See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/7185228

Test protocols for evaluation of spinal implants

ARTICLE in THE JOURNAL OF BONE AND JOINT SURGERY MAY 2006
Impact Factor: 4.31 DOI: 10.2106/JBJS.E.01363 Source: PubMed

CITATIONS

DOWNLOADS

VIEWS

108

74

303

5 AUTHORS, INCLUDING:
Vijay K Goel

Manohar M Panjabi

University of Toledo

Yale University

125 PUBLICATIONS 1,715 CITATIONS

346 PUBLICATIONS 17,109 CITATIONS

SEE PROFILE

SEE PROFILE

Avinash G Patwardhan

Hassan Serhan

Loyola University Medical Center

Johnson & Johnson

128 PUBLICATIONS 2,547 CITATIONS

50 PUBLICATIONS 696 CITATIONS

SEE PROFILE

SEE PROFILE

Available from: Hassan Serhan

Retrieved on: 14 August 2015

This is an enhanced PDF from The Journal of Bone and Joint Surgery
The PDF of the article you requested follows this cover page.

Test Protocols for Evaluation of Spinal Implants

Vijay K. Goel, Manohar M. Panjabi, Avinash G. Patwardhan, Andrew P. Dooris and Hassan Serhan
J Bone Joint Surg Am. 88:103-109, 2006. doi:10.2106/JBJS.E.01363

This information is current as of April 5, 2006

Reprints and Permissions

Click here to order reprints or request permission to use material from this
article, or locate the article citation on jbjs.org and click on the [Reprints and
Permissions] link.

Publisher Information

The Journal of Bone and Joint Surgery

20 Pickering Street, Needham, MA 02492-3157
www.jbjs.org

Downloaded from www.ejbjs.org on April 5, 2006

103
COPYRIGHT 2006

BY

THE JOURNAL

OF

BONE

AND JOINT

SURGERY, INCORPORATED

Test Protocols for

Evaluation of Spinal Implants
BY VIJAY K. GOEL, PHD, MANOHAR M. PANJABI, PHD, AVINASH G. PATWARDHAN, PHD,
ANDREW P. DOORIS, PHD, AND HASSAN SERHAN, PHD

Prior to implantation, medical devices are subjected to rigorous testing to ensure safety and efficacy. A full battery of
testing protocols for implantable spinal devices may include many steps. Testing for biocompatibility is a necessary
first step. On selection of the material, evaluation protocols should address both the biomechanical and clinical performance of the device. Before and during mechanical testing, finite element modeling can be used to optimize the
design, predict performance, and, to some extent, predict durability and efficacy of the device. Following bench-type
evaluations, the biomechanical characteristics of the device (e.g., motion, load-sharing, and intradiscal pressure) can
be evaluated with use of fresh human cadaveric spines. The information gained from cadaveric testing may be supplemented by the finite element model-based analyses. Upon the successful completion of these tests, studies that
make use of an animal model are performed to assess the structure, function, histology, and biomechanics of the device in situ and as a final step before clinical investigations are initiated.
The protocols that are presently being used for the testing of spinal devices reflect the basic and applied research experience of the last three decades in the field of orthopaedic biomechanics in general and the spine in particular. The innovation within the spinal implant industry (e.g., fusion devices in the past versus motion-preservation devices at present)
suggests that test protocols represent a dynamic process that must keep pace with changing expectations. Apart from
randomized clinical trials, no single test can fully evaluate all of the characteristics of a device. Due to the inherent limitations of each test, data must be viewed in a proper context. Finally, a case is made for the medical community to converge toward standardized test protocols that will enable us to compare the vast number of currently available devices,
whether on the market or still under development, in a systematic, laboratory-independent manner.

