 

 

.1.)
..|l

.l
v.

.. .. bl.
Talliw‘. bx .
f . ,.
xxx}?
.
a. .3: ...

 

E.
$3...
...;_uz.ﬁ<
i. .. . .
1.133“.

.735
El";
ii... 1
1.1: 3...
21.2.1;
.1 ..

«331..
.51 4

 

¢ ., . .
...

a...‘.m...w.‘_§.\._..: . , ,

 

\ M w ' LIBRARIES
HIGAN STATE UNIVERSITY
Q EXIST LANSING, MICH 48824-1048

This is to certify that the
thesis entitled

INVESTIGATION OF THE ACUTE INJURY RESPONSE
OF ARTICULAR CARTILAGE IN VITRO AND IN VIVO:
ANALYSIS OF VARIOUS THERAPEUTIC
TREATMENTS

presented by
Steven Anthony Rundell

has been accepted towards fulﬁllment
of the requirements for the

MS. degree in Mechanical Engineen’nL

 

 

ﬂmczy/md'

Major Professors Signature
MM 9, 2005’
d I

Date

MSU is an Affirmative Action/Equal Opportunity Institution

- F‘- -

- ._.. -_ ~ ‘ . .

—v— v ﬂ v ‘- —

—' v v ‘

-v v‘ -‘ 'v v‘, 'Vv‘.’

.- v - ‘v’ - v - v

-v-‘

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

alas—cit “WWW .l 15‘

 

INVESTIGATION OF THE ACUTE INJURY RESPONSE OF ARTICULAR
CARTILAGE IN VITRO AND IN V1 V0: ANALYSIS OF VARIOUS THERAPEUTIC
TREATMENTS

By

Steven Anthony Rundell

A THESIS

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
For the degree of

MASTER OF SCIENCE
Department of Mechanical Engineering

2005

ABSTRACT
INVESTIGATION OF THE ACUTE INJURY RESPONSE OF ARTICULAR
CARTILAGE IN VIT R0 AND IN V1 V0: ANALYSIS OF VARIOUS THERAPEUTIC
TREATMENTS
By
Steven Anthony Rundell

Excessive mechanical loading to a joint can lead to matrix damage and
chondrocyte death in articular cartilage. These injuries have been associated with the
subsequent development of osteoarthritis. Understanding the mechanisms responsible for
acute damage of cartilage is essential in the development of therapeutic methods to either
prevent or treat these early alterations. The research presented in the current thesis uses
both the in vitro chondral explant and in viva rabbit models to examine the acute response
of articular cartilage to blunt impact loading. Chapter 1 addressed the issue of injury
severity in chondral explants exposed to a culture medium prior to mechanical loading
versus explants taken directly from the joint. This study hypothesized that excess ﬂuid
present in explants allowed to bathe in a culture medium would result in an artiﬁcially
high amount of surface ﬁssuring and associated chondrocyte death. Chapter 2 describes
experiments in which knee joints of giant Flemish rabbits were subjected to a 6 Joule
intensity impact and retro-patellar cartilage was studied in terms of surface ﬁssuring as
well as chondrocyte death. This study also evaluated the efﬁcacy of a mild non-ionic
surfactant, poloxamer 188, which has been found to reduce chondrocyte death in vitro by
repairing damaged cells after an impact. Chapter 3 evaluated the efﬁcacy of the
nutraceutical glucosamine in enhancing the mechanical integrity of cartilage explants

when pre-treated for a period of 6 days prior to an injurious unconﬁned compression test.

DEDICATION

I would like to thank my parents who have always supported me in every decision I have
made. I would also like to thank them for instilling a good work ethic in me and
exposing me to so many great opportunities.

iii

ACKNOWLEDGMENTS

I would like to acknowledge my professor, mentor and good friend Dr. Roger
Haut.

I am extremely grateﬁil to Dr. Mike Orth and Dr. Neil Wright for serving on my
committee.

I would like to express my gratitude towards Clifford Becket who not only
contains a vast amount of knowledge but was always willing to share it with me.

I would also like to thank Jane Walsh and Jean Atkinson for all their hard work
and dedication.

I would like to thank the undergrads that supplied me with great data and did a lot
of hard work: Austin McPhillamy, Aaron Stewart, Jo Ewen, Michelle Foncannon, and
Zach Kaltz

Last but not least I would like to acknowledge my fellow graduate students for
their help and friendship: Derek Baars, Jill Krueger, Dan Phillips, Eric Meyer, Lynn

Martin, Mike Sinnot, and Eugene Kepich.

iv

TABLE OF CONTENTS

LIST OF TABLES ...................................................................................................... vii
LIST OF FIGURES ..................................................................................................... viii
RESEARCH PUBLICATIONS ................................................................................... .ix

INVESTIGATIONS INTO THE ACUTE INJURY RESPONSE OF ARTICULAR
CARTILAGE IN VI T R0 AND IN VIVO: ANALYSIS OF VARIOUS THERAPEUTIC

TREATMENTS
Introduction ............................................................................................................ 1
References .............................................................................................................. 6
CHAPTER 1

TISSUE EQUILIBRATION ALTERS THE RESPONSE OF CARTILAGE EXPLANTS
TO UNCONFINED COMPRESSION

Abstract .................................................................................................................. 9

Introduction ............................................................................................................ 10
Methods ................................................................................................................. 13
Results .................................................................................................................... 19
Discussion .............................................................................................................. 21
References .............................................................................................................. 28
Figure Captions ...................................................................................................... 33
Figures ................................................................................................................... 37

CHAPTER 2

THE LIMITATION OF ACUTE NECROSIS IN RETRO-PATELLAR CARTILAGE
AFTER A SEVERE BLUNT IMPACT TO THE IN VI V 0 RABBIT PATELLO-
FEMORAL JOINT

Abstract .................................................................................................................. 45
Introduction ............................................................................................................ 46
Methods ................................................................................................................. 49
Results .................................................................................................................... 52
Discussion .............................................................................................................. 54
References .............................................................................................................. 61
Figure Captions ...................................................................................................... 64
Figures ................................................................................................................... 66
CHAPTER 3

GLUCOSAMINE SUPPLEMENTATION CAN HELP LIMIT MATRIX DAMAGE
AND ADJACENT CELL DEATH IN TRAUMATIZED EXPLANT S
Abstract .................................................................................................................. 7O

Introduction ............................................................................................................ 71

Methods ................................................................................................................. 74
Results .................................................................................................................... 79
Discussion .............................................................................................................. 82
References .............................................................................................................. 88
Tables ..................................................................................................................... 91
Figure Captions ...................................................................................................... 92
Figures ................................................................................................................... 95

CONCLUSIONS AND RECONHVIENDATIONS FOR FUTURE WORK .................. 103

APPENDIX A

RAW DATA FROM CHAPTER 1 .............................................................................. 107
Mechanical Data ..................................................................................................... 107
Equilibrated Weight Data ....................................................................................... 109
Surface Fissure Lengths .......................................................................................... 110
Cell Viability Data .................................................................................................. 111
Increased Variability Mechanical Data ................................................................... 113

APPENDIX B

RAW DATA FROM CHAPTER 2 .............................................................................. 116
Mechanical Data ..................................................................................................... 116
Surface Fissure Data ............................................................................................... 117
Cell viability counts for Time 0 .............................................................................. 118
Cell viability counts for 4 Day No Poloxamer ......................................................... 120
Cell viability counts for 4 Day Poloxamer .............................................................. 122

APPENDIX C

RAW DATA FROM CHAPTER 3 .............................................................................. 124
Mechanical Data ..................................................................................................... 124
Surface Fissure Data ............................................................................................... 128
Cell Viability Data .................................................................................................. 130
Theoretical Material Property Table ....................................................................... 134

APPENDIX D

RABBIT IMPACT SOP .............................................................................................. 135

vi

LIST OF TABLES

CHAPTER 3
Table 1 ................................................................................................................... 91

vii

LIST OF FIGURES

CHAPTER 1
Figure 1 .................................................................................................................. 37
Figure 2 .................................................................................................................. 37
Figure 3 .................................................................................................................. 38
Fgmm4. .................................................................................................................. 38
Figure 5 .................................................................................................................. 39
Figure 6 .................................................................................................................. 39
Figure 7 .................................................................................................................. 40
Figure 8 .................................................................................................................. 40
Figure 9 .................................................................................................................. 41
Figure 10 ................................................................................................................ 41
Figure 11 ................................................................................................................ 42
Figure 12 ................................................................................................................ 42
Figure 13 ................................................................................................................ 43
Figure 14 ................................................................................................................ 44

CHAPTER 2
Figure l .................................................................................................................. 66
Figure 2 .................................................................................................................. 66
F1gure3 .............................................................................. 67
Figure 4 .................................................................................................................. 67
Figure 5 .................................................................................................................. 68
Figure 6 .................................................................................................................. 69
Figure 7 .................................................................................................................. 69

CHAPTER 3
Figure l .................................................................................................................. 95
Figure 2 .................................................................................................................. 96
Figure 3 .................................................................................................................. 96
Figure 4 .................................................................................................................. 97
Figure 5 .................................................................................................................. 97
Figure 6 .................................................................................................................. 98
Figure 7 .................................................................................................................. 99
Figure 8 .................................................................................................................. 100
Figure 9 .................................................................................................................. 100
Figure 10 ................................................................................................................ 101
Figure 11 ................................................................................................................ 101
Figure 12 ................................................................................................................ 102
Figure 13 ................................................................................................................ 102

viii

LIST OF PUBLICATIONS
Peer Reviewed Manuscrimts

Rundell, S., Haut RC, 2004. Tissue equilibration alters the response of cartilage explants
to unconﬁned compression. Journal of Biomechanics, In Revision

Jex, C.T., Rundell, S., Wan, C., MacDonald, B., Haut, RC, 2004. The effect of ﬁxation
types on the biomechanical properties of the Well osteotomy. Journal of Foot and Ankle
Surgery, In Revision

Rundell, S., Baars, D., Phillips, D., Haut, RC, 2004. Repair of damaged chondrocytes in
the in vivo traumatized joint. Journal of Orthopaedic Research, In Revision.

Baars, D., Rundell, S., Haut, RC, 2005. Treatment with the non-ionic surfactant
Poloxamer pl 88 reduces TUNEL positive cells in bovine chondral explants exposed to
injurious unconﬁned compression. Biomechanics and Modeling in Mechanobiology, In
Review

Peer Reviewed Abstrﬂ

Rundell, S., Haut R.C., Tissue equilibration alters the response of cartilage explants to
unconﬁned compression. 5 lst Annual Meeting of the Orthopaedic Research Society,
2005

Rundell, S., McPhilamy, A., Orth, M., Haut, R.C., Glucosamine supplementation can

help limit matrix damage and adjacent cell death in traumatized explants. 5 lst Annual
Meeting of the Orthopaedic Research Society, 2005

ix

Introduction

Lower extremity injuries account for approximately $21.5 billion each year in
treatment, rehabilitation, and lost work day expenses (Miller, 1995). These injuries are a
frequent outcome of automobile accidents, comprising nearly 25% of the total injtu'ies
(Luchter et al., 1995). A more recent analysis of the National Accident Sampling System
database has found that 10% of these injuries involve trauma to the knee joint, while only
5% of these injuries result in fracture. The Federal Safety Standard requires that femur
forces generated during a knee-instrument panel collision must not exceed 10 kN, a force
that has been shown to result in gross fracture of the patella, femur, or pelvis (Melvin et
al., 1975; Patrick et al., 1965; Powell et al., 1975). However, the presence of chronic joint
disease has been found in patients that have only suffered from a sub—fracture knee injury
(States, 1970).

Osteoarthritis (0A) is a chronic joint disease characterized by the loss of articular
cartilage. Articular cartilage is the connective tissue that lines the ends of bones in
diarthroidial joints. Its ﬁrnction is to provide a near frictionless surface on which bones
can glide over, as well as absorb the shock from physical motion. Cartilage consists of a
solid organic matrix and free interstitial ﬂuid, which is mostly water. The solid matrix
mainly consists of collagen (type II) and proteoglycans (PGs) (Mow et al., 1990). The
PGs attract water and repel each other due to their electro-negativity. The network of
collagen ﬁbers resists this swelling (Grodzinsky et al., 1978), and gives cartilage its
compressive stiffness. Chondrocytes (cartilage cells) are imbedded in the solid matrix and

are responsible for synthesis and degradation of the solid matrix constituents. It has been

suggested that the pathogenesis of 0A is the result of an imbalance in solid matrix
turnover with degradation exceeding synthesis (Goldring, 2000; Malemud 1999).

Clinically 0A is characterized by joint pain and narrowing of the joint, as
diagnosed by radiological examination (Flores and Hochber, 1998). Pathologically, the
disease exhibits a loss of cartilage and sclerosis of underlying bone. While the etiology
of DA is not fully understood the risk of developing this disease is increased signiﬁcantly
in joints suffering a major injury (Felson et al., 2004). However, establishing a cause and
effect relationship between joint injury and CA can be difﬁcult as after an osteochondral
injury, joint disease may not be diagnosed for 2 to 5 years, while less severe joint injuries
may not be diagnosed for 10 or more years after the trauma (Wright, 1990).

Animal models have been developed in order to study the associations between
blunt impact trauma to a joint and the subsequent development of chronic joint disease.
In a recent study a rigid impact mass was dropped with a 6 Joule intensity onto the ﬂexed
patello-femoral joint of a giant Flemish rabbit (Ewers et al., 2002). This study
documented a progressive increase in the degradation of retro-patellar surface cartilage
and the thickening of underlying subchondral bone after 3 years. A separate study that
used a New Zealand white rabbit model and a 10 Joule impact intensity, found
signiﬁcantly advanced OA-like changes in the patello-femoral joint, with ﬁbrillation,
ulceration and erosion of retro-patellar cartilage within 6 months post-contusion
(Mazieres et al., 1987)). Early OA-like changes have also been described using the
ﬂexed canine patello-femoral joint subjected to approximately 2.2 kN of impact force

delivered with a gravity-dropped rigid mass (Thompson et al., 1991). While these studies

indicate a link between blunt impact trauma to a joint and a subsequent degradation of
joint tissue, the mechanisms responsible are still unclear.

Studies that examine the acute response of articular cartilage after a blunt impact
to a joint both in viva and in situ document gross surface lesions on the cartilage surface
(Atkinson and Haut, 2001; Newberry et al., 1998). Also, In vitra chondral and
osteochondral explant models have been developed in order to examine the acute
response of injurious loading on articular cartilage. Studies that have subjected either
chondral or osteochondral explants to an injurious level of unconﬁned compression have
documented surface ﬁssuring across the surface, as well as associated chondrocyte death
(Ewers et al., 2001; Krueger et al., 2003). Acute chondrocyte death, or necrosis, has also
been linked to degradation of articular cartilage after 1 year in viva (Simon et al., 1976).
These studies found that a severe impact to a joint will result in gross ﬁssuring of the
articular cartilage, as well as a reduction in chondrocyte viability. While the exact
mechanisms responsible for long term degradation of articular cartilage after a blunt
impact are not fully understood, the acute development of surface ﬁssuring and
chondrocyte death are among the suspected factors.

Matrix damage in the form of surface ﬁssuring has been suggested to occur when
the interstitial ﬂuid pressurization in cartilage developed during loading becomes too
great and exceeds the restraining capacity of the collagen network (Morel and Quinn,
2004; Pins et al., 1995). A study that exposed cartilage explants to varying rates of
injurious loading documents an increased likelihood of surface ﬁssuring at rates of
loading that exceed the gel diffusion rate (Morel and Quinn, 2004). Theoretical models

that simulate the effects of a blunt impact on an intact joint document the presence of

high shear (distortional) strains near the articular surface, when cartilage is modeled as
transversely isotropic and biphasic (Garcia et al., 1998; Donzelli et al., 1999).
Interestingly, acute chondrocyte death has been found to occur predominately around
surface cracks when cartilage is loaded at a high rate (Ewers et al., 2001; Lewis et al.,
2003). However, when the rate of loading is slower and closer to the gel diffusion rate a
more diffuse pattern of chondrocyte death throughout the thickness has been found
(Morel and Quinn, 2004: Quinn et al., 2001). In a recent study that performed conﬁned
compression on chondral explants it was documented that cell death in the superﬁcial
zone occurred only when water was allowed to ﬂow out (Milentijevic and Torzilli, 2005).
This study suggests that chondrocyte death can either occur by compaction of the
superﬁcial zone, which occurs when water ﬂows out, or by collagen tensile failure where
chondrocyte death occurs around cracks. This study also suggests that hydrostatic
pressure generated during a high rate of loading has a protective effect on cells, where as
at slower rates water is allowed to ﬂow which gives rise to distortional strains.
Understanding the mechanisms responsible for acute damage of cartilage and
associated chondrocyte death are essential in the ﬁrture development of therapeutic
methods to either prevent or treat these early alterations. The research presented in the
current thesis uses both the in vitra chondral explant and in viva rabbit models to examine
the acute response of articular cartilage to blunt impact loading. Chapter 1 addressed the
issue of injury severity in chondral explants exposed to a culture medium prior to
mechanical loading versus explants taken directly from the joint. This study
hypothesized that excess ﬂuid present in explants allowed to bathe in a culture medium

would result in an artiﬁcially high amount of surface ﬁssuring and associated

chondrocyte death. Chapter 2 describes experiments in which knee joints of giant
Flemish rabbits were subjected to a 6 Joule intensity impact and retro-patellar cartilage
was studied in terms of surface frssuring as well as chondrocyte death. No previous
studies have attempted to document acute chondrocyte death as a result of a blunt impact
in an in viva model. This study also evaluated the efﬁcacy of a mild non-ionic
surfactant, poloxamer 188, which has been found to reduced chondrocyte death in vitra
(Phillips and Haut, 2004), in ‘repairing’ damaged cells after an impact in viva. Chapter 3
evaluated the efficacy of the nutraceutical glucosamine in enhancing the mechanical
integrity of cartilage explants when pre-treated for a period of 6 days prior to an injurious
unconﬁned compression test.

The research presented in this thesis provides useful data in regards to the
response of articular cartilage to blunt impact loading in both an in vivo and an in vitra
setting. Furthermore, potential therapeutic treatments have been investigated and found
to be effective in preventing damage that may result in long term degradation of articular
cartilage. Future studies can utilize the data presented in this thesis to design experiments
that will further help explain the mechanisms responsible for the efﬁcacy of these

treatments .

References

Atkinson PJ, Haut RC, 2001. Injuries produced by blunt trauma to the human
patellofemoral joint vary with ﬂexion angle of the knee. J Orthop Res
Sep;19(5):827-33

Donzelli PS, Spilker RL, Ateshian GA, Mow VC, 1999. Contact analysis of biphasic
transversely isotropic cartilage layers and correlations with tissue failure. J Biomech
32(10):]037-47

Ewers BJ, Dvoracek-Driksna D, Orth MW, Haut RC, 2001. The extent of matrix damage
and chondrocyte death in mechanically traumatized articular cartilage explants
depends on rate of loading. J Orthop Res 192779-84

Ewers BJ, Weaver BT, Sevensma ET, Haut RC, 2002. Chronic changes in rabbit retro-

patellar cartilage and subchondral bone after blunt impact loading of the
patellofemoral joint. J Orthop Res 20:545-50

Felson DT, 2004. An update on the pathogenesis and epidemiology of osteoarthritis.
Radiol Clin North Am 42: 1-9

Flores R, Hochber M, 1998. Deﬁnition and classiﬁcation of osteoarthritis. In: Brandt K,
Doherty M, Lohmander S, editors. Osteoarthritis. Oxford: Oxford University Press:
p. 1—12.

