycles
without collapsing, disconnecting, or deforming permanently.
They also may be used as wear tests to determine how much and
what kind of debris are produced under expected in vivo loads
and motions.
These mechanical tests should be performed between two polyacetal or equivalent test blocks. The test blocks will eliminate the
effects of the variability of bone properties and morphology for the
fatigue tests. These tests should be performed in a defined testing
environment (e.g. in a 0.9% saline environmental bath at 37C).
Other tests such as hydrating tests may also be important with
implants made of hydrogel.
Biomechanical tests
Functional in vitro flexibility tests
PRECLINICAL MECHANICAL AND BIOMECHANICAL

TESTING OF NUCLEUS DEVICES
Before being put into clinical practice, nucleus implants and the
surgical approaches for implantation should be tested. Some
of the new biological solutions may not be testable in all test
In contrast to the pure mechanical tests, another category of tests

with the goal to determine the in situ performance is flexibility
tests. These are ideally performed using human cadaveric spine
specimens. These tests require mechanical testing machines that
allow loads that simulate the physiological motion of the specimens
with the implants in place.
15
They should be carried out in flexion/extension, lateral bending,

and axial rotation. It is recommended for standardization that
they are tested under pure moments without preload.12 Eventually,
tests under shear loading, compression, muscle forces, and other
representative in vivo loads should also be carried out. To decide
the best approach in a given clinical situation, in vitro evaluation
involving intact, nucletomized, or degenerated specimens with the
device implanted provide the most realistic option.
The parameters which should be determined are the range of
motion and the neutral zone in the different motion planes, shear
translations, and height changes. These parameters, however, do
not represent the full information about the kinematics. Therefore,
the center of rotation or helical axis could also be important information about the load sharing between the different structures of
the spinal segment.
Additional biomechanical tests

A hard implant leads to a high stress concentration on the endplate (seen as Modic changes) which may lead to remodeling,
subsidence, or even to failure of the endplate, particularly with
poor bone quality. For this reason, other specific set-ups may be
required to determine the endplate deformations.
One of the goals of nucleus replacement is to re-establish the
physiological strain on the anulus, which is assumed to be one of
the prerequisites in order to maintain healthy tissue. This may be
indirectly determined by comparing the bulging of the intact,
nucleotomized, treated disc.
In vitro test with cyclic loading

Depending on the type of device, expulsion or subsidence of the
implant may be a problem. In order to evaluate these types of
biomechanical failures the specimens should be subjected to
cyclic loading as well. Because of degradation of the cadaveric
tissue the testing time may be limited. Thus 100,000 cycles with
possibly exaggerated loads should at least be attempted.
Following this cyclic test or several times during the test secondary
stability tests should be performed as described above.12
Animal experiments
In addition to the tests described above, animal experiments may
be useful to evaluate the efficacy and reliability of nucleus
implants. However, the ability to design a scaled version for animal implantation, coupled with the validity of the available animal species, has potential limitations for extrapolating expected
human performance.
CONCLUSION
Nucleus replacement is an exciting technology and may be a
promising alternative to other non-fusion technologies. Many
different ideas are available or in the development stages. Besides
the mechanical challenges presented to the implant, the implant
has to re-establish the physiological biomechanics of a spinal
segment, plus remain in the disc space and not expulse or subside through the endplate. However, the ultimate judge of the
implant is not the biomechanical data, but the clinical outcome.
REFERENCES
1. Stokes, I. A. (1988). Bulging of lumbar intervertebral discs: non-contacting
measurements of anatomical specimens. J Spinal Disord 1(3): 189-93.
2. Brinckmann, P. (1986). Injury of the anulus fibrosus and disc protrusions. An
in vitro investigation on human lumbar discs. Spine 11(2): 149-53.
3. Hamby, W. B. and H. T. Glaser (1959). Replacement of spinal intervertebral discs
with locally polymerizing methyl methacrylate: experimental study of effects
upon tissues and report of a small clinical series. J Neurosurg 16(3): 311-3.
4. Nachemson, A. (1962). Some mechanical properties of the lumbar intervertebral discs. Bull Hosp Joint Dis 23: 130-43.
5. Carl, A., E. Ledet, et al. (2004). New developments in nucleus pulposus
replacement technology. Spine J 4(6 Suppl): 325S-329S.
6. Fernstrm, U. (1966). Arthroplasty with intercorporal endoprothesis in
herniated disc and in painful disc. Acta Chir Scand Suppl 357: 154-9.
7. Wilke, H. J., S. Kavanagh, et al. (2001). Effect of a prosthetic disc nucleus on
the mobility and disc height of the L4-5 intervertebral disc postnucleotomy.
J Neurosurg 95(2 Suppl): 208-14.
8. Nishimura, K. and J. Mochida (1998). Percutaneous reinsertion of the nucleus
pulposus. An experimental study. Spine 23(14): 1531-8; discussion 1539.
9. Okuma, M., J. Mochida, et al. (2000). Reinsertion of stimulated nucleus pulposus cells retards intervertebral disc degeneration: an in vitro and in vivo
experimental study. J Orthop Res 18(6): 988-97.
10. Neidlinger-Wilke, C., K. Wurtz, et al. (2005). A three-dimensional collagen
matrix as a suitable culture system for the comparison of cyclic strain and hydrostatic pressure effects on intervertebral disc cells. J Neurosurg Spine 2(4): 457-65.
11. Wilke, H. J., F. Heuer, et al. (2006). Is a collagen scaffold for a tissue engineered nucleus replacement capable of restoring disc height and stability in
an animal model? Eur Spine J 15 Suppl 3: S433-8.
12. Wilke, H.-J., K. Wenger, et al. (1998). Testing Criteria for Spinal Implants:
Recommendations for the Standardization of In Vitro Stability Testing of
Spinal Implants. European Spine Journal 7: 148-154.
16
Chapter 9
Kinematic Demands of
Nucleus Arthroplasty Technology
Denis J. DiAngelo, PhD

ASSOCIATE PROFESSOR
Department of Biomedical Engineering and Imaging

The University of Tennessee Health Science Center
Memphis, TN 38138
Brian P. Kelly, PhD

ASSISTANT PROFESSOR
Department of Biomedical Engineering and Imaging

The University of Tennessee Health Science Center
Memphis, TN 38138
KEY POINTS
The instantaneous axis of rotation (IAR) is an important factor
of spinal segment kinematics; however, there is no consensus on
where the IAR of the lumbar disc is during flexion/extension.
Alignment of device IAR and spinal segment IAR is important for
optimal performance of interbody motion preservation devices.
Previous testing of non-compliant interbody motion technologies demonstrated non-concentric IARs may lead to an over
constrained condition, resulting in failure of the device to
provide adequate motion restoration.
Compliant nucleus replacement technologies do not have a prescribed axis of rotation. A new test methodology is proposed for
evaluating the kinematic restorative effect for such devices.
17
DEFINITIONS AND TERMINOLOGY
NUCLEUS ARTHROPLASTY DESIGNS
Kinematics: describes the movement between two rigid bodies

with no consideration to the forces involved. Movement of a
body can be described in Cartesian coordinates as having three
orthogonal translations and three rotations about each translational axis. The motion can be described in two-dimensions
(2D) or three dimensions (3D). For 2D motion, two translations
and one rotation are required. Additional kinematic parameters
can be calculated that describe motion and include the center of
rotation (CR) in 2D, the helical axis of motion (HAM) in 3D, or
their instantaneous components (ICR or IHAM).
Various types of nucleus arthroplasty (NA) devices exist which

can be categorized into three groups: Void fillers (NuCore, Spine
Wave; BioDisc, CryoLife), kinematically-constrained mechanical
devices (NUBAC, Pioneer; Regain, Biomet), and load sharing
devices (HydraFlex, Raymedica, LLC; NeuDisc, Replication
Medical; DASCOR, Disc Dynamics).1, 2, 3 Further, each device has
an associated surgical procedure for preparing the nucleus site
and placing the NA device that alters the physical properties of
the treated spinal level. Hence, in addition to studying the biomechanical properties and function of NA devices, one should also
understand the impact that the various surgical techniques have
on the stability and function of the treated disc. The influence of
these techniques include facet disruption, bony removal, the surgical approach (anterior versus posterior), affect of an annular
incision (which depressurizes the nucleus), and the amount of
nucleus material removal: micro (associated with discectomies)
or complete. All of these incremental surgical alterations introduce different degrees of instability to the spinal joint that must
be compensated for by the NA device itself.
Load
STIFFNESS
Low Load
High Load
Displacement
Figure 1
Typical stiffness curve. The curve is generally nonlinear and has
two regions: a low load region and a high load region.
AS TISSUE IS EXPOSED TO INCREASING DISPLACEMENTS,

STIFFNESS IS GREATLY INCREASED AND SMALL CHANGES
IN DISPLACEMENT INDUCE LARGE LOAD RESPONSES.
Stiffness (inverse of flexibility): the ability of a structure to

deform per unit displacement. The typical load-displacement
curve for spinal MSU displays a non-linear relationship having
two different regions: a low load region and a high load region
(Figure 1). As tissue is exposed to increasing displacements,
stiffness is greatly increased and small changes in displacement
induce large load responses.
Coordinate System: a reference system used to define the position and orientation of a body in space or relative to another body.
Motion Segment Unit (MSU): two adjacent vertebrae and
interconnecting disc and surrounding ligamentous tissues.
Intervertebral Disc: an inner nucleus pulposus core surrounded
by an anulus fibrosus tissue.
18
HOW TO STUDY THE KINEMATICS

OF NUCLEUS ARTHROPLASTY
The nucleus pulposus is a pressurized gelatinous region consisting of proteoglycans (glycosaminoglycans), loose Type II collagen fibrils, mineral salts, and water that is surrounded by an
anulus fibrosus structure. Together, the nucleus and anulus display nonlinear material properties that influence the biomechanics
of a spinal MSU. The goal of nucleus arthroplasty is to restore the
compliant function of the disc and spinal MSU and prevent overloading of the adjacent bony and soft tissue structures.
The design rationale for nucleus arthroplasty differs from that of
total disc arthroplasty. Total disc arthroplasty devices are noncompliant devices designed to restore motion to the degenerative
disc. However, there is no consensus on where the IAR of the lumbar disc is during flexion/extension activities, with various locations reported in the literature. Since most total disc replacement
devices (TDR) are mechanical joints, if the IAR of the device does
not coincide with that of the native spine an over constrained condition within the MSU may occur, leading to overloading of adjacent structures or failure to provide adequate motion restoration.4
The goal of nucleus arthroplasty is to restore the compliancy of the
native intervertebral disc and recreate a flexible load bearing MSU
system. A more flexible system serves to reduce the occurrence of

an over constrained, non-mobile condition. Moreover, if the IAR
of the MSU varies, the NA device may dynamically deform to
accommodate the positional changes. An improved method for
evaluating nucleus arthroplasty devices and their ability to restore
segment compliancy and motion is to impose a series of different
kinematic motion profiles and measure their reaction loads. The
different motion profiles force the implanted MSU to adjust to the
prescribed kinematic pattern since the axes of rotation would be
selected such that they are not co-centric with the disc center.
Thus, the reactive forces needed to follow the prescribed motion
represent how well the NA device performs. This chapter discusses
a new approach for studying in vitro, the capacity of nucleus
arthroplasty to restore the kinematic of a spinal MSU in a human
cadaveric model. This new testing methodology is a paradigm shift
from the conventional displacement or load control methods.5
CURRENT BIOMECHANICAL TESTING METHODS

Limited biomechanical studies exist that evaluate NA devices.
Although the conventional testing method of applying a pure or
constant bending moment across the spinal construct and measuring the motion response to that loading condition can be
done, there are significant limitations with this testing methodology. Physiologically, the spine is not loaded with a constant
bending moment, but rather experiences a moment distribution
that varies across all spinal levels as you go down the spine.
Although pure moment methods provide a standard approach
Combined MSU Flexion/Extension

Rotation of the Lumbar Spine
In Vitro
Motion Segment Unit Level
L1-L2
In Vivo (4 sources*)
L2-L3
L3-L4
L4-L5
for comparing different lumbar spinal devices and may be

acceptable for testing fusion instrumentation,5 it is not well suited
for studying any type of spinal device that permits motion and/or
has a variable stiffness or modulus (i.e., is not a rigid metal structure). Alternatively, eccentric compressive load test methods have
been used to study both fusion and non-fusion spinal instrumentation.6 A compressive load is typically applied eccentric to the
long axis of the spine causing the spine to flex or extend under
a combined compressive load and bending moment. Using this
testing method, a more physiologic response in the rotational
involvement of each MSU occurs throughout the lumbar spine
(Figure 2).7, 8, 9, 10 However, even though the eccentric loading
method induces a physiologic rotational response across the
intact lumbar spine, this method may not have the sensitivity to
study the compliant properties and kinematic requirements of
NA devices or different disc conditions. With either load control
or displacement control methods limited information is available
about the loads acting on the disc and/or NA device, or the
amount of load sharing that occurs between the NA and adjacent
supporting structures as the MSU moves through a functional
range of motion.
NEW KINEMATIC TESTING PROTOCOL FOR

STUDYING COMPLIANT STRUCTURES
Human joints move under a state of minimum energy; they follow the path of least resistance. A new testing protocol is proposed that involves prescribing a known kinematic input to a
spinal MSU and measuring the capacity of the intact MSU to
accommodate the motion. The effects of changing the MSU
properties via surgery (nucleotomy) or placement of a nucleus
arthroplasty device, changes the effort or work required to move
the altered spine condition through a prescribed motion path
relative to the intact spine condition. The closer the loading
mechanics of the altered spine are to the intact spine condition,
the better the likelihood the device will restore the native properties. Further, the kinematic path can be a simple rotation about
a fixed point in space, or a coupled movement (displacement and
rotation) along a path.
L5-S1
0
10
15
20
Range of Rotation (Degrees)
Figure 2
In Vitro versus In Vivo MSU rotations of the lumbar spine.6, 7, 8, 9
19
Figure 4
Schematic of the
Spine Robot used
to move a vertebral body about a
fixed center of
rotation relative
to an adjacent
vertebral body.
Figure 3
Programmable
multi-axis
Spine Robot.
A custom-designed spine robot (Figure 3) was used that consisted

of four programmable degrees of freedom that can each be independently operated under displacement control, force feedback
control, and combinations thereof.11 Using the spine robot, the
kinematic profile of an intact MSU can be programmed to follow
a specified path or to rotate about a fixed point in space (Figure
4). More advanced kinematic analyses are possible that map out
the motion response to a given multi-directional force profile.
to establish the orientation of the MSU (Figure 5B) when mounted

in the spine robot (Figure 5C). The MSUs were tested under three
fixed points of rotation along the center line of the disc in the anterior-posterior (A-P) direction: 1) the mid point of the disc (C), 2)
half way between the mid-point and anterior aspect of the disc
(A), and 3) half way between the mid-point and posterior aspect of
the disc (P) (Figure 5B). The MSUs were rotated about the designated fixed points of rotation until a target bending moment of
8Nm of flexionextension was reached or the shear or compressive forces exceeded 400N. For all test conditions, MSU axial force
(+Fz net MSU tissue tension, -Fz net MSU tissue compression), A-P shear force (+Fx net MSU posterior shear, -Fx net
MSU anterior shear), sagittal rotation (+y flexion, - y extension),
and sagittal bending moment were measured and compared using
a one-way ANOVA (P=0.05).
A preliminary series of kinematic tests were performed on the

spine robot to study the flexion and extension mechanics of three
lumbar MSUs. Each specimen was tested in three different spine
conditions: the intact harvested condition, post-nucleotomy condition, and post-implanted nucleus arthroplasty condition. The
hydrated HydraFlex device (Raymedica, LLC) was used for the
implanted condition. The orientation of each MSU in a neutral
alignment was measured on a radiograph (Figure 5A) and used
Figure 5
A) Radiograph showing neutral
alignment of lumbar spine MSU.
B) Maintenance of neutral alignment of MSU in mounting pots
and location of fixed points of
rotation. C) Potted MSU specimen
mounted in Spine Robot.
A
20
Compressive Force (Fz): Flexion
Compressive Force (Fz): Extension
250
250
150
A
C
P
50
-50
-150
Force
Force
150
A
C
P
50
-50
-150
-250
-250
-350
-350
Harvested
Nucleotomy
Implant
Harvested
Spine Conditions
100
100
50
50
Force
Implant
Shear Force (Fx): Extension
Shear Force (Fx): Flexion
A
C
P
-50
-100
-150
-50
-100
-150
-200
-200
-250
-250
Harvested
Nucleotomy
Implant
Harvested
Spine Conditions
Nucleotomy
Implant
Spine Conditions
Rotation (0y) Extension
Rotation (0y) Flexion