rior to implantation, medical devices are subjected to

rigorous testing to demonstrate safety. Efficacy (i.e.,
the demonstrated and proven beneficial effects of the
device on the involved pathology), is equally important in the
modern and aggressive field of medical technology. As our understanding of the structure and function of the human body
increases, the performance expectations for medical devices
also rise. Consequently, test protocols, which probe the efficacy of medical devices, need to keep pace with the innovation
in the medical device industry so that a particular revised
characteristic of a device can be properly evaluated. Not long
ago, the main purpose of implantable spinal devices was to
provide rigid fixation to enhance fusion across a spinal segment; this characteristic is in direct contrast to the present
drive for devices that will preserve or restore motion across
the affected spinal segment. This philosophical reversal requires revised experimental techniques.
A full battery of testing protocols for implantable spinal
devices includes many steps. Testing for biocompatibility is a
necessary first step toward ensuring acceptance of the materials
used in the manufacture of any medical device. On selection of
the material, evaluation of the implant should address both the
biomechanical and clinical performances. Initial mechanical

tests should be designed to ensure safety and durability. Additional biomechanical studies can be performed by implantation of the device within human cadaveric spines. This aspect
also can be supplemented with the finite element model-based
analyses. When these steps have been successfully completed,
studies using an animal model are suggested to assess the structure, function, histology, and biomechanics of the device in situ
as a final step before clinical investigations.
A basic understanding of several aspects of the spine is
necessary to fully appreciate this overview. These areas include
spine anatomy (including the musculature), spinal biomechanics (including the contributions of various structures, such as
the vertebral body, facet joints, intervertebral discs, and ligaments in supporting the axial, bending, and torsional loads on
the skeleton), spinal kinematics (e.g., quality and quantity of
motion and the instantaneous axis of rotation), associated pathologies that may impair active and passive structures (disc
degeneration, facet arthritis, and spondylolisthesis), and the
prevailing techniques to treat these pathologies. Presuming this
background, the following sections briefly describe the testing
strategies in use and/or under development for the evaluation
of spinal instrumentation, under the two broad categories of
safety and efficacy.

Downloaded from www.ejbjs.org on April 5, 2006

104
THE JOUR NAL

BONE & JOINT SURGER Y JBJS.ORG

VO L U M E 88-A S U P P L E M E N T 2 2006

OF

TE S T P RO TO CO L S F O R E V A L U A T I O N
SPINAL IMPLANTS

OF

Safety
afety testing may be largely divided into two categories
tests to ensure that the device tolerates the body, and tests
to ensure that the body tolerates the device.

Fig. 1-A

The screw is mounted in a foam block and is being pulled out

along its axis in a servohydraulic materials testing machine
(model #810; MTS Systems, Eden Prairie, Minnesota).

Biocompatibility
One of the causes and/or results of medical implant failure is
the release of particulate material debris from the device because of wear, physical deterioration, or chemical attack by
the harsh physiologic environment. Biocompatibility testing
may be categorized according to bulk material biocompatibility or particulate biocompatibility. In its bulk form, the
implants surface may be leached or corroded. This would be
applicable to devices subject to fatigue failure such as from
fretting at screw junctions, corrosion, and elastomeric fatigue.
Although some materials may be biocompatible in bulk form,
as particles they may incite adverse reactions from cells. The
presence of particles or any evidence of acute inflammation,
chronic inflammation, granulation tissue, fibrosis of the
meninges, foreign body reaction, arachnoiditis, or transdural migration of new material particles should be assessed.
Because of the proximity of the disc replacement devices to
sensitive neural tissues, the effects of the new material particles on the dura and neural tissue should also be evaluated.
To evaluate biocompatibility, a twelve-week rat subcutaneous pouch model might be used. This test is customarily recommended when a new material without a previous clinical
history is introduced. Typically, the cellular response of the
test material is compared with the response to high-density
polyethylene, a biocompatible material used in many joint
prosthetic devices. Neurotoxicity might be evaluated with
use of a sheep or a rabbit model1.

Fig. 1-B

View showing the pneumatic system for applying compression load to the device held between
foam blocks or vertebrae while being pulled or pushed out. (Clockwise from right: actuator, load
cell, readout display, pneumatic regulator, and lever).

Downloaded from www.ejbjs.org on April 5, 2006

105
THE JOUR NAL

BONE & JOINT SURGER Y JBJS.ORG

VO L U M E 88-A S U P P L E M E N T 2 2006

OF

TE S T P RO TO CO L S F O R E V A L U A T I O N
SPINAL IMPLANTS

OF

the axial compression (Fig. 1-B). Although device migration is

prevented by the surface shear strength of the end plate, subsidence is resisted by surface compression3. For devices with
osteoconductive coatings, which may increase the interface
strength over time due to bone-device integration, evaluation
of implant-bone interfaces over time is best performed with
the use of animal models, which allow for tissue ingrowth.