Garcia JJ, Altiero NJ, Haut RC, 1998. An approach for the stress analysis of transversely
isotropic biphasic cartilage under impact load. J Biomech Eng 120(5):608-13

Grodzinsky AJ, Lipshitz H, Glimcher MJ, 1978. Electromechanical properties of articular
cartilage during compression and stress relaxation. Nature 275(5679):448-50

Goldring MB, 2000. The role of the chondrocyte in osteoarthritis. Arthritis Rheum
43:1916-26

Krueger JA, Thisse P, Ewers BJ, Dvoracek-Driksna D, Orth MW, Haut RC, 2003. The
extent and distribution of cell death and matrix damage in impacted chondral
explants varies with the presence of underlying bone. J Biomech Eng 125:114-9

Lewis, J.L, Deloria, LB, Oyen-Tiesma, M, Thompson, RC, Jr., Ericson, M, Oegema TR,
Jr., 2003. Cell death after cartilage impact occurs around matrix cracks. J Orthop
Res 21, 881-7

Luchter S, Walz MC, 1995. Long-term consequences of head injury. J Neurotrauma.
12(4):517-26

Malumed CJ, 1999. Fundamental pathways in osteoarthritis: an overview. Front Biosci
15: D659-61

Mazieres B, Blanckaert A, Thiechart M, 1987. Experimental post-contusive osteo-
arthritis of the knee. Quantitative microscopic study of the patella and the femoral
condyles. J Rheum 14:1 19-21

Melvin J, Stalnaker R, Alem N, Benson J, Mohan D, 1975. Impact response and tolerance
of the lower extremities. 19th annual Stapp Car Crash Conf. 192543-559

Milentijevic D, Torzilli PA, 2005. Inﬂuence of stress rate on water loss, matrix

deformation and chondrocyte viability in impacted articular cartilage. J Biomech
38, 493-502

Miller TR, Levy DT, 1995. The effect of regional trauma care systems on costs. Arch
Surg 130(2):]88-93.

Morel V, Quinn TM, 2004. Cartilage injury by ramp compression near the gel diffusion
rate. J Orthop Res 22:145-151

Mow VC, 1990. Fundamentals of articular cartilage and meniscus biomechanics, in:
Articular Cartilage and Knee Joint Function: Basic Science and Arthroscopy, J.W.
Ewing, ed., Raven Press, Ltd., New York, ppl-18.

Newberry WN, Garcia JJ, Mackenzie CD, Decamp CE, Haut RC, 1998. Analysis of acute
mechanical insult in an animal model of post-traumatic osteoarthrosis. J Biomech
Eng 120(6):704-9

Patrick LM, Koell CK, Mertz HJ Jr., 1965. Forces on the human body in simulated
crashes. 9th Annual Stapp Car Crash Conf. 12:237-259

Phillips DM, Haut RC, 2004. The use of a non-ionic surfactant (P188) to save
chondrocytes from necrosis following impact loading of chondral explants. J
Orthop Res 22:1135-42

Pins GD, Huang EK, Christiansen DL, Silver FH, 1995. Effects of static axial strain on
the tensile properties and failure mechanisms of self-assembled collagen ﬁbers. J
Appl Polym Sci 63:1429—40

Powell WR, Ojala SJ, Advani SH, 1975. Cadaver femur responses to longitudinal
impacts. 19'h Annual Stapp Car Crash Conf. 19:561-579

Quinn TM, Allen RG, Schalet BJ, 2001. Matrix and cell injury due to sub-impact loading
of adult bovine articular cartilage explants: effects of strain rate and peak stress. J
Orthop Res 19(2):242—9

Simon WH, Richardson S, Herman W, Parsons JR, Lane J, 1976. Long-term effects of
chondrocyte death on rabbit articular cartilage in vivo. J Bone Joint Surg Am
58:517-26

States JD, 1970. Traumatic arthritis: a medical dilemma, In: Proceedings of the 14th
Annual Conference of the American Association for Automotive Medicine. 14:21-
28

Thompson RC, Oegema T, Lewis J, Wallace L, 1991. Osteoarthritic changes after acute
transarticular load. J Bone Joint Surg Am 73:990-1001

Wright V, 1990. Post-traumatic osteoarthritis- A medico-legal mineﬁeld. Br J Rheum
29:474-8

CHAPTER 1: TISSUE EQUILIBRATION ALTERS THE RESPONSE OF
CARTILAGE EXPLANTS TO UNCONFINED COMPRESSION

Abstract:

Excessive mechanical loading can lead to matrix damage and chondrocyte death in
articular cartilage. Previous studies on chondral explants have not clearly distinguished to
what extent the degree and the distribution of cell death are dependent on the amount of
free swelling seen during tissue equilibration. The current study hypothesized that
increased ﬂuid content inside equilibrated chondral explants, when subjected to injurious
compression, would lead to greater matrix damage. Equilibrated and non-equilibrated
chondral explants were loaded to 3OMPa at a fast rate of loading (~600MPa/s) in an
unconﬁned compression experiment. Stress-strain curves were created for each explant,
and ﬁt with a hyperelastic Ogden ﬁrnction using a least squares method. After 24 hours in
a culture medium, matrix damage was assessed by the total length of surface ﬁssures. The
explants were also sectioned and stained for cell viability in the various layers of the
cartilage. A separate group of explants was exposed to a displacement controlled
unconﬁned compression experiment at a fast rate of loading (1200%c/s). These explants
were dried out immediately after loading in order to separate contributions of the solid
and ﬂuid phase to the mechanical response. The stiffness of the equilibrated specimens
was less than non-equilibrated specimens, and it correlated with the amount of ﬂuid taken
in during equilibration. More matrix damage in the form of surface ﬁssuring and cell
death in the superﬁcial zone was documented in equilibrated explants. Matrix damage
correlated positively with ﬂuid gained during equilibration. The amount of cell death in

the deep layer of the cartilage was less in equilibrated versus non-equilibrated explants,

and also depended on the amount of ﬂuid the specimens took in during equilibration.
Correlations between peak stress and the percentage of water in the explant as well as dry
weight in the displacement controlled tests were documented to decrease when explants
were allowed to equilibrate. This study indicated that equilibration alters explant response
to mechanical loading in terms of stifﬁress, matrix damage and cell viability. The
alterations in response depend on percentage uptake of ﬂuid during tissue equilibration.
These data may have relevance to the applicability of experimental data from equilibrated
chondral explants to the in viva condition.

Introduction:

In early stages of osteoarthritis (OA), matrix damage is often seen as ﬁssures on
the articular surface (Brandt et al., 1986). Chondrocyte death, resulting from a reduction
in tissue cellularity or chondrocyte malfunction, has been suggested to facilitate
development of OA (Blanca et al., 1998; Buckwalter 1995). Excessive levels of
mechanical load can generate acute matrix damage and chondrocyte death in articular
cartilage (Ewers et al., 2001). However, establishing a cause and effect relationship
between impact trauma and development of CA has been difﬁcult as joint injuries with
bone ﬁ'acture may require 2-5 years to develop clinical symptoms, while less severe
injuries may not be diagnosed for 10 or more years after trauma (Wright, 1990).
Therefore, animal models have been used to study the potential association of trauma and
disease. For example, blunt impact to the rabbit patella-femoral joint has indicated acute
surface ﬁssuring and a subsequent chronic degradation of the traumatized articular
cartilage with thickening of the underlying subchondral bone (Newberry et al., 1998).

Computational models of the joint have been developed to help investigate correlations

10

between the acute state of stress and strain with the subsequent changes in joint tissues
(N ewberry et al., 1998), but these models are complex and not yet validated.

To help clarify potential associations between mechanical events and early
alterations in joint tissue, chondral (Ewers et al., 2001; Milentijevic et al., 2003; D’Lima
et al., 2001) and osteochondral (Kreuger et al., 2003; Clements et al., 2001) explant
models have also been developed for the laboratory setting. While these experimental
models provide more ideal geometries for mechanical analyses, they are in vitro models
that may not adequately represent the in vivo situation. In a study that compared the
amounts of matrix damage and cell death in chondral versus osteochondral explants when
exposed to 30 MPa of pressure, the presence of underlying bone reduced matrix damage
and cell death (Krueger et al., 2003). Similarly in situ cadaveric human joint studies
show minimal surface ﬁssuring at contact pressures in the range of 20 to 30 MPa
(Atkinson and Haut 2001, Atkinson and Haut 1995). Also, an in vivo rabbit study
documented minimal or no occurrence of surface ﬁssuring on the retropatellar surface at
pressures of approximately 30 MPa delivered in roughly 50 milliseconds (Ewers et al.,
2002). A study which used similar rates and amplitudes of loading on chondral explants
documented excessive ﬁssuring throughout the whole explant surface (Ewers et a1, 2001).
Even studies using osteochondral explants exposed to the same level of high rate
unconﬁned compression experienced substantial surface damage (Krueger et al., 2003).
One possible explanation for differences in the extent of surface ﬁssuring in in situ and in
vitro models may be the amount of tissue ﬂuid present in the in situ joint cartilage versus
that of explanted tissue bathed for a period of time in a standard culture medium. This

procedure is common in cartilage explant studies (Borelli et al., 1997; Bush and Hall,

11

2003; DiMicco et al., 2004: Sauerland et al., 2003) and is typically referred to as tissue
equilibration.

Interstitial ﬂuid pressurization has been shown to contribute to the functional
response of articular cartilage (Mansour and Mow, 1976; Mow et al. 1980; Zarek and
Edward, 1963). This pressurization is the result of polyanionic glycosaminoglycans
which create an electrostatic tissue swelling and resistance to compression. When
collagen ﬁbers are damaged, they lose their ability to resist swelling. This phenomenon
has been documented in a number of studies. Grushko and co-workers ( 1989) noted that
damaged or ﬁbrillated human cartilage plugs swelled when placed in physiological
saline. Sah and co-workers (1989) showed a swelling of excised bovine articular
cartilage when placed in physiologic saline. Swelling of articular cartilage has been
observed to be non-uniform and anisotropic (Myers et al., 1984, Maroudas et al., 1986)
which characterizes the non-uniform organization of collagen (Clark, 1991; Clark, 1990)
and the non-uniform distribution of negatively charged proteoglycans (Maroudas a al
1969). It follows that when explanting cartilage from a joint, the collagen matrix is
disrupted allowing for changes in ﬂuid content that arises when the tissue is bathed in a
standard culture medium.

Articular cartilage consists of a solid and ﬂuid phase which, due to their
interation, determines the mechanical characteristics of the tissue. Theoretical and
computational analyses have separated the contributions of the ﬂuid and solid phase in
load bearing, and found that the ﬂuid phase is responsible for approximately 90% of the
load bearing (Ateshian et al., 1994; Ateshian and Wang, 1995; Macirowski et al., 1994).

It is believed that interstitial ﬂuid pressurization is the dominant mechanism of load

12

support (Park et al., 2003). Current studies, which induce or inhibit cartilage swelling by
varying levels of NaCl concentrations in tissue baths (Maroudas, 1975; Lee et al., 1981),
ﬁnd signiﬁcant effects occur in the tissue’s mechanical response characteristics. Given
that the ﬂuid content and the associated interstitial ﬂuid pressurization has such a large
effect on the load bearing capacity of this tissue, any changes in these parameters may
result in an altered mechanical response during blunt immct. In addition, matrix damage
in the form of surface ﬁssuring, has been attributed to ﬂuid pressurization and a
subsequent tensile damage to the collagen network when the tissue is subjected to blunt
impact loading (Morel and Quinn, 2004). Increases in cartilage explant ﬂuid
contentresulting from tissue equilibration, could potentially result in increased surface
ﬁssuring during high rates of unconﬁned compression. A previous study has shown that
cell death in an intact bovine patella occurs exclusively around impact induced surface
cracks (Lewis et al., 2003) suggesting that if in fact increased ﬂuid results in a higher
incidence of surface ﬁssuring there will likely be an associated increase in chondrocyte
death.

The hypothesis of the current study was that excised chondral explants would
swell in the medium during equilibration and the increase in ﬂuid content would then
alter the mechanical response of the explants resulting in greater matrix damage for
equilibrated explants versus non-equilibrated tissue when exposed to a high rate of
injurious compression.

Methods and Materials:
Three pairs of skeletally mature (12-24 months) bovine forelegs were obtained

ﬁ'om a local abattoir within 3 hours of slaughter. The legs were cut proximal to the

13

metacarpal surface leaving the joint intact. The legs were rinsed with distilled water,
skinned, and rinsed again prior to opening the joint under a laminar ﬂow hood. A 6-mm
biopsy punch was used to make 52 cartilage plugs. These plugs were removed ﬁom the
underlying bone with a scalpel and divided evenly into two groups, equilibrated and non-
equilibrated. All explants were weighed immediately after removal from the joint.

Twenty-six explants were assigned to a so-called “equilibrated” group. These
were placed in Dulbecco’s Modiﬁed Eagle’s Medium (DMEM): F12 (Gibco, USA
#12500-039) supplemented with 10 % fetal bovine serum, additional amino acids, and
antibiotics (penicillin 100 units/ml, streptomycin lug/ml, amphotericin B 0.25 rig/ml).
They were then washed three times, and allowed to equilibrate for one day in a humidity-
controlled incubator (37° C, 7.2% C02, NuAire, Plymouth, MN). After equilibration
these specimens were subjected to mechanical loading and retrn'ned to incubation for
another 24 hours. They were then evaluated in terms of surface damage and cell
viability.

Another group of twenty-six were assigned to the “non-equilibrated” group.
These explants were taken directly to a servo-hydraulic machine (Instron, model 1331
with 8500 upgraded electronics) for immediate mechanical testing of the explants. Tests
were performed immediately after removal from the bone for each explant in order to
avoid potential drying out. These specimens were then placed in supplemented
DMEM:F12 and incubated for twenty-four hours before being examined for cell viability
and matrix damage. Six equilibrated and non-equilibrated specimens were chosen as non-
impact controls. These explants followed the same procedure as their respective groups

with the exception of receiving impact.

14

The unconﬁned compression experiments were performed with a high rate of
loading (50 ms to peak, ~600MPa/s). Each explant was placed between two highly
polished stainless steel plates (ﬁgure 1). Prior to the test, the plates were pressed together
at 100 N and the location of the machine actuator was recorded. The thickness of each
specimen was determined by ﬁnding the difference in the actuator’s location at 5 N of
load during the test and when the plates were pressed together prior to loading. Each
specimen received a 0.5 N preload before being compressed to a peak load of 857N
(~30MPa) using a single haversine load-time pulse. This protocol has been established in
previous experiments from this laboratory (Ewers et al., 2001; Krueger et al., 2003). The
load, time, and actuator displacement were recorded during each experiment with an
accuracy of 0.1 N, 0.001 s, and 0.01 mm, respectively. After loading, each explant was
washed three times in media before being returned to pre-assigned wells with one ml of
supplemented media and incubated for an additional 24 hours.

Twenty-four hours after the mechanical compression experiment, each explant
was examined for matrix damage and cell viability. The surface of each explant was
wiped with India ink. The explants were immediately photographed at 25X under a
dissection microscope (Wild MSA, Wild Heerbrugg Ltd., Switzerland) to determine the
total length of the surface ﬁssures (ﬁgure 2). The total ﬁssure length was measured with
digital imaging software (Sigma Scan, SPSS Inc., Chicago, IL). One observer (AM)
digitally recorded the length of the surface ﬁssures in each photograph.

For cell viability, two 0.5-mm slices were taken through the thickness at the
center of each explant using a customized cutting tool (Ewers et al., 2001). The sections

were stained with a kit containing calcein and ethidium bromide homodimer (Live/Dead

15

&Viabﬂity/Cytotoxity, Molecular Probes, Oregon). Each section was viewed under a
ﬂuorescence microscope at 100X (Lecia DM LB (frequency: ,50-60 Hz), Lecia
Mikroskopie and Systeme GmgH, Germany). Full thickness, digital images were taken of
a 2.5-mm length at the center of each explant. These images were partitioned into the
superﬁcial zone (top 20%), intermediate zone (middle 50%), and deep zone (bottom
30%). The viable (green) and dead (red) cells were manually counted by one observer
(AS) using image software (Sigma Scan, SPSS Inc., Chicago, IL) (Figure 3). The percent
of cell death was computed for each layer.

Stress-strain curves were determined using the collected load and displacement
data from the unconﬁned compression tests of the equilibrated and non-equilibrated
explants (Figure 4). Compressive stress and corresponding strain were determined at 1,
5, 10, 15, and 20 MPa and used to generate an average response curve for each group of
explants. To help quantify the overall responses of each group of explants during
unconﬁned compression the stress-strain data were curve ﬁt to a hyperelastic Ogden

function using a least squares method,
O'(.ua a, 8) : Z’L—lkl + g)a—1 _(1+ 8)-0.5a—1]
a

where c is the compressive stress, a is compressive strain, p. is a shear modulus
parameter, and or is the stiffening factor. In order to ensure that the values for p. and
a converged, the initial guesses, determined from a preliminary analysis, were multiplied
by a random constant between 0.1 and 10 before conducting the curve ﬁtting procedure.
The process was repeated 100 times for each curve ﬁt in order to obtain a guess that

resulted in the minimum amount of error (Abramowitch and Woo, 2004). All guesses

16

tried in this range resulted in the same values for p. and a. R squared values were
calculated for each curve in order to quantify the goodness of ﬁt. The overall error (SSE)
was calculated by subtracting the theoretical stress from the actual stress at l, 5, 10, 15,
and 20 MPa and summing all the values. In order to quantify the goodness of ﬁt of the
curve, the r squared value was calculated. The r squared value (R2) was calculated by

taking the ratio of the sum of squares of the regression (SSR) and the total sum of squares

(SST).

SSR = 2(ath(ﬂaaa£)i —0-avg 2

i=1

n
_ 2
SST - Z (0(actual)i — aavg
i=1

R 2 _ SSR
SST

Unpaired t-tests were used to compare the peak strain, ﬁssure length, n, or, and
total cell death for equilibrated and non-equilibrated explants. Repeated factors AN OVA
was used to compare cell death between the different layers through the thickness.
Pearson correlation tests were performed in order to determine statistical signiﬁcance of
correlations between parameters within each group of explants. All data were reported as
mean :l: one standard deviation. Statistical signiﬁcance was indicated at p<0.05.
Controlled Displacement Tests

A separate group of a total of 96 (48 equilibrated and 48 un-equilibrated) explants

from 4 bovine forelegs were subjected to a displacement controlled unconﬁned

l7

compression experiment. The unconﬁned compression experiments were performed with
a high rate of loading (50 ms to peak displacement, ~1200%a/s). Each explant was placed
between two highly polished stainless steel plates (Figure 1). Prior to the test, the plates
were pressed together at 100 N and the location of the machine actuator was recorded.
The thickness of each specimen was determined by ﬁnding the difference in the
actuator’s location at a preload of -0.5 N and when the plates were pressed together.
Using the thickness of each explant the actuator arm was programmed to travel a total
displacement corresponding to 60% strain (a) for each individual explant. Each specimen
received a 0.5 N preload before being compressed to a peak strain of 60% using a single
haversine load-time pulse. The load, time, and actuator displacement were recorded
during each experiment with an accuracy of 0.1 N, 0.001 s, and 0.01 mm, respectively.
After loading, each explant was dried out in a vacuum oven for 24 hours in order to
determine the dry weight (DW).