12
12
7
7
A
C
P
2
-3
Degrees
Degrees
Nucleotomy
Spine Conditions
A
C
P
2
-3
-8
-8
Harvested
Nucleotomy
Implant
Spine Conditions
Harvested
Nucleotomy
Implant
Spine Conditions
Figure 6
Comparison of the three fixed points of rotations. Mean values of the axial compressive load, A-P shear force, and MSU rotation
during flexion and extension.
PRELIMINARY FINDINGS USING KINEMATIC

TESTING PROTOCOL
The mean values of the axial load, shear force, and MSU rotation
for the three different fixed axes of rotation conditions were calculated and graphed (Figure 6). Using data for the harvested
spine condition, significant differences in the MSU rotation, axial
force, and shear force values occurred between the three different
points of rotation (P, C, A). MSU rotations were significantly different between all points of rotation in flexion and extension,
except between points C and P in flexion. The axial forces were
significantly different between P versus A and P versus C in flexion. A similar trend occurred with the axial force values during
extension but the differences were not significant (likely due to
the small sample size). Shear forces were significantly different
between points P versus A and P versus C during flexion. In

extension the shear forces were significantly different between all
points of rotation (P versus A, P versus C, and A versus C). In
general, for the harvested intact MSU, when rotated in flexion
about the mid-point of the disc (location C), MSU posterior soft
tissue tension and resistance to posterior directed shear provided
the stabilization effect. When the point of rotation was shifted
posterior (location P), MSU anterior tissue compression and
resistance to anterior directed shear provided the stabilization
effect. Shifting the rotation point anterior (location A) had minimal effect on tissue stabilization response, but decreased segmental rotation compared to point C. The data demonstrates
that the point about which a single MSU is rotated has a significant effect on the rotational range of motion as well as the soft
tissue stabilization response.
21
Compressive Force (Fz): Flexion
Compressive Force (Fz): Extension
250
250
150
Harvested
50
Nucleotomy
-50
Implant
-150
Force
Force
150
-250
Harvested
50
Nucleotomy
-50
Implant
-150
-250
-350
-350
A
Point of Rotation
Shear Force (Fx): Flexion

100
50
50
Harvested
-50
Nucleotomy
-100
Implant
-150
-200
0
Force
0
Force
Shear Force (Fx): Extension
100
Harvested
-50
Nucleotomy
-100
Implant
-150
-200
-250
-250
A
Point of Rotation
Nucleotomy
Implant
Point of Rotation
Degrees
Harvested
Rotation (0y) Extension
14
12
10
8
6
4
2
0
-2
-4
-6
-8
A
C
Point of Rotation
Rotation (0y) Flexion
Degrees
C
Point of Rotation
14
12
10
8
6
4
2
0
-2
-4
-6
-8
Harvested
Nucleotomy
Implant
Point of Rotation
Figure 7
Comparison of the three different spine conditions at specific points of rotations. Mean values of the axial compressive load, A-P shear
force, and MSU rotation during flexion and extension.
When comparing between the different spine conditions (harvested, nucleotomy, and implanted) there was a trend in the data
that demonstrated the nucleotomy increased the MSU rotation,
altered the load response and appears to have more variation in
response (e.g. less stable). Following implantation of the
HydraFlex device, the MSU rotation returned to the harvested
condition and had less variation compared to the denucleated
condition (Figure 7). Alterations to the harvested MSU via the
nucleotomy or NA implant tended to reduce/lessen the A-P
shear forces, with the denucleated condition having a large variation. At posterior point P, the denucleated and implanted states
demonstrated less tissue compression. In extension, the implant
condition tended to reduce the compressive response of the
posterior elements at Point C and P.
22
In general, for all test conditions, the rotational range of motion

and the tissue loading response differed depending on the selected
kinematic axis of rotation (point C, P or A). Denucleating the
MSU led to more rotation and increased variation in the test data,
indicative of a less stable/predictable response. After implantation
of a hydrated HydraFlex device, variation was reduced and the
response profile trended more towards the intact state for all test
points. However, the small sample size and large variation noted
in the denucleated specimens may have limited significance from
occurring. Increasing the sample size should further confirm significant differences between the spine conditions or points of
rotation, but the test method utilized demonstrates the importance of understanding the constraints of a test setup and how
results may vary as the constraints are changed.
AS NEW TECHNOLOGIES EVOLVE, THE METHODS UTILIZED TO ANALYZE THEIR

BIOMECHANICAL PERFORMANCE MUST EVOLVE. THE PRELIMINARY DATA FROM THIS
STUDY DEMONSTRATED THAT UTILIZING A ROBOT TO CHANGE THE LOCATION OF
THE MSU AXIS OF ROTATION HAD AN EFFECT ON THE KINEMATIC RESPONSE OF
THE HARVESTED, DENUCLEATED AND IMPLANTED MSU.
CONCLUSION
REFERENCES
As new technologies evolve, the methods utilized to analyze

their biomechanical performance must evolve. The preliminary
data from this study demonstrated that utilizing a robot to
change the location of the MSU axis of rotation had an effect
on the kinematic response of the harvested, denucleated and
implanted MSU. Since the in vivo MSU axis of rotation is not
believed to be a single point of rotation and since compliant
nucleus replacement technologies do not have a prescribed axis
of rotation, evaluating the kinematic response at multiple locations of rotation may more effectively characterize the restorative effect of these technologies compared to more traditional
test methods. Additional testing is being performed to further
evaluate the utility of this method as well as evaluating the
restorative effect of the implant on a path that matches the natural motion of the MSU. Overall, this method looks very promising for thoroughly understanding the kinematic response of
Nucleus Arthroplasty technologies.
1. Davis R. J., and Girardi F. (eds.) 2006. Nucleus Arthroplasty in Spinal Care:
Book 1 Fundamentals. Minneapolis (MN): Raymedica, LLC.
ACKNOWLEDGMENT
Elizabeth Sander and Nephi Zufelt for assistance with the biomechanical tests and data processing.
2. Di Martino A., Vaccaro A., Lee J., et al. Nucleus pulposus replacement: basic
science and indications for clinical use. Spine. 2005;30 (suppl 16):16-22.
3. Klara P., Ray C. Artificial nucleus replacement: clinical experience. Spine.
2002;27:1374-77.
4. DiAngelo DJ, Foley KT, Morrow B, Kiehm KJ, Gilmour L, The Effects of OverSizing of a Disc Prosthesis on Spine Biomechanics, Proceedings of the Sixth
Annual Spine Arthroplasty Society Meeting, 2006.
5. Goel VK, Wilder DG, Pope MH, and Edwards WT. Biomechanical testing of
the spine. Load controlled versus displacement-controlled analysis. Spine
1995;20:2354-7.
6. DiAngelo DJ, Scifert JL, Kitchel S, Cornwall GB, McVay BJ, Bioabsorbable
Anterior Lumbar Plate Fixation in Conjunction with Cage Assisted Anterior
Interbody Fusion, J Neurosurg. Nov;97(4 Suppl):447-55, 2002.
7. Dvorak J., Panjabi M.M., Chang D.G., et al. Functional radiographic diagnosis of the lumbar spine: flexion/extension and lateral bending. Spine 16
(5):562-71, 1989.
8. Pearcy M.J., Portek I., and Shepherd J.. Three-dimensional x-ray analysis of
normal movement in the lumbar spine. Spine 9 (3):294, 1984.
9. Froning E.C. and Frohman B. Motion of the lumbosacral spine after laminectomy and spine fusion: Correlation of motion with the result. Journal of Bone
and Joint Surgery 50-A (5):897-918, 1968.
10. White A.A. and Panjabi M.M. Clinical biomechanics of the spine,
Philadelphia:J.B. Lippincott Co., 1990.
11. Kelly BP, DiAngelo DJ, Foley KT, Design Of A Multi-Axis Programmable
Spine Robot For The Study Of Multi-Body Spinal Biomechanics, Proceedings
of the 32th Annual Meeting of the Cervical Spine Research Society, Dec. 2004.
23
Chapter 10
Device Stiffness vs.

Load-Sharing with Nucleus
Arthroplasty Devices
Brian P. Beaubien, BME, MS
LEAD ENGINEER
Midwest Orthopaedic Research Foundation

and the Gustilo Medical Education Center
Minneapolis, MN 55415
Andrew L. Freeman, BME, MS

LEAD ENGINEER
Midwest Orthopaedic Research Foundation

and the Gustilo Medical Education Center
Minneapolis, MN 55415
KEY POINTS
The engineering term elastic modulus is used to characterize
the uni-axial stress-strain response of linear-elastic materials.
Biologic tissues and advanced polymeric biomaterials are typically
not linear-elastic and the stress-strain relationship is complex.
Physiologic interactions significantly impact the response characteristics of devices made from advanced biomaterials.
Cadaver evaluation of polymeric devices demonstrated differences in the compressive load transfer through a device based on
bulk device properties (i.e. device modulus) as well as changes in
the level of constraint provided by physiologic tissue.
It is proposed that device design differences, as well as physiologic
interactions, require characterization of a Nucleus Arthroplasty
devices load-sharing properties in simulated use conditions,
which provides a more thorough characterization of performance
with increased clinical understanding.
24
INTRODUCTION
he engineering term modulus is perhaps the most widely

used descriptor of the mechanical properties of a material.
For the most basic conditions and materials, the modulus simply
relates the amount of stress and strain seen in a material and is
useful in predicting and describing the load-deformation behavior. However, with biological tissues and advanced biomaterials
the stress-strain response is more complex and is dependent on
the surrounding environment. This chapter aims to highlight
important considerations in describing the stress-strain relationship of tissues and biomaterials and to discuss the effects of these
properties, implant design and spine mechanics on the in vitro
performance of the native nucleus pulposus, anulus fibrosus and
Nucleus Arthroplasty (NA) devices.
MATERIAL PROPERTIES:
DEFINITIONS AND EXAMPLES
The modulus of elasticity (E)
The modulus of elasticity is used in Hookes Law (Robert
Hooke, 1635-1703) to describe a linear relationship between
stress and strain:
In this equation, stress (, force per unit area) and strain (, elongation per unit length) are linearly related by the modulus of
elasticity (E), which is often called the elastic modulus or Youngs
Modulus (Thomas Young, 1773-1829). The elastic modulus is
measured in the linear region of the stress-strain curve and represents the slope of the curve. Each material generally has a unique
elastic modulus (Figure 1). In general, a low modulus equates to
a softer, more easily deformable material.
The elastic modulus provides a simple but powerful relationship
between the stress and recoverable strain in materials as diverse
as bone, rock and steel. However, this limited version of Hookes
Law accurately describes only tension and compression behavior
of linearly elastic materials in a uni-axial direction. It should be
noted that modulus may be direction-dependent or anisotropic.
This is important to consider for composite materials (e.g. carbon fiber implants, reinforced concrete) as well as biologic tissues
with internal structured arrangement (e.g. tendon, ligament).
Tangent modulus
For other materials, such as many used in NA devices, the relationship between stress and strain is nonlinear. In other words, at
small strains many polymers and biological tissues exhibit a relatively low elastic modulus, whereas the same materials exhibit a
higher elastic modulus at higher strains (Figure 2). Two or more
Figure 1
Diagram of a simple tension test (left) and typical stress-strain curves and elastic modulus values for materials
commonly encountered in orthopaedics.
25
distinct regions may be observed, but the transition is generally

gradual, and the stress-strain relationship can thus be described
as nonlinear. For non-linear materials, the modulus is determined by the slope of a line tangent to the stress-strain curve at a
given point; this is called the tangent modulus. The tangent modulus provides an intuitive measure of a materials non-linear
stress-strain response, analogous to the elastic modulus.
FOR BIOLOGICAL TISSUES AND ADVANCED POLYMERIC

BIOMATERIALS, THE STRESS-STRAIN RELATIONSHIP IS
USUALLY DEPENDENT UPON THE DURATION OF LOADING.
Viscoelasticity: Modulus dependence with time
Figure 2
Stress-strain curve for a typical soft tissue or nonlinearly-elastic biomaterial.
Tangent lines for points A and B are indicated and the tangent modulus
calculation is shown for Point B (EB).
For biological tissues and advanced polymeric biomaterials, the

stress-strain relationship is usually dependent upon the duration
of loading. When these materials are exposed to a given stress,
the resulting strain increases with time at a rate dependent upon
the material; this phenomenon is known as creep. Similarly, an
applied strain results in a steadily decreasing amount of stress
over time (Figure 3), and this stress relaxation may or may not
proceed at a rate proportional to creep.1 Materials exhibiting
time-dependent material properties are considered viscoelastic.
An important consideration in defining the elastic modulus
for a viscoelastic material is the rate and duration of loading. The
stress-strain-time relationship can be simplified from static tests
by describing the apparent modulus upon immediate loading or
deformation (instantaneous modulus) and after a long duration
of loading or deformation (equilibrium modulus). It is important
to recognize the various methods for calculating modulus and
the potential variation in outcomes with viscoelastic materials.
Fluid flow
For biologic tissue and some biomaterials (e.g. hydrogels), fluid
flow in and out of the material is important in determining the
time-dependent properties of the material. For example, the load
bearing capacity of the nucleus is attributed to its hydration
level.2 However, as hydration levels change as a result of mechanical pressure, the permeability of the anulus and endplates allows
an outflow of water, altering the mechanical response of the
disc.3,4 Thus, the disc response to load is viscoelastic in nature.
Figure 3
Results for the stress-relaxation response of a porcine spinal ligament. An immediate strain step was applied and held (top) and the resulting stress response was
observed (bottom). Stress-relaxation experiments are performed to explore the
stress-strain-time relationship.
26
Radial deformations and the effect of constraint

The three-dimensional behavior of a material is only partially
described by the uni-axial stress-strain and time dependency.
Equally important is the behavior of a material in a plane perpendicular to the applied force, or stated otherwise, the radial
direction. For the simple bar shown in Figure 1, the radial deformation can be described as a proportion of the axial strain as
shown in the following relationship:
where is a constant known as Poissons ratio. Poissons ratio is

important not only in describing radial deformation, but also
the amount of volume change exhibited by a material for a given
load. The unit change in volume (V) of a cube of solid material
can be described using the following equation:
Note that for a Poissons ratio of 0.5, this change in volume is zero.
Such a material is considered incompressible. Water is the most
common example of a nearly incompressible material. Water can
easily undergo a shape change, but imposing a volume change is
nearly impossible if the water is constrained radially. Biological tissues, rubber and other polymeric materials are examples of materials that are often considered nearly incompressible with sufficient
radial constraint.
Unconfined
Compression
Confined
Compression
Sample
Ring
12
Tangent Modulus (MPa)
Over a 24 hour period the balance of mechanical and osmotic

pressure causes the disc volume and height to decrease such that
a ~20mm decrease in height is seen over the entire length of the
spine.5 In the healthy disc this change is balanced by swelling
during the discs nightly intake of water while the spine is
unloaded.6 Factors governing the rate of fluid flow include
osmotic and physical pressure gradients, porosity, and the availability and viscosity of the surrounding fluid. While inherent
viscoelastic properties of a hydrogel or soft tissue and fluid-flow
appear similar in the discs response to loading, the respective
contributions of these variables to material properties are at least
partially independent.7
10
8
6
4
2
0
Unconfined
Compres s ion
Confined
(S ilicone R ing)
Confined
(PE ring)
Figure 4
Unconfined vs. confined compression of a hydrogel. Confined testing data from
this study were obtained with either silicone or polyethelyne rings. Data adapted
from Joshi, et al.18, 25
The apparent modulus of a water-containing material is dependent upon the rigidity and permeability of its physical constraints.
If constrained radially in an undeformable and impermeable container, such a material would behave in a relatively incompressible
manner and could thus withstand high loads with negligible
deformation. If the constraining container were made permeable,
then volume change of the same material would occur over time
as water seeped from the material. In a permeable container, the
material would appear to have a high modulus under rapid
dynamic loading, but would have a relatively low apparent modulus after many hours of loading as the water content decreased
with time. Stress or time-dependent deformation of the constraining container would similarly result in decreased rigidity.
This phenomenon is illustrated by the varying tangent modulus
of a hydrogel material subjected to various constraints (Figure 4).
Thus, material modulus is dependent on both the constraint
rigidity and the constraint permeability for water-containing
materials such as cartilage, the nucleus pulposus, the anulus fibrosus and hydrogel materials. Relatively incompressible materials
not containing water would have moduli dependent on the constraint rigidity but not the permeability of the constraint.
27
Apparent Modulus and Device Behavior