Fig. 2-A

Schematic showing follower setup to apply preloads during testing of multisegment specimens.

Implant-Tissue Interface
In addition to the chemical and biologic reactions occurring at
the interface between the device and the body, potential failure
modes of device interfaces may also be mechanical in nature. As
with biocompatibility testing, this testing focuses on the device
(and procedure) safety rather than efficacy. Depending on the
spinal instrumentation, the implant-tissue interface may include interactions of the implant with any of the functional spinal unit elements, including the laminae, pedicles, spinous
processes, vertebral body, vertebral end plates, anulus layers,
and nucleus. Quasistatic and cyclic pullout/bending tests of
screws placed in vertebrae or foam blocks are fairly common
(Fig. 1-A)2. Although the pullout test is simple and popular, it
does not represent the complex in vivo loads at the bone-screw
interface. Interlaminar hooks are also used to secure the device
to the spine, similarly to screws. A hooks ability to hold on to
the lamina without slippage has been investigated in a manner
similar to pullout strength testing of screws.
Performance of an interbody device also depends to
some degree on the interaction with the adjoining vertebral
end plate. An evaluation of this interface may include compression of the device between vertebrae or foam blocks followed by
push or pullout of the device in the direction perpendicular to

Mechanical Testing
Because the primary function of most orthopaedic and/or spinal implants is structural, tests to investigate whether the device
tolerates the in situ environment are often mechanical in nature. Generally, the device should be capable of withstanding
maximum expected static loads and average daily dynamic
loads throughout its predicted life without functional compromise or major permanent deformation. The goal of this characterization is to provide sufficient evidence to support the use of
the device in clinical trial applications. Quasistatic mechanical
tests can be used to characterize the ultimate strength and deformation of the device as well as failure modes in axial compression, compressive shear, torsion, lateral bending, flexion,
and extension4. Dynamic test results can be used to establish endurance limits and associated long-term performance of the device under difficult, yet realistic, conditions.
Static and dynamic mechanical tests for interbody fusion
devices and artificial discs are very similar, with two key exceptions: (1) artificial discs are intended to last for the life of the patient, and thus the total number of cycles performed in the
fatigue test is larger; and (2) artificial discs are intended to move
and/or rotate and/or deform (depending on the nature of the

Fig. 2-B

Radiograph of a specimen showing follower preload path during testing

of an actual specimen.

Downloaded from www.ejbjs.org on April 5, 2006

106
THE JOUR NAL

BONE & JOINT SURGER Y JBJS.ORG

VO L U M E 88-A S U P P L E M E N T 2 2006

OF

TE S T P RO TO CO L S F O R E V A L U A T I O N
SPINAL IMPLANTS

OF

TABLE I ASTM Proposed Standards5 That May Be Adapted for Device Evaluation*
Standard or Guide

Status

Focus

Use

ASTM F 1582-98 Standard Terminology

Relating to Spinal Implants

Reapproved 2003

General spinal
device testing

Defines basic terms and considerations for

spinal implant devices and their mechanical
analyses

ASTM F 2193-02 Standard Specifications

and Test Methods for Components Used in
the Surgical Fixation of the Spinal Skeletal
System

Approved 2002

Fusion devices

Static and fatigue testing of components

screws, rods, and plates

ASTM F 1717-96 Standard Test Methods

for Spinal Implant Constructs in a Vertebrectomy Model

Reapproved 2004

Fusion devices

Materials and methods for static and

fatigue testing of constructs (implants in
UHMWPE blocks). Focus on estimation of
short-term stability while arthrodesis takes
place

ASTM F 2077-01 Test Methods For Intervertebral Body Fusion Devices

Approved 2001

Fusion devices

Axial compression, compressive shear, and

torsion of interbody devices in static and
fatigue testing

ASTM F 2267-04 Standard Test Method

for Measuring Load Induced Subsidence
of an Intervertebral Body Fusion Device
Under Static Axial Compression

Approved 2004

Fusion devices

Materials and methods for the axial compressive subsidence testing of nonbiologic
intervertebral body fusion devices; spinal
implants designed to promote arthrodesis
at a given spinal motion segment

ASTM Draft Method WK455 Draft Standard

Test Methods for Occipital-Cervical and
Occipital-Cervical-Thoracic Spinal Implant
Constructs in a Vertebrectomy Model