Besides differences in mechanical loading equilibrated and non-equilibrated
explants used in these tests underwent the same procedure as previously described. Dry
weights were obtained for these specimens in order to calculate the percentage of ﬂuid
present in the explant at the time of mechanical loading. Data from these tests was used
to separate contributions of the ﬂuid and solid phase to the mechanical response of
cartilage explants. Chondrocyte viability and matrix damage were not assessed because
this does not allow for the ﬁrll explant to be dried out after mechanical loading. Pearson
correlation tests were performed to ﬁnd correlations between peak stress and explant
thickness, explant weight, dry weight, percent ﬂuid content, and percent ﬂuid gained

during equilibration. Statistical signiﬁcance was indicated at p<0.05.

l8

Results:
The peak stress generated during loading for the equilibrated group (28.86 i 0.71

MPa, n=20) was not signiﬁcantly different than that for the non-equilibrated group
(29.08i 0.32 MPa, n=20). Nonlinear stress-strain curves were generated for explants
with and without equilibration. Equilibration decreased the explant stiffness by
producing a shift in the response curves to the right (Figure 5). The maximum strain for
the equilibrated explants (62 i 13 %, n=20) was signiﬁcantly different than that for the
non-equilibrated explants (56 i 6 %, n=20). There was no signiﬁcant differences in the
average thickness of the equilibrated explants (0.56 i 0.11 mm, n=20) versus that of the
non-equilibrated explants (0.59 i 0.09 mm, n=20). The Ogden function ﬁt the data with a
high degree of accuracy. The r2 values for both the equilibrated and non-equilibrated
explant response curves were 0.99 i 0.01 (n=20). All r2 values for individual tests were
above 0.95. The shear modulus parameter, p, for the equilibrated group (2.0 i 0.9 MPa,
n=20) was signiﬁcantly different than for the non-equilibrated group (2.9 i 0.9 MPa,
n=20). The value of at for the equilibrated group (-2.1 i 0.8 MPa, n=20) was not
signiﬁcantly different than that for the non-equilibrated group (-2.5 i 0.7 MPa, n=20).
The average ﬂuid gain for the equilibrated group was 13 i 5 % (n=20). All
values were positive. The shear modulus parameter, p, signiﬁcantly correlated inversely
(r2 = 0.44, p = 0.001) with ﬂuid gain for the equilibrated group (Figure 6). The peak
strain, on the other hand, signiﬁcantly correlated positively (r2 = 0.53, p < 0.001) with

ﬂuid gain in the equilibrated group (Figure 7).

l9

Equilibration of the chondral explants also signiﬁcantly increased the degree of
surface ﬁssuring for a 30 MPa unconﬁned compression compared to non-equilibrated
specimens (Figure 8). The total length of surface ﬁssuring for the equilibrated group was
57.6 :1: 15.2m (n=20) versus 45.1 i 17.8mm (n=20) for the non-equilibrated group
(p=0.022). In the equilibrated group, there was a signiﬁcant positive correlation (r'2 =
0.40, p = 0.003) between the length of surface ﬁssuring and the amount of ﬂuid gained by
the explant (Figure 9). There were no visible surface ﬁssures observed on the surfaces of
the non-impacted specimens for either group of chondral explants.

There was no signiﬁcant difference (p = 0.53) in the percentage of total cell death
in equilibrated (17 i 6 %, n=20) and non-equilibrated groups of explants (19 i 9 %,
n=20). Signiﬁcantly (p=0.006) less cell death, however, was noted in the deep zone of
the equilibrated explants (12 i 10 %, n=20) (Figure 10) versus the non-equilibrated
explants (25 i 17 %, =20) (Figure 11). In contrast, a signiﬁcant increase (p=0.016) in
the percentage of cell death was documented in the superﬁcial zone of the equilibrated
group of explants (36 i 12 %, n=20) versus the non-equilibrated group (27 i 10 %,
n=20). There was also a signiﬁcant inverse correlation (r2 = 0.36, p = 0.005) between
ﬂuid gain and the percentage of cell death in the deep zone for the equilibrated group of
explants (Figure 12). No cell death was documented in the non-impacted explants from
either group.

Controlled Displacement Tests
The peak strain generated during loading for the equilibrated group (62.5 i 3.7 %,
=48) was not signiﬁcantly different than that for the non-equilibrated group (61.4 i 3.2

%, n=48). The average ﬂuid gain for the equilibrated group was 12 i 9 % (n=48). The

20

average thickness of the equilibrated explants (0.71 i 0.11 mm, n=48) was signiﬁcantly
higher (p<0.001) versus that of the non-equilibrated explants (0.62 i 0.10 mm, n=48).
Signiﬁcant correlations were found between peak stress and thickness, explant weight,
dry weight, and percent ﬂuid content for the non-equilibrated explants (Figure 13).
Signiﬁcant correlations were found between peak stress and thickness, dry weight,
percent ﬂuid content, and percent ﬂuid gain for the equilibrated explants (Figure 14). The
strength of the correlations, as determined by the r value,» decreased in all instances for
the equilibrated group of explants. The signiﬁcance (p<0.001) of the correlation between
explant weight and peak stress documented in the non-equilibrated explants was lost
(p=0. 167) in the equilibrated explants.

Discussion:

The objective of this study was to document differences in the biomechanical
responses of chondral explants subjected to 30 MPa of unconﬁned compression that were
either taken directly from the joint or equilibrated in a standard culture medium for 24
hours before impact loading. The hypothesis of the study was that excised chondral
explants would swell in the medium during equilibration, and the increase in ﬂuid content
would then alter the mechanical response of the explants resulting in greater matrix
damage for equilibrated explants versus non-equilibrated tissue during a high rate of
unconﬁned compression. The current study conﬁrmed that equilibration of chondral
explants in a standard culture medium resulted in a signiﬁcant uptake of ﬂuid over 24
hours. The study also conﬁrmed that the increased ﬂuid content resulted in more matrix
damage resulting ﬁ'om a 30 MPa unconﬁned compression, as documented by an

increased extent of surface ﬁssuring in equilibrated versus non-equilibrated explants.

21

All equilibrated explants gained weight during the 24 hour period of equilibration
despite that the standard culture media used was hypertonic (300 mosM) with regards to
an isotonic osmolarity of 150 mosM. Parsons and Black (1979) documented an increase
in cartilage thickness in intact rabbit femoral condyles in hypotonic solutions (<150
mosM) and a corresponding decrease in hypertonic solutions (>150 mosM). However,
many explant studies document swelling at both physiological and hypertonic
osmolarities. A study by T orzilli et a1. (1997), for example, shows that bovine chondral
explants swell 10.2 i 11.0% in the thickness direction and approximately 3.31 i 2.67%
radially when exposed to physiological saline (~150 mosM) for 30 minutes. In another
study on calf cartilage disks swelling was observed axially 25-40% above the cut tissue
thickness of 1.0 mm, when incubated in culture media (~300 mosM) for 6 days (Sah et
al., 1989). In the current study the controlled displacement test equilibrated specimens
experienced approximately a 14% increase in thickness during 24 hours in culture media.
The current study documented a range in the amount of weight from 3 — 22% of the
original wet weight. This large range of swelling was similar to the variations mentioned
above in the previous studies by others. A study by Appleyard et al. (2003) documents
site-speciﬁc trends in the contents of collagen and proteoglycans across the surface of
articular cartilage on the ovine tibia. Similar ﬁndings were documented by Brama et al.
(2000) for mature equine metacarpophalangeal joints. Areas of cartilage rich in collagen
may better restrain tissue swelling in the bovine explant, but additional studies are needed
to more completely explain variations in the percentage of ﬂuid gain for various explants

taken from different areas on the same joint surface.

22

The percent of dead cells in the deep zone following impact loading for
equilibrated explants was substantially lower than that for non-equilibrated explants. This
difference in deep zone cell death contrasted with a higher percentage of cell death in the
superﬁcial zone for the equilibrated versus non-equilibrated explants. Previous studies
have documented the appearance of dead cells predominately around surface cracks
(Ewers et al., 2001; Lewis et al., 2003) resulting from high rates of injurious
compression. In the current study the increased length of surface ﬁssures documented in
the equilibrated explants helps explain the increase in the percentage of dead cells noted
in the superﬁcial zone. This, however, does not explain the seemingly protective effect of
an increase in ﬂuid volume on cells in the deep zone of equilibrated explants.
Unfortunately, the mechanisms responsible for cell damage resulting from a blunt impact
are yet unclear. The current data, in combination with previous studies that document
zonal dependencies on the amount of free swelling in articular cartilage (Narmoneva et
al., 1999), infer that susceptibility to cell death throughout the thickness of chondral
explants during blunt impact may, in part, depend on local ﬂuid content (Morel and
Quinn, 2004). However, more research in this area is required in order to validate this
hypothesis. The current study did suggest that changes in ﬂuid content arising ﬁom 24
hours of equilibration in culture media had an effect on the levels of cell death throughout
the thickness of chondral explants, but additional studies on mechanisms of cell death via
blunt impact loading are warranted.

On average, the total length of surface ﬁssuring on equilibrated explants was
greater than that on non-equilibrated explants. Previous studies correlate the onset of

surface ﬁssures with the development of high tensile strains in the collagen ﬁbril network

23

(Morel and Quinn, 2004; Askew and Mow, 1978; Eberhardt et al., 1991). Studies that
have modeled in situ blunt impact loading have documented high tensile strains near the
surface of the cartilage (Li et al., 1995; Atkinson et al., 1998). The compressive strain at a
given stress (Figure 1), and consequently the in-plane tensile strain, based on the Poisson
effect, was less for non-equilibrated cartilage possibly explaining the occurrence of more
surface ﬁssuring in the equilibrated explants. This, however, does not explain why
equilibrated explants exhibited a less stiff response to high rates of mechanical loading.
The precise reasons as to why an increased ﬂuid content would result in softening of the
tissue is not clear, however, it is possible that swelling created geometric differences in
the collagenous microstructure. A study by Wilson et al. (2004) used a poroviscoelastic
ﬁbril reinforced model to show that local stress and strain ﬁelds in cartilage are highly
inﬂuenced by collagen orientation. Interestingly the softening of the equilibrated explants
was observed to occur in the initial response to loading. This was exhibited by a
signiﬁcant decrease in the instantaneous shear modulus (u) with no signiﬁcant difference
in the stiffening factor (at) for equilibrated versus non-equilibrated explants. A possible
explanation may be that ﬂuid taken on by the explant during equilibration was more
easily expelled during the initial phase of loading. Although theoretical models show that
under impact loading cartilage responds as an incompressible elastic material (Armstrong
et al., 1984; Mak et al., 1987), i.e., ﬂuid cannot ﬂow out, ﬂuid taken on by chondral
explants during equilibration may have substantially increased the initial permeability of
the cartilage. Future studies may wish to observe swelling in the thickness direction and
how it correlates to deformation during the initial loading phase. The current study

suggests that the increased length of surface ﬁssures documented in the equilibrated

24

versus non-equilibrated explants may have resulted from greater tensile stresses
generated near the surface of equilibrated explants as a result of ﬂuid gained during
equilibration.

A high ﬂuid content, as a result of a damaged collagen network, is one of the
early alterations documented in osteoarthritis (Buckwalter, 1995; Bank et al., 2000).
Equilibrated explants displayed a reduction in dynamic stifﬁress exhibited by a greater
peak strain and a lower shear modulus parameter p. Both of these quantities were
correlated with the amount of ﬂuid gained during tissue equilibration. Osteoarthritic
cartilage has also been shown to exhibit a softening effect (Ateshian et al., 1994; Mow et
al., 1992; Setton et al., 1993) associated with tissue “edema” (Maroudas et al., 1986;
Mankin and Thrasher 1975). The present results may suggest that damage to the
collagenous network and associated tissue matrix swelling that occurs in early stages of
OA make diseased tissue more vulnerable to a blunt mechanical injury. Blunt mechanical
loads on a joint exhibiting early stages of disease may help accelerate the disease process
in these individuals. While additional studies are needed to validate this hypothesis, the
result may have important implications in forensic biomechanics. Additionally, tissue
swelling could make joint tissues relatively more susceptible to blunt impact injury
following a surgical procedure.

In the controlled displacement tests signiﬁcant correlations were documented
between the peak stress generated during unconﬁned compression and various
parameters. In all cases the correlations were stronger for non-equilibrated versus
equilibrated explants. In the case of explant weight, a signiﬁcant correlation was lost in

the equilibrated group of explants. Previous studies that have determined the mechanical

25

properties of cartilage across the surface of an intact joint document changes in stiffness
parameters that correlate with site-speciﬁc collagen and proteoglycan content (Appleyard
et al., 2003; Brama et al., 2000). In the current study the decrease in the strength of
correlations in the equilibrated explants may indicate that relationships between solid
matrix constituents and the mechanical response of articular cartilage are shrouded by an
artiﬁcially high amount of ﬂuid acquired during ‘tissue equilibration’. Furthermore, the
peak stress was found to signiﬁcantly decrease as ﬂuid gained during equilibration
increased suggesting that increased ﬂuid leads to a less stiff response. These data agree
with the data obtained from the load controlled explant group.

In conclusion, cartilage explant studies that allow for a period of tissue
equilibration in standard culture media should consider the possible effects this may have
on the tissue’s mechanical characteristics. Severe levels of matrix damage and associated
cell death occurring in chondral explants exposed to injurious levels of compression may
be in part due to an artiﬁcially high level of interstitial ﬂuid due to the experimental
protocol. Changes in the explant stiffness, resulting from tissue swelling, have an effect
on chondrocyte viability throughout the thickness, as well as an increased susceptibility
for surface ﬁssuring. In order to help understand the large amount of variation in ﬂuid
gain between explants from the same joint surface noted in this study, additional studies
are needed.

Acknowledgements: This study was supported by grants from The National Center for
Injury Control and Prevention, The Centers for Disease Control and Prevention
(R49/CCR503607) and The TRW Automotive Fund. Its contents are solely the

responsibility of the authors and do not necessarily represent the ofﬁcial views of the

26

CDC or The TRW Automotive Fund. The authors wish to grateﬁrlly acknowledge the
help of Clifford Beckett, Austin McPhillamy, and Aaron Stewart (AS) for technical

assistance during this study.

27

References

Abramowitch, S.D., Woo, S.L.Y., 2004. An improved method to analyze the stress
relaxation of ligaments following a ﬁnite ramp time based on the quasi-linear
viscoelastic theory. Journal of Biomechanical Engineering 126, 92-97.

Appleyard, R.C., Burkhardt, D., Ghosh, P., Read, R., Cake, M., Swain, M.V., Murrell,
GA. 2003. Topographical analysis of the structural, biochemical and dynamic
biomechanical properties of cartilage in an ovine model of osteoarthritis.
Osteoarthritis and Cartilage 11, 65-77.

Armstrong, C.G., Lai, W.M., Mow, V.C., 1984. An analysis of the unconﬁned
compression of articular cartilage. Journal of Biomechanical Engineering 106,
165-173.

Askew, M., Mow, V.C., 1978. The biomechanical function of the collagen ﬁbril
ultrastructure of articular cartilage. Journal of Biomechanics 100, 105-1 15.

Ateshian, G.A., Lai, W.M., Zhu, W.B., Mow, V.C., 1994. An asymptotic solution for the
contact of two biphasic cartilage layers. Journal of Biomechanics 27, 1347-1360.

Ateshian, G.A., Wang, H., 1995. A theoretical solution for the ﬁictionless rolling contact
of cylindrical biphasic articular cartilage layers. Journal of Biomechanics 28,
1341-1355.

Atkinson, P.J., Haut, RC, 2001. Injuries produced by blunt trauma to the human

patellofemoral joint vary with ﬂexion angle of the knee. Journal of Orthopaedic
Research 19, 827-833.

Atkinson, P.J., Haut, RC, 1995. Subfracture insult to the human cadaver patellofemoral
joint produces occult injury. Journal of Orthopaedic Research 13, 936-944.

Atkinson, T.S., Haut, R.C., Altiero, N.J., 1998. Impact-induced ﬁssuring of articular
cartilage: an investigation of failure criteria. Journal of Biomechanical
Engineering 120, 181-187.

Bank, R.A., Soudry, M., Maroudas, A., Mizrahi, J., TeKoppele, J.M., 2000. The
increased swelling and instantaneous deformation of osteoarthritic cartilage is
highly correlated with collagen degradation. Arthritis and Rheumatism 43, 2202-
2210.

28

Blanco, F .J ., Guitian R., Vazquez-Martul E, de Toro F.J., Galdo F., 1998. Osteoarthritis
chondrocytes die by apoptosis: A possible pathway for osteoarthritis pathology.
Arthritis and Rheumatology 41, 284-289.

Borrelli, J., Torzilli, P.A., Grigiene, R., Helfet, DL, 1997. Effect of impact load on
articular cartilage: development of an intra-articular fracture model. Journal of
Orthopaedic Trauma 11, 319-326

Brama, P.A., Tekoppele, J.M., Bank, R.A., Barneveld, A., van Weeren, RR, 2000.
Functional adaptation of equine articular cartilage: the formation of regional

biochemical characteristics up to age one year. Equine Veterinary Journal 2000
May;32(3):217-21.

Brandt, KD., Mankin H.J., Shulman LE, 1986. Workshop on etiopathogenesis of
osteoarthiritis. Journal of Rheumatology 13, 1126-1160.

Buckwater, J.A., 1995. Osteoarthritis and articular cartilage use, disuse, and abuse:
Experimental studies. Journal of Rheumatology Supplement 43,13-15.

Bush, P.G., Hall, AC, 2003. The volume and morphology of chondrocytes within non-
degenerate and degenerate human articular cartilage. Osteoarthritis and Cartilage.
11, 242-251.

Clark, J .M., 1990. The organisation of collagen ﬁbrils in the superﬁcial zones of articular
cartilage. Journal of Anatomy. 171, 117-130.

Clark, J .M., 1991. Variation of collagen ﬁber alignment in a joint surface: a scanning
electron microscope study of the tibial plateau in dog, rabbit, and man.

Journal of Orthopaedic Research 9, 246-257.

Clements, KM., Bee, Z.C., Crossingham, G.V., Adams, M.A., Sarif, M., 2001. How
severe must repetitive loading be to kill chondrocytes in articular cartilage?
Osteoarthritis and Cartilage 9, 499-507.

DiMicco, M.A., Patwari, P., Siparsky, P.N., Kumar, S., Pratta, M.A., Lark, M.W., Kim,
Y.J., Grodzinsky, A.J., 2004. Mechanisms and kinetics of glycosaminoglycan
release following in vitro cartilage injury. Arthritis and Rheumatology. 50, 840-
848.

D’Lima, D.D., Hashimoto, S., Chen, P.C., Colwell, C.W. Jr., Lotz, MK, 2001. Impactof
mechanical trauma on matrix and cells. Clinical Orthopaedics and Related
Research 391, 890-894

Eberhardt, AW., Lewis, J.L., Keer, LM, 1991. Normal contact of elastic spheres with

two elastic layers as a model of joint articulation. Journal of Biomechanical
Engineering 113, 410-417.

29

Ewers, B.J., Dvoracek-Driksna, D., Orth, M.W., Haut, RC, 2001. The extent of matrix
damage and chondrocyte death in mechanically traumatized articular cartilage
explants depends on rate of loading. 19(5), 779-784.

Ewers, B.J., Jayaraman, V.M., Banglamaier, R.F., Haut, RC, 2002. Rate of blunt impact

loading affects changes in retropatellar cartilage and underlying bone in the rabbit
patella. Journal of Biomechanics 35, 747-755.

Krueger, J.A., Thisse, P., Ewers, B.J., Dvoracek-Driksna, D., Orth, M.W., Haut, RC,
2003. The extent and distribution of cell death and matrix damage in impacted
chondral explants varies with the presence of underlying bone. Journal of
Biomechanical Engineering 125, 114-119.

Lee, R.C., Frank, E.H., Grodzinsky, A.J., Roylance, D.K, 1981. Oscillatory
compressional behaviour of articular cartilage and the associated
electromechanical properties. Journal of Biomechanical Engineering 103, 280-
292.

Lewis, J.L., Deloria, L.B., Oyen-Tiesma, M., Thompson, R.C., Jr., Ericson, M., Oegema,
T.R., Jr., 2003. Cell death after cartilage impact occurs around matrix cracks.

Journal of Orthopaedic Research 21, 881-887.

Li, X., Haut, R.C., Altiero, N.J., 1995 . An analytical model to study blunt impact
response of the rabbit P-F joint. Journal of Biomechanical Engineering 117, 485-
491.

Macirowski, T., Tepic, 8., Mann, R.W., 1994. Cartilage stresses in the human hip joint.
Journal of Biomechanical Engineering 116, 10-18.

Mak, A.F., Lai, W.M., Mow, V.C., 1987. Biphasic indentation of articular cartilage-I.
Theoretical analysis. Journal of Biomechanics 20, 703-714.

Mankin, H.J., Thrasher, AZ, 1975. Water content and binding in normal and
osteoarthritic human cartilage. Journal of Bone and Joint Surgery 57A, 76-80.

Maroudas, A., 1975. Biophysical chemistry of cartilaginous tissues with special reference
to solute and ﬂuid transport. Biorheology 12, 233-248.

Maroudas, A., Mizrahi, J., Katz, E.P., Wachtel, E.J., Soudry, M., 1986. Physiochemical
properties and functional behavior of normal and osteoarthritic human cartilage.
In: Kuettner, K., et al. (Eds), Articular Cartilage Biochemistry. Raven Press, New
York, pp. 311-329.

Maroudas, A., Muir, H., Wingham, J., 1969. The correlation of ﬁxed negative charge
with glycosaminoglycan content of human articular cartilage.Biochimica et
Biophysica Acta 177, 492-500.