While the 3-dimensional material properties are important to
consider in a detailed analysis of materials and implants,
assuming homogeneity by assessing bulk material properties
can allow the engineer or surgeon to more effectively consider
overall performance.
Bulk material properties: Stiffness and apparent modulus

Uni-axial deformation of uniform materials and shapes can be
described using the equations above, but the deformation of heterogeneous materials is more complex. Because the load
response is often proportional to the cross-sectional area and
geometric shape, the apparent modulus of structures can also be
calculated by making the assumption of uniform bulk material
properties and cross-sectional areas. While not perfect, these
assumptions allow the scaling of material properties to different
geometries. In this way we can compare the properties of anulus
fibrosus and nucleus, for example, although these are highly
inhomogeneous composites of numerous constituents.
Stiffness of Nucleus Arthroplasty devices

For NA devices, the structural device stiffness depends upon the
bulk material properties of the device, the device geometry and
constraining effect of the design. For example, a spherical device
may have a low stiffness as the device contacts the adjacent endplates, but as deformation of the spherical device continues, the
device may stiffen with the changing shape. This geometric nonlinearity, compounded with material behavior and device design
yield a highly nonlinear load-deflection response. In addition, as
discussed above, the device bulk material properties can be
affected by environmental factors, such as temperature, hydration permeability constraints or rigidity constraints. Thus, characterization of NA devices with routine engineering material
properties may not provide insight to the clinical relevance of
the anticipated in vivo performance.
NA device performance and spinal load sharing

Understanding of the bulk material and structural properties of a
NA device is important but must be described in the context of
the intended device in vivo performance. While one device may
28
employ a material having a lower bulk modulus, other factors,

such as device geometry, device constraints, surgical technique,
and the constraining effect of the native anatomy (e.g. anulus)
may impact its 3-dimensional in vivo behavior. In order to further
understand NA device function one must consider the in vivo
loading conditions, load sharing among spinal segment structures
and tissues, and the associated impact of surgical procedures.
Disc loading
In vivo the disc is subjected to shear forces, moments and axial
loads and all of these loading mechanisms engage the nucleus and
anulus to some degree. In vitro data suggest that at the extremes of
motion, where the outer anulus and ligamentous structures
engage, neither nucleus removal (nuclectomy) nor NA have a substantial effect on the stiffness or strength of the disc.8-10 However, at
low displacements and rotations, removal of the nucleus significantly decreases rotational and compressive stiffness.10 Intradiscal
pressure measurements suggest that direct load bearing of the
nucleus is greatest in compressive loading modes.8, 9
Estimating compressive disc loads is important in understanding
the loading environment in which NA devices must maintain disc
height and avoid significant endplate remodeling. Compressive
loads have been estimated in vivo by measuring intradiscal pressures. In healthy thoracic and lumbar discs, pressure has been
found to range from ~0.1 - 0.2 MPa in the prone position to ~0.5 1.0 MPa while standing and up to 2.5 MPa in dynamic and weight
bearing motions.6, 11-13 The corresponding compressive loads are
estimated to range from ~400 800 N in relaxed standing to over
2300 N while bending with weights.12, 13
Loads on the posterior elements have been estimated to be 10 16% of the applied compressive load at a neutral position.14, 15
Load through the facets increases with reduced disc height,
degenerative changes, increased extension angle and prolonged
axial loading, and can be as high as 50% of the applied load.14, 15
Substantial load remains in the disc even with half of the load
borne by the posterior elements.
Load-bearing in the healthy disc involves both the anulus and
nucleus. In the healthy disc a well-hydrated nucleus represents a
nearly incompressible material with a relatively low unconstrained
Youngs modulus.16, 17 With such a material elastic containment of
the nucleus by the anulus is important in enabling load-bearing
by healthy nuclear material while allowing some deformation of
the disc.18 Because of its high tensile stiffness and low permeability compared to the nucleus the anulus is an ideal radial constraint and allows pressurization of the nuclear material with
low permeability to maintain nucleus hydration.19 As degeneration progresses, the swelling pressure of the nucleus decreases
and its ability to transport water (i.e., permeability) increases.2
Degeneration may thus shift the load-bearing in the disc from a
fluid pressurization mode to solid matrix load-bearing of the
anulus fibrosus.20
Direct load-bearing of the anulus may occur with degeneration
and nuclectomy. In vitro studies have found that applied compressive loads dramatically increase the axial stress of both the nucleus
and central endplate, and the anulus and peripheral endplate.9, 21
Loads in both structures may in fact be important to their survival
as suggested by increased matrix synthesis at certain levels of
hydrostatic pressure in the nucleus and anulus.22 The balance of
nucleus vs. anulus loads may vary over the duration of a day with
changes in nucleus volume and pressure. A further shift is noted
with nuclectomy where the disc pressure and endplate strains
decrease, but do not disappear in the nuclear space and the resulting strains in the anulus may be harmful.9, 23 Finite element models
(FEM) have found similar results and suggest that nucleus
removal leads to high interlaminar shear stresses that may result
in separation of the laminae and eventual disc degeneration.23, 24
Compressive load sharing in the NA reconstructed spine

There are numerous NA devices being developed to replace the
diseased nucleus with the goal of preventing additional disc
degeneration while maintaining the compressive and motion
function of the segment. Understanding the relationship between
the device loads and the segment loads is key to characterizing
NA device performance.
Compressive loads in the spine reconstructed with NA devices
are distributed across the posterior elements, the anulus and
the NA device. One important factor is the constraining effect
of the anulus on the NA device. A close anulus-device fit
(direct or via scar tissue) applies radial constraint to a NA
device allowing the increase in stiffness of the disc to be much
greater than that expected with a given device.18, 25 Meakin, et
al23 loaded sagittal cross-sections of ovine motion segments
with the cut face of the disc sealed with a transparent window.
With load the anulus of the intact disc bulged outward at both
Figure 5
Bulge of the anulus and nucleus with the application of compressive loads.
Inward bulge of the anulus is noted after nuclectomy; implantation of a well-fit
NA device prevents inward bulge of the anulus. Adapted from Meakin, et al.23
its inner and outer margins, whereas removal of the nucleus

allowed the anulus to bulge inward. Implanting a silicone NA
device (~0.5) eliminated this inward bulge regardless of the
elastic modulus of the device, which ranged from 0.2 to 40
MPa (Figure 5).
The importance of anular radial constraint is further illustrated
in studies by Joshi, et al where it was shown that disc compressive properties appear more dependent on device sizing than on
NA material modulus. They compared the stiffness of intervertebral discs in the intact state, after a cylindrical trans-endplate
nuclectomy and after insertion of a cylindrical hydrogel NA
device.25 The devices were sized to match the nucleotomy closely
or to be oversized by 1mm in height or diameter and had elastic
moduli varying from 0.05 to 1.5 MPa. Denucleation reduced the
compressive spinal segment stiffness by 52%. Subsequent device
implantation increased the stiffness to a level near that of the intact
state. Interestingly, each 1% increase in device elastic modulus
resulted in a motion segment stiffness increase of only 0.04
N/mm, whereas a 1% increase in implant height or diameter
resulted in disc stiffness changes of 15 N/mm and 21 N/mm,
respectively. In a similar study by the same group hydrogel
implants having structural compressive stiffness of 2.5 N/mm
at 15% strain resulted in an increase in motion segment stiffness
of 680 N/mm at the same strain.18
These studies highlight the importance of evaluating in vitro NA
device performance to account for the effect of anular radial constraint and/or effects of device/nuclear space sizing in determining
reconstructed disc performance properties.
29
Ongoing studies on NA devices and spinal load sharing

The above sections highlight the importance of characterizing a
NA device both on the bench top and in the motion segment.
While the stiffness of motion segments implanted with NA
devices has been investigated previously, the relative load-bearing
of a NA device compared to that of the anulus and posterior elements has not been thoroughly studied.8, 10, 25 A study at our institution is under way to determine the relative load-sharing of
motion segments reconstructed with: 1) a high stiffness interbody fusion device (representative PEEK spacer) and 2) a low
stiffness NA device (hydrated HydraFlex device, Raymedica,
LLC). The objective of these tests is to determine the relevance of
device stiffness (PEEK versus NA device) and NA device sizing on
the implanted segment load-deflection response, the load-sharing
characteristics and the device subsidence characteristics.
This ongoing study is conducted in three phases. In the first
phase, the load-deflection response of each cadaveric motion
segment was evaluated in the intact, denucleated and post
implanted condition by applying a 1600 N compressive load at a
rate of 0.5 Hz. Each test state was pre-conditioned with 30 load
-200 -6
-5
-4
-3
-2
-1
cycles 1600 N prior to testing. Specimen stiffness was determined for each test condition utilizing the tangent stiffness in
the low-load primary stiffness region (<400 N) and in the high
load secondary stiffness region (>800 N). In addition, lateral
radiographic images were obtained at compressive loads of 20 N
(unloaded) and 800 N (loaded, representative of relaxed
standing).12, 13 Radiographic images were scaled to the actual vertebral body depth and the disc height measurement is measured
digitally for each condition.
In the second phase, the load carried by the implant, anulus and
posterior elements was determined. The specimens from Phase 1,
which were reconstructed with the NA device, were tested in
displacement control to the corresponding peak displacement as
recorded at 1600N in Phase 1 (i.e., each specimen had a unique
displacement). Displacement-controlled tests were then conducted for the implanted condition, after removal of the anulus
and after removal of the posterior elements. Removal of the anulus and posterior elements was conducted in the test frame
under rigid constraint to maintain relative segment position for
each test state. Testing in displacement control in this fashion and
-200 -6
Load (N)
Load (N)
-4
-3
-1000
-1400
-1000
-1400
Intact
Nuclectomy
PEEK implanted
-1800
Displacement (mm)
-200 -6
-5
-4
-3
Intact
Nuclectomy
HydraFlex undersized
-2
Displacement (mm)
-1
Load (N)
-600
-1000
-1400
-1800
Intact
Nuclectomy
HydraFlex appropriately sized
Displacement (mm)
Figure 6
Load-Deflection response for intact, denucleated and post-implanted conditions.
(A) PEEK spacer (B) Undersized HydraFlex device (C) Appropriately sized HydraFlex device.
30
-2
-600
-600
-1800
-5
-1
Discheight
height at
at 20
20 N
N
Disc
Disc
Disc height
heightatat800
800NN
22
20
20
18
18
Disc height (mm)
Disc height (mm)
22
16
14
12
10
16
14
12
10
8
Intact
Nuclectomy
Implant
Intact
Nuclectomy
Implant
Figure 7
Segment disc height for the intact, denucleated and PEEK device states (a) unloaded (20N) state (b) 800N load
Disc Height and Load-Deflection Results

AFTER NUCLECTOMY, THE PRIMARY STIFFNESS
DECREASED BY 65 17% OF THE INTACT VALUE AND
THE SECONDARY STIFFNESS DECREASED BY 15 11%.
recording peak loads after sectioning allows the loads supported

by each component of the reconstructed segment to be estimated. The sequence of anulus removal and posterior element
removal was alternated to minimize order effects.
In the third phase, the specimens from Phase 2 with the anulus
and posterior elements removed are tested in load control until
endplate failure was radiographically observed. The segmental
stiffness and load sharing results are discussed below; subsidence
results are included in Chapter 11.
In this study, sizing of the interbody devices is achieved with
spacers provided by the manufacturers and is performed in the
unloaded specimen. All PEEK devices were selected to provide a
tight fit and are tapped gently into place as is done clinically.
Equal distribution of the NA devices in the oversized, undersized
and correctly sized groups was attempted, but in these relatively
young patients (age range 30 to 58) most discs were healthy and
many were above the clinically-indicated range. This resulted in
more frequent undersizing of the HydraFlex devices.
The load-deflection response was recorded and plotted for a PEEK

device, an undersized HydraFlex device, and an appropriately
sized HydraFlex device (Figure 6). Removing the nucleus caused
substantial decrease in disc height (mean standard deviation:
2.2 1.0 mm vs. intact). After nuclectomy, the primary stiffness
decreased by 65 17% of the intact value and the secondary stiffness decreased by 15 11%. The greater reduction in primary
stiffness with nuclectomy illustrates how the nucleus, which has a
low elastic modulus in the unconstrained state, is responsible for a
large proportion of the discs stiffness in the low-load region. After
the initial disc space collapse, the compressed anulus achieved a
secondary stiffness close to that of the intact state.
Implantation of a PEEK spacer restored the disc height to within
0.1 1.8mm of the intact disc height in the unloaded condition.
When the specimen was subjected to 800 N the PEEK spacer
maintained a loaded height similar to that of the intact state
under the same load (Figure 7). The PEEK spacer also restored
segmental stiffness towards that of the intact group in the primary stiffness region (103 39% vs. intact) and to a lesser
degree in the secondary stiffness region (83 9%).
31
20
Device Load-Sharing Results
Disc height at 20 N
Disc height (mm)
18
16
14
12
10
8
6
Intact
20
Nuclectomy
Implant
At peak displacement the estimated posterior element resistance

to the applied load averaged 12 6% across both implant groups.
This result is similar to values previously reported.14, 15 Removal of
the anulus resulted in an 18 15% decrease in the peak load in
specimens with the PEEK spacer and 45 20% decrease in load
for specimens with the HydraFlex device. The lower stiffness
HydraFlex device was estimated to have sustained a mean 44
20% of the applied compressive load compared with 70 17%
for the PEEK spacer at peak displacement. These results show the
load-sharing through the device at both low displacement and
peak displacement (Figure 9).
Disc height at 800 N
Disc height (mm)
18
Dependence of NA device sizing on load sharing
16
14
12
10
8
6
Intact
Nuclectomy
Implant
Figure 8
Segment disc height for the intact, denucleated and HydraFlex device
states (a) HydraFlex unloaded (20N) (b) HydraFlex at 800N.
Both disc height and segment stiffness are not effectively restored
with an undersized HydraFlex device (Figures 6B & 6C). For
implanted HydraFlex devices determined to be representative of
clinical sizing (appropriately sized, based on HydraFlex specific
sizing instrumentation), disc height was restored to within 0.1
0.2mm of the intact value in the unloaded condition. The stiffness
of the segment with appropriately sized devices was 53 15% of
the intact value in the primary stiffness region and 67 13% in
the secondary stiffness region. With undersized devices the disc
height was restored to a lesser degree at low loads (-1.2 1.0mm).
The primary stiffness for these devices was 39 15% and the secondary stiffness was 71 12% of the respective intact state. At 800
N, which approximates standing loads, the implanted segment
height was on average 0.4mm less than the respective intact height
for appropriately sized devices and 2.0mm less for undersized
devices (Figure 8). Appropriate device sizing thus appears critical
for approximation of intact segmental properties.
32
In addition to the load-sharing tests conducted with properly

sized HydraFlex implants, the effect of under-sizing (n=8) and
over-sizing (n=2) of the device was also investigated utilizing the
same test protocol. The posterior element loads did not change
with device sizing. Under-sized devices also bore the least load
with as little as 15% of the applied load transferred through the
device (mean 38%), compared with 46 22% for correctly-sized
and 68 5% in the over-sized conditions. Linear regression of
the estimated compressive load support by the device versus
device sizing is represented as the difference between measured
intact disc height and hydrated device height (Figure 10).
The dependence of load-sharing on device and disc height may
be explained by the greater constraint and load-bearing contribution given by a larger device. The larger device is pre-compressed to a greater degree and applied loads shift the response
of the device further along the load-deflection curve resulting in
a higher apparent modulus. Additionally, the volume of the
vacant space in the nuclectomy is decreased, thereby increasing
radial constraint on the device. Together, these factors combine
to yield a significantly higher load through the NA device.
The data collected to date demonstrate that sizing of a NA device
is critical to device performance. Clinically, undersizing may
result in insufficient segment restoration, whereas oversizing may
create additional load and resulting contact stress at the implantendplate interface. Previous studies considering over-sized NA
devices have found similar trends.25
Percent of applied load
100%
51st Annual Meeting of the Orthopaedic Research Society, Washington, DC,