Draft

Fusion

Materials and methods for the static and

fatigue testing of OC and OCT spinal
implant assemblies in a vertebrectomy
model (UHMWPE). OCT load and geometry
specific. (May be incorporated into F 1717
in future)

ASTM Method F 2346-05 Standard Test

Methods for Static and Dynamic Characterization of Spinal Artificial Discs

Approved 2005

Artificial discs

Materials and methods for static and

dynamic testing of artificial discs. Compression, compressive shear, and torsion
are included. Similar to its fusion device
counterpart

ASTM WK7479 Draft Standard Test

Method for the Functional, Kinematic,
and Wear Assessment of Extra-Discal
Spinal Motion Preserving Implants

Draft

Artificial discs

For wear assessment of lumbar and cervical disc replacements, including methods
for loads, angles, moments, test environment, and data analysis

ISO ISO/TC 150 / SC 5 Implants for

surgery Wear of total intervertebral
spinal disc prostheses Part 1: Loading
and displacement parameters for wear
testing and corresponding environmental
conditions for tests

En route to
approval 2005

Artificial discs

For wear assessment of lumbar and cervical disc replacements, including methods
for loads, angles, moments, test environment, and data analysis, more applicable to
sliding implants (nonelastomeric)

ASTM WK4863 Draft Standard Practice/

Guide for the Mechanical Characterization
of Lumbar Nucleus Devices

Draft

Nucleus
replacements

This guide gives general guidance on

testing methods for various forms of
nucleus replacement and nucleus augmentation devices, outlining the types of
testing that are recommended

*ASTM = American Society for Testing and Materials, UHMWPE = ultra-high molecular weight polyethylene.

implant) for the life of the patient, and so should resist tearing,
denting, cracking, or wearing. Wear tests for artificial discs have
particularly benefited from three decades of experience with hip
replacements. The artificial disc wear-testing has been recommended to run ten million cycles, which is twice that of hips.
Typical outcomes of this testing are gravimetric wear rate and

particulate analysis. Particulate analysis is complicated because

the effects of debris in a nonsynovial joint have not been well
characterized. The presumption is that, barring neurotoxic effects, achievement of wear rates lower than those of hips provides a sufficient safety factor for artificial discs.
Although the American Society for Testing and Materials

Downloaded from www.ejbjs.org on April 5, 2006

107
THE JOUR NAL

BONE & JOINT SURGER Y JBJS.ORG

VO L U M E 88-A S U P P L E M E N T 2 2006

OF

TE S T P RO TO CO L S F O R E V A L U A T I O N
SPINAL IMPLANTS

OF

Fig. 3

Three-dimensional finite element model of the ligamentous L3-S1 segment. (A): Intact. (B): Modified to simulate placement of a Charit artificial disc (DePuy Spine, Raynham, Massachusetts) across the L5-S1 segment.

test standards (Table I) represent to a large extent the mechanical characterization requested by the United States Food and
Drug Administration, it may be necessary to perform many
additional tests to assess safety and durability 5. In all motion
preservation systems, including artificial discs, nuclei, dynamic
posterior screws, and interspinous spacers, the test loading
modalities include compression, shear, flexion and extension, lateral bending, and torsion; however, they all have different failure loads, failure criteria, and necessary fatigue strengths.
Nucleus replacements may require the performance of many
material characterization tests that are not associated with spine
geometry but that are necessary to interpret the effects of body
loads over time. A number of researchers and companies have
proposed testing nucleus devices inside an artificial anulus fixture or within a fixture mimicking vertebral end plates. This
proposal brings with it many questions concerning the implantfixture interface, the load-sharing between the nucleus and anulus, and ideal disc space geometries. Much research is needed in
those areas and in posterior dynamic stabilization systems, such
as Dynesys (Zimmer Spine, Minneapolis, Minnesota), DIAM
(Device for Intervertebral Assisted Motion; Medtronic, Minneapolis Minnesota), and X STOP (St. Francis Medical Technologies, Alameda, California). The complicated biomechanical
loading environment of the Dynesys system and the potential
failure modes of the DIAM and X STOP systems have not yet
been fully elucidated and require additional research. Artificial
facet systems are even more complicated due to complex in vivo
loads and potential implant wear issues.
Efficacy
lthough an understanding of the device function and design are necessary for safety testing, a clear description of