30

Mansour, J .M., Mow, V.C., 1976. The permeability of articular cartilage under
compressive strain and at high pressures. Journal of Bone and Joint Surgery 58,
509-16.

Milentijevic, D., Helfet, D.L., Torzilli, RA, 2003. Inﬂuence of stress magnitude on water

loss and chondrocyte viability in impacted articular cartilage. Journal of
Biomechanical Engineering 125, 594-601.

Morel, V., Quinn, T.M., 2004. Cartilage injury by ramp compression near the gel
diffusion rate. Journal of Orthopaedic Research 22, 145-151.

Mow, V.C., Kuei, S.C., Lai, W.M., Armstrong, CG, 1980. Biphasic creep and stress
relaxation of articular cartilage in compression? Theory and experiments. Journal
of Biomechanical Engineering 102, 73-84.

Mow, V.C., Ratcliffe, A., Poole, AR, 1992. Cartilage and diarthrodial joints as
paradigms for hierarchical materials and structures. Biomaterials 13, 67-97.

Myers, E.R., Lai, W.M., Mow, V.C., 1984. A continuum theory and an experiment for
the ion-induced swelling behavior of articular cartilage. Journal of Biomechanical
Engineering 106, 151-158.

Narmoneva, D.A., Wang, J.Y., Setton, LA, 1999. Nonuniform swelling-induced residual
strains in articular cartilage. Journal of Biomechanics 32, 401-408.

Newberry, W.N., Garcia, J.J., Mackenzie, C.D., Decamp, C.E., Haut, RC, 1998.
Analysis of acute mechanical insult in an animal model of post-traumatic
osteoarthrosis. Journal of Biomechanical Engineering 120, 704-709.

Newberry, W.N., Mackenzie C.D., Haut RC, 1998. Blunt impact causes changes in bone
and cartilage in a regularly exercised animal model. Journal of Orthopaedic
Research 16, 348-354.

Park, S., Krishnan, R., Nicol], S.B., Ateshian, GA, 2003. Cartilage interstitial ﬂuid load
support in unconﬁned compression. Journal of Biomechanics 36, 1785-1796.

Parsons, J .R., Black, J ., 1979. Mechanical behavior of articular cartilage: quantitative
changes with alteration of ionic environment. Journal of Biomechanics 12, 765-
773.

Sah, R.L., Kim, Y.J., Doong, J.Y., Grodzinsky, A.J., Plaas, A.H., Sandy, JD, 1989.
Biosynthetic response of cartilage explants to dynamic compression. Journal of
Orthopaedic Research 7, 619-636.

Sauerland, K, Raiss, R.X., Steinmeyer, J ., 2003. Proteoglycan metabolism and viability

of articular cartilage explants as modulated by the frequency of intermittent
loading. Osteoarthritis and Cartilage. 11, 343-350.

31

Setton, L.A., Zhu, W.B., Mow, V.C., 1993. The biphasic poroviscoelastic behavior of
articular cartilage: role of the surface zone in governing the compressive behavior.
Journal of Biomechanics 26, 581-592.

Torzilli, P.A., Grigiene, R., Huang, C., Friedman, S.M., Doty, S.B., Boskey, A.L., Lust,
G., 1997. Characterization of cartilage metabolic response to static and dynamic
stress using a mechanical explant test system. Journal of Biomechanics 30, 1-9.

Wilson, W., van Donkelaar, C.C., van Rietbergen, B., Ito, K., Huiskes, R., 2004. Stresses
in the local collagen network of articular cartilage: a poroviscoelastic ﬁbril-
reinforced ﬁnite element study. Journal of Biomechanics 37, 357-366.

Wright, V. 1990. Post-traumatic osteoarthritis—A medico-legal mineﬁeld. British
Journal of Rheumatology 29, 474-478.

Zarek, J .M., Edward, J ., 1963. The stress-structure relationship in articular cartilage.
Medical Electroncs and Biological Engineering 1, 497-507.

32

Figgre Captions

Figure l

(a) Photograph of servo-hydraulic testing machine set up. Explant specimens were
placed between the actuator arm and the lower platen which is attached to a load cell. (b)
A two dimensional drawing showing the loading scenario created by the servo-hydraulic
testing machine.

Figure 2

Photographs taken at 25X magniﬁcation were analyzed using digital imaging software.
Manual digital pixel overlays were drawn along the length of all visible surface cracks.
The software counted the total number of pixels. Pixel amounts were converted into
units of millimeters by calibrating the software with a photograph of a ruler at the same
magniﬁcation.

Figure 3

Fluorescence images taken at 100X were separated into superﬁcial, middle, and deep
zones. The dead (red) and viable (green) cells were manually counted in digital imaging
software. Dead cell counts were divided by total cell counts in each zone in order to
yield a percentage of dead cells.

Figure 4

Non-linear stress versus strain curves were generated for all of the impacted explants.
Values of strain were interpolated from stress values of l, 5, 10, 15, and 20 mega Pascals.

The curves were ﬁt with a hyper-elastic Ogden ﬁmction.

33

Figure 5

Average stress-strain plots for equilibrated and non-equilibrated chondral explants were
constructed following blunt impact mechanical loading. Non-equilibrated explants
exhibited a stiffer response resulting in a shift to the left on the graph.

Figure 6

A plot of the shear modulus parameter p vs. water gain for the equilibrated explants ﬁt
with a least squares linear regression line showing a statistically signiﬁcant negative
correlation between this mechanical parameter and ﬂuid intake during equilibration using
a Pearson correlation coefﬁcient test.

Figure 7

A plot of peak strain vs. water gain for the equilibrated explants ﬁt with a least squares
linear regression line showing a statistically signiﬁcant positive correlation between
explant strain and ﬂuid intake during equilibration using a Pearson correlation coefﬁcient
test.

Figure 8

Gross photographs were used to determine the average total ﬁssure length for cartilage
explants following blunt impacts. Less ﬁssuring was documented in non-equilibrated
explants (a) compared to equilibrated specimens (b).

Figure 9

A plot of the length of surface ﬁssuring vs. water gain during 24 hours of equilibration ﬁt
with a least squares linear regression line showing a statistically signiﬁcant positive
correlation between the total length of surface cracks and ﬂuid intake during equilibration

using a Pearson correlation coefﬁcient test.

34

Figure 10

Gross inspection of ﬂouresence cell viability pictures taken at 100X show a higher
occurrence of deep zone cell death in the non-equilibrated (b) versus equilibrated group
(a).

Figure 11

Comparison bar graph showing percentages of cell death in the various zones of
equilibrated and non-equilibrated chondral explants after 30 MPa of unconﬁned
compression. Signiﬁcant differences (*) in cell death were noted in both deep and
superﬁcial zones between goups (p<0.05).

Figure 12

A plot of the percentage of dead cells in the deep zone vs. water gained in the
equilibrated explants ﬁt with a least squares linear regression line showing a statistically
signiﬁcant negative correlation between cell death in the deep zone and ﬂuid intake
during equilibration using a Pearson correlation coefﬁcient test.

Figure 13

Pearson correlation tests revealed signiﬁcant correlations between the peak compressive
stress generated during unconﬁned compression and thickness (a), explant weight (b), dry

weight (c), and the percentage of water content ((1) in the non-equilibrated explants.

35

Figure 14

Pearson correlation tests revealed signiﬁcant correlations between the peak compressive
stress generated during unconﬁned compression and thickness (a), dry weight (c), the
percentage of water content ((1), and the percentage of ﬂuid gained during equilibration.
A signiﬁcant correlation was not documented between explant weight and peak

compressive stress (b).

36

Figures

 

.2” Hydraulic
Actuator

 

 

 

 

 

,4. Lower
Platen

 

 

 

 

 

 

a» Load
Cell

 

 

 

 

   

Figure l

Drawn on
_, digital
., '. m 1' ment
Explant ' easu e
_ lmes
Specrmen

(25X) \

 

 

Figure 2

37

 

559?“ 95’s” / '1’».
.teﬁmﬁvmﬁgﬂusémasm Me

 

    
  

 

Cross section
slice divided
into sections

 

 

. 1‘.\..§'=.l§.%§.‘15. 2'

Dead Cells
(red)

 

—|~ )N N
01 O 01

Stress (MPa
8

OUT

1

o 0.2 0.4 0.6 0.8
Strain

Figure 4

38

Stress (MPa)

 

25

 

 

 

 

20 ~

+ Non-Equil
15 - + Equil
10 a

 

  
 
  
 
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O T l l
O 0.1 0.2 0.3 0.4 0.5 0.6
Strain (mm/mm)
Figures
4.5
o
1’ 3.5
3 1\
R 3
E
a 2.5 c
o. . .
3 2 . x .
3
1: 1.5 A
o
E o o\
h 1 .
a a
2 0.5 o_
in
o I l I l
0.00% 5.00% 10.00% 15.00% 20.00% 25.00%
Fluid Gain (°/o)
Figure6

39

0.7

 

 

9
~1
01

Peak Strain (mm/mm)
9
N

.0
A
01

_o
A

0.00%

5.00%

10.00% 15.00%
Fluid Gain (%)

20.00%

 

25.00%

 

Figure 7

 

Figure 8

40

 

 

1 00
90
8O
70
60
50
40
30
20
1 0

0
0.00% 5.00% 10.00% 15.00% 20.00% 25.00%

Fluid Gain (%)

 

 

Length of ﬁssuring (mm)

 

 

 

~,___l

 

Fiure9

 

Figure 10

41

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

g 40%
3 [ISupiedicial
'6 30% .DMiddle
g llDeep
g 20% -
10% -
0% -
Non-Equilibrated Equilibrated
Figure 11
l 0.35 A7
I .
l 0.3 —_ ‘I—I , _e
‘ ii 0.25 ° ’74—-
O
a
3 0'2 \ .
:0 0.15
0 o
a o
g 0.1 -
. . O \
0.05 0 0 g n
O
l o \
0 T I I c I
0.00% 5.00% 10.00% 15.00% 20.00% 25.00%
L Fluld Gain (%)
Figure 12

42

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1 O 30 O
E 0.0 . a 25 1
E -
~07 0.6 5 2°
in E 15 j
‘” 0.4 i or
E 0.2 ~ 3 5 a
p—

o T a e 0 f e t

0 20 40 60 8o 0 20 40 60 80
3) Peak Stress (MPa) b) Peak Stress (MP3)
0 75% . -. .
3, 0 , ° .. 70% — ° '
E 0 . g 65% -
E , g 60% -
g 0 g 55% -
ED ~ g 50% ~ r=-0.719
5 0- “- 45% l p<0.001
0 TIT I l 40% U T f

0 20 40 60 80 O 20 40 60 80

C) Peak Stress (M Pa) d) Peak Stress (MPa)
Figure 13

43

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.2 35 T .
A 1 q . . 3o . . .. ,
E00- . 225] M
V ‘ 0 O . . v _ o. . DO 0 .
g 0.6 T .0 00 9 .. H 20 o o .0 .. .
Q) o . a 15 - o
g 0.4 — ’ r=0.313 '6 10 . r= 0.207
E 0.2 . p = 0.034 3 5 , p = 0.167
0 T T I o I I T
0 20 40 60 80 0 20 40 60 80
8) Peak Stress (MPa) b) Peak Stress (M Pa)
12 80%
’5 10 ~ . ,_.
E 8 _ . o: . 3:3 700‘ ‘1
E ° 8
.g 6 '1 o .0 ° 0 60% 4 .
3 4 i . ° r= 0.539 3 50% 4 r--0.576
E 2 - p < 0.001 H- p < 0.001
0 r r r 40% r r t
0 20 40 60 80 0 20 40 60 80
C) Peak Stress (M Pa) d) Peak Stress (MPa)
50%
40% i '

.E

8 30% ~

g zoo/ti

“- 10% -

0%
0 20 40 60 80
6)
Peak Stress (MPa)
Figure 14

44

 

CHAPTER 2: THE LIMITATION OF ACUTE NECROSIS IN RETRO-
PATELLAR CARTILAGE AFTER A SEVERE BLUNT IMPACT TO THE IN
VIVO RABBIT PATELLO-FEMORAL JOINT

Abstract:

Our laboratory has previously shown that severe levels of blunt mechanical load,
generating contact pressures greater than 25 MPa, on chondral and osteochondral
explants produce surface lesions and acute necrosis of chondrocytes. In vivo studies by
our laboratory have also found surface lesions and chronic degradation of retro-patellar
cartilage within 3 years following a 6 Joule impact intensity with an associated average
pressure of 25 MPa in the rabbit patella-femoral joint. A hypothesis of the current study
was that cellular necrosis is produced acutely in the retro-patellar cartilage of the
aforementioned rabbit model as a result of a 6 Joule impact. Another hypothesis of the
study was that an early injection of the non-ionic surfactant, poloxamer 188 (P188) would
signiﬁcantly reduce the percentage of necrotic cells in the traumatized retro-patellar
cartilage.

In the current study eighteen rabbits were equally divided into 3 groups. One
group was termed ‘time zero’, and the other groups were carried out for 4 days. One ‘4
day’ group was administered a 1.5 m1 injection of P188 into the impacted joint
immediately after trauma, while the other was injected with a placebo solution. Impact
trauma produced surface lesions on retropatellar cartilage in all groups. Approximately
15% of retro-patellar chondrocytes suffered acute necrosis in the ‘time zero’ and ‘4 day
no poloxamer’ groups. In contrast, signiﬁcantly fewer (7%) cells suffered necrosis in the

‘4 day poloxamer’ group. The effect of this treatment was most signiﬁcant in the

superﬁcial layer of the cartilage.

45

The study indicated the potential use of P188 surfactant in the acute repair of
mechanically damaged cell membranes in the in vivo setting. Its use early after severe
trauma to articular cartilage may allow sufﬁcient time for damaged cells to heal, which
may in turn mitigate the potential for a post-traumatic osteoarthritis in the joint.
Additional studies are needed to improve the efficacy of this surfactant and to determine
the long-term health of joint cartilage after P188 intervention.

Introduction:

Osteoarthritis (OA) affects over 21 million Americans and is the leading cause of
disability in the United States (Wright, 1990). While the mechanisms responsible for this
disease are unknown, the risk of CA is increased signiﬁcantly in joints suffering a major
injury (Felson, 2004). However, establishing a cause and effect relationship between joint
injury and OA can be difﬁcult as after an osteochondral injury joint disease may not be
diagnosed for 2 to 5 years, while less severe joint injuries may not be diagnosed for 10 or
more years after the trauma (U .5. Census Bureau, 2000). Clinically this disease is
characterized by joint pain and narrowing of the joint, as diagnosed by radiological
examination (Flores and Hochber, 1998). Pathologically, the disease exhibits a loss of
cartilage and sclerosis of underlying bone.

Experimental studies with animal models have been conducted in attempts to
understand the potential association between acute joint injury and the chronic
pathogenesis of OA. This laboratory has developed a model of post-traumatic OA using
the Flemish Giant rabbit (Haut et al., 1995). In a recent study a rigid impact mass was
dropped with 6 Joule of energy onto the ﬂexed patello-femoral joint, using this model

(Ewers et al., 2002). The study found surface ﬁssures acutely with a progressive increase

46

in the degradation of retro-patellar surface cartilage and the thickening of underlying
subchondral bone after 3 years. Another study found that during this impact event
patella-femoral contact pressures reach approximately 25 MPa (Newberry et al., 1998).
A study by another laboratory, using the New Zealand white rabbit subjected to a 10
Joule impact intensity, found signiﬁcantly advanced OA-like changes in the patello-
femoral joint, with ﬁbrillation, ulceration and erosion of retro-patellar cartilage within 6
months post-contusion (Mazieres et al., 1987). Early OA-like changes have also been
described using the ﬂexed canine patello-femoral joint subjected to approximately 2.2 kN
of impact force delivered with a gravity-dropped rigid mass (Thompson et al., 1991).
The study acutely found surface ﬁssures on retro-patellar cartilage with clefts in the
underlying subchondral bone and calciﬁed cartilage. After 6 months the surface ﬁssures
had loss of saﬁ'anin-O staining and new bone formation in the underlying subchondral
bone. While the studies described above found an association between blunt impact
trauma to a joint and the subsequent development of OA-like changes in the traumatized
cartilage, the mechanisms of acute damage leading to these pathological changes are yet
unclear.

Excessive mechanical loads resulting from a single blunt impact have been found
to result in surface ﬁssures and necrotic cell death in chondral (Ewers et al., 2001;
Torzilli et al., 1999) and osteochondral explants (Krueger et al., 2003; Morel and Quinn,
2004). Forty megapascals (MPa) of unconﬁned compression on bovine chondral explants
results in signiﬁcant surface ﬁssuring of the tissue with approximately 50% of cells
acutely necrotic (Ewers et al., 2001). Necrotic cells have been found in the superﬁcial

zone (upper 20%) located near ﬁssures and in the middle zone (next 50% of thickness)

47

zone, with a few dead cells found in the deepest zone of the cartilage, especially in
osteochondral explants (Krueger et al., 2003). A critical threshold stress of 15-20 MPa
was found for cell death and collagen matrix damage in unconﬁned compression using
the bovine chondral explant (Torzilli et al., 1999)). A later study using osteochondral
explants found that signiﬁcant surface ﬁssuring with adjacent cellular necrosis occurs
when impact rates of loading exceed the gel diffusion rate of the cartilage (Morel and
Quinn, 2004). This study also documented cell death largely in the superﬁcial zone
adjacent to ﬁssures. Necrotic cell death has also been found in situ using the bovine
patella subjected to a single impact of 53 MPa (Lewis et al., 2003)). In that study cell
death occurred around surface cracks, but not in impacted areas away from these cracks.
The authors suggested “that early stabilization of damaged areas of the cartilage may
prevent late sequelae that lead to OA.” Other studies have also associated the
physiopathology of articular cartilage in CA with the death of chondrocytes (Hashimoto
et al., 1998).

Few in vivo experimental studies, however, have been conducted that speciﬁcally
associate cell death with the development of OA. In one study signiﬁcant necrosis of
chondrocytes was induced in the patella-femoral joint of rabbits by localized freezing of
the tissue in vivo (Simon et al., 1976). After 1 year the affected cartilage had changes
that were consistent with early signs of OA. These tissue changes were similar to those
documented by our laboratory using the impacted joint model (Ewers et al., 2001). A
hypothesis of the current study was that a 6 Joule blunt impact onto the rabbit patello-

femoral joint would induce acute necrosis of chondrocytes in the retro-patellar cartilage.

48

A deﬁning feature of cell death by necrosis is damage to the plasma membrane,
and the inability of the cell to maintain ionic gradients across its membrane resulting in
swelling with subsequent rupture of the cell (Duke et al., 1996). Due to the nature of
necrotic cell death, mild surfactants have been used to restore integrity to cells after
physical and chemical stresses (Clarke and McNeil, 1992; Papoutsakis, 1991).
Speciﬁcally, poloxamer 188 (P188) has been found to ‘save’ neurons from early necrotic
death after severe mechanical loading (Marks et al., 2001; Borgens et al., 2004). A recent
study by our laboratory also found that P188 surfactant was able to ‘save’ chondrocytes
from acute necrotic death in bovine chondral explants subjected to 25 MPa of unconﬁned
compression (Phillips and Haut, 2004). A second hypothesis of the current study was
that injection of P188 surfactant into the in vivo patella-femoral joint capsule shortly
after a 6 Joule blunt impact would signiﬁcantly reduce the percentage of necrotic cells in

retro-patellar cartilage in the acute setting (4 days post trauma).

Methods:

Eighteen skeletally mature, Flemish Giant rabbits (aged 6-8 months) were used in
this study. The project was approved by the All-University Committee on Animal Use
and Care. The blunt impact experiments have been described previously (Haut et al.,
1995). Brieﬂy, a 1.33 kg mass with a ﬂat 25 mm diameter aluminum impact interface
was dropped from 0.46 m (6 Joules of impact energy) onto the right patellofemoral joint
of anesthetized animals (2% Isoﬂurane and oxygen) (Figure l). The opposite limb was
not impacted and used as a paired, un-impacted control. A load transducer (2.225-kN

capacity) (model 31/432; Sensotec, Columbus, OH, USA.) was attached behind the rigid

49

interface to record impact loads. Peak contact load, time to peak, and total contact
duration were collected at 10 kHz. The mass was arrested electronically after the ﬁrst
impact, preventing multiple impacts.