February 20-23, 2005. 2-10-2005.
Implant
Anulus
Posterior elements
80%
4. Dunlop RB, Adams MA, Hutton WC. Disc space narrowing and the lumbar
facet joints. J Bone Joint Surg Br. 1984;66:706-10.
40%
5. Roberts N, Hogg D, Whitehouse GH et al. Quantitative analysis of diurnal

variation in volume and water content of lumbar intervertebral discs. Clin
Anat 1998;11:1-8.
20%
6. Wilke HJ, Neef P, Caimi M et al. New in vivo measurements of pressures in the
intervertebral disc in daily life. Spine 1999;24:755-62.
60%
0%
HydraFlex
PEEK Spacer
HydraFlex
20% of Peak Displacement
PEEK Spacer
Peak Displacement
Figure 9
The percentage of an applied compressive load estimated to pass through various
structures. Reconstructed motion segments were loaded to a given displacement
and the resulting load was recorded. Repeated testing after anulus and posterior
element removal allowed estimation of the load through these structures.
Intact Specimen
8. Bertagnoli R, Sabatino CT, Edwards JT et al. Mechanical testing of a novel

hydrogel nucleus replacement implant. Spine J 2005;5:672-81.
9. Frei H, Oxland TR, Rathonyi GC et al. The effect of nucleotomy on lumbar
spine mechanics in compression and shear loading. Spine 2001;26:2080-9.
10. Wilke HJ, Kavanagh S, Neller S et al. Effect of a prosthetic disc nucleus on
the mobility and disc height of the L4-5 intervertebral disc postnucleotomy.
J Neurosurg. 2001;95:208-14.
11. Polga DJ, Beaubien BP, Kallemeier PM et al. Measurement of in vivo intradiscal pressure in healthy thoracic intervertebral discs. Spine 2004;29:1320-4.
100%
12. Schultz A, Andersson G, Ortengren R et al. Loads on the lumbar spine.

Validation of a biomechanical analysis by measurements of intradiscal
pressures and myoelectric signals. J Bone Joint Surg Am 1982;64:713-20.
80%
Load sharing
7. Stokes I, Laible J, Iatridis J. Time dependent mechanical behavior of intervertebral

disc only partly explained by fluid flow at slow loading rate. 53rd Annual Meeting
of the Orthopaedic Research Society, San Deigo, CA, February 11-14. 2007.
60%
13. Nachemson A, Elfstrom G. Intravital dynamic pressure measurements in

lumbar discs. A study of common movements, maneuvers and exercises.
Scand.J Rehabil.Med.Suppl 1970;1:1-40.
40%
20%
14. Yang KH, King AI. Mechanism of facet load transmission as a hypothesis for
low-back pain. Spine 1984;9:557-65.
0%
-8
-6
-4
-2
Estimated Sizing (mm)

Intact Disc Height
Figure 10
The load borne by a NA device is linearly related to device sizing.
CONCLUSION
In situ, load-sharing between nucleus arthroplasty devices and the
anulus fibrosus is believed to be critical for clinical success. The
data from these studies demonstrates that sizing is an important
variable for lower modulus devices. Undersizing may result in
insufficient segment restoration, whereas oversizing may result in
excessive load-bearing by the NA device. Ideally, balancing these
goals will yield an arthroplasty capable of maintaining disc height
with an appropriate amount of load-sharing and stress in the anulus and posterior elements. Surgical considerations associated with
such a balance include disc height and nuclectomy volume, the
size and composition of the device and the surgical approach.
REFERENCES
1. Provenzano P, Lakes R, Keenan T et al. Nonlinear ligament viscoelasticity.
Ann.Biomed Eng 2001;29:908-14.
2. Johannessen W, Elliott DM. Effects of degeneration on the biphasic material
properties of human nucleus pulposus in confined compression. Spine
2005;30:E724-E729.
3. Laffosse J-M, Ambard D, Accadbled F et al. Influence of location, fluid flow
direction and bone tissue maturity on the permeability of vertebral endplates.
15. Adams MA, Hutton WC. The effect of posture on the role of the apophysial
joints in resisting intervertebral compressive forces. J Bone Joint Surg Br.
1980;62:358-62.
16. Yang K, Kish V. Compressibility measurement of human intervertebral
nucleus pulposus. J Biomech 1988;21:865.
17. Cloyd J, Malhotra N, Mauck R et al. Defining unconfined compression modulus and poisson's ratio for human nucleus pulposus and potential nucleus
replacement materials. 53rd Annual Meeting of the Orthopaedic Research
Society, San Deigo, CA, February 11-14. 2007.
18. Joshi A, Fussell G, Thomas J et al. Functional compressive mechanics of a
PVA/PVP nucleus pulposus replacement. Biomaterials 2006;27:176-84.
19. Perie D, Korda D, Iatridis JC. Confined compression experiments on bovine
nucleus pulposus and anulus fibrosus: sensitivity of the experiment in
the determination of compressive modulus and hydraulic permeability.
J Biomech 2005;38:2164-71.
20. Iatridis JC, Setton LA, Foster RJ et al. Degeneration affects the anisotropic
and nonlinear behaviors of human anulus fibrosus in compression.
J Biomech 1998;31:535-44.
21. Edwards WT, Ordway NR, Zheng Y et al. Peak stresses observed in the posterior lateral anulus. Spine 2001;26:1753-9.
22. Ishihara H, McNally DS, Urban JP et al. Effects of hydrostatic pressure on
matrix synthesis in different regions of the intervertebral disk. J Appl Physiol
1996;80:839-46.
23. Meakin JR, Reid JE, Hukins DW. Replacing the nucleus pulposus of the
intervertebral disc. Clin Biomech (Bristol., Avon.) 2001;16:560-5.
24. Goel VK, Monroe BT, Gilbertson LG et al. Interlaminar shear stresses and
laminae separation in a disc. Finite element analysis of the L3-L4 motion
segment subjected to axial compressive loads. Spine 1995;20:689-98.
25. Joshi A, Mehta S, Vresilovic E et al. Nucleus implant parameters significantly
change the compressive stiffness of the human lumbar intervertebral disc.
J Biomech Eng 2005;127:536-40.
33
Chapter 11
Endplate Mechanics
Joshua D. Auerbach, MD
Philip M. Maurer, MD
RESIDENT
CLINICAL ASSISTANT PROFESSOR OF
Department of Orthopaedic Surgery

The University of Pennsylvania
Philadelphia, PA 19107
ANESTHESIA AND ORTHOPEDIC SURGERY
Heather L. Guerin, PhD
Pennsylvania Hospital
ENGINEER
Richard A. Balderston, MD
Biomechanics Practice
Exponent, Inc.
CLINICAL DIRECTOR OF SPINE SERVICE

SECTION ON ORTHOPEDIC SURGERY
Pennsylvania Hospital
KEY POINTS
Vertebral endplates play an integral role in the mechanical and
biological function and degeneration of the intervertebral disc.
Intra-operative factors to minimize risk of interbody subsidence are implant sizing, endplate preparation and quality of
the host bone.
Results from an evaluation of the loads required to initiate device
subsidence of a new Nucleus Arthroplasty device compared to a
representative interbody fusion device look promising.
Augmentation of vertebral bodies with low bone density may
be a viable clinical technique for reducing the risk of device
subsidence in patients at risk of developing osteoporosis.
34
INTRODUCTION
he intervertebral discs serve a largely mechanical role of

absorbing and transmitting the loads of the spine. The adjacent vertebral endplates are integral to this role. Both the
mechanical and biological functions of the endplates are related
to their micro and macrostructure. This chapter will discuss
these relationships of endplate structure, their mechanical and
biological functions, and the overall impact on intervertebral
disc mechanics as it relates to Nucleus Arthroplasty devices.
Understanding endplate mechanics is important when developing interbody and Nucleus Arthroplasty (NA) devices in order
to avoid subsidence and restore the load-sharing characteristics
similar to those of the normal intervertebral discs.
Figure 2
Graphic depiction of human vertebral development showing the vascular structures
extending from the centrum of the vertebral body to the endplate and disc.
endplate is thickest adjacent to the nucleus pulposus, whereas the

cartilaginous endplate is thinnest adjacent to the nucleus pulposus.1, 2 The neurovascular structures and bone marrow contact
channels retreat so that they terminate in the bony endplate layer
(Figure 3). Nerve fibers are most concentrated in regions adjacent
to the nucleus pulposus, with innervation density similar to the
outer disc anulus fibrosus.
Figure 1
Schematic representation of the normal histology. A=anulus fibrosus;
B=vertebral body (bone); V=blood vessels; C=calcified cartilage; D=disc;
EP=endplate; M=marrow; N=nucleus pulposus; T=tidemark.
ANATOMICAL CONSIDERATIONS OF THE ENDPLATE

The endplates of the intervertebral disc serve as borders between
the fibrocartilaginous intervertebral discs, whose fibers anchor
into the endplates, and the vertebral bodies. The endplate is comprised of two layers; a layer of calcified cartilage adjacent to the
vertebral bone, and a layer of hyaline cartilage adjacent to the
intervertebral discs (Figure 1). Early in life, the cartilaginous endplate is thick and serves as the growth plate for the vertebral bodies. During this time, vascular and marrow contact channels
extend through the cartilaginous endplate into intervertebral disc
(Figure 2). With skeletal maturation, the endplates become calcified adjacent to the vertebral bone, which develops into a bony
endplate layer with a thickness of approximately 0.5mm.1 The
remaining cartilaginous endplate layer adjacent to the intervertebral discs has a thickness of approximately 0.6mm.2 The bony
The endplates are thought to be the primary mode for fluid flow
into and out of the disc. The semi-permeable nature of the endplates is integral to the creep response, and the diurnal variation
in pressure, of the intervertebral disc. The resistance of the endplates to fluid flow is direction-dependent, with resistance to fluid
flow out of the disc much greater than resistance to fluid returning into the disc.3 Thus, the endplates provide a physical barrier
Figure 3
Sagittal section through mature
vertebral bodies and disc,
showing the distribution of the
nutrient arterioles throughout
the body, without vascular
communication with the disc
space or cartilaginous endplate.
35
enabling hydrostatic pressurization of the intervertebral disc as

compressive load is applied, and, in conjunction with osmotic
pressure, enable re-hydration of the disc as load is removed. Fluid
flow occurs across the bony endplate through marrow contact
channels and in the cartilaginous endplate by diffusion.
In addition to fluid flow, the endplates (especially the central
regions adjacent to the nucleus pulposus) are the primary route for
metabolite and waste product transport. Nutrients reach the cartilage endplates via the marrow contact channels and capillary buds
of the bony endplates. Diffusion through the adult cartilage endplates occurs and depends both on the properties of the metabolite
solute including molecular size and charge, and the cartilage
matrix. Larger sized molecules, such as growth factor-binding protein complexes, are excluded from diffusion into the intervertebral
disc by the large, negatively charged aggregating proteoglycans.2
ALTHOUGH THE ETIOLOGY OF INTERVERTEBRAL DISC

(IVD) DEGENERATION IS LIKELY MULTIFACTORIAL,
THERE IS INCREASING EVIDENCE THAT POINTS TO
THE ENDPLATE AS HAVING A PRIMARY ROLE IN THE
DEGENERATIVE CASCADE.
CHANGES WITH DEGENERATION

Although the etiology of intervertebral disc (IVD) degeneration is
likely multifactorial, there is increasing evidence that points to the
endplate as having a primary role in the degenerative cascade.4
Several important and notable changes to the endplate take place
in disc degeneration, including matrix disorganization, decrease in
cell density in the endplate, cracks in the endplate cartilage, and
further reduction in endplate vascularity. IVD degeneration is also
associated with calcification of the endplate cartilage and a reduction in the number of marrow contact channels, which results in
decreased endplate permeability. These changes have been shown
to correlate with biochemical degeneration of the disc.5 Structural
endplate changes that appear related to the presence of disc degeneration are also associated with a reduction in the diffusion across
the endplate. It has recently been shown that endplate cartilage has
the most limiting effect on diffusion, and thereby controls the sole
mechanism for nutritional transport to the nucleus pulposus.
Therefore, endplate damage, whether in the form of cracks, fissures, fractures, or Schmorls nodes, which disrupt the nutritional
delivery to the disc, may contribute to the progression of the
degenerative cascade of the IVD.4
36
Degeneration causes somewhat paradoxical changes to the cartilage and bony endplates. With degeneration, the proteoglycans
of the cartilage endplate break down, enabling larger molecules,
both beneficial and detrimental, to diffuse in and out of the disc.
Ultimately, age and other degeneration-related changes to endplate vascularity and permeability are thought to be detrimental
to the nutritional status of the intervertebral disc, and directly
contribute to the IVD degenerative cascade. These endplate
changes precede age-related changes to the nucleus pulposus,
and may be the harbinger of intervertebral disc degeneration.4
VERTEBRAL ENDPLATE BIOMECHANICS

Compared with the more widely studied nucleus pulposus and
anulus fibrosus, relatively little is known about the mechanical
properties of the vertebral endplate. The endplates distribute
spinal loads across the intervertebral disc. The individual mechanical properties of the bony and cartilaginous endplates act in conjunction with vertebral body and intervertebral disc mechanical
properties to govern overall motion segment mechanics.
The calcified cartilage of the bony endplate provides stiffness and
strength to the endplate construct. The trabecular bone adjacent
to the endplates is more dense and rodlike compared to the trabecular architecture further removed from the endplate. Bony
endplate mechanical properties vary with spatial location; the
greatest strength and stiffness in the postero-lateral regions, and
lowest values in the central and antero-central regions (where
interbody devices are commonly placed).1 One study showed that
the points closest to the pedicles were about 2.5 times stronger
than more centrally-located points.1 It has also been shown that
the inferior endplate is stiffer and 40% stronger than the superior
endplate. Sacral endplates are stronger than superior lumbar endplates, and among sacral endplates, posterior test sites were
3 times stronger than anterior sites.1 Decrease in bone mineral
density of the vertebral bodies causes these variations to become
more pronounced.1, 6
In the healthy intervertebral disc, the cartilaginous endplate is
pivotal in maintaining a uniform stress distribution between the
intervertebral disc and the vertebral body. The fibers of the
intervertebral disc are anchored into the cartilaginous endplate,
directly transmitting intervertebral disc loads to the endplate
and ultimately the vertebral body. Furthermore, hydrostatic
pressurization of the nucleus pulposus results in axial bulging of
the endplates. The collagen fibers of the cartilaginous endplate
are oriented parallel to the axial surface of the tissue; thus,
ADEQUATE ANTERIOR COLUMN SUPPORT IS CRITICAL FOR MOST SPINAL RECONSTRUCTIVE

PROCEDURES AS IT FACILITATES THE ACHIEVEMENT OF APPROPRIATE SAGITTAL ALIGNMENT,
CORONAL BALANCE, AND LOAD-SHARING BETWEEN ANTERIOR AND POSTERIOR COLUMNS.
bulging results in tension in the endplate tangential to the bulge.