the device purpose and the associated pathology are critical in

the testing of device efficacy.
In Vitro Spinal Segment Testing
Multisegment testing is performed to assess the function of a
spinal construct (the spine segment with the implant in place)6.
Change in motion due to instrumentation with respect to the
intac case can be described in terms of three rotations of the vertebra and three translations of a point on the vertebra, the instantaneous axis of rotation of the segment (or its equivalent
helical axis of motion in three-dimensional space), the quality of
the motion (nonlinearity, neutral zone, and coupled motions),
and the quantity of the motion (e.g., range of motion)7. Several
other parameters of interest are changes during testing of the
construct in disc height, disc bulge, intradiscal pressure, spinal
alignment, vertebral body strains, creep and viscoelastic behavior of the construct, and bone-implant interface interactions.
Another focus of this testing is quantification of load-sharing between the device and functional spinal unit structures. Besides
the use of a motion measurement system, additional transducers, such as the strain gauges applied to vertebrae and pressure
transducers inserted into the nucleus, may be needed to document these parameters in response to the applied loads.
For such studies, fresh osteoligamentous spines of appropriate length are procured, screened (e.g., bone mineral
density above a certain value), and prepared for testing6. Test
data from single functional spinal units and multispinal segments have been reported in the literature. A consensus seems
to be emerging that the multispinal segments should be used
for testing and should include at least one free functional
spinal unit on either side of the construct length essential for
evaluation of the device. The total segment length should

Downloaded from www.ejbjs.org on April 5, 2006

108
THE JOUR NAL

BONE & JOINT SURGER Y JBJS.ORG

VO L U M E 88-A S U P P L E M E N T 2 2006

OF

also account for assessment of the effects of the device on the

adjacent segments, if needed. Loads are applied to a loading
frame attached to an end (usually the cephalad) vertebra,
and the resulting motion of each vertebra is recorded with
the help of transducers. The data are obtained sequentially
for the intact case and treatment cases, such as the placement
of an artificial disc.

TE S T P RO TO CO L S F O R E V A L U A T I O N
SPINAL IMPLANTS

OF

that have included the use of pure moments in which the

specimens are free to move in an unconstrained manner. The
moments may range from 6 to 10 Nm for the lumbar region
and 1.5 to 3 Nm for the cervical region. These moments may
be applied in a quasistatic manner with three load-unload cycles. The data are recorded on the third load cycle.
Testing of Adjacent Level Effects

Role of Muscles

In vitro tests cannot incorporate the effect of the musculature.

Furthermore, the biomechanical role of muscles in general,
and especially following surgery and until healing occurs, is
not well documented. During testing, an application of compressive preload with accompanying moments in clinically
relevant loading modes may simulate the effect of musculature. Data in the earlier literature were based on tests undertaken with no preload. The current consensus is that, in the
lumbar region, a preload of 400 N (about 60% of the body
weight above the L3-L4 disc level of an average person) with a
maximum of 1000 N should be included in the tests, using the
follower load concept8; the test setup used by Patwardhan et
al.8 for applying follower-type preload is shown in Figure 2.
The effect of the preload on the outcome is highly dependent
on its location with respect to the center of rotation of the individual segments. For the cervical region, the follower preload may range from 50 to 100 N (one to two times the weight
of the head). How will this concept apply during sequential
tests such as decompression and instrumentation of the segment? Additionally, at present, the application of the follower
load does not yield satisfactory results in modes other than
flexion and extension. The quality of the measured data with
and without preloads is similar, and thus tests without preload
may also be acceptable for comparative evaluation of spinal
instrumentation7. However, construct testing should be conducted with a follower load whenever possible.
Load Types and Magnitudes

Load testing may be the most controversial issue facing the industry and its research. Some tests apply a known displacement at the free end and quantify the resulting loads and
motions across various segments (displacement control or
stiffness protocol). Others measure the motion in response to
a known load (load control or flexibility protocol). In the load
control testing protocol, one could apply forces such as anterior and posterior shear, pure moments, or a complex load.
Application of forces results in a varying amount of bending
moment across the segment, depending on the location
within the cantilever specimen and the deflection of the specimen itself. Furthermore, the deflection of the specimen following stabilization changes significantly compared with the
intact case. These could present difficulties in computing the
load displacement curves of the specimen for different treatment simulations (e.g., intact, decompression, and fusion).
This issue is mitigated if one applies pure moments and the
resulting unconstrained motions are measured7,9,10. Consensus
seems to be forming that investigators should provide data