Six rabbits were randomly selected as ‘time zero’ animals. These were sacriﬁced
immediately after impact. The remaining 12 animals were sacriﬁced 4 days post impact.
During these 4 days the animals were exercised 10 minutes a day at 0.3 mph on a
treadmill (Oyen-Tiesma et al., 1998). The exercise protocol was initiated on the ﬁrst day
approximately 4 hours after impact. Animals were housed in individual cages (122 cm x
61 cm x 49 cm) when not exercising. Six of the animals (4 day poloxamer group)
received a single 1.5 m1 injection of an 8mg/ml concentration of P1 88 surfactant in sterile
phosphate-buffered saline (PBS) injected into the traumatized patella-femoral joint
capsule shortly (within approximately 2 minutes) after impact. The remaining six
animals (4 day no poloxamer group) received a sham injection of 1.5 mL sterile PBS into
the impacted joint shortly after trauma The combination P188 in PBS and the PBS sham
solutions were ﬁlter sterilized prior to injection in the joints using a 0.2 mm vacuum ﬁlter
(Nalgene, Nalge Nunc Int., Rochester, New York, USA). After injection, the animal’s
limb was exercised manually to help distribute the P188 surfactant and PBS solutions in
the joint.

Patellae from the impacted and the opposite un-impacted limbs were excised
immediately after sacriﬁce from each animal. The retro-patellar surface was wiped with
India ink to highlight surface defects and was photographed using a digital camera

(Polaroid DMC2, Polaroid Corporation, Waltham, MA, USA) under a dissecting

50

microscope (Wild TYP 374590, Heerbrugg, Switzerland). Each patella was then
wrapped in PBS soaked gauze and prepared for cell viability analyses.

Osteochondral sections ﬁ'om the impacted and opposite un-impacted patella from
each animal were prepared for cell viability analyses. A low speed bone saw (model# 11-
1180-170, Buehler, Inc., Lake Bluff, IL) was used to remove trabecular bone from the
overlying retro-patellar cartilage, leaving approximately 0.5 mm of subchondral bone
(Figure 2). Full depth sections (0.5 mm thick) of retropatellar cartilage and subchondral
bone were cut using a specialized cutting device (Ewers et al., 2001) from the area of
known patella-femoral contact during blunt impact to the joint (Haut et al., 1995). The
sections were stained with Calcein AM and Ethidium Homodimer, according to the
manufacturer’s speciﬁcations (Live/Dead Cytotoxicity Kit, Molecular Probes, Eugene,
OR). Approximately 8 sections from each patella were viewed near the midline of the
patella using a ﬂuorescence microscope (Leitz Dialux 20 (frequency: 50-60 Hz), Leitz
Mikroskopie und Systeme GmgH, Wetlzar, Germany). The sections were photographed
with a digital camera (Polaroid DMC2, Polaroid Corporation, Waltham, MA, USA). Each
photographic section was divided into two zones: superﬁcial (top 20% of the cartilage
depth) and a deep zone (bottom 80%), using an established protocol (Phillips and Haut,
2004). A blinded observer (A.S.) manually counted live and dead cells of in each section
using image analysis software (Sigma Scan, SPSS INC., Chicago, IL, USA). The
percentage cell death in each zone and the total percentage cell death were determined for
each section. The individual section data was averaged to yield the percentage of cell
death in each zone and the total percentage of cell death for each limb. The length of

surface ﬁssuring was measured on 4 randomly selected impacted and un-impacted pairs

51

of patellae from each group using the same image analysis software, according to an
established protocol (Ewers et al., 2001).

Two factor repeated measures AN OVA and Student-Newman-Keuls post-hoe
tests were used to detect statistical differences in the percentage of cell death, percentage
of cell death in each zone, and the average total ﬁssure length between impacted and un-
impacted limbs, as well as between treatment groups. A two factor (time and limb)
ANOVA was performed to detect differences between and within the ‘time zero’ and ‘4-
day no poloxamer’ groups, while a separate two factor (P188 treatment and time)
ANOVA was performed to detect differences between and within the ‘4-day no
poloxamer’ and ‘4-day poloxamer’ groups. A single factor ANOVA was used to detect
differences in peak impact loads and time to peak loads between treatment groups.
Statistical signiﬁcance in these tests was set at p<0.05. All experimental data are

presented as mean 1- one standard deviation.

Results:

There were no statistical differences in the times to peak impact load or the
magnitudes of peak load between the treatment groups. The peak impact load developed
on the patella-femoral joint was 6757:1344 N, and the time to peak load was 3.53:1.0
ms for this study. The animals did not appear to favor one limb over the other during the
4 days of treadmill exercise, per observations by the licensed veterinary animal

technician‘(J.A.).

52

Gross inspections of the impacted and un-impacted patellae indicated surface
ﬁssures on the retro-patellar surfaces (Figure 3). On the impacted patellae there were
more ﬁssures with a proximal to distal orientation that were located near the mid-line of
the patella. This resulted in a statistically signiﬁcant difference in the length of surface
frssuring on the impacted versus the un-impacted patellae for each group (Figure 4). No
signiﬁcant differences in the length of surface ﬁssures were detected between treatment
groups for either impacted or un-impacted patellae, however with the small sample size
of the current study the statistical power between groups for this analysis was below the
desired level of 0.8.

Magniﬁcation of the stained chondral sections showed the presence of necrotic
cells in both the superﬁcial and deep zones of impacted and the opposite, un-impacted
retro-patellar cartilage (Figure 5). Analyses of the cell viability across the depth of the
retro-patellar cartilage showed statistically signiﬁcant differences in the percentages of
dead cells between the impacted and un-impacted patellae for the ‘time 0’ (p<0.001) and
for the ‘4 day no poloxamer’ treatment groups (p=0.002)(Figure 6). In contrast, there
was not a statistically signiﬁcant difference (p=0.072, power = 0.947) in the percentage
of total cell death between the impacted and un-impacted patellae for the ‘4 day
poloxamer’ group. The effect of P188 injections into the joint shortly after impact
trauma was also evident in a signiﬁcant difference (p=0.027) in the percentage of total
cell death between the impacted limbs of the ‘4 day no poloxamer’ group and the ‘4 day
poloxamer’ group. In contrast, there was not a signiﬁcant difference in the percentage of
total cell death in the Mpacted (p=0.195) or the un-impacted (p=0.899) retro-patellar

cartilage between the ‘time 0’ and the ‘4 day no poloxamer’ groups.

53

Gross inspection of the cartilage sections indicated a large amount of cell death in
the superﬁcial zone (Figure 5b). While there were statistically signiﬁcant differences
between the percentages of cell death in the superﬁcial zones of impacted versus un-
impacted retropatellar cartilage for the ‘time zero’ (p=0.002) and the ‘4 day no
poloxamer’ (p=0.008) groups, there was no signiﬁcant difference (power = 0.994)
recorded in the percentage of cell death in this zone for the ‘4 day poloxamer’ (p=0.095)
group (Figure 7). In contrast, however, there were signiﬁcant differences in the
. percentages of cell death in the deep zones between impacted and un-impacted patellae in
all treatment groups. Statistical differences were not detected for the superﬁcial or deep
zones between groups in this study, but with the small sample size the statistical power
between groups was again less than 0.8.

Discussion

An objective of this study was to test the hypothesis that 6 Joules of impact
energy delivered to the ﬂexed rabbit patello-femoral joint, via a rigid impact interface,
would cause a signiﬁcant amount of cell death in the retro-patellar cartilage. Impact to
the joint produced surface lesions, or ﬁssures, on the retro-patellar cartilage. The
proximal to distal orientation of the impact-induced surface ﬁssures have previously been
associated with the impact event (Haut et al., 1995). In the current study impact trauma
resulted in the acute death of chondrocytes throughout the depth of the retro-patellar
cartilage by approximately 15% more in the impacted versus the opposite un-impacted
patella. A second objective of the study was to test the hypothesis that an immediate post-
trauma injection of P188 surfactant into the joint would signiﬁcantly reduce the

percentage of dead cells in the retro-patellar cartilage. The current study found that the

54

percentage of total dead cells in the impacted versus the un-impacted retro-patellar
cartilage was not statistically different for the ‘4 day poloxamer’ group. While the zonal
layer data showed statistically signiﬁcant differences in the percentage of dead cells in
the superﬁcial and deep zones of impacted versus un-impacted patella for the ‘time zero’
and ‘4-day no poloxamer’ groups, in the ‘4-day poloxamer’ group the effect was only
statistically signiﬁcant in the deep zone. These data suggest that P188 was most effective
in ‘saving’ cells from necrotic death in the superﬁcial zone in the current study.

The current study veriﬁed that a blunt impact to the rabbit patella-femoral joint
with a 6 Joule intensity produced statistically signiﬁcant cell death in the retro-patellar
cartilage. A deﬁning feature of cell death by necrosis is swelling, due to the injured cell
being unable to maintain ionic gradients across a damaged plasma membrane (Duke et
al., 1996). In the current study the percentage of dead cells was measured by membrane
disruption, as membrane damage, was documented by the ability of EthD-l (ethidium
homodimer) to only pass through [a damaged cell membrane. This particular mechanism
of damage was also supported by the efﬁcacy of P188 surfactant to repair these cells in
this study. A previous study showed that this surfactant speciﬁcally inserts into only
damaged areas of a cell membrane (Marks et al., 2001).

In a similar, in vivo study by D’Lima et al. (2001) which used New Zealand
White rabbits, a 3 Joule intensity blunt impact to the patello-femoral joint resulted in a
14% increase in apoptotic (TUNEL positive) cells in retro-patellar cartilage versus the
un-impacted limb at 4 days post trauma (D’Lima et al., 2001). The study veriﬁed cellular
apoptosis by examining attributes of nuclear morphology indicative of apoptosis.

Chondrocyte apoptosis has also been shown in human biopsy tissue near sites of chondral

55

 

fracture (Kim et al., 2002). Canine cartilage explants subjected to 5 MPa of cyclic
compression at 0.3 Hz exhibited the development of both cellular necrosis and apoptosis
progressively increasing with load duration and time. Necrosis was observed 2 hours
after cessation of loading, whereas apoptosis (TUNEL-positive cells) was not signiﬁcant
until 48 or more hours after loading stopped. Apoptosis was veriﬁed in some cells using
transmission electron microscopy (Chen et al., 2001). A study that induced
osteochondral wounding of a joint also found signiﬁcant percentages of necrotic and
apoptotic cells in the tissue (Tew et al., 2000). These data suggest that mechanical injury
to a joint may result in both necrotic and apoptotic cell death. The D’Lima et al. study
(2001) also found 34% of chondrocytes in human chondral explants die via apoptosis
when exposed 1014 MPa of unconﬁned compression. Administration of z-VAD.fmk, a
pan-caspase inhibitor, reduced cell death to 25% in these explants. A limitation of the
current study was that cell death by other mechanisms, for example by apoptosis, was not
examined following P188 intervention. An inﬂux of Ca2+, for example, into the
chondrocyte prior to membrane rescaling by P188 may trigger an early programmed cell
death (Aigner and Kim, 2002). A previous study on neuronal cells has also found P188 to
be effective in limiting apoptosis as detected by TUNEL staining (Serbest, 2003). It is
likely that after trauma, interventions like z-VAD.fmk could be used in combination with
P188 to more effectively reduce chondrocyte death or dysfunction in the longer term.

The current study found that early administration of P188 surfactant into the joint
resulted in a decrease in the percentage of total dead cells in retro-patellar cartilage from
approximately 18% to 7%. In a previous study by our laboratory using bovine chondral

explants, 25 MPa of unconﬁned compression applied in 1 second resulted in

56

 

approximately 34% of the cells necrotic 24 hours after impact (Phillips and Haut, 2004).
Immediate treatment of these explants after impact with an 8 mg/ml concentration of
P188 surfactant reduced cell death to approximately 14 % at 24 hours post trauma.
Differences in total cell death between the previous and current studies may have been
due to the presence of underlying bone in the in vivo model that stiffened the articular
cartilage and prevented excessive deformation during impact to the joint (Krueger et al.,
2003). Furthermore, in contrast to the current study in which P188 surfactant appeared
more effective in ‘saving’ cells in the superﬁcial zone of the retro-patellar cartilage, the
treatment of chondral explants in the previous study with this same concentration of
surfactant signiﬁcantly reduced cell death in all layers of the tissue. One explanation for
the more limited efﬁcacy of the treatment in the current study may relate to the
penetration of the P188 surfactant. In the previous study the chondral explants were
‘pumped’ immediately after administration of the surfactant and approximately 22 hours
after impact with 10 cycles of unconﬁned compression at 1 MPa pressure and a
frequency of 1 Hertz. In the current study the animal joint was ﬂexed approximately 10
times immediately after injection of the surfactant, and the animals were exercised daily
beginning on the day of blunt impact to the joint. It was assumed that post trauma
exercise would help ‘pump’ the surfactant into the cartilage. The ‘pumping’ of surfactant
into the tissue, while not veriﬁed in the previous or current studies, may have been less
effective in the in vivo joint due to differences in the intensity of the pressure and the
unknown loading of the joint during treadmill exercise. Penetration of the surfactant into
the cartilage may have also been limited by the underlying subchondral bone in the in

vivo joint. The efﬁcacy of this treatment, on the other hand, may be enhanced by a

57

 

reduction in the concentration of the surfactant solution, after the relationship between
efﬁcacy and solution concentration is established in future in vitro studies with both
chondral and osteochondral explants. Another limitation of the current study was that the
quantity of surfactant solution was limited to 1.5 ml. There was no basis for choosing
this quantity other than to limit the amount of ﬂuid in the small rabbit knee, and that in a
previous study this quantity of polysulfated glycosaminoglycan solution was injected into
the rabbit joint (Ewers and Haut, 2000). The relationships between optimization of the
surfactant concentration, the quantity of ﬂuid, and the timing of the intervention post
impact should be determined in future studies using the established rabbit model, or
possibly another larger, yet undeveloped in vivo animal model.

The current study used a small sample size (n=6) and an in vivo model which
yielded large variations in both surface ﬁssure and chondrocyte viability data. This
resulted in an insufﬁcient amount of statistical power between treatment groups. The
average total length of surface ﬁssures in the ‘4-day poloxamer’ group was
approximately 4 times greater in the un-impacted and approximately 2 times greater in
the impacted patellae than the ‘4-day no poloxamer’ group on average. A statistically
signiﬁcant difference was not detectable due to the low power. The greater length of
surface ﬁssuring in the ‘4-day poloxamer group’ resulted from two animals that exhibited
large amounts of base-line surface ﬁssuring on their un-impacted retropatellar surface.
These animals were not disregarded due to the already small sample size. Furthermore,
previous studies have documented chondrocyte death around surface cracks (Ewers et al.,
2001; Lewis et al., 2002), but the ‘4-day poloxamer’ group exhibited signiﬁcantly less

percentage of dead cells on average in the impacted patellae versus the ‘4-day no

58

poloxamer’ group. These data suggest that the increased, baseline surface ﬁssuring
documented in the ‘4-day poloxamer’ group did not adversely affect the results of the
study.

In summary, 6 Joules of blunt impact to the ﬂexed rabbit patello-femoral joint
was found to result in a signiﬁcant percentage of necrotic cells in the retro-patellar
cartilage immediately after insult. It is hypothesized that the chronic degradation of this
cartilage, which has been documented by our laboratory in a 3 year post trauma study
(Ewers et al., 2002), may be due, at least in part, to the death of these chondrocytes. The
current study also found, in concert with earlier studies on chondral explants (Phillips and
Haut, 2004), that immediate administration of the traumatized joint with P188 surfactant
resulted in a statistically signiﬁcant decrease in the percentage of necrotic cells. The
above hypothesis on a mechanism of post traumatic osteoarthritis may be tested in the
future with injection of P188 surfactant into the joint immediately after blunt insult. The
long term consequences of ‘saving’ these cells ﬁ'om early necrotic death, in terms of
them ultimately becoming apoptotic and producing degradation enzymes into the joint
tissue (Pickvance et al., 1993) must be investigated in future studies. This intervention
with P188 surfactant, however, may also allow sufﬁcient time to evaluate the biological
condition of these and other cells in the traumatized tissue and utilize additional
pharmacological treatments, such as the caspase inhibitor mentioned earlier, if needed,

for the long term survival of the joint cartilage.

Acknowledgements: This study was supported by grants ﬁ'om The Centers for Disease

Control and Prevention, The National Center for Injury Control and Prevention

59

(R49/CCR503607) and The TRW Automotive Fund. Its contents are solely the
responsibility of the authors and do not necessarily represent the ofﬁcial views of the
CDC or the TRW Corporation. The authors wish to gratefully acknowledge the help of
Clifford Beckett, Aaron Stewart (AS), and Jean Atkinson (J.A.) for technical assistance

during this study.

60

References

Aigner T, Kim HA. Apoptosis and cellular vitality: issues in osteoarthritic cartilage
degeneration. Arthritis Rheum 2002 ;46:1986-96

Borgens RB, Bohnert D, Duerstock B, Spomar D, Lee RC. Subcutaneous tri-block
copolymer produces recovery from spinal cord injury. J Neurosci Res
2004;76: 141 -54.

Chen CT, Burton-Wurster N, Borden C, Huegger K, Bloom SE, Lust G. Chondrocyte
necrosis and apoptosis in impact damaged articular cartilage. J Orthop Res.
2001;19:703-11

Clarke MSF, McNeil PL. Syringe loading introduces macromolecules into living
mammalian cell cytosol. J Cell Sci 1992;102:533-41

Costouros JG, Dang AC, Kim HT. Inhibition of chondrocyte apoptosis in vivo following
acute osteochondral injury. Osteoarthritis Cartilage 2003;11:756-9

D'Lima DD, Hashimoto S, Chen PC, Colwell CW, Lotz MK Impact of mechanical
trauma on matrix and cells. Clin Orthop Rel Res 2001;39IS:90-9

Duke RC, Ojcius DM, Young JDE. Cell suicide in health and disease. Sci Am
1996;275:80-7

Ewers BJ, Dvoracek-Driksna D, Orth MW, Haut RC. The extent of matrix damage and
chondrocyte death in mechanically traumatized articular cartilage explants
depends on rate of loading. J Orthop Res 2001 ;19:779-84

Ewers BJ, Haut RC. Polysulphated glycosaminoglycan treatments can mitigate decreases
in stiffness of articular cartilage in a traumatized animal joint. J Orthop Res
2000;18:756-61

Ewers BJ, Weaver BT, Sevensma ET, Haut RC. Chronic changes in rabbit retro-patellar
cartilage and subchondral bone after blunt impact loading of the patella femoral
joint. J Orthop Res 2002;20:545-50

Felson DT. An update on the pathogenesis and epidemiology of osteoarthritis. Radio]
Clin North Am 2004 ;42:1-9

Flores R, Hochber M. Deﬁnition and classiﬁcation of osteoarthritis. In: Brandt K,
Doherty M, Lohmander S, editors. Osteoarthritis. Oxford: Oxford University
Press: 1998. p. 1—12.