The biphasic mechanical properties (permeability and aggregate
modulus) of the cartilaginous endplates have been shown to be
similar to those of articular cartilage, and likely vary with strain.
Structural disruptions of endplate integrity such as macroscopic
and microscopic anomalies, bony intrusions, and thinning of
the endplate can occur early in life, and may precede clinically
apparent endplate failure that occurs years later. Failure may
occur through acute or chronic mechanisms. Perey (1957)
described three major forms of endplate failure: 1) central (nondegenerate discs), 2) peripheral, and 3) complete endplate. In
central failures, compressive loads increased the nuclear hydrostatic pressures, causing excessive loading and failure in the endplate. In contrast, degenerate discs, characterized by a loss of
proteoglycan and hydrostatic pressures, displayed peripheral failure patterns since most of the load under compression is then
transmitted through the anulus. As a result, the endplate is
loaded in the periphery and fails accordingly.8
The mechanical properties of both the endplate and the vertebral body have been shown to be rate dependent, with failure
load increasing with increasing load application rate (i.e. quasistatic vs. impact loading). Removal of the bony endplateoften
required prior to placement of an interbody fusion device to
facilitate bone-implant osseointegrationhas been shown to
detrimentally affect stiffness and endplate failure load.9 Thus,
intact endplates appear to be an important factor in increasing
the failure loads of devices in contact with the endplate surfaces.
Furthermore, while failure stress increases with increasing load
rate for the vertebral body, the endplates are not similarly
affected. At high loading rates, the vertebral bodies may fail via a
bursting mechanism initiated by the intrusion of intervertebral
disc material through an endplate fracture.10 Fatigue testing of
motion segments has shown that the endplates typically fail first,
accompanied by micro-fracture of the trabecular architecture of
the vertebral body.11
ENDPLATE MECHANICS IN
DEGENERATIVE DISC DISEASE
No study to date has quantified the degree of endplate mechanical compromise that accompanies disc degeneration, which is
the subject of current research efforts at our institute. As intervertebral disc degeneration causes a reduction in water content
in the nucleus pulposus, the mechanism for compressive loadbearing then changes from hydrostatic pressurization in the
nucleus pulposus and tensile hoop stress in the anulus fibrosus
to direct compressive force transmission to both the nucleus pulposus and the anulus fibrosus.12, 13 Endplate failure may also contribute to the alteration in load-sharing in the degenerated IVD
because it decompresses the pressurized state of the nucleus pulposus, which may result in a transfer of compressive load-bearing from the nucleus pulposus to the anulus fibrosus. Ultimately,
alterations to the normal loading behavior of intervertebral disc
tissues is thought to initiate structural damage to those tissues.14
With these changes, the spatial variations in vertebral architecture and mechanical properties become more evenly distributed
across the disc. In many cases of endplate failure, depressurized
disc material is allowed to penetrate the vertebral body (i.e.
Schmorls nodes), and it has been hypothesized that this may
initiate vertebral body fracture.10
CLINICAL CONSIDERATIONS
Endplate Subsidence and Failure with Interbody Devices
Adequate anterior column support is critical for most spinal
reconstructive procedures as it facilitates the achievement of
appropriate sagittal alignment, coronal balance, and load-sharing
between anterior and posterior columns, thus enabling the optimal biomechanical environment for spinal fusion or, alternatively, motion preservation. Subsidence of an interbody device is
a common and potentially devastating problem as it may lead to
segmental kyphosis, coronal plane deformities, compression of
neural elements, pseudarthrosis, and failure of instrumentation.
The most common factors associated with endplate subsidence
37
2.5
2
1.5
A Stress (MPa)
p=0.08
Control
Treatment
25
20
15
10
0.5
Modulus (MPa)
p=0.07
Control
Treatment
0
Sub-Failure
Max
Sub-Failure
Max
Figure 4
Average (std dev) sub-failure and max A) stress and B) apparent modulus in Treatment (vertebral bodies
augmented with Cortoss in lumbar TDR) and controls, *p0.05.
and failure in the setting of an interbody device are 1) implant

geometry and size, 2) degree of endplate removal/preparation,
and 3) regional endplate strength. Each of these factors is
addressed below.15
WITH PROPER SIZING, A NUCLEUS IMPLANT

(HYDRAFLEX DEVICE BY RAYMEDICA, LLC) DEMONSTRATED NEARLY EQUAL LOAD SHARING WITH THE
ANULUS, THUS REDUCING THE RISK FOR CENTRAL
Intra-operative Methods to Prevent Subsidence
IMPLANT SUBSIDENCE.
There is an increasing body of literature that has focused on

methods to optimize the integrity of the bone-implant interface
at the time of initial surgery. Such attention may maximize the
longevity of the first procedure, and therefore minimize the risk
of requiring a potentially more involved revision surgery.
support by the device. With proper sizing, a nucleus implant

(HydraFlex device by Raymedica, LLC) demonstrated nearly
equal load sharing with the anulus, thus reducing the risk for
central implant subsidence.
The first and easiest technique to maximize the strength of the

bone-implant interface is appropriate implant sizing. Whether
the interbody device to be implanted is a fusion cage, total disc
replacement (TDR), or nucleus replacement, it is critical to avoid
relying solely on the central, weakest portion of the vertebral
body.16 When it is technically feasible, the largest size device that
can be safely implanted should usually be used in order to
achieve maximum fit. Devices that require removal of a substantial portion of the anulus fibrosus for placement of an interbody
device (e.g. total disc arthroplasty devices) will engage more of
the apophyseal rim, which in turn should minimize the risk of
device subsidence. As has been shown in the sizing of total disc
replacement devices, larger devices rest on more peripheral
regions of bone with a higher stiffness, and therefore produce
less stress at the endplate surfaces, which results in less device
displacement under compressive loads. As shown in Chapter 10
of this publication, implant sizing is critical for nucleus replacement devices and can have a significant impact on the load
The second intra-operative consideration is endplate preparation.

While this intra-operative step may receive more attention with
fusion or total disc replacement devices, which aim to achieve an
osseointegration between the device and host bone, this is less of
an issue in nucleus arthroplasty implantation. Nucleus replacement devices allow preservation of the cartilaginous endplate
which in turn provides additional mechanical support, and minimizes the risk of device subsidence and endplate fracture. For
these reasons, there is no formal endplate preparation and no tissue removal with Nucleus Arthroplasty devices. Instead, the diseased nucleus pulposus is completely removed to remove the pain
generator and to create space for proper functioning of the
Nucleus Arthroplasty device. Although nucleus pulposus swelling
pressures and flow states influence the mechanics of the underlying vertebral endplate, no study to date has quantified the biomechanical effect of nuclear tissue removal and implantation of a
Nucleus Arthroplasty device on vertebral endplate mechanics. It
stands to reason, that Nucleus Arthroplasty devices made from
38
materials that allow continued fluid movement through the endplate may have physiologic benefits. However, this needs to be
further investigated in a clinical setting.
improved design features, including a more anatomically-shaped

implant design, a larger contact footprint, and a faster hydrating
and softer hydrogel core which reduces device stiffness.
The final contributing factor to bone-implant interface strength

is the quality of the host bone. Recently, vertebroplasty has
gained popularity as a relatively safe, minimally invasive, effective treatment for osteoporotic vertebral fractures. Filling of
fractured vertebrae with injected bone cement has been shown
to reduce stresses in the anulus and neural arch, resulting in a
more even distribution of compressive forces on the disc and
adjacent vertebral bodies. Other biomechanical studies have
shown that vertebral augmentation results in an improvement in
adjacent motion segment stiffness, and compressive load-bearing, the effects of which are more pronounced in vertebral bodies with low bone density. Recently, the clinical use of
vertebroplasty in conjunction with total lumbar disc replacement (TDR) has been described. Furthermore, the authors have
shown in a biomechanical cadaveric spine study that vertebral
augmentation with Cortoss (Orthovita, Malvern, PA) in lumbar
TDR significantly improves the sub-failure mechanical properties of the implant-endplate interface, with more dramatic differences in low bone density specimens (Figure 4).17 These
findings demonstrate that vertebral augmentation potentially
reduces the risk for implant subsidence and endplate failure in
patients who have or are at risk for the development of osteoporosis. Future studies will evaluate further the potential role
for vertebral augmentation with Nucleus Arthroplasty devices.
Subsidence tests were conducted in cadaveric motion segments

(L2-3 or L4-5). The specimens were potted in resin after the
bone mineral density (BMD) had been measured using DEXA.
Devices were implanted into the cadaveric segments according to
manufacturers recommendations. Next, the anulus and posterior elements were removed to isolate the device-endplate interface. For each specimen, compression tests were performed
starting at 800 N and increasing in 200 N increments until failure. Fluoroscopic imaging was utilized to identify initiation of
device subsidence, defined as radiographic endplate deformation
compared to the baseline image. Testing was stopped after occurrence of gross device subsidence (>2mm subsidence). The BMD
versus failure load relationship was investigated using linear
regression. The test methods utilized are considered worst case
since all load passes through the device and does not account for
the soft tissue load sharing reported in Chapter 10.
Normalized to the HydraFlex device, the authors reported the
PDN-SOLO device was 1.7 times stiffer, and the PEEK cage was
20 times stiffer. The BMD was similar for all test groups with the
device subsidence load correlating with BMD. Loads to initiate
subsidence utilizing fluoroscopic imaging demonstrated not only
higher loads for the HydraFlex device compared to the PDNSOLO and PEEK devices (3200 1224 N versus 2050 191 N,
and 2700 837 N, respectively), but also differing failure mechanisms (Figure 5). It is hypothesized that the lower stiffness of the
Interbody nuclear replacement is emerging as an alternative to

interbody fusion in the treatment of degenerative disc disease
(DDD). Device subsidence is an inherent clinical risk of any
interbody device. However, despite promising clinical results
with nucleus replacement devices, concerns related to endplate
remodeling or endplate subsidence exist.
Beaubien, et al18 evaluated the cadaveric subsidence characteristics
of two nucleus replacement devices (PDN-SOLO and
HydraFlex, Raymedica, LLC, Minneapolis, Minnesota) to a
PEEK interbody fusion device. The PDN-SOLO nucleus replacement device has the largest and longest worldwide human clinical
experience. PEEK interbody devices have demonstrated good long
clinical outcomes for interbody fusion. The HydraFlex device is
the latest product offering by Raymedica, LLC and has several
Radiographic failure load (N)
Endplate Subsidence Characteristics

of Interbody Devices
4000
3000
2000
1000
0
HydraFlex
PDN-SOLO
PEEK
Figure 5
Load at initiation of subsidence of 3 interbody devices.
39
HydraFlex device allowed for more device deformation with

increasing loads, improving endplate conformity and effectively
distributing endplate stress, resulting in a less pronounced initial
endplate failure and slow progression of subsidence with initial
endplate failure noted approximately 600-800 N prior to gross
(>2mm) device subsidence. Conversely, subsidence of the high
stiffness PEEK spacer occurred simultaneously with fracture of
the endplates.
The results from this study demonstrated that the HydraFlex
device has greatly improved the in vitro subsidence performance.
While the intended function of the Nucleus Arthroplasty devices
and a PEEK spacer are decidedly different, comparisons between
their subsidence characteristics offers insight into the expected
clinical performance.
THE CASCADE OF INTERVERTEBRAL DISC DEGENERATION INFLUENCES THE BIOMECHANICAL INTEGRITY

OF THE ENDPLATE.
REFERENCES
1. Grant JP, Oxland TR, and Dvorak MF. Mapping the structural properties
of the lumbosacral vertebral endplates. Spine 2001;26:889-96.
2. Roberts S, Menage J, and Urban JP. Biochemical and structural properties
of the cartilage endplate and its relation to the intervertebral disc. Spine
1989;14:166-74.
3. Ayotte DC, Ito K, and Tepic S. Direction-dependent resistance to flow in the
endplate of the intervertebral disc: an ex vivo study. J Orthop Res.
2001;19:1073-7.
4. Rajasekaran S, Babu JN, Arun R, Armstrong BR, Shetty AP, and Murugan S.
ISSLS prize winner: A study of diffusion in human lumbar discs: a serial magnetic resonance imaging study documenting the influence of the endplate on
diffusion in normal and degenerate discs. Spine 2004;29:2654-67.
5. Benneker LM, Heini PF, Alini M, Anderson SE, and Ito K. 2004 Young
Investigator Award Winner: vertebral endplate marrow contact channel
occlusions and intervertebral disc degeneration. Spine 2005;30:167-73.
6. Grant JP, Oxland TR, Dvorak MF, and Fisher CG. The effects of bone density
and disc degeneration on the structural property distributions in the lower
lumbar vertebral endplates. J Orthop Res 2002;20:1115-20.
7. Guerin, H. L., Baxter S, Nguyen A, Elliott, D. M., Villaraga ML, and Kurtz S.
Human intervertebral disc cartilaginous endplate biphasic mechanical properties. Trans of Orthop Res Soc, 1135. 2007.
8. Perey O. Fracture of the vertebral endplate in the lumbar spine; an experimental biochemical investigation. Acta Orthop Scand.Suppl 1957;25:1-101.
9. Oxland TR, Grant JP, Dvorak MF, and Fisher CG. Effects of endplate removal
on the structural properties of the lower lumbar vertebral bodies. Spine
2003;28:771-7.
CONCLUSION
The cascade of intervertebral disc degeneration influences the
biomechanical integrity of the endplate. One clinical concern
with interbody devices is endplate failure. Intra-operative factors
to minimize risk of interbody subsidence are implant sizing, endplate preparation and quality of the host bone. For fusion and
total disc arthroplasty devices, using the largest size device that
can be safely implanted engages more of the apophyseal rim and
reduces the segment displacement under compressive loads. In
side-by-side testing of a PEEK fusion device versus the HydraFlex
Nucleus Arthroplasty device, the HydraFlex device demonstrated
higher loads required to initiate endplate subsidence. For patients
at risk of developing osteoporosis, initial biomechanical and clinical results from vertebral body augmentation used to ameliorate
interbody device subsidence and catastrophic failure are promising. However, additional clinical studies of these technologies and
techniques are needed to determine if they are viable.
10. Ochia RS, Tencer AF, and Ching RP. Effect of loading rate on endplate and
vertebral body strength in human lumbar vertebrae. J Biomech.
2003;36:1875-81.
11. Hansson TH, Keller TS, and Spengler DM. Mechanical behavior of the
human lumbar spine. II. Fatigue strength during dynamic compressive loading. J Orthop Res 1987;5:479-87.
12. Meakin JR, Redpath TW, and Hukins DW. The effect of partial removal of
the nucleus pulposus from the intervertebral disc on the response of the
human anulus fibrosus to compression. Clin.Biomech.(Bristol., Avon.)
2001;16:121-8.
13. Seroussi RE, Krag MH, Muller DL, and Pope MH. Internal deformations of
intact and denucleated human lumbar discs subjected to compression, flexion, and extension loads. J Orthop Res 1989;7:122-31.
14. Stokes IA and Iatridis JC. Mechanical conditions that accelerate intervertebral
disc degeneration: overload versus immobilization. Spine 2004;29:2724-32.
15. Lowe TG, Hashim S, Wilson LA et al. A biomechanical study of regional endplate strength and cage morphology as it relates to structural interbody support. Spine 2004;29:2389-94.
16. Auerbach JD, Ballester CM Hammond F, Carine ET, Balderston RA, and
Elliott DM. Effect of implant size on lumbar vertebral biomechanics in total
disc replacement. Trans of Orthop Res Soc, 324. 2007.
17. Auerbach JD, Yoder JH, Maurer PM, Erbe EM, Entrekin D, Balderston RA,
Bertagnoli R, and Elliott DM. Vertebral body augmentation with Cortoss
improves compression biomechanics for lumbar disc arthroplasty. Proceedings
of the International Society for the Study of the Lumbar Spine. 2007.
18. Beaubien BP, Freeman AL, Sherman TN, Gullickson AL, and Seme SJ.
Subsidence characteristics of two nucleus replacement devices compared to
a PEEK interbody fusion device. Proceedings of the 7th Annual Spine
Arthroplasty Society Meeting. 2007.
40
Chapter 12
Repair of the Anulus Fibrosus

of the Lumbar Disc
Steven L. Griffith, PhD

VICE PRESIDENT, SCIENTIFIC AFFAIRS
Anulex Technologies Inc.