The flexibility protocol using pure moments may be modified

to address the effects of a device on the adjacent segments. For
example, it would be advantageous to use a protocol that
would achieve the same overall range of motion for the intact
specimen as well as for the instrumented construct by applying pure moments that distribute evenly down the column.
This is the hybrid protocol7,9,10.
Functional Animal Studies

Although spinal segment testing can be an effective tool, it is

limited by the lack of bone and soft-tissue remodeling (that
may include osteophyte generation, end-plate settling, and
scar-tissue formation), the biochemical response, and by limited replication of in vivo loads. These restrictions limit functional spinal unit testing results to immediate postoperative
conditions. To account for such shortcomings and for the biologic effects, spectrum, magnitude, and mode of in vivo
loads, animal models are necessary. Animal models provide a
dynamic biologic and mechanical environment in which the
implant can be evaluated. Temporal changes in both the host
biologic tissue and the instrumentation can be assessed with
selective incremental sacrificing of the animals. Common
limitations of animal studies include dissimilarities between
human and animal spines with respect to the spinal loads,
spinal motions, anatomy, and the difficulties in adjusting the
device to properly fit the animal spine. Animal models can
range from rats to rabbits, sheep, goats, dogs, pigs, and baboons. Functional evaluation of a disc replacement requires
use of an animal model that simulates the structure, function,
and biomechanical properties of the human spine. Use of
comparable surgical techniques and approaches as in humans
is also desirable. For evaluating interbody devices, the sheep
and baboon are often used for the cervical and lumbar regions, respectively11. Smaller primates can be used to approximate load modes, but larger primates such as baboons are
necessary to simulate both load magnitude and direction.
Another complication of using functional animal models
may be in appropriately sizing or shaping the device to fit the
anatomy of the animal. Not all device features may be reducible to animal-sized discs; some mechanisms (such as diffusion across hydrogels) are size-dependent. Likewise, a sheep
cervical disc displays a very narrow posterior disc height and
a caudally sloping anterior that pose potential issues for interbody devices. Thus, animal studies have limitations for
evaluating the function of a device.
Both implant and implant-tissue interface characteristics may be evaluated in the animal model. Implant characteristics include such features as resorption (such as with

Downloaded from www.ejbjs.org on April 5, 2006

109
THE JOUR NAL

BONE & JOINT SURGER Y JBJS.ORG

VO L U M E 88-A S U P P L E M E N T 2 2006

OF

polylactic acid and/or polyglycolic acid devices), static compressive strength, wear, cracking, and deterioration. The interface may be investigated for subsidence (osseous end-plate
deterioration), fixation (migration of the device), ingrowth
(into osteoconductive coatings), and possible wear-debris effects on neural elements and surrounding tissues. In most animal studies, a systemic analysis is also commonly performed,
including the histopathologic response in local and systemic
tissues to device material and possibly wear debris generated
in nonfailure and failure modes (if applicable). Pathologic assessment for all tissues should include but not be limited to
comments on the architecture of the tissues and the presence
of wear debris, as well as any signs of foreign-body giant-cell
and/or granuloma inflammatory reactions, degenerative changes,
or autolysis. For motion-preserving devices, the segmental
stiffness properties through the normal range of motion may
be investigated.
Finite Element Models

In vitro investigations and in vivo animal studies contain

numerous limitations, including that they are both timeconsuming and expensive. The most important limitations
of in vitro studies are that muscle contributions to loading
are not usually incorporated; in addition, the quality of the
cadaveric specimens is highly variable. In vivo animal studies
involve quadruped animals, and the implant sizes usually need
to be scaled according to the animal size. In an attempt to
complement the above protocols, several experimentally validated finite element models of the ligamentous spine have
been developed (Fig. 3)10. Such studies can be used to help
design the device itself or study the effects of the device on
the load-sharing, stresses, and strains in the spinal structures
under simple and complex loading scenarios. Finite element
modeling has been coupled with adaptive bone remodeling
algorithms to investigate the temporal changes associated with
interbody fusion devices such as cages.