Hashimoto S, Ochs RL, Komiya S, Lotz M. Linkage of chondrocyte apoptosis and
cartilage degradation in human osteoarthritis. Arthritis Rheum 1998;41:1632-8

61

Haut RC, Ide TM, DeCamp CE. Mechanical responses of the rabbit patella-femoral joint
to blunt impact. J Biomech Eng 1995;117:402-8

Kim HT, Lo MY, Pillarisetty R. Chondrocyte apoptosis following intraarticular fracture
in humans. Osteoarthritis Cartilage 2002;10:747-9

Krueger JA, Thisse P, Ewers BJ, Dvoracek-Driksna D, Orth MW, Haut RC. The extent
and distribution of cell death and matrix damage in impacted chondral explants
varies with the presence of underlying bone. J Biomech Eng 2003;125:114-9

Lewis JL, Deloria LB, Oyen-Tiesma M, Thompson RC, Ericson M, Oegema TR. Cell
death after cartilage impact occurs around matrix cracks. J Orthop Res
2003;21:881-7

Marks JD, Pan CY, Bushell T, Cromie W, Lee RC. Amphiphilic, tri-block copolymers
provide potent membrane-targeted neuroprotection. FASEB 2001;15:1 107-9

Mazieres B, Blanckaert A, Thiechart M. Experimental post-contusive osteo-arthritis of
the knee. Quantitative microscopic study of the patella and the femoral condyles.
J Rheum 1987;14:119-21

Morel V, Quinn TM. Cartilage injury by ramp compression near the gel diffusion rate. J
Orthop Res 2004;22:145-51

Morel V, Quinn TM. Short-term changes in cell and matrix damage following

mechanical injury of articular cartilage explants and modelling of microphysical
mediators. Biorheo logy 2004 ;4 1 2509-19

Newberry WN, Garcia JJ, Mackenzie CD, DeCamp CE, Haut RC. Analysis of acute
mechanical insult in an animal model of post-traumatic osteoarthrosis. J Biomech
Eng 1998;120:704-9

Oyen-Tiesma M, Atkinson J, Haut RC. A method for promoting regular rabbit exercise in
orthopaedics research. Contempy Top Lab Anim Sci 1998;37:77-80

Papoutsakis ET. Media additives for protecting freely suspended animal cells against
agitation and aeration damage. TIBTECH 1991 ;9:316-24

Phillips DM, Haut RC. The use of a non-ionic surfactant (P188) to save chondrocytes

ﬁom necrosis following impact loading of chondral explants. J Orthop Res
2004;22:] 135-42

Pickvance EA, Oegema TR, Thompson RC. Immunolocalization of selected cytokines
and proteases in canine articular cartilage after transarticular loading. J Orthop
Res 1993;11:313-23

Serbest G. In vitro neuronal cell injury model: characterization and treatment strategies.
Ph.D. dissertation, Drexel University, 2003;103-42

62

Simon WH, Richardson S, Herman W, Parsons JR, Lane J. Long-term effects of
chondrocyte death on rabbit articular cartilage in vivo. J Bone Joint Surg Am
1976;58:517-26

Tew SR, Kwan APL, Hann A, Thomson BM, Archer CW. The reactions of articular
cartilage to experimental wounding. Role of apoptosis. Arthritis Rheum
2000;43:215-25

Thompson RC, Oegema T, Lewis J, Wallace L. Osteoarthritic changes after acute
transarticular load. J Bone Joint Surg Am 1991;73:990-1001

Torzilli PA, Grigiene R, Borrelli J, Jr., Helfet DL. Effect of impact load on articular
cartilage. Cell metabolism and viability, and matrix water content. J Biomech Eng
1999;121:433-41

U.S.Census Bureau. State and County Quickfacts. 2000. US. Public Information Office.

Wright V. Post-traumatic osteoarthritis- A medico-legal mineﬁeld. Br J Rheum
1990;29:474-8

63

Figgge Captions

Figure 1

(a) Photograph of the rabbit impact set-up (b) Sketch of the setup depicting the load
alignment with the knee joint.

Figure 2

Patellae were glued to cylindrical block (a,b) in order to align the surface parallel with the
Isomet gravity saw blade (0). The saw was used to create osteochondral sections of the
retro-patellar surface.

Figure 3

Digital photographs (25X) of un-impacted (a) and impacted (b) retro-patellar surfaces
were taken for each rabbit. Impact induced ﬁssures were stained with India ink. The
ﬁssures were digitally measured and recorded with digital imaging software.

Figure 4

Digitally measured and recorded surface ﬁssure lengths were averaged and depicted in a
bar chart. Brackets indicate signiﬁcant differences between groups using a 2 factor
repeated measures AN OVA . There was a statistically signiﬁcant greater average length
of surface ﬁssuring in impacted versus un-impacted patellae in all of the groups.
Figure 5

Impacted (a) and un-impacted (b) patellar cross sectional slices were stained for cell
viability. Live cells were stained green and dead cells were stained red. The impacted
patellae were subjected to a 6 Joule impact and stained with calcein AM and ethidium

homodimer either immediately following impact or 4 days post-impact. Images (a) and

64

(b) show patellae that were untreated and viewed immediately after impact. The impacted
patella (b) shows extensive areas of cell death in the surface zone after trauma.

Figure 6

Statistically greater amounts of cell death were documented in the impacted versus un-
impacted patellae of both the “time 0” and the “4-day no poloxamer” groups. A
signiﬁcantly greater amount of cell death was recorded in the impacted patellae of the ‘4
day no poloxamer’ group versus the ‘4 day poloxamer’ group. The brackets denote
statistical signiﬁcance by a 2 factor repeated measures ANOVA.

Figure 7

There was a statistically greater percentage of cell death in the superﬁcial and deep zones
of the impacted versus un-impacted patellae in both the “time 0” and “4 day no
poloxamer” groups. There was also a signiﬁcant difference in the total percentage of cell
death in the deep zones of impacted versus un-impacted patellae in the “4 day
poloxamer” group. The brackets denote statistical differences using a 2 factor repeated

measures AN OVA performed separately for the superﬁcial and deep zones.

65

Fr ures

 

 

 

 

 

 

 

l

igure

F

 

 

 

 

)

 

 

2

Figure

66

Average Total Length of Fissures (mm)
r—I —a N N DJ w ah A M
o u- o u: o u. o u. o u- o

I Un-impacted

El Impacted

time zero

Figure 3

4-Day No Poloxamer

Figure 4

67

 

 

4-Day Poloxamer

 

Figure 5

68

40.00

 

a I I
l 'i transacted?
* * ” ' * "'lqtmpacteg ;—

 

 

 

 

  

 

 

 

 

 

 

 

time zero 4-Day No Poloxamer 4-Day Poloxamer

 

 

Figure 6
70.00 ﬁ , t — _
l I Superﬁcral
l——| .

 

60.00 T WW W lIEIDeep

 

 

 

 

 

Average % Cell Death
N w J: U:
.0 .0 .0 .O
o o o o
o o o o

| l I

10.00 -

 

 

 

 

 

0.00 -

 

 

Empacted impacted Bunnacted impacted unimpacted impacted

time 230 —-‘ 4-Day No Poloxama I '— 4-Day Poloxamer --]

Figure 7

69

 

CHAPTER 3: GLUCOSAMINE SUPPLEMENTATION CAN HELP LIMIT
MATRIX DAMAGE AND ADJACENT CELL DEATH IN TRAUMATIZED
EXPLANTS

Abstract

Severe blunt mechanical trauma has been linked with the onset of osteoarthritis.
Matrix damage in the form of surface ﬁssuring and necrotic cell death has been
documented in cartilage explants following blunt mechanical trauma. Surface ﬁssuring
and cell death are both believed to be pathways in the pathogenesis of osteoarthritis. Our
laboratory has previously shown that bovine chondral explants treated with glucosamine
(GlcN) will experience less compressive deformation when exposed to a high rate of
impact. The current study hypothesized that cartilage explants exposed to GlcN for a
period of 6 days will develop an extracellular matrix richer in glycosaminoglycans
(GAG). These GAGs will reduce the permeability which will in turn result in hydrostatic
pressure generated during loading to help limit distortional strains which have been
suggested to be responsible for chondrocyte death. A ﬁnite element model, using a
transversely isotropic biphasic material model for articular cartilage was developed in
order to test the effects of permeability on the stress ﬁeld in a chondral explant under
unconﬁned compression. The results indicate that GlcN supplementation help reduce the
percentage of dead cells in impacted cartilage explants at both a high (~50ms to peak
pressure) and low (~1 s to peak pressure) rate of loading. The study also showed, using a
ﬁnite element model, that a decrease in permeability reduced the area of high Von Mises

stresses around a crack during unconﬁned compression. The current study suggests that

pretreating chondral explants with GlcN can help limit impact induced cell death.

70

Introduction:

Excessive mechanical loading to a joint has been linked to the long term chronic
degradation of articular cartilage (Ewers et al., 2000). In situ studies that examine the
acute response of a blunt impact to human knee joints document the appearance of
surface lesions on the retro-patellar cartilage resulting ﬁom pressures of approximately
30 MPa (Atkinson and Haut, 2001). In viva studies using the giant Flemish rabbit also
document the appearance of surface ﬁssures on retro-patellar cartilage as the result of a 6
Joule blunt impact to the knee at pressures of approximately 25 MPa (Newberry et al.,
1998). This same in viva rabbit model has been used to document signiﬁcant degradation
of articular cartilage 3 years post impact (Ewers et al., 2002). A similar study that looked
at the acute response of a 6 Joule impact to the rabbit knee found surface lesions and
associated chondrocyte death acutely in the retro-patellar cartilage (Baars et al., 2005). In
vitro studies that subject bovine chondral explants to 30 MPa of unconﬁned compression
ﬁnd excessive surface ﬁssuring and associated chondrocyte death (Ewers et al., 2000;
Krueger et al., 2003; Phillips and Haut 2004). In an in viva study signiﬁcant necrosis of
chondrocytes was induced in the patella-femoral joint of rabbits by localized ﬁeezing of
the tissue in viva (Simon et el., 1976). After 1 year the affected cartilage shows changes
that are consistent with early signs of OA. The tissue changes appear similar to those
documented by our laboratory using the impacted joint model (Ewers et al., 2002).
These data suggest that the association between blunt impact trauma to a joint and a
subsequent degradation of articular cartilage may be the result of impact induced surface

damage and associated chondrocyte death.

71

Previous studies correlate the onset of surface ﬁssures with the development
of high tensile strains in the collagen (Morel and Quinn 2004, Askew and Mow1978,
Eberhardt et al., 1991). One study found that surface ﬁssures in cartilage explants could
not be induced when the loading rate was equal or less than the gel diffusion rate
suggesting that ﬂuid pressurization generated during loading causes collagen ﬁber failure
(Morel and Quinn, 2004). While chondrocyte death has been documented to occur
predominately around surface cracks at high rates (Ewers et al., 2000; Lewis et al., 2003),
other studies have found chondrocyte death to occur throughout the thickness without the
presence of surface cracks (Millentejivic and Torzilli, 2005; Morel and Quinn, 2004). A
study by Ewers et a1. (2000) compared the effects of high and low rates of loading and
found chondrocyte death to occur around surface cracks at both rates, but with a greater
diffusion of cell death away from cracks in lower rate of loading experiments. It has been
suggested that chondrocyte death may be limited in areas away from cracks if ﬂuid ﬂow
is prevented, thus increasing the hydrostatic pressure and limit cell deformations during
loading (Millentejivic and Torzilli, 2005). These data suggest that an increase in
glycosaminoglycan content, and thus an increase in interstitial ﬂuid pressurization and
decrease in tissue permeability could potentially limit chondrocyte deformation during
loading.

Some studies have documented an upregulation of glycosaminoglycan production
both in viva (Oegema et a1, 2002) and in cultured human chondrocytes (Dodge and
Jimenez 2003) as a result of treatment with the nutraceutical glucosamine (GlcN). A
study by this laboratory found that treating bovine chondral explants with GlcN prior to a

high rate of blunt impact loading signiﬁcantly reduced cartilage deformation, matrix

72

damage, and nitric oxide production post impact (Kreuger et al., 2000). A study by
Lippiello (2003) found bovine cartilage explants exposed to GlcN experienced a 1000%
increase in cell metabolism. In clinical trials GlcN has been shown to signiﬁcantly reduce
the progression of joint space narrowing (Reginster et al., 1999), decrease pain, and
improve joint mobility in patients suffering from GA (Camarada and Dowless, 1998;
Rovat, 1992; Vaz, 1982; Pujalte et al., 1980). In vitro studies, which chemically induce
cartilage degradation via exposure to interleukin-1 (IL-1) or lipopolysaccharide (LPS),
have documented signiﬁcant decreases in the tissue’s subsequent production of
degradation enzymes when exposed to GlcN (Fenton et al., 2000; Fenton et al., 2002).
These data suggest that GlcN can improve the quality of the cartilage matrix by reducing
the presence of degradation enzymes, as a well as promoting the synthesis of tissue
glycosaminoglycans.

The aim of the current study was to further investigate the effects of
preconditioning chondral explants with GlcN prior to a severe blunt impact. The
hypothesis was that chondral explants pretreated with GlcN and exposed to injurious
unconﬁned compression will possess a less permeable matrix richer in
glycosaminoglycans which will in turn resist ﬂuid movement and result in less
chondrocyte death around surface cracks. To test the hypothesis a transversely isotropic
biphasic model of cartilage was developed in order to determine whether tissue
permeability affects stress ﬁelds around the area of a crack in a chondral explant exposed
to unconﬁned compression. Also, bovine chondral explants were exposed to GlcN for a
period of six days prior to an unconﬁned compression of 30 MPa at both at high and low

rates. These explants were then studied in regards to their mechanical response under

73

compression, the presence of surface cracks, and chondrocyte death both throughout the
thickness as well as number of dead cells adjacent to surface lesions.

Methods

Experimental

Four pairs of bovine forelegs were obtained from a local abattoir within 6 hours of
slaughter. The legs were cut just above the knee joint leaving the metacarpal joints intact.
The forelegs were cleaned and skinned prior to opening the joint under a laminar ﬂow
hood. A 6.35 mm punch with a smooth edge was used to make 120 plugs. Fifteen plugs
were taken from each limb. The cartilage plugs were removed from the underlying bone
with a scalpel. Sixty explants were placed in individual wells containing Dulbecco’s
Modiﬁed Eagle’s Medium (DMEM):F12 (Gibco, USA #12500-039) supplemented with
10% fetal bovine serum and antibiotics (penicillin 100 units/ml, streptomycin lug/ml,
amphotericin B 0.25 pig/ml), while the remaining sixty specimens were placed in the
same media supplemented with 1 mg/ml glucosamine (GlcN) (FCHG49®).

After 48 hours of equilibration, all explants were subjected to manual cyclic loads
of 0.5 MPa at 1 Hz. Ten consecutive cycles were applied once each day for four days.
The media was replaced every 48 hours. Manual cyclic loading was applied with a
custom solid stainless steel cylinder placed in series with a 50 pound load cell (model
31/1432: Sensotec, Columbus, OH). The load cell was connected to a computer with an
analog to digital converter. LabView (National Instruments Corp, Austin, TX) software
was used to monitor the amplitude and frequency of the intermittent loading.

After four days of intermittent loading, 56 explants from each group were

subjected to unconﬁned compression. Twenty-eight low- and 28 high-rate of loading

74

experiments were conducted to a peak of 30 MPa in ls or 50ms, respectively, using a
haversine impact load function. Each explant was placed between two highly polished
stainless steel plates. Prior to the test, the plates were pressed together at 100 N and the
location of the machine actuator was recorded. The thickness of each specimen was
determined by ﬁnding the difference in the actuator’s location at 5 N of load during the
test and when the plates were pressed together prior to loading. Each specimen received
a 0.5 N preload before being compressed. This protocol has been established in previous
experiments from this laboratory (Ewers et al., 2001, Krueger et al., 2003). The load,
time, and displacement were recorded during each experiment with an accuracy of 0.1 N,
0.001 s, and 0.01 mm, respectively. Four explants from each media group were not
impacted and used for positive and negative controls of cell viability. After the load
cycled to approx. 30 MPa each explant was washed three times in media before being
returned to pro-assigned wells with one ml of supplemented media and incubated for
another 24 hours.

Approximately 24 hours after each mechanical compression experiment, each
explant was examined for matrix damage and cell viability. The surface of each explant
was wiped with India ink. The explants were immediately photographed with a digital
camera (Polaroid DMC2, Polaroid Corporation, Waltham, MA, USA) at 25X under a
dissection microscope (Wild M5A, Wild Heerbrugg Ltd., Switzerland) to determine the
total length of the surface ﬁssures. The total ﬁssure length was calculated with digital
imaging software (Sigma Scan, SPSS Inc., Chicago, IL). One observer (AM) digitally

measured the length of the surface ﬁssures in each of the photographs.

75

For cell viability, two 0.5-mm slices were taken through the thickness at the
center of each explant using a customized cutting tool (Ewers et al., 2001). The sections
were cut perpendicular to the preferred orientation of surface ﬁssures. The sections were
stained with a kit containing calcein and ethidium bromide homodimer (Live/Dead
&Viabi1ity/Cytotoxity, Molecular Probes, Oregon). Positive controls were put through
three freeze-thaw cycles in order to kill all of the cells. Each section was viewed under a
ﬂuorescence microscope at 100X (Lecia DM LB (frequency: 50-60 Hz), Lecia
Mikroskopie and Systeme GmgH, Germany). Full thickness, digital images were taken of
a 2.5-mm length at the center of each explant. These images were partitioned into the
superﬁcial zone (top 20%), intermediate zone (middle 50%), and deep zone (bottom
30%). The viable (green) and dead (red) cells were manually counted by one observer
(SR) using image software (Sigma Scan, SPSS Inc., Chicago, IL). The percent of cell
death was computed for each layer. Also, in order to test whether or not potential
differences in cell death were a result of differences in surface ﬁssuring, a blinded
observer (AM) manually measured the depth of each surface ﬁssure on each section and
counted the number of dead cells in the zone adjacent to each ﬁssure. The ratios of dead
cells per ﬁssure and per unit ﬁssure depth were documented.

Unpaired t-tests were used to compare the total length of surface ﬁssures, peak
strain, peak stress, time to peak, cartilage thickness, and strains at given stresses between
“with GlcN” and “no GlcN” groups for both high and low rates of loading tests. Two
factor ANOVAs were performed to evaluate effects of glucosamine treatment and rates
of loading on the percentage of cell death for each cartilage tissue layer (superﬁcial,

middle, and deep) as well as for the total overall percentage of cell death. Two factor

76

ANOVAs were also used to evaluate signiﬁcant differences in the amount of dead cells
recorded per ﬁssure as well as per unit depth of ﬁssure. The two factors or independent
variables were glucosamine treatment and rate of loading while the dependent variable
was percentage of cell death or number of dead cells. All data were reported as mean i
one standard deviation. Statistical signiﬁcance was indicated at p<0.05.

Theoretical

A transversely isotropic biphasic model of articular cartilage was developed using
a commercial ﬁnite element software package (Abaqus v 6.1, Abaqus Inc., Providence,
Rhode Island). The cartilage material model consisted of six independent material
constants including Young’s modulus in the one and three directions (E1 and E3), shear
modulus in the one-three plane (613), Poisson’s ratio for the one-three plane (013), and
hydraulic permeability in both the one and three directions (kt and k3) (Figure la). For
transverse isotropy, material properties in the two direction were set equal to properties in
the one direction. Material properties were assigned based on data from indentation
testing of rabbit retro-patellar cartilage (Newberry et al., 1998) (Table 1).

An axisymmetric model of unconﬁned compression of a chondral explant was
constructed (Figure lb). An arbitrary load of 10 Newtons was applied to a rigid platen in
contact with the top surface of the explant. In order to ensure a uniform boundary
pressure the platen was not allowed to rotate during loading. Boundary conditions for
explant displacement were set to be zero in the axial direction along the bottom side of
the explant. Transverse displacement was set at zero along the symmetric axis. Pore

pressure was set to zero along the unrestricted edge of the explant in order to allow water

flow at the edges.

77

Two separate models of the unconﬁned compression experiment were created.
One model used an intact cartilage explant and the other had a pre-existing crack, in order
to evaluate stress ﬁelds around a crack during loading (Figure 1c). Validation of the
model was performed by compressing an intact explant at two different rates of loading.
The high rate of loading experiment required peak load to be reached in 50ms, and a low
rate required peak load to be reached at 1 second. Loading was applied using a ramp. A
previous study that applied these rates of loading to bovine chondral explants
documented a higher overall strain in the low rate of loading experiments (Ewers et al.,
2001). In the current study total distance traveled by the platen was documented and
regarded as axial deformation and compared between the high and low rate experiments.

In order to simulate the effects of GlcN treatment the material properties of the
cartilage were altered. The permeability for GlcN treated cartilage was decreased an
amount that the permeability of rabbit retro-patellar articular cartilage decreased in a
separate study where the rabbits were orally administered GlcN (unpublished data)
(Table 1). In this study rabbits were fed GlcN for a period of two months prior to
indentation testing of the retro-patellar cartilage. Unconfmed compression experiments
were ran at both high and low rates for GlcN treated cartilage with a pre-existing crack.
Von Mises stress patterns were documented around the crack for an explant with normal
and reduced permeability. Von Mises stress, is used to estimate yield criteria for ductile
materials. It is calculated by combining stresses in two or three dimensions, with the
result compared to the tensile strength of the material loaded in one dimension. The
formula for the Von Mises stress is given below

(http://en.wikipedia. org/wiki/Von_Mis es_stress).