Minnetonka, MN 55343

Baltimore Neuro Surgical Associates, PA

Baltimore, MD 21204

ORTHOPEDIC RESEARCH
Emory University
Atlanta, GA 30322
HERNIATED INTERVERTEBRAL DISCS AND DEGENERATIVE

DISC DISEASE (DDD) ARE TWO DISTINCT BUT RELATED CONDITIONS. THEY BOTH CAN BE RESPONSIBLE FOR SYMPTOMS
THAT ULTIMATELY REQUIRE SURGICAL TREATMENT.
KEY POINTS
The anulus fibrosus has many functions, one of which is to contain
the intradiscal material (nucleus pulposus or prostheses).
There are variations in both the biomechanics of the lumbar disc
and associated pathology as a result of mechanical functions and
failures of the anulus fibrosus that can vary circumferentially.
Advanced techniques to surgically close the anulus fibrosus
have been developed that facilitate its repair.
Anular closure may be important in the surgical treatment of
the intervertebral disc.
41
INTRODUCTION
erniated intervertebral discs and degenerative disc disease

(DDD) are two distinct but related conditions. They both can
be responsible for symptoms that ultimately require surgical treatment. Because of its location and function, the anulus fibrosus of
the intervertebral disc is a key determinant of outcomes with certain potential surgical solutions, yet the anulus has often been
viewed as inconsequential in many of these surgical treatments.
The purpose of this chapter is to emphasize the significance of
the anulus by describing its anatomy and structure, its biomechanical and physiological functions, and its role in symptomproducing pathology. Attempts to preserve, repair, reinforce, or
heal the anulus are discussed, both as a focused treatment strategy and in combination with other surgical techniques such as
nucleus replacements.
THE INTERVERTEBRAL DISC, IN PARTICULAR, IS A VITAL

STRUCTURE OF HIGHLY SPECIALIZED AND ORGANIZED
DEFORMABLE CARTILAGE WITHOUT WHICH THE SPINE
WOULD BE INCAPABLE OF SUPPORTING LOADS OR
MAINTAINING FLEXIBILITY.
FORM AND FUNCTION OF THE ANULUS FIBROSUS

Central to our ability to function in an upright stature is the
structure of the multi-segmental spinal column that both supports our bodies and allows freedom to move. Many anatomical
tissues are important in maintaining and generating the overall
mobility, flexibility, and stability of the spine. The intervertebral
disc, in particular, is a vital structure of highly specialized and
organized deformable cartilage without which the spine would
be incapable of supporting loads or maintaining flexibility. In
the healthy intervertebral disc, the nucleus pulposus in the central portion of the structure is capable of sustaining and transmitting load while the outer anulus fibrosus surrounding the
nucleus provides both resistance to stresses and movements of
the spine. The combination of the nucleus pulposus and the
anulus fibrosus working together in response to imposed loading
conditions is important to the overall function of the spine.
Pathology, injury, and surgery can disrupt the form and function
of the intervertebral disc, and more specifically the anulus fibrosus. Understanding the structure and physiology of the anulus
fibrosus is key in ultimately providing appropriate treatment for
spinal problems with specific origin in the intervertebral disc.
42
The anulus fibrosus is a specialized dense connective tissue

composed of collagen fibrils organized in approximately 15 to
25 lamellar layers encircling the nucleus pulposus.1 The fibers of
the anulus are, in the main, made of both non-extensible Type I
collagen and elastin.2 The fibrils in the outer edges of the anulus
are highly organized and provide much of the tensile strength
needed to contain disc material. Each fiber layer is organized as
collagen fibrils at alternating 70 degree angles relative to each
other and this geometric configuration allows for resistance to
imposed passive stresses (e.g., torsion, axial compression,
intradiscal pressure, etc.). The forces acting on the nucleus pulposus causes an overall radial expansion of the anulus.3, 4 The
transition area between the highly organized anulus layers and
the gelatinous nucleus pulposus is characterized by wispy fibrocartilage. Because of the anulus fibrosus function of retaining
the nucleus pulposus through its tensile strength and load
resistance, the outer layers of the anulus have the highest collagen content with the lowest water and proteoglycan content relative to the rest of the disc.5 In contrast, the nucleus pulposus is
composed primarily of water-binding proteoglycans and Type II
collagen.6 These hydrophilic glycoproteins in the nucleus are
responsible for the hydration function of the disc.
Although sparsely distributed, elastic fibers are distributed in
the deeper anular layers and attach to the endplates. Also, elastic
fibers in the outer anulus rim attach to the endplates of adjacent
bony vertebrae and are referred to as Sharpeys fibers; these
micro-structural tissue elements are well demonstrated with
high-resolution magnetic resonance imaging.7 Unlike collagen
fibers, elastic fibers can be stretched and when released can return
to their original length. The elastic fibers in the anulus may be
important in maintaining the anulus in its undistended position.2
There are other ligamentous tissues in close proximity to the anulus fibrosus that play a role in stabilizing the spinal segment. On
the ventral side of the spine, the anterior longitudinal ligament
(ALL) extends between adjacent vertebral bodies.8, 9 The longitudinally-oriented collagen fibers of the ALL resist extension loading.
The posterior longitudinal ligament (PLL) is situated on the dorsal
aspect of the intervertebral disc spanning adjacent vertebral bodies.
A sizable number of elastic fibers have been seen in the ALL as well
as in the interlaminar ligamentum flavum2 and these structures
probably assist in the maintenance of an erect posture. Schendel et
al10 demonstrated that the ALL and PLL function predominantly to
resist extension and flexion loading, respectively.
Biomechanics of the Anulus Fibrosus

Regional and circumferential variations in the structure and function of the ring-like structure of the anulus fibrosus have been
described. At the tissue level, Ebara et al11 studied the positional
dependence of the tensile properties of human anulus fibrosus
and concluded that the mechanical role of the anulus fibrosus
varies widely with the position around the disc. Specifically, the
lower modulus and failure stress in the posterolateral region compared with the anterior region was suggestive of one mechanism
by which preferential failure of the posterior anulus fibrosus
occurs resulting in disc herniation. In a similar study, Best et al12
showed that the anulus is inhomogeneous with both regional
(anterior vs. posterior) and radial differences (inner vs. outer) in
mechanical material properties and biomechanical composition.
At a functional level, Edwards et al13 examined the distribution of
stresses within the lumbar disc under different loading condition.
Their data suggested that for all loading conditions, the largest
stresses occurred in the anulus rather than the nucleus and furthermore, the posterolateral anulus saw the highest peak stress
during extension-compression bending.
Its apparent that the local biomechanics of the disc and the anulus fibrosus are such that the posterior lateral anulus nearest the
intervertebral neuroforamen is vulnerable and is thus more susceptible to tissue injury than other areas around the disc. The
regions of high pressure in the posterior anulus correlate well
with the common locations for herniated discs needing repair.14, 15
This data also suggests that the anterior and lateral portions
of the anulus are less inclined to herniation due to its relative
increased thickness and tensile strength and are probably easier
to repair should it be needed.
Role of the Anulus Fibrosus in Lumbar Disc Pathology

The anulus fibrosus undergoes age-related changes and is affected
during the cascade of events leading to degenerative spinal diseases.16, 17, 18 Frank mechanical failure of the anulus fibrosus itself,
independent of collapsing degeneration, can manifest as anular
tears, fissures, and ultimately herniations.19, 20, 21 Defects such as
tears, fissures, weakening, or incompetence in the anulus fibrosus
are major contributors to symptomatic pathology whose origin is
the spinal disc.22, 23, 24 Both circumferential delamination and radial
fissures in the anulus fibrosus can result in extrusion of otherwise
contained nucleus pulposus tissue.25 Particularly in the posterior
region, disc protrusions or fragments of nucleus pulposus can
cause pain and other radicular symptoms as they displace the
nerve root exiting from the dural sac. There have also been
DURING THE PROCESS OF DISC COLLAPSE, AND AS THE

INTERVERTEBRAL DISC OVERALL LOSES ITS ABILITY TO
SUSTAIN THE LOADING IMPOSED UPON IT BECAUSE OF
NUCLEUS DEHYDRATION AND TISSUE DESTRUCTION, THE
ANULUS ULTIMATELY BEGINS TO BEAR MUCH OF THE
LOAD THAT WOULD ORDINARILY BE BORNE BY THE
NUCLEUS PULPOSUS.
chemical mediators identified from this intradiscal tissue that

may play a role in pain production and inflammation.26, 27
The PLL also has its own pathophysiological role. This ligament
is intimately associated with the posterior fibers of the anulus
fibrosus and it is often difficult to clinically determine herniations that are contained below the PLL or have migrated.28
Ohshima et al29 have suggested morphologic variations in the
PLL can be related to the various types of disc herniation. In
their study, the size of the PLL diminished significantly the lower
the disc level and there was a loose attachment of the central
rhomboidal expansion of the PLL. These anatomic features correlate well with the common posterior-lateral extrusion seen in
the L4-L5 and L5-S1 area.
In contrast to herniated discs that are a result of anular defects
specifically, destructive forces of the overall functional spinal unit
can lead to degenerative disc disease (i.e., disc collapse).30, 31 During
the process of disc collapse, and as the intervertebral disc overall
loses its ability to sustain the loading imposed upon it because of
nucleus dehydration and tissue destruction, the anulus ultimately
begins to bear much of the load that would ordinarily be borne by
the nucleus pulposus. Adams et al32 has shown that as a result of
degeneration, structural changes in the anulus and endplate lead
to a transfer of load from the nucleus to the posterior anulus, in
particular. In response to these added stresses, there is a concomitant broadening of the anular wall as it attempts to carry the
increased load often resulting in significant protrusion. This anular protrusion manifests as either circumferential disc bulging or
focal herniation (when isolated in the posterior disc and spinal
canal area). Tissue remodeling, often in the posterior column, is
often accompanied by calcification of the anulus, osteophyte formation in the areas of Sharpey's fibers, and other detrimental
characteristics that pose a significant challenge during treatment.
Historically, surgical treatment to reverse degenerative disc collapse
has included re-establishment of disc height and geometry while
removing flexibility (i.e., arthrodesis). But more recently, surgical
43
techniques have begun to move away from those that eliminate

spinal segmental motion to those that retain or restore motion.
Non-fusion techniques with a focus at the disc, frequently the
source of pain,33 include replacing the total disc (i.e., nucleus pulposus and a significant portion of the anulus fibrosus) with a prosthetic implant,34, 35 or alternatively, replacing the central nucleus
pulposus while retaining a significant portion of the anulus.36, 37, 38
It is important to consider the role of the anulus fibrosus in the
latter, less-invasive surgical technique of nucleus pulposus
replacement for treatment of mild to moderate degeneration.
Surgical Considerations and

Healing Response of the Anulus Fibrosus
In addition to naturally-occurring defects in the anulus, surgical
techniques can also iatrogenically affect the ultimate result.
An incision through the anulus can alter the mechanics of the
involved disc level and the amount of nucleus removal or lack of
replacement can affect post-surgery biomechanics and ultimately
clinical outcome.39 Until nucleus replacement technologies are
available for commercial use, less-aggressive nucleus pulposus
removal generally is perceived to be advantageous in mitigating
resultant low back pain,40 presumably due to disc collapse. For
treatment of pathology requiring more aggressive nucleus
removal, nucleus replacement devices implanted through a controlled anular incision or flap, combined with anular fibrosus
repair or reinforcement, should be a relatively intuitive goal.
Differences in anulotomy techniques and reparative techniques
have been investigated by only a few authors. More than 20 years
ago, Hampton et al41 showed that posterior oriented anular incisions had limited healing potential and that the persistent defect
could provide a pathway for irritating nuclear fluid escape onto
perineural tissue. Further refinement to this thought was made
by Ahlgren et al42 and Ethier et al43 who used animal models to
suggest that a box-type or window anulotomy, without direct
repair, produced a significantly weaker healing response than
either a slit or a cruciate incision. Furthermore, these studies suggested that a trocar technique might reasonably preserve the anular tissue as a result of stretching rather than lacerating. In any of
these situations, histologic evidence of direct primary healing of
the anulus fibrosus could not be demonstrated, rather a superficial fibrous tissue capping was noted, particularly after a fissure
produced with the trocar technique.
Ahlgren44 later reported on the healing potential of the anulus
after direct suture repair of defects that simulated those that
44
might be seen or made at the time of surgery. In this study, it

was difficult to ascertain whether or not any particular repair
technique that was used changed the rate or strength of the anular healing response. They concluded that preservation of as
much anular tissue as possible should be the goal during surgery.
Although repairing the anulus fibrosus has seemed intuitive
to many, until recently only a few surgeons have attempted this
difficult task particularly given the limitations of the current
microscopic and endoscopic techniques used in many surgical
approaches. Yasargil45 reported on using a 7-0 suture to close the
anulus after nucleus removal during microdiscectomy claiming
the benefit of future adhesion prevention. Lehmann et al46
reported on a technique to close the posterior longitudinal ligament, peridural membrane, and outer anular layers with 4-0
silk. He suggested that a greater percentage of patients who
were sutured had less post-surgery pain than those not sutured.
Cauthen,47, 48 studied his own group of patients in whom he
carefully microsutured the anulus after microdiscectomy and
suggested over a 50% reduction in the recurrent herniation rate.
Improvement in techniques to close the anulus fibrosus are now
being suggested as a result of advanced engineering. Placement
of intradiscal prostheses that are contained by the anulus fibrosus can also benefit from advanced closure techniques. Since
there are circumferential variances in the structure and function
of the anulus, optimizing the repair techniques based on the
type of defect, the location of the closure site, and the desired
outcome should be considered. Two distinctive repair methods
are presented below.
Advanced Closure Methods

for Repair of the Anulus Fibrosus
In the past, surgeons have considered and tried suturing the anulus closed using standard suture, needles, and microsurgical techniques. In a posterior approach, exposure limitations during a
microdiscectomy make standard suturing difficult to perform.
Often the relatively thin posterior anulus (as compared to the
anterior or lateral anulus) does not have sufficient integrity to
hold the sutures, particularly as the needles pass through small,
thin tissue flaps. Two experimental data sets are described below,
each using a different anular closure technique. The first approach
uses standard suturing techniques with curved surgical needles.
The second approach uses an advanced system that allows placement of tissue anchors away from the defect into stronger, thicker
tissue; these tissue anchors are connected with a tension band
Figure 1
Anular Incisions Used in Suture Experiments. An anterior lateral anular
flap incision created immediately lateral to the anterior longitudinal ligament (ALL) with the flap opening laterally (Panel A). An anular incision
made horizontal to the disc space is shown in Panel B.
Figure 2
Anulus Fibrosus Suture Repairs. Direct suture repair of an anular flap (Panel A)
or a horizontal incision (Panel B).
capable of drawing together tissue surrounding a defect; using a

pre-tied knot eliminates this sometimes difficult manual step.
at a rate of 1mm/sec. Comparison of pull-out loads was made

between anular incisions without suture repair (Figure 1) and
after suture repair (Figure 2). Anular repair was accomplished
using a simple running or simple interrupted suture technique
using 2-0 suture. When a slit incision was used, repairing the anulus with sutures increased the PDN-SOLO extraction force 56%
+/- 36%. And when a flap incision was used, the extraction force
increased 63% +/- 14% after suture repair. These results suggest
that directly repairing either the slit or the flap incision with
sutures in the anterior-lateral anulus had a positive effect (i.e., a
higher extraction force) on retention of the nucleus replacement
device, thus reducing the potential risk of device migration.
Nucleus replacement devices may benefit from closing the anulus

fibrosus by increasing resistance to migration. Simple suture closure of the antero-lateral anulus in combination with a nucleus
replacement has been studied in the laboratory using cadaveric
spines.49 The purpose of these experiments was to evaluate if
direct suture repair of the anterior anulus could affect nucleus
replacement device pullout forces. Each disc from functional
spinal units (n=8) was tested in the following manner to evaluate
two different types of anular incisions and subsequent suture closure. On one side of the disc, a flap opening was made laterally
using an anterior lateral retroperitoneal approach similar to that
which could be used clinically.50 On the opposite side of each disc,
a single slit incision was made. In each case, sufficient nucleus
material was removed to allow placement of PDN-SOLO devices
(Raymedica, LLC, Minneapolis, Minnesota) using the manufacturers recommendations. A stainless steel wire was connected to
the implant allowing controlled extraction from the disc space
Direct manual suture closure of the anulus fibrosus, although

tedious and time-consuming with standard surgical needles and
sutures, has been shown to be beneficial in both animal studies
and clinical experience. Further engineering refinements in techniques and tools have allowed anular closure to become a quick
and repeatable reality. One such commercially-available product
is described below.
Figure 3
Xclose Tissue Repair System (Anulex Technologies, Minnetonka,
MN). Panel A shows a surface view of two tension bands in a cruciate
repair pattern. Panel B shows one tension band in cross-section.
45
Figure 4
Repair and Tissue Re-Approximation Patterns. The Xclose system can be used to effect a repair using either
parallel tension bands (A) or a cruciate pattern (B).
The Xclose Tissue Repair System (Anulex Technologies Inc,