TE S T P RO TO CO L S F O R E V A L U A T I O N
SPINAL IMPLANTS

OF

Summary
rigorous testing methodology has been presented that can
be used to evaluate the function of a spinal device. Because
of the complex role of a device, the sensitivity of the surrounding structures, and the unique biomechanical role of the spine,
testing requires considerable planning and resources. Care must
be taken to consider the limitations of each test while evaluating
the test results. Although no test except for pilot clinical studies
can sufficiently account for all variables or thoroughly evaluate
efficacy, it is imperative to produce a full battery of test results
that can reasonably predict safety to the candidate patient. This
review also underscores the fact that additional innovative test
protocols are needed to address the newer designs that are
about to come on the market, such as artificial facets and nucleus replacements. The orthopaedic community must work
hard to agree on standardized test protocols that will enable the
clinician to compare the vast number of devices currently available on the market as well as those under development on a
laboratory-independent basis. 

Corresponding author:
Vijay K. Goel, PhD
Department of Bioengineering, 5051 C Nitschke Hall, College of
Engineering, University of Toledo, Toledo, OH 43606. E-mail address:
vijay.goel@utoledo.edu
The authors did not receive grants or outside funding in support of
their research for or preparation of this manuscript. They did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid
or directed, or agreed to pay or direct, any benefits to any research fund,
foundation, educational institution, or other charitable or nonprofit
organization with which the authors are affiliated or associated.

doi:10.2106/JBJS.E.01363

References
1. Cunningham BW, Orbegoso CM, Dmitriev AE, Hallab NJ, Sefter JC, Asdourian P,
McAfee PC. The effect of spinal instrumentation particulate wear debris. An in
vivo rabbit model and applied clinical study of retrieved instrumentation cases.
Spine J. 2003;3:19-32.
2. Lim TH, An HS, Hasegawa T, McGrady L, Hasanoglu KY, Wilson CR. Prediction
of fatigue screw loosening in anterior spinal fixation using dual energy x-ray absorptiometry. Spine. 1995;20:2565-9.
3. Steffen T, Tsantrizos A, Aebi M. Effect of implant design and endplate preparation on the compressive strength of interbody fusion constructs. Spine. 2000;
25:1077-84.
4. Spenciner DB, Paiva JA, Crisco JJ. Testing of human cadaveric functional spinal
units to the ASTM draft standard, Static and fatigue methods for spinal artificial
disc. In: Melkerson MN, Kirkpatrick J, Griffith S, editors. Spinal implants: are we
evaluating them appropriately? West Conshohocken, PA: ASTM International;
2003. p 114-26.
5. Annual book of ASTM standards. Volume 13.01: Medical and surgical materials
and devices; anesthetic and respiratory equipment; pharmaceutical application of
process analytical technology. West Conshohocken, PA: ASTM International; 2005.

6. Panjabi MM. Biomechanical evaluation of spinal fixation devices: I. A conceptual framework. Spine. 1988;13:1129-34.
7. Goel VK, Panjabi MM, editors. Roundtables in spine surgery; spine biomechanics; evaluation of motion preservation devices and relevant terminology. St.
Louis: Quality Medical Publishing; 2005.
8. Patwardhan AG, Carandang G, Ghanayem AJ, Havey RM, Cunningham B, Voronov
LI, Phillips FM. Compressive preload improves the stability of anterior lumbar interbody fusion cage constructs. J Bone Joint Surg Am. 2003;85:1749-56.
9. Panjabi MM. Biomechanical testing to quantify adjacent-level effects. World
Congress of Biomechanics; 2002 Aug 4-9; Calgary, Alberta, Canada. Abstract
5079.
10. Goel VK, Grauer JN, Patel TC, Biyani A, Sairyo K, Vishnubhotla S, Matyas A,
Cowgill I, Shaw M, Long R, Dick D, Panjabi MM, Serhan H. Effects of Charite artificial disc on the implanted and adjacent spinal segments mechanics using a hybrid testing protocol. Spine. 2005;30:2755-64.
11. Wilke HJ, Kettler A, Claes LE. Are sheep spines a valid biomechanical model
for human spines? Spine. 1997;22:2365-74.

Downloaded from www.ejbjs.org on April 5, 2006