78

 

 

a. = (on —a.)'2 +(a. —a.)2 +(o-3 —a.)2
2

Where 0‘1, 0'2, and o; are the principal stresses at a point in the material. The current
study evaluated Von Mises stresses since it is a measure of the distortional strain energy
and distortional strains have been suggested as a potential mechanism for impact induced
chondrocyte death (Millentejivic and Torzilli, 2005).
Results:
Experimental

In high rate of loading experiments there was no difference in the peak pressures
between the “no GlcN” group (28.5 i 0.5MPa, n=28) and the “with GlcN” group (28.4 i
0.6MPa, n=28). The time to reach peak load for the “with GlcN” group (44 :h 4ms, n=28)
was not signiﬁcantly different from that of the “no GlcN” group (43 i 4ms, n=28) in the
high rate experiments. There was, however, a signiﬁcant difference in peak strain for the
“no GlcN” group (55 i 6%, =28) versus the “with GlcN” group (51 :1: 6%, n=28)
(p=0.043). There were no signiﬁcant differences in cartilage explant thickness between
the “with GlcN” group (0.68 i 0.1mm, n=28) versus the “no GlcN” group (0.67 n:
0.1mm, n=28). From the data recorded during the high rate tests, nonlinear stress-strain
curves were generated for explants from both groups. The “with GlcN” group
experienced an increase in explant stiffness by producing a shift in the response curves to
the left. There were significant differences in the amount of strain seen by the explant at
20 (p = 0.022), 15 (p = 0.025), 10 (p = 0.029), and 5 (p = 0.033) MPa between the “with

GlcN” and “no Glc ” groups (Figure 2).

79

In the low rate of loading experiments there was also no difference in peak
pressures between the “no GlcN” group (28.9 :t 0.5MPa, n=28) and the “with GlcN”
group (28.8 i 0.5MPa, n=28). The time to reach peak load for the “with GlcN” group
(0.89 :1: 0.053, n=28) was not signiﬁcantly different from that of the “no GlcN” group
(0.89 i 0.045, n=28) in the low rate experiments. There were no signiﬁcant differences
seen in the peak strain between the “with GlcN” group (61 i 6%, n=28) versus the “no
GlcN” group (59 :1: 6%, n=28). There were no signiﬁcant differences in cartilage explant
thickness between the “with GlcN” group (0.65 :t 0.1mm, n=28) versus the “no GlcN”
group (0.66 3: 0.1mm, n=28). The non-linear stress-strain curves were very similar for
both groups during low rate loading, and showed no signiﬁcant differences in strain at 5,
10, 15, or 20 MPa.

Supplementation of the culture medium with GlcN signiﬁcantly reduced the
amount of surface ﬁssuring in the “with GlcN” group for explants exposed to high rates
of loading (Figure 3). Digital analysis of stained explant photographs indicated that for
explants in the “no GlcN” group, the total length of ﬁssuring in the high rate tests was
42.8 :1: 15.3m (n=28). In contrast, for “with GlcN” explants the total length of ﬁssures
was 31.5 :t 13.3mm (n=28) (p =3 0.003) (Figure 4). The low rate explants showed no
signiﬁcant differences in total surface ﬁssuring between “with GlcN” (48.3 i 19mm,
n=28) versus “no GlcN” (46.4 3: 11mm, n=28).

For high rates of loading there was a signiﬁcant difference (p < 0.001) in the
percentage of total cell death in the “with GlcN” (20 i ll %, n=28) group versus “no
GlcN” (35 i 17 %, n=28) group of explants (Figure 5a,b). There was also a signiﬁcant

difference (p = 0.017) in the total percentage of cell death in the low rate of loading

80

experiments between the “with GlcN” (25 i 10 %, n=28) and “no GlcN” (34 i 12 %,
n=28) groups (Figure 5c,d). Zonal analysis revealed that signiﬁcantly less cell death was
recorded in both the superﬁcial and middle zones in the “with GlcN” groups versus the
“no GlcN” groups in both the high and low rates of loading (Figure 6). There were no
signiﬁcant differences found in the percentage of cell death of the deep zone as a result of
treatment or rate of loading. The percentage of cell death in the superﬁcial zone for the
high rate of loading “with GlcN” group (32 i 16 %, n=28) was signiﬁcantly less (p <
0.001) than that for the low rate of loading “with GlcN” group (49 i 15 %, =28). Cell
death typically appeared adjacent to surface lesions in a butterﬂy wing-like pattern
(Figure 7). There was a signiﬁcantly greater amount of dead cells per ﬁssure in the “no
GlcN” group versus the “with GlcN” group at both high and low rates of loading (Figure
8). There was also a signiﬁcantly greater amount of dead cells per millimeter ﬁssure
depth in the “no GlcN” group versus the “with GlcN” group at both rates of loading
(Figure 9). The positive and negative controls showed virtually all dead and all live cells,
respectively (Figure 10).
Theoretical

The total axial deformation of the explant in the ﬁnite- element, unconﬁned
compression model was greater for a low rate versus a high rate of loading. The peak
axial deformation for the low rate experiment was 6.354 x 10'5 m and 6.554 x 10'5 m for
the high rate. These data display a rate effect on tissue stiffness that is consistent with
previous studies (Ewers et al., 2001; Krueger et al., 2003). The pattern of Von Mises
stress gradients around the crack tip were similar to the patterns of chondrocyte death

observed in explants around surface lesions (Figure 11). For the high rate of loading

81

simulation the explant with a decreased permeability, or rather the ‘with GlcN’ explant,
experienced a higher range of Von Mises stresses (0.259 — 0.766 MPa) versus the explant
with normal permeability (0.257 — 0.741 Wa). However, the stress ﬁeld was different
between explants with different permeabilities. Thee explant with with a smaller
permeability experienced higher Von Misses stresses concentrated locally around the
crack as to where the normal permeability explant experienced a more diffuse pattern of
high Von Mises stress around the crack (Figure 11). Similar trends were observed for the
low rate of loading simulation, however, the differences in both stress range and pattern
were not as strong (Figure 12).
Discussion:

This study conﬁrmed some of the results of a previous study (Krueger et al.,
2000), showing that supplementation of the culture medium with glucosamine reduced
peak explant strain and the degree of matrix damage during a high rate of unconﬁned
compression. The current study was able to show these results with a concentration of 1
mg/mL versus 2.5 mg/mL, which was used in the previous study (Krueger et al., 2000).
The current study documented stiffening in the high rate of loading ‘With GlcN” explants
as observed by a shifting to the left of the stress versus strain response curve (Figure 2).
A separate study found the compressive stiffness of articular cartilage to correlate with
glycosaminoglycan (GAG) content as determined by magnetic resonance imaging (MRI)
techniques (Kurkijarvi et al., 2004). A previous study, which measured the dynamic shear
modulus of ovine articular cartilage found it to correlate with cartilage thickness,
collagen organization, and GAG content (Oakley et al., 2004). Previous studies have

documented an upregulation of GAGs in cartilage exposed to GlcN both in viva and in

82

vitro (Lippiello et al., 2000; Lippiello 2003).. These data, as well as data from the
previously mentioned studies, suggest that GlcN treatment may have had an affect on
proteoglycan content and/or collagen organization that in turn affected the mechanical
response of explants during a high rate of loading. A limitation of the current study was
that collagen organization and proteoglycan content was not quantiﬁed. Further studies
are needed to examine the affects of GlcN treatment on collagen organization and
proteoglycan content.

The current study reported a signiﬁcant reduction in the average total length of
surface ﬁssures in GlcN treated explants exposed to a high rate of loading. This
phenomenon was not, however, documented in explants exposed to a lower rate of
loading. Previous studies correlate the onset of surface ﬁssures with the development of
high tensile strains in the collagen ﬁbril network (Morel and Quinn 2004; Askew and
Maw, 1978; Eberhardt et al., 1991). Theoretical models of in situ cartilage loading
scenarios suggest that a potential mechanism for surface ﬁssuring could be the
development of high tensile strains near the surface of cartilage during loading (Li et al.,
1995; Atkinson et al., 1998). It has been documented that the development of surface
ﬁssures occurs at fast rates of loading thatdon’t allow for ﬂuid movement through the
semi-permeable solid matrix (Morel and Quinn, 2004). This inability of ﬂuid to escape
ﬁom the matrix results in interstitial ﬂuid pressurization generated during loading that
exceeds the restraining capacity of the collagen network (Pins et al., 1995). The
maximum compressive strain was less, on average, in the high rate of loading “with
Glc ” group of explants suggesting that the in-plane strain, based on the Poisson affect,

was also less. There were no signiﬁcant differences in maximum compressive strain

83

documented in the low rate of loading explants between the “with GlcN” versus the “no
GlcN” explants, possibly helping to explain no signiﬁcant differences in average surface
ﬁssure length between groups. The data from the current study suggest that the reduction
in surface ﬁssuring in the high rate of loading “with GlcN” group may have resulted from
the stiffened response as documented by a shift to the left of the stress versus strain curve
(Figure 2).

A signiﬁcantly less amount of total cell death was documented in explants
pretreated with GlcN for both high and low rates of loading tests. These ﬁndings are in
contrast with those from a previous study which did not document any signiﬁcant
differences in cell death between GlcN treated and untreated explants exposed to a blunt
impact (Krueger et al., 2000). Potential causes for the contrast in data may have been
ﬁ'om differences in the GlcN pretreatment incubation period, GlcN concentration, and the
introduction of intermittent cyclic loading prior to impact. The previous study exposed
explants to a concentration of 2.5 mg/ml of GlcN, as opposed to the current study which
used 1 mg/mL GlcN. Recent studies have found that high concentrations of GlcN have
cytotoxic effects at high concentrations (Mello et al., 2004). Furthermore, the previous
study pretreated explants for a period of 48 hours, as opposed to the current study which
used a 6 day incubation period. Other in vitro studies that have shown an increase in
cartilage matrix synthesis as a result of GlcN treatment have used incubation periods as
long as 10 days (Lippiello, 2003; Fenton et al., 2004). Also, the introduction of
intermittent low level cyclic loading throughout the incubation period may have also
contributed to the differences in cell viability in the current versus the previous study. In

a separate study that examined the effects of a mild surfactant on chondrocyte viability

84

low level cyclic loading after treatment was necessary in order for the treatment to be
effective. The protocol was suggested to have allowed for the surfactant to be ‘pumped’
into the tissue (Phillips and Haut, 2004). A separate study found a 1000% increase in
cartilage metabolism in GlcN treated explants exposed to 24 hours of static stress, as
opposed to a 40% increase in those that were not stressed. The data from the current
study suggests that GlcN dosage concentration, pretreatment time duration, and
mechanical stress may play a role in GlcN’s ability to prevent chondrocyte death as a
result of a blunt impact in vitro. The increased beneﬁts of GlcN on traumatized cartilage
in explants that were exposed to low level cyclic loading prior to injury may also suggest
co-beneﬁts to the cartilage that may arise ﬁom exercise in combination with GlcN in an
in viva situation.

A reduction in impact induced surface cracks, as a result of less compressive
strain in the GlcN treated, high rate of loading group of explants may have been
responsible for the reduction in chondrocyte death in the superﬁcial zone. This agrees
with previous studies that document the presence of cell death primarily around surface
cracks in explant studies (Lewis et al., 2003; Ewers et al., 2000). However, there was also
a signiﬁcant decrease in chondrocyte death in the “with GlcN” versus “no GlcN”
explants for low rates of loading. In order to eliminate the effects of surface ﬁssuring on
chondrocyte viability the number of dead cells per surface ﬁssure, and dead cells per unit
depth of surface ﬁssure were quantiﬁed. Results from this data showed a signiﬁcant
reduction in dead cells per ﬁssure and dead cells per unit ﬁssure depth in GlcN
supplemented groups for both rates of loading versus their respective “no GlcN” groups.

While the mechanism for the action of GlcN on cell death near surface cracks is yet

85

unclear, upregulation of tissue PGs in the pericellular matrix (Quinn et al., 1998) may
help limit shearing strains in chondrocytes near the cracks. The effect could, as well, be
that GlcN increases the yield stress and reduces post-yield strains in articular cartilage.
Both of these alterations can reduce the zone of plastic strain adjacent to surface ﬁssures
(Mendelson 1968). The zone of plastic strain around a crack tip resembles the zone of
dead cells in an impacted cartilage explant (Figure 13). The data from the current study
indicate that pro-treatment with GlcN can help reduce the zone of plastic strain perhaps
by increasing the yield strength of the tissue. Previous studies have documented cell
death to occur predominately around surface cracks when loaded at a high rate (Lewis et
al., 2003; Ewers et al., 2001) and away ﬁ'om cracks and further into the depth of the
cartilage when loaded at slower rates (Morel and Quinn 2004). It has been suggested that
large distortional strains developed in the cartilage matrix during loading may contribute
to acute cell death (Milentejivic and Torzilli, 2005). The same study suggested that
hydrostatic pressure developed during loading could help limit cell deformations. A
decrease in permeability, to model GlcN treatment, was shown to reduce the zone of high
Von Mises stresses around a crack in the theoretical model of unconﬁned compression.
These data may suggest a mechanism whereby GlcN treatment may help limit the
potential for developing a post-traumatic osteoarthritis by decreasing cartilage
permeability, perhaps as a result of an up-regulation of tissue GAGs, to increase a
hydrostatic pressure effect in the cartilage during impact loading and limit chondrocyte
death. More studies need to be performed in order to validate this hypothesis. Further
experimental and theoretical analyses are needed to better explain the chondroprotective

effect of glucosamine for blunt impact situations.

86

In conclusion, in the current study bathing cartilage explants in GlcN
supplemented culture media prior to a blunt impact had some signiﬁcant beneﬁcial
effects. Impact induced chondrocyte death was found to be signiﬁcantly reduced in
explants that were pretreated with GlcN. This reduction in cell death was found to occur
predominately around surface lesions with fewer dead cells per lesion in the GlcN
supplemented explants. However, the mechanism by which GlcN was able to reduce cell
death around these traumatized areas cannot be deduced based on the collected data from
the current study. Future investigations should explore potential mechanisms of this
phenomenon.

Acknowledgement: This research was supported by a grant from the Cendter for Disease
Control (R49/ CCR503607). Nutramax Laboratories, Inc., Edgewood, MD provided the
supplement. The Authors would like to thank Clifford Beckett for his technical

assistance.

87

References

Askew M, Mow V: The biomechanical function of the collagen ﬁbril ultrastructure of .
articular cartilage. Journal of Biomechanical Engineering 100:105-115, 1978

Atkinson TS, Haut RC, Altiero NJ: An investigation of biphasic failure criteria for
impact-induced ﬁssuring of articular cartilage. Journal of Biomechanical
Engineering 120:536-537, 1998

Atkinson PJ, Haut RC: Injuries produced by blunt trauma to the human patello femoral

joint vary with ﬂexion angle of the knee. Journal of Orthopaedic Research 19,
827-833, 2001

Baars D, Phillips DM, Haut RC: Repair of damaged chondrocytes in the in viva
traumatized joint. Transactions of the Slst Annual meeting of Orthopaedic
Research Society 2005

Camara, CC, Dowless GV: Glucosamine sulfate for osteoarthritis. Annals of
Pharmacotherapeutics 32(5), 580-587, 1998

Dodge GR, Jimenez SA: Glucosamine sulfate modulates the levels of aggrecan and
matrix metalloproteinase-3 synthesized by cultured human osteoarthritis articular
chondrocytes. Osteoarthritis and Cartilage 11(6): 424-432, 2003

Eberhardt AW, Lewis JL, Keer LM: Normal contact of elastic spheres with two elastic
layers as a model of joint articulation. Journal of Biomechanical Engineering 113,
410-417, 1991

Ewers B, Dvoracek-Driksna D, Orth M, Altiero N, Haut R: Matrix damage and
chondrocyte death in articular cartilage depends upon loading rate. Transactions
of the 46th Annual meeting of Orthopaedic Research Society. 2000

Ewers BJ, Dvoracek-Driksna D, Orth MW, Haut RC: The extent of matrix damage and
chondrocyte death in mechanically traumatized articular cartilage explants
depends on rate of loading. Journal of Orthopaedic Research 19:779-784, 2001

Ewers BJ, Weaver BT, Sevensma ET, Haut RC: Chronic changes in rabbit retro-patellar
cartilage and subchondral bone after blunt impact loading of the patello femoral
joint. Journal of Orthopaedic Research 20:545-550, 2002

http://en.wikipedia. org/wiki/Von_Mises_stress

Krueger J, Dvoracek-Driksna D, Ewers B, Haut R, Orth M. Effects of pretreatment with
glucosamine on mechanically traumatized cartilage explants. 47, 257. 2001.
Transactions of the 47th Annual Meeting of the Orthopaedic Research Society.
2001.

88

Krueger JA, Thisse P, Ewers BJ, Dvoracek-Driksna D, Orth MW, Haut RC: The extent
and distribution of cell death and matrix damage in impacted chondral explants
varies with the presence of underlying bone. Journal of Biomechanical
Engineering 125(1):]14-119, 2003

Kurkijarvi JE, Nissi MJ, Kiviranta I, Jurvelin JS, Nieminen MT: Delayed gadolinium-

. enhanced MRI of cartilage (dGEMRIC) and T2 characteristics of human knee
articular cartilage: topographical variation and relationships to mechanical
properties. Magnetic Resonance in Medicine 52(1):41-6, 2004

Lewis JL, Deloria LB, Oyen-Tiesma M, Thompson RC, Ericson M, Oegema TR: Cell
death after cartilage impact occurs around matrix cracks. Journal of Orthopaedic
Research 21 :881-887, 2003

Li X, Haut RC, Altiero NJ: An analytic model to study blunt impact response of the
rabbit P-F joint. Journal of Biomechanical Engineering 117:485-491, 1995

Lippiello L, Han MS. Dose and stress response characteristics of articular cartilage
chondrocytes to glucosamine and chondroitin sulfate. 49th Annual Meeting of the
Orthopaedic Research Society, 2003

Lippiello L, Woodward J, Karpman R, Hammad T: In vivo chondroprotection and
metabolic synergy of glucosamine and chondroitin sulfate. Clinical Orthopaedics
and Related Research 381:2292240, 2000

Mello DM, Nielsen BD, Peters TL, Caron JP, Orth MW: Comparison of inhibitory
effects of glucosamine and mannosamine on bovine articular cartilage
degradation in vitro. American Journal of Veterinary Research 65(10): 1440-5,
2004

Milentijevic, D., Helfet, D.L., Torzilli, RA, 2005. Inﬂuence of stress rate on water loss,

matrix deformation and chondrocyte viability in impacted articular cartilage.
Journal of Biomechanics 38, 493-502.

Morel V, Quinn TM: Short-term changes in cell and matrix damage following
mechanical injury of articular cartilage explants and modeling of microphysical
mediators. Biorheology 41 (3-4):509-5 l 9, 2004

Newberry, W.N., Garcia, J.J., Mackenzie, C.D., Decamp, C.E., Haut, RC, 1998.
Analysis of acute mechanical insult in an animal model of post-traumatic
osteoarthrosis. Journal of Biomechanical Engineering 120, 704-709.