Minnetonka, Minnesota) is a commercial product that can be
used to reapproximate soft tissue such as the anulus fibrosus.
The system places tension bands that are held in place with tissue T-anchors (Figure 3). Each tension band is comprised of a
pair of non-absorbable braided suture loops each attached to a
T-anchor. These loops are then connected together with another
loop of suture; a pre-tied knot on this connecting loop is used to
facilitate tightening of the tension band, drawing the construct
together and thereby closing the tissue. All implanted components of this construct are made from polyethylene terephthalate
(PET). The device is provided sterile and preloaded on disposable delivery instruments. The Xclose system has been cleared
for use by the Food and Drug Administration and is indicated
for use in soft tissue approximation for procedures such as general and orthopedic surgery.
An Xclose system kit consists of two tension band devices and
one knot pusher. To reapproximate the anulus, surgeons have the
option of using a cruciate pattern or a parallel pattern using the
two tension bands (Figure 4). On each tension band delivery
tool, there is a tissue stop 12mm from the needle tip that facilitates appropriate depth placement of the T-anchors in and
around the transition zone between the anulus and nucleus pulposus (Figure 5A & B). Once the T-anchors are in place, the
46
defect is closed by advancing the pre-tied knot using the knot

pusher cinching the lines together with the T-anchors providing
resistance (Figure 5C).
The mechanical integrity of repairing the posterior anulus fibrosus with the Xclose system has been studied in cadaveric experiments.51 In this study, the systems ability to remain intact and in
place after cyclic loading and thus providing an effective tissue
closure was determined. After partial laminectomy, posterior
anular incision, and limited nucleus removal (~ 0.2cc), the anulus of each functional spinal unit tested (n=9) was repaired with
a single Xclose tension band. Periodic video imaging was performed during cyclic fatigue loading consisting of 20K cycles of
flexion-extension, 15K cycles of lateral bending, and 5K cycles of
axial rotation. Calibrated digital macrophotography was also
used to determine the systems ability to retain tissue re-approximation and closure of the defects; measurement of knot slippage
(indicative of system loosening) was used to assess repair
integrity. In this study, all anulotomy defects were effectively
closed and remained closed after 40K cycles. Upon completion
of cyclic loading, the Xclose system loosening was minimal with
average knot slippage of 0.8 +/- 0.7mm and the system did not
migrate out of the anulus or pull through the tissue. Thus, tension in the system was maintained throughout testing and further confirmed when it was manually cut at the conclusion of
Figure 5
Placement of the Xclose Tssue Repair System. The Xclose delivery needle after placement of the first t-anchor is shown
in Panel A. The needle insertion of the second t-anchor is shown in Panel B; needle depth is gauged by a tissue stop 12mm
from the needle tip. A disposable knot pusher (included in the system kit) is used to remove slack from the system (Panel
C), thus tensioning the device until the tissue is re-approximated.
the experiment. These experiments suggested that the Xclose

system can effectively reapproximate posterior anulus fibrosus
tissues and this closure remains after cyclic loading.
REFERENCES
1. Marchand F and Ahmed AM. Investigation of the laminate structure of
lumbar disc anulus fibrosus. Spine 15(5):402-410, 1990.
2. Johnson EF, Chetty K, Moore IM, Stewart A, Jones W. The distribution and
arrangement of elastic fibers in the intervertebral disc of the adult human.
J Anat 135(2):301-309, 1982.
CONCLUSION
The anulus fibrosus of the intervertebral disc is an important
structure that plays a significant role in spinal pathology and its
treatment. Discectomy surgery is generally considered a successful standard of care today to treat herniated intervertebral discs.
Nonetheless, surgically repairing defects in the anulus in conjunction with and at the conclusion of discectomy procedures
may enhance both the technical and clinical outcomes after these
procedures. Nucleus Arthroplasty technologies are being evaluated for treatment of patients requiring more aggressive nucleus
removal. Their primary objectives are to maintain segmental
motion while preventing disc space collapse typically associated
with aggressive nucleus removal. Cadaveric testing has demonstrated the potential benefit of an effectively repaired anular
incision. Thus, the structure and function of the anulus fibrosus
should most certainly not be overlooked during surgical treatments involving the intervertebral disc. The Xclose tissue repair
system is an advanced system that provides a safe and simple
anular closure technique.
3. Brinckmann P, Horst M. The influence of vertebral body fracture, intradiscal

injection, and partial discectomy on the radial bulge and height of human
lumbar discs. Spine 10(2):138-145, 1985.
4. Brinckmann P and Grootenboer H. Change of disc height, radial disc bulge,
and intradiscal pressure from discectomy- An in vitro investigation on human
lumbar discs. Spine 16(6):641-646, 1991.
5. Urban J. Biochemistry:disc biomechistry in relation to function. In:Wiesel SW,
Weinstein JN, Herkowitz H, et al, eds. The Lumbar Spine, 2nd edition.
Philadelphia:WB Saunders, 285-310, 1996.
6. Schollmeier G, Lahr-Eigen R, Lewandrowski KU. Observations on fiberforming collagens in the anulus fibrosus. Spine 25(2):2736-2742, 2000.
7. Schiebler ML, Grenier N, Fallon M, Camerino V, Zlatkin M, Kressel HY.
Normal and degenerated intervertebral disk: In vivo and in vitro MR imaging
with histopathologic correlation. AJR 157:93-97, 1991.
8. Neumann P, Keller T, Ekstrom L, Hult E, Hansson T. Structural properties of
the anterior longitudinal ligament- Correlation with lumbar bone mineral
content. Spine 18(5):637-645, 1993.
9. Neumann P, Keller TS, Ekstrom L, Perry L, Hansson TH, Spengler DM.
Mechanical properties of the human anterior longitudinal ligament. J Biomech
25(10):1185-1194, 1992.
10. Schendel MN, Wood KB, Buttermann GR, Lewis JL, Ogilvie JW. Experimental
measurement of ligament force, facet force, and segment motion in the
human lumbar spine. J Biomech 26(4-5):427-438, 1993.
11. Ebara S, Iatridis J, Setton L, Foster RJ, Mow VS, Weidenbaum M. Tensile
properties of nondegenerate human lumbar anulus fibrosus. Spine 21(4):452461, 1996.
47
12. Best BA, Guilak F, Setton LA, Zhu W, Saed-Nejad F, Ratcliffe A, Weidenbaum M,
Mow VC. Compressive mechanical properties of the human anulus fibrosus and
their relationship to biomechemical composition. Spine 19(2):212-221, 1994.
33. Kuslich SD, Ulstrom CL, Michael CJ. The tissue origin of low back pain and sciatica: A report of pain response to tissue stimulation during operation on the
lumbar spine using local anesthesia. Orthop Clin North Am 22:181-187, 1991.
13. Edwards WT, Ordway NR, Zheng Y, McCullen G, Han Z, Yuan HA. Peak
stresses observed in the posterior lateral anulus. Spine 26(16):1753-1759, 2001.
34. Guyer RD, Ohnmeiss DD. Intervertebral disc prostheses. Spine 28(Suppl 15):
S15-S23, 2003.
14. Knop-Jergas BM, Zucherman JF, Hsu KY, DeLong B. Anatomic position of
a herniated nucleus pulposus predicts the outcome of lumbar discectomy.
J Spine Disord 9(3):246-250, 1996.
35. Bertagnoli R, Kumar S. Indications for full prosthetic disc arthroplasty:

A correlation of clinical outcome against a variety of indications. Eur Spine
J 11:5131-5136, 2002.
15. Spangler E. The lumbar disc herniationa computer-aided analysis of 2,504

operations. Acta Orthop Scand Suppl 142:1-95, 1972.
36. Bao QB, Yuan HA. New technologies in spine: Nucleus replacement. Spine
27:1245-1247, 2002.
16. Farfan HF, Huberdeau RM, Dubow HI. Lumbar intervertebral disc degeneration. J Bone Joint Surg 54-A(3):492-510, 1972.
37. Carl A, Ledet E, Yuan H, et al. New development in nucleus pulposus

replacement technology. Spine J 4(Suppl 6):S325-S329, 2004.
17. Videman T, Nurminen M. The occurrence of anular tears and their relation to
lifetime back pain history: A cadaveric study using barium sulfate discography.
Spine 29:23):2668-2676, 2004.
38. Klara PM, Ray CD. Artificial nucleus replacement. Spine 27:1374-1377, 2002.
18. Nerlich AG, Schleicher E, Boos N. Immunohistological markers for age-related

changes of human lumbar intervertebral disc. Spine 22(24):2781-2795, 1997.
39. Mochida J, Nishimura K, Nomura T, Toh E, chiba M. The importance of

preserving disc structure in surgical approaches to lumbar disc herniation.
Spine 21(13):1556-1563, 1996.
19. Coventry MN, Ghormley RK, Kernohan JW. The intervertebral disc: Its
microscopic anatomy and pathology: Part II. Changes in the intervertebral
disc concomitant with age. J Bone Joint surg 27:233-247, 1945.
40. Carragee EJ, Spinnickie AO, Alamin TF, Paragioudakis S. A prospective controlled study of limited versus subtotal posterior discectomy: Short-term outcomes in patients with herniated lumbar intervertebral discs and large
posterior anular defect. Spine 31(6):653- 657, 2006.
20. Hickey DS, Hukins DWL. Relation between the structure of the anulus fibrosus
and the function and failure of the intervertebral disc. Spine 5:106-116, 1979.
41. Hampton D, Laros G, Mccarron R, Franks D. Healing potential of the anulus

fibrosus. Spine 14(4):399-401, 1989.
21. Osti OA, Vernon-Roberts B, Fraser RD. Anulus tears and intervertebral disc
degeneration. Spine 15:762-767, 1990.
42. Ahlgren BD, Vasavada V, Brower RS, Lydon C, Herkowitz HN, Panjabi MM.
Anular incision technique on the strength and multidirectional flexibility of
the healing intervertebral disc. Spine 19(8):948-954, 1994.
22. Farfan HF, Huberdeau RM, Dubow HI. Lumbar intervertebral disc degeneration. J Bone Joint Surg 54-A(3):492-510, 1972.
23. Brinckmann P and Porter RW. A laboratory model of lumbar disc protrusion. Fissure and fragment. Spine 19(2):228-235, 1994.
43. Ethier DB, Cain JE, Yaszemski MJ, Glover JM, Klucznik RP, Pyka RE,
Lauerman WC. The influence of anulotomy selection on disc competence.
Spine 19(18):2071-2076, 1994.
24. Hadjipavlou AG, Simmons JW, Pope MH, Necessary JT, Goel VK.
Pathomechanics and clinical relevance of disc degeneration and annular tear:
a point-of-view review. Am J Orthop 28(10):561-571, 1999.
44. Ahlgren BD, Lui W, Herkowitz HN, Panjabi MM, Guiboux JP. Effect of anular
repair on the healing strength of the intervertebral disc- a sheep model. Spine
25(17):2165-2170, 2000.
25. Kim Y. Prediction of peripheral tears in the anulus of the intervertebral disc.
25(14):1771-1774, 2000.
45. Yasargil MG. Microsurgical operation of herniated lumbar disc. Adv

Neurosurg 4:81, 1977.
26. Kawakami M, Tamaki T, Hayashi N, Hashizume H, Nishi H. Possible mechanism of painful radiculopathy in lumbar disc herniation. Clin Orthop Relat
Res 351:241-251, 1998.
46. Lehmann TR, Titus MK. Refinements in technique for open lumbar discectomy. Proceedings of the International Society for the Study of the Lumbar
Spine (ISSLS) June 1997.
27. Specchia N, Panotta A, Toesca A, et al. Cytokines and growth factors in the protruded intervertebral disc of the lumbar spine. Eur Spine J 11:1145-1151, 2002.
47. Cauthen J. Microsurgical annular reconstruction (annuloplasty) following lumbar microdiscectomy: preliminary report of a new technique. Proceedings of the
AANS/CNS Joint Section on Spine and Peripheral Nerves. Orlando, FL, 1999.
28. Dellerud R, Johansen JG, Johnsen UL, Magnaes B. Differentiation between

contained and noncontained lumbar disk hernia by CT and MR imaging.
Acta Radiol 36(5):491-496, 1995.
29. Ohshima H, Hirano N, Osada R, Matsui H, Tsuji H. Morphologic variation
of lumbar posterior longitudinal ligament and the modality of disc herniation. Spine 18(16):2408-2411, 1993.
30. Harris RI and MacNab I. Structural changes in the lumbar intervertebral
discs- Their relationship to low back pain and sciatica. J Bone Joint Surg 3bB(2):304-322, 1154.
31. Yong-Hing K, Kirkaldy-Willis WH. The pathophysiology of degenerative disease of the lumbar spine. Orthop Clin North Am 14(3):491-504, 1983.
32. Adams MA, McNally DS, Dolan P. 'Stess' distributions inside intervertebral discs.
The effects of age and degeneration. J Bone Joint Surg Br 78:965-972, 1996.
48
48. Cauthen J. Chapter 11: Microsurgical anular reconstruction (anuloplasty)

following lumbar microdiscectomy. In: Guyer RD, Zigler JE, editors. Spinal
Arthoplasty: A New Era in Spine Care. St. Louis MO: Quality Medical
Publishing. pp 157-177, 2005.
49. Hutton W, Davis R. Cadaver evaluation of PDN-SOLO device pullout with
direct suture repair. Unpublished, internal data, PDE #377, Raymedica, LLC,
Minneapolis, MN, 2006.
50. Bertagnoli R, Vazquez RJ. The Anterolateral TransPsoatic Approach (ALPA): a
new technique for implanting prosthetic disc-nucleus devices. J Spinal Disord
Tech 16(4):398-404, 2003.
51. Bourgeault C. Beaubien BP, Freeman AL, Bentley I, Houfburg R, Griffith SL.
Biomechanical asessement of anulus fibrosus repair with suture-tethered tissue anchors. Prooceedings of the 2007 Annual Meeting of the Spine
Arthroplasty Society (SAS7), Berlin Germany, May 2007.
Chapter 13
Integration of Nucleus
Arthroplasty Technology
into the Continuum of Care
ASSISTANT PROFESSOR

New York, NY 10021
Andrew A. Sama, MD
ASSISTANT PROFESSOR OF ORTHOPEDIC SURGERY
Weill Medical College of Cornell University

New York, NY 10021


New York, NY 10021
Dr. med. univ. Rudolf Bertagnoli

FOUNDER
Pro-Spine Medical Consulting

Straubing, Germany 94315
INTRODUCTION
here is a fine line between what is called degenerative disc

disease (DDD) and simply the normal aging process of
individuals. In the near future, with the assistance of gene
expression and molecular markers, this line may be able to be
drawn more accurately.
Degenerative disc disease, as the name implies, is a syndrome of
symptoms primarily related to pain caused by some sort of biomechanical or biochemical alteration of the normal function
and physiology of the spinal column. It can cause an abnormal
interaction between the soft tissues (e.g. ligaments, capsules) and
the bony elements. This abnormal interaction might result in
49
loss of function and/or normal loading-motion coupling of the

spinal unit.
There have been several classifications and attempts, both radiographically and clinically, to describe the degenerative process.
Unfortunately, radiographic classification does not necessarily
correlate with a patients clinical symptoms nor does it correlate
a patients clinical outcomes with an attempted treatment.
However, over the last few years, more research has been done to
try to better understand the natural course of this disease, as well
as the physical, emotional, social and economical impact of this
process which could be self-limited or, on the contrary, continue
to affect the patients quality of life. Health generic and disease
specific outcome instruments have been utilized more often to
understand the starting and end points of this complex disease
and its different treatments.
DEGENERATIVE DISC DISEASE, AS THE NAME IMPLIES,

IS A SYNDROME OF SYMPTOMS PRIMARILY RELATED TO
PAIN CAUSED BY SOME SORT OF BIOMECHANICAL OR
BIOCHEMICAL ALTERATION OF THE NORMAL FUNCTION
AND PHYSIOLOGY OF THE SPINAL COLUMN.
RADIOGRAPHIC ASSESSMENT
Findings seen on plain radiographs can be helpful in determining
the stage of the disease and possible treatment options. These
include: endplate shapes (concave, convex, flat, or irregular);
endplate irregularities (presence of Schmrl nodes or cysts);
motion quantity and quality (with fluoroscopic views); alignment
and sagittal profiles. However, it is also paramount to rule-out
other associated pathologies that could confuse the diagnosis,
such as the presence of spondylolysis, spondylolisthesis or coronal
deformities. Bone scans or nuclear medicine, can be useful to ruleout these pathologies, as well as to identify particular endplate
problems or occult, benign or malignant tumors. A CT scan, alone
or used in conjunction with discography, can provide a fairly
good description of the anulus, endplates, facets, and pars interarticularis. Sagittal and coronal reformats can provide a more
detail evaluation. An MR scan can be complementary by helping
to better assess the disc height, nucleus hydration, anular tears,
endplate inflammation and subchondral bony changes.1
50
In the past, we looked at the radiographic differences between

patients with degenerative disc disease and herniated nucleus
pulposus. Using MRI, patients diagnosed with earlier stage
degenerative disc disease had primarily concave endplates.
Similarily, patients with a herniated nucleus pulposus and concave endplates demonstrated less degeneration. Conversely,
patients with a diagnosis of herniated nucleus pulposus and
flat endplates had more disc degeneration. In addition, differences in endplate concavity were noted in patients when comparing degenerated and non-degenerated levels on both MRIs
and plain radiographs. Therefore, endplate concavity seems to
be a good predictor of the level of disc degeneration.1
CLINICAL SCENARIO
When patients present to clinicians, the main underlying symptoms are generally low back pain with possible associated leg and
neurologic complaints, and disability caused by these symptoms.
By a combination of the radiographic imaging techniques discussed above, physical examination, as well as the patient history, the diagnosis of chronic low back pain secondary to
degenerative disc disease can be made. However, in addition,
there could be associated pathologies, such as psychological,
social or emotional with significant socio-economical impacts
that can affect the presentation and possible outcome of any
attempted treatment.
TREATMENT OPTIONS
The majority of patients presenting with low back pain secondary to degenerative disc disease, without any progressive or
significant neurologic findings, should first be treated by nonsurgical options. This attempt should be tried for a significant
amount of time, (three to twelve months) and could combine
strategies that involve, but are not limited to, physical therapy,
pain management, physiatry, rehabilitation, medications,
acupuncture and other modalities. The majority of such
patients will benefit from these therapies and may never require
surgical intervention. However, there is a small percentage that
will not be helped with these techniques. Therefore, once all
these conservative modalities have been tried and there has
been no improvement in quality of life due to pain, surgical
intervention becomes an option.
Until recently, spinal fusion was the only option for the treatment
of this population, regardless of disease stage. However, this surgery has several associated factors. Many patients will try to avoid
this fearful surgical procedure and try to live with the pain for
several years. This may entail the need for significant narcotic medication that, as we know, can affect their pain perception. This
chronic pain syndrome at some point could become an irreversible
situation. In addition, when a patient is excluded from their obligations and responsibilities for a long period of time, their likelihood
of returning to their previous level is not high. This creates a significant social and economical impact that becomes a major public
health issue in many countries. However, the newer options now
available to the patient mean that less invasive surgery could come
sooner, thus negating the need for a fusion.2, 3
Over the last ten years, there have been numerous ideas presented
to obviate the need for a fusion. These ideas involve restoring
motion when it is lost, stabilizing motion when it is unstable,
or limiting motion in situations where motion is excessive or
abnormal. All these techniques and technologies have an intrinsic
major goal of decreasing the associated surgical morbidity and
allowing a faster patient recovery.2, 4-6, 8-13
One such technology emerging as a treatment option for patients
with mild to moderate degenerative disc disease is the use of
Nucleus Arthroplasty systems. These systems might be an excellent treatment option for patients with primary discogenic back
pain, with or without leg pain. Currently, most devices in this field
are typically contraindicationed for patients with excessive body
mass index, irregular endplates, incompetent anulus, or significant
collapse of more than 50% of the disc space. Furthermore,
patients with spinal instability or deformity might not be a good
candidate for stand-alone nucleus replacement.
As discussed in Chapter 8, Nucleus Arthroplasty systems can be
described as mechanical (e.g. non-load sharing motion devices),
polymer (e.g. load-sharing motion devices) or tissue engineered
NUCLEUS ARTHROPLASTY SYSTEMS CAN BE

DESCRIBED AS MECHANICAL, POLYMER OR TISSUE
ENGINEERED IMPLANTS.
implants (e.g. non-structural tissue scaffolds). Each may play a

vital role in the treatment algorithm. For the initial stages of
DDD, where there are no significant anular tears, no segment
instability, no significant nucleus removal and no significant collapse, in situ formed void filling technology and tissue engineered
implants might be an option. For mild to moderate stages of
DDD, with more noteable anular disruption and disc space collapse, load-bearing devices that can provide structural support
and help stabilize segment motion, such as pre-formed polymer
and structural in situ formed devices, might be an option. For
later-stage DDD, with more significant collapse and significantly
altered load-sharing, mechanical devices may serve as an alternative to fusion.2, 4, 6, 7, 11 These are general recommendations based
on what is known today. Understanding the expected in situ biomechanical performance of a nucleus replacement implant, combined with clinical data, will be essential to selecting technologies
that match the patients need.
The need for Nucleus Arthroplasty sytems is supported by data
presented in Chapter 5. When we looked at patients treated in
our office in the past, 40% of the surgical group that underwent
a fusion procedure or a total disc replacement for back pain
secondary to degenerative disc disease, could, potentially, have
benefited from Nucleus Arthroplasty treatment. Therefore, this
technology would seem marked for success provided we can
match the proper patient with the appropriate implant,
approach, and postoperative rehabilitation program. All these
factors are believed to be of equal importance and integral
to a successful outcome.
WHEN WE LOOKED AT OUR PATIENTS TREATED IN OUR OFFICE IN THE PAST, 40% OF THE
PATIENTS THAT UNDERWENT A FUSION PROCEDURE OR A TOTAL DISC REPLACEMENT
FOR BACK PAIN SECONDARY TO DEGENERATIVE DISC DISEASE, COULD, POTENTIALLY,
HAVE BENEFITED FROM NUCLEUS ARTHROPLASTY TREATMENT.
51
THE NUCLEUS REPLACEMENT DEVICES CURRENTLY BEING DEVELOPED VARY GREATLY

IN DESIGN AND FUNCTION. UNDERSTANDING THEIR BIOMECHANICAL AND CLINICAL
PERFORMANCE IS KEY TO ENSURING PROPER DEVICE UTILIZATION.
CONCLUSION
REFERENCES
Nucleus replacement will most likely become a major option for

patients with mild to moderate degenerative disc disease. Ideal
candidates will be patients with discogenic back pain that present
without significant anular tears or endplate irregularities, preservation of disc height, good bone quality, and normal body mass
index (BMI).
1. Pappou IP, Girardi FP, Cammisa FP: Correlation of end plate shape on MRI
and disc degeneration in surgically treated patients with degenerative disc disease and herniated nucleus pulposus.Spine J.Jan-Feb; 7(1):32-8, 2007.
The nucleus replacement devices currently being developed

vary greatly in design and function. Understanding their biomechanical and clinical performance is key to ensuring proper
device utilization.
Nucleus Arthroplasty technologies provide the opportunity to
intervene sooner in the degenerative cascade and could prolong
or obviate the need for the next step (e.g. posterior dynamic stabilization or a total disc replacement). This extension of the algorithm should hopefully delay or obviate the need for radical
procedures, such as fusion, in the future.
2. Tropiano P, Huang RC, Girardi FP, Cammisa FP Jr, Marnay T. Lumbar total
disc replacement. Seven to eleven-year follow-up. J Bone Joint Surg Am.
Mar;87-A(3):490-6, 2005.
3. Huang RC, Girardi FP, Lim MR, Cammisa FP Jr.: Advantages and disadvantages of nonfusion technology in spine surgery. Orthop Clin North Am.
Jul;36(3):263-9, 2005.
4. Tropiano P, Huang RC, Girardi FP, Marnay T. Lumbar Disc Replacement:
Preliminary Results with ProDisc II After a Minimum Follow-Up Period of 1
Year. Spine. 28 Suppl:362-8, 2003.
5. Huang RC, Girardi FP, Cammisa Jr FP, Wright TM: The implications of constraint in lumbar total disc replacement. Spine. 28 Suppl:412-7, 2003.
6. Huang RC, Girardi FP, Cammisa Jr FP, Tropiano P, Marnay T: Long-term
flexion-extension range of motion of the prodisc total disc replacement.
J Spinal Disord Tech. Oct;16(5):435-40, 2003.
7. Huang RC, Girardi FP, Lim M, Cammisa, Jr FP: The Prevalence of
Contraindications to Total Disc Replacement in a Cohort of Lumbar Surgical
Patients.Spine. Nov 15; 2 9(22):2538-41, 2004.
8. Lim M, Huang R, Zhang K, Girardi FP, Peterson M and Cammisa FP:
Measurement of Total Disc Replacement Radiographic Range-of-Motion: A
Comparison of Two Techniques. J Spinal Disord Tech. Jun;18(3):252-6, 2005.
9. Huang RC, Girardi FP, Cammisa FP Jr, Lim MR, Tropiano P, Marnay T:
Correlation between range of motion and outcome after lumbar total disc
replacement: 8.6-year follow-up. Spine. Jun 15;30(12):1407-11, 2005.
10. Frelinghuysen P, Huang RC, Girardi FP, Cammisa FP Jr Lumbar total disc
replacement part I: rationale, biomechanics, and implant types. Orthop Clin
North Am. Jul;36(3):293-9, 2005.
11. Tropiano P, Huang RC, Girardi FP, Cammisa FP Jr, Marnay T. Lumbar total
disc replacement. J Bone Joint Surg Am. Mar;88-(1 Suppl 1):50-64, 2006.
12. Huang RC, Lim M, Tropiano P, Girardi FP, Cammisa FP and Marnay T.
Range of Motion and Adjacent Level Degeneration After Lumbar Total Disc
Replacement. The Spine Journal 2006 May - June;6(3):242-247.
13. Lim MR, Loder RT, Huang RC, Lymans S, Zhang K, Sama A, Papadopoulos EC,
Warner K, Girardi FP, Cammisa FP Jr: Measurement Error of Lumbar Total
Disc Replacement Range of Motion. Spine 2006 May 1;31(10):E291-E297.
52
Conclusion
Baltimore Neurosurgical Associates, PA

Baltimore, MD 21204

New York, NY 10021
ASSISTANT PROFESSOR
ORTHOPEDIC RESEARCH

New York, NY 10021
Emory University
Atlanta, GA 30322
lthough much remains to be learned about degenerative

disc disease (DDD), we are on the precipice of a new treatment stratagem for this problem effecting a wide-ranging patient
population which has been inadequately treated with available
treatment options. Nucleus Arthroplasty is poised to address
mild to moderate DDD and may be a successful treatment
option for later-stage disease as well.
Pioneering spine surgeons have successfully used nucleus
replacement to treat DDD. However, nucleus arthroplasty takes
this promising therapy one step further. Nucleus arthroplasty is
a total system approach that encompasses patient selection, indications, device sizing, surgical technique and approach, and
post-operative patient care. By adopting this system approach,
nucleus arthroplasty is expected to offer an effective treatment
option for patients who are currently left untreated.
FUTURE PUBLICATIONS ON
NUCLEUS ARTHROPLASTY TECHNOLOGY
We hope this monograph, Volume IIBiomechanics &
Development, provides valuable guidance to spine surgeons as
they seek to provide more effective and less invasive options for
the treatment of DDD. This volume is part of a continuing series
on nucleus arthroplasty that can be used individually or collectively. A number of well-known authors will be represented in
future monographs on nucleus arthroplasty.
Upcoming scheduled publication dates are as follows:
Emerging Technology is scheduled to be released at the
North American Spine Society (NASS) meeting in Austin,
Texas, October 2007.
Please note that each one of these volumes will be available at:
www.nucleusarthroplasty.com
This series has been made possible through the financial support of Raymedica, LLC.
www.raymedica.com
Part No. 55124-001 Rev. A

Nucleus Arthroplasty Volume II

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Nucleus Arthroplasty Volume II

Uploaded by

Copyright:

Available Formats

Nucleus

Volume II: Biomechanics &

Deputy Editorial Board

Biomechanics of the Degenerated Disc

Principles and Mechanical Requirements of Nucleus Implants

Kinematic Demands of Nucleus Arthroplasty Technology

Device Stiffness vs. Load-Sharing with Nucleus Arthroplasty Devices

Repair of the Anulus Fibrosus of the Lumbar Disc

Integration of Nucleus Arthroplasty Technology into the Continuum of Care

his monograph series is a groundbreaking project in the

Reginald J. Davis, MD, FACS

Frank P. Cammisa, Jr., MD, FACS

ASSOCIATE PROFESSOR OF CLINICAL SURGERY

Baltimore Neurosurgical Associates, PA

Hospital for Special Surgery

William C. Hutton, DSc

PROFESSOR AND DIRECTOR OF

Hospital for Special Surgery

he first documented works describing the diagnosis and

disorders. This sub-discipline represents the coalescence of many

in the forefront of Nucleus Arthroplasty technology development

Reginald J. Davis, MD, FACS

Frank P. Cammisa, Jr., MD, FACS

William C. Hutton, DSc

Deputy Editorial Board

Reginald J. Davis, MD, FACS

Dr. Davis is founder of Baltimore Neurosurgical Associates, chief

Dr. Girardi is assistant professor of orthopedic surgery, Weill

He has broad experience in advanced procedures such as spinal

Frank P. Cammisa, Jr., MD, FACS

William C. Hutton, DSc

Dr. Cammisa is associate professor of clinical surgery, Weill

Dr. Hutton is professor and director of orthopedic research,

His clinical interests include non-fusion and motion preservation

His major area of interest is biomechanics with a particular

Weill Medical College of Cornell University

Hospital for Special Surgery

s people age, the lumbar discs undergo a progressive

anulus. The hydrophilic proteoglycans in the nucleus control the

THERE IS A COMPLEX ARRANGEMENT OF BONES AND

AGING AND LUMBAR DISC DEGENERATION

THE DEGENERATIVE CASCADE

THE KIRKALDY-WILLIS STAGES DESCRIBED

as a mechanical low back pain that is episodic. During this stage

Stage II (Instability) is manifested on MRI with further loss

Stage III (Stabilization) is manifested on MRI by worsening of

CAN BE RADIOGRAPHICALLY CHARACTERIZED

THE PAIN GENERATOR

PROVIDING RESISTANCE TO COMPRESSIVE LOADS

According to Buckwalter it appears that the degenerative process

CONSIDERING THE IMPORTANT BIOMECHANICAL ROLE OF THE NUCLEUS

7. Rohlmann A, Zander T, Schmidt H, Wilke HJ, Bergmann G. Analysis of the

Principles and Mechanical

Prof. Dr. Hans-Joachim Wilke

Institute of Orthopaedic Research and Biomechanics

BASIC BIOMECHANICAL CONSIDERATIONS

on-fusion technologies in spinal surgery gain more and

In many cases, disc degeneration or disc prolapse is treated just by

Regain, Biomet Inc.

NUBAC, Pioneer Surgical

IPD, Dynamic Spine

knitted titanium, Buck

Newcleus, Zimmer Spine

Dual Disc cylinders,