Oakley SP, Lassere MN, Portek I, Szomor Z, Ghosh P, Kirkham BW, Murrell GA, Wulf
S, Appleyard RC: Biomechanical, histologic and macroscopic assessment of
articular cartilage in a sheep model of osteoarthritis. Osteoarthritis and Cartilage.
12(8):667-79, 2004

89

Oegema TR, Deloria LB, Sandy JD, Hart DA: Effect of oral glucosamine on cartilage
and meniscus in normal and chymopapain-injected knees of young rabbits.
Arthritis and Rheumatism 46(9):2495-2503, 2002

Phillips DM, Haut RC: The use of a non-ionic surfactant (P188) to save chondrocytes

from necrosis following impact loading of chondral explants. Journal of
Orthopaedic Research 22:1135-42, 2004

Pins GD, Huang EK, Christiansen DL, Silver FH: Effects of static axial strain on the
tensile properties and failure mechanisms of self-assembled collagen ﬁbers.
Journal of Applied Polymer Science 63: 1429—40, 1995

Pujalte JM, Llavore EP, Ylescupidez FR: Double-blind clinical evaluation of oral
glucaosamine sulphate in the basic treatment of osteoarthrosis. Current Medical
Research and Opinion 7:110-114, 1980

Quinn TM, Grodzinsky AJ, Hunziker EB, Sandy JD: Effects of injurious compression on
matrix turnover around individual cells in calf articular cartilage explants. Journal
of Orthopaedic Research 16:490-499, 1998

Reginster J, Dacre J, Rovati L, Gosset C: Glucosamine sulfate signiﬁcantly reduces
progression of knee osteoarthritis over 3 years: A large randomized, placebo-
controlled, double-blind, prospective trial. Osteoarthritis and Cartilage 8(2): 151-
152, 2001

Rovati LC: Clinical research in osteoarthritis: Design and results of short-term and long-
term trials with disease-modifying drugs. International Journal of Tissue Reaction
14:243-251, 1992

Simon WH, Richardson S, Herman W, Parsons JR, Lane J: Long-term effects of
chondrocyte death on rabbit articular cartilage in vivo. Journal of Bone and Joint
Surgery 58(4):517-526, 1976

Vaz AL: Double-blind clinical evaluation of the relative efﬁcacy of ibuprofen and

glucosamine sulphate in the management of osteoarthrosis of the knee in out-
patients. Crurent Medical Research and Opinion 8: 145-149, 1982

90

Tables

Table 1. Transversely isotropic and biphasic material properties values for the
theoretical cartilage material model

 

 

 

 

 

 

 

 

 

 

 

“no GlcN” “with GlcN”
E1(MPa) 7.835 7.835
133 (MPa) 1.53 1.53
R] (*10'13 ml) 5.95 3.149
R3 (*10'13 m2) 1.222 0.858
01 0.5 0.5
U3 0 O
6,, (MPa) 0 0

 

91

 

Figure Captions

Figure 1. An axisymmetric ﬁnite element model of an unconﬁned compression of a
chondral explant was developed. Cartilage was modeled as being transversely isotropic
biphasic (a). Two separate geometries were used; one was an intact cartilage explant (b)

while the other was cartilage with a pre-existing crack(c)

Figure 2. Stress vs. Strain response curves for high rate of loading tests for both the
“with GlcN” and “no Glc ” groups. Stress and strain data was collected during
unconﬁned compression tests. Stars indicate statistically signiﬁcant different amounts of

strain at the same pressure (p<0.05

Figure 3. Gross photographs were used to determine the average total ﬁssure length for
cartilage explants following blunt impacts. In the high rate of loading groups, less
ﬁssuring was documented in GlcN treated explants (3) compared to non-treated

specimens (b).

Figure 4. The total length of surface ﬁssuring was quantiﬁed using digital imaging
software. Bar graphs of average values of total ﬁssure length both high rate and low rate
tests were made. Error bars represent the standard deviation. “*” indicates statistically

signiﬁcant (p<0.05) difference from corresponding “no GlcN” group

Figure 5. Explants cross sections were stained with calcein and ethidium homodimer.
Photographs were taken at 100x with a digital camera attached to a ﬂouresence

microscope. Viable cells are indicated via a green stain and dead cells a red stain. (a) low

92

rate “with GlcN” (b) high rate “with GlcN” (c) low rate “no Glc ” ((1) high rate “no

GlcN”

Figure 6. Cross section chondrocyte viability photographs were divided into superﬁcial,
middle, and deep zones. Amount of dead and live cells were manually counted in each
zone using digital imaging software. Percentages of chondrocyte death were placed in bar
charts for both (a) high rate and (b) low rate of loading tests. A * indicates a statistical
difference versus respective ‘No GlcN’ group. A ** indicates a statistical difference in %

cell death between rates of loading.

Figure 7. Fluorescence photographs of explant cross sections reveal a greater area of
chondrocyte death around impact induced cracks for both rates of loading. (a) “No GlcN”

high rate, (b) “With GlcN” high rate (c) “No GlcN” low rate ((1) “With G1cN”Low rate.

Figure 8. The number of dead cells around individual ﬁssures were manually counted,
recorded, and averaged for each group. “*” indicates a statistical difference from

corresponding “no GlcN” groups.

Figure 9. The number of dead cells around individual ﬁssures were manually counted
and divided by the depth of the ﬁssure in millimeters. “*” indicates a statistical difference

from corresponding “no GlcN” groups.

Figure 10. Half of the un-impacted explants were put through three freeze-thaw cycled
and revealed virtually all dead cells (a). Other un-impacted explants showed virtually no

cell death (b).

93

Figure 11. The Van Mises stress patems were examined after a high rate of load
application to a cracked explant in a ﬁnite element model of unconﬁned compression.
GlcN modeled cartilage was given a lower permeability based on a hypothesized increase
in glycosaminoglycan content. The area of increased Von Mises stresses appeared to be

greater in the “No GlcN” modeled cartilage versus the “with GlcN” modeled cartilage.

Figure 12. The Von Mises stress patterns were examined after a low rate of load
application to a cracked explant in a ﬁnite element model of unconﬁned compression.
There did not appear to be any differences in Van Mises stress patterns between the ‘with

GlcN’ and ‘no GlcN’ groups.

Figure 13. (a) The area of plastic strain around a crack tip resembles the shape of a
butterﬂy wing. (b) The pattern of plastic strain around a crack superimposed over a
ﬂuorescence image of a cartilage explant cross section exposed to 30 MPa of unconﬁned

compression. The pattern of cell death is falls within the bounds of the plastic strain.

94

Figures

«I. «<1.

 

Figure l

95

 

 

Stress (MPa)

 

25

 

    

 

 

 

+"With GlcN" High
20 - Rate . *
—I—"No GlcN" ngh Rate
15 —
10 -
5 _
O I l l l l T

 

0 0.1 0.2 0.3 0.4 0.5 0.6
Strain (mm/mm)

 

Figure 2

 

96

 

Surface Fissure Length (mm)

80
7O
60
50
40
30
20

1O

    

7L5, Rate

WW-W—thighRate

With GlcN No GlcN

 

Figure 5

97

80%

70%

60%

50%

40%

“/o Cell Death

30%

20%

10%

0%

80%

70%

60%

50%

% Cell Death
on A
O O
o\° e\°

20%

1 0%

0%

No GlcN

No GlcN

(b)

Figure 6

98

Superﬁcial
I Middle

ll Deep

I Total

 

With GlcN

Superﬁcial
I Middle

I Deep

I Total

 

GlcN

 

Figure 7

99

i

in

ll High Rate
it! Low Rate

l_

 

 

 

 

 

 

 

 

 

180 ,

 

160

 

m=oo been no Lon—:32

 

No GlcN

With GlcN

Figure 8

:1
64

4‘

I High Rat

LolRat
i
N
l
l
l
*1
_l

 

 

E,

i
l

 

 

*

 

 

 

 

 

 

I ..

i

O O O O 0

I

I

c

_

fl

0 O O

O 0 0 O 0 O
m 2 O 8 6 4 2
1 1 1

EE can m=oo been co con—:32

  

No GlcN

With GlcN

Figure 9

100

 

Figure 10

 

Figure 11

101

 

Figure 12

    

Crack Tip

‘t
‘ed

(3.2 -

 

4:} ('3‘

Figure 13

102

Conclusions and Recommendations for Future Work

The previous chapters describe the results of an injurious blunt impact load to
articular cartilage in terms of matrix damage and chondrocyte viability using both in vitro
and in viva models. Potential therapeutic treatments that can help limit cell death, either
by membrane rescaling or matrix strengthening were studied.

In Chapter 1 ‘tissue equilibration’ of chondral explants prior to an injurious
unconﬁned compression experiment was found to result in a greater overall length of
surface ﬁssuring versus explants that were impacted immediately after removal from the
joint. This corresponded to a greater percentage of cell death in the superﬁcial zone for
the equilibrated explants. In contrast, the percentage of chondrocyte death in the deep
zone of non-equilibrated explants was signiﬁcantly greater then for equilibrated explants.
It was hypothesized that increased ﬂuid present in the equilibrated explants resulted in
higher deep zone hydrostatic pressures generated during loading, which in turn limited
cell deformation. Future studies should consider the presence of underlying bone, and
determine the effect on deep zone cell death. Additionally, it was observed that
correlations between stiffness and various parameters, indicative of water and solid
matrix content, were diminished when explants were allowed to equilibrate. Future
studies should include biochemical assays in order to determine correlations between
mechanical characteristics and various matrix constituents, and whether or not these
correlations are also diminished as a result of equilibration. Future studies that wish to
compare theoretical models with experimental data should be aware of the effects that

‘tissue equilibration’ might have on the mechanical properties of the tissue.

103

Chapter 2 described a study on rabbit patello-femoral joints subjected to a 6 Joule
intensity impact. The major ﬁndings of this study were the presence of acutely necrotic
cells in the retro-patellar cartilage and a reduction in the percentage of these dead cells
with the administration of poloxamer 188 (P188) directly into the joint immediately after
impact. The methods used in this chapter did not account for the long term viability of
cells ‘saved’ by P188. The possibility that these cells, which were saved within the ﬁrst 4
days after impact, may not function properly or die by apoptosis at a later time needs
further investigation. Future studies should acknowledge this possibility and document
chondrocyte viability over a longer time period. Methods for determining apoptosis, such
as TUNEL staining, should also be included in future studies. Also, a small sample size
(n=6) was used in this study, which resulted in a lack of statistical power in numerous
analyses. Unfortunately, biologically meaningful differences may have existed, but they
were unable to be detected due to the lack of statistical power in the current study.
Furthermore, two rabbits in the ‘4-day poloxamer’ group had very high base-line levels
of surface ﬁssuring, resulting in average total ﬁssure lengths approximately twice as great
as that documented in the other treatment groups. This compromised the current database.
The concentration of P188 and the amount of exercise of the rabbit were chosen
arbitrarily and should be studied more speciﬁcally in ﬁrture studies. The ability of the
P188 to penetrate into the deep zone of the tissue was not recorded in the current study.
Future studies may wish to ‘tag’ the P188 and track the presence of it throughout the
thickness of the cartilage.

In Chapter 3 the effects of bathing chondral explants in culture media

supplemented with the nutraceutical glucosamine (GlcN) were examined. This study

104

found that explants treated with GlcN experienced less compressive strain when impacted
at a high rate. This limited the amount of surface ﬁssuring. The study also found that the
percentage of chondrocyte death was reduced in the group of explants treated with GlcN,
in both the superﬁcial and middle zones of the tissue. This affect appeared due to a
smaller dispersion of dead cells in the vicinity of impact induced cracks with GlcN
supplementation of the medium. A transversely isotropic, biphasic theoretical model of
the cartilage explant exposed to unconﬁned compression found the area of high Von
Mises stresses around a crack reduced if the permeability of the tissue was reduced. My
study was limited by the fact that no independent measure of permeability was performed
to determine if GlcN supplementation, indeed, decreased tissue permeability. Future
studies should incorporate mechanical testing of explants both treated and not treated
with GlcN to deduce hydraulic permeability. Furthermore, it was suggested that GAGs
produced by chondrocytes as the result of GlcN treatment might have been deposited in
the pericellular matrix, in turn having a protective effect on the cells. However, in a
companion study no signiﬁcant increase in GAG was documented in explants treated
with the same concentration of GlcN used in the current study over a 6 day period.
Future studies may wish to examine GAG contents in the immediate vicinity of the
chondrocyte, as there may be a rather small, localized effect yielding these results. Also,
the current study exposed all explants to manual cyclic loading during pre-treatment with
GlcN. Future studies should ﬁrrther examine the potentially synergistic affects of cyclic
mechanical loading and GlcN treatment on cartilage explants.

While the current studies ﬁnd effects of tissue equilibration, P188, and GlcN pre-

treatment on the injury response of articular cartilage, they do not ﬁrlly explain their

105

mechanisms. Future studies should incorporate other experimental methods such as
biochemical assays, molecular magnetic resonance imaging techniques, and apoptotic

cell death detection methods to ﬁrrther investigate these ﬁndings.

106

Appendix A: Chapter 1 Raw Data

Lopd Controlled Egperiments
Table 1. Mechanical data for the non-equilbrated explants

 

 

 

Time to Peak Peak
Specimen Peak Stress Strain Thickness p. or
[81 IMPal ImMI

1 0.041 29.55 0.5932 0.70 2.723 -2.408
2 0.033 29.13 ' 0.5798 0.59 2.258 -2.5
3 0.041 29.00 0.4294 0.67 5.333 -3.632
4 0.042 29.24 0.5203 0.70 3.422 -3.08
5 0.04 29.15 0.6268 0.66 2.078 -2.313
6 0.041 28.60 0.5427 0.51 3.336 -2.672
7 0.035 28.84 0.5346 0.56 3.813 -2.318
8 0.04 29.05 0.5475 0.73 2.403 -3.492
9 0.037 28.82 0.5532 0.73 1.816 -3.853
10 0.042 28.73 0.5376 0.66 3.57 -2.668
1 1 0.029 28.88 0.4994 0.50 3.149 -2.749
12 0.025 29.48 0.5595 0.51 2.831 -2.462
13 0.038 29.06 0.5755 0.40 3.579 -1.779
14 0.026 28.85 0.6334 0.42 1.794 -1.803
15 0.036 28.91 0.6703 0.45 1.815 -1.532
16 0.034 28.75 0.6622 0.50 1.963 -1.418
17 0.021 29.36 0.5386 0.40 3.336 -2.313
18 0.013 29.96 0.5931 0.46 2.002 -2.218
19 0.025 29.10 0.6654 0.40 1.614 -1.816
20 0.026 29.1 1 0.4423 0.64 4.274 -3.68
Average 0.03325 29.08 0.5652 0.56478597 2.85545 -2.5353
Std. Dev. 0.00824541 3 0.323 0.0665 0.1 1 829986 0.9870341 0.71 4273

107

Table 2. Mechanical data fro the equilibrated explants

 

 

 

 

 

 

 

 

 

Time to Peak
Specimen Peak Stress Peak Strain Thickness p a
[S] [MPal me}

21 0.028 28.76 0.483135075 0.6860243? 3.696 -3.457
22 0.03 28.39 0.636599281 0.591 1 175 2.086 -1.746
23 0.024 29.1 1 0.551089931 0.63070628 4.124 -1.897
24 0.032 28.13 0.597155084 0.51655235 2.624 -1.874
25 0.028 28.38 0.573146548 0.74151165 2.013 -3.17
26 0.028 28.60 0.659859868 0.68541583 1 .289 -2.638
27 0.019 27.96 0.652667409 0.82900104 1 .572 -1.968
28 0.026 28.08 0.525510225 0.60141453 3.071 -2.957
29 0.029 27.86 0.602277073 0.61 1 17848 1.817 -3.04
30 0.018 29.04 0.6428061? 0.6410789 2.53 -0.896
31 0.028 28.01 0.585485859 0.4954436 2.228 -2.517
32 0.028 28.83 0.6137 0.43598885 2.126 -1.775
33 0.022 29.76 0.6684 0.5751 1755 1.905 -0.866
34 0.017 29.46 0.6765 0.44708819 0.813 -2.015
35 0.026 29.63 0.731 1 0.6264876 1.329 -0.822
36 0.023 28.59 0.7135 0.62121262 0.863 -1.418
37 0.026 29.16 0.5268 0.61 10233 2.292 -2.946
38 0.016 30.43 0.6321 0.66077322 0.554 -2.745
39 0.027 29.35 0.6692 0.53320697 1 .927 -1 .032
40 0.025 29.52 0.6554 0.45653088 1.277 -1.916

Average 0.025 28.84 0.6198221 44 0.59984369 2.0068 -2.08475

Std. Dev. 0.004507304 0.701 0.06459351 1 0.09968538 0.91 38422 0.820396

108

Table 3. Equilibrated explant weights and ﬂuid gain

 

 

 

 

 

 

 

Time 0 Time 24
Specimen Weight Weight Fluid Gain
Lg'1OA-5L [g‘10"-5]
21 2090 2253 7.80%
22 1420 1650 16.20%
23 1768 1834 3.73%
24 1418 1603 13.05%
25 2106 2401 14.01%
26 1504 1767 17.49%
27 2116 2374 12.19%
28 1850 1896 2.49%
29 1523 1623 6.57%
30 1762 2105 19.47%
31 1453 1620 11.49%
32 1519 1660 9.28%
33 2028 2479 22.24%
34 1445 1720 19.03%
35 1885 2294 21 .70%
36 1966 2266 1 5.26%
37 2021 2246 11.13%
38 2236 2647 18.38%
39 1749 1930 10.35%
40 1666 1921 15.31%
Average 1776.25 2014.45 13.36%
Std. Dev. 271 .215757 332.5921281 5.62%

109

 

Table 4. Surface ﬁssure data in pixels and millimeters for non-equilibrated (specimens
1-20) and equilibrated (specimens 21-40).

 

 

 

 

 

 

Matrix Matrix Matrix Matrix
Specimen Damage Damagg_ Specimen Damage Damage
Pixels mm Pixels mm

1 7071 75.35 21 4209 44.85612789
2 5573 59.39 22 5229 55.72646536
3 2604 27.75 23 3308 35.25399645
4 5383 57.36 24 5320 56.69626998
5 5528 58.91 25 7266 77.43516874
6 3190 33.99 26 5386 57.39964476
7 6983 74.41 27 7098 75.64476021
8 3707 39.50 28 41 1 8 43.88632327
9 5137 54.74 29 4179 44.53641208
10 6129 65.31 30 6290 67.03374778
11 1184 12.61 31 3432 36.57548845
12 5203 55.44 32 5360 57.12255773
13 3387 36.09 33 6496 69.22912966
14 3285 35.00 34 6351 67.68383659
15 3964 42.24 35 6005 63.9964476
16 3721 39.65 36 8105 86.37655417
17 2881 30.70 37 2786 29.69094139
18 3350 35.70 38 6609 70.43339254
19 5016 53.45 39 4759 50.71758437
20 1299 13.843 40 5711 60.86323268
Average 4229.75 45.077 Average 5400 57.55790409
Std. Dev. 1667.918964 17.775 Std. Dev. 1425 15.18769892

110

Table 5. Cell viability data for non-equilibrated (specimens 1-20) and equilibrated
(specimens 21-40). The percentage of dead cells was recorded in each zone.

 

 

 

% %

Specimen % Dead % Dead Dead Dead

Super Middle Deep ALL
1 12.24% 2.70% 5. 56% 5. 61 %
2 13.22% 7.20% 14.04% 10.70%
3 22.64% 4.40% 17.76% 12.03%
4 24.76% 8.31% 15.02% 14.37%
5 37.93% 30. 38% 52.73% 37.56%
6 30.34% 12.20% 6.76% 13. 74%
7 18.12% 10.33% 53.68% 20. 08%
8 23.47% 23.43% 24.30% 23.68%
9 23.0870 4.54% 24.04°/o 12.82%
10 26.57% 13.37% 31.17% 19.97%
1 1 34.95% 6.56% 44.86% 22.03%
12 16.00% 5.12% 5.93% 7.39%
13 23.01% 21.97% 52.94% 29.05%
14 39.08% 20.09% 37.37% 28.10%
15 21.80% 11.93% 14.81% 15.90%
1 6 43. 88% 1 6.74% 24. 19% 24.83%
17 16.48% 3.85% 11.61% 8.21%
18 44. 55% 29.21% 52 .94% 36.47%
19 47.92% 1 3. 86% 7.97% 18.44%
20 24.53% 3.13% 1 1 .88% 9.40%
Average 27.23% 12.47% 25.48% 1 8.52%
Std. Dev. 1 0.73% 8.63% 17.51% 9.27%

111

 

 

 

% % %
Specimen Dead % Dead Dead Dead
Super Middle Deep All
21 25.96% 1 1 .45% 24.00% 16.99%
22 35.06% 12.73% 12.00% 15.99%
23 46.58% 14.51% 30.87% 26.01%
24 22.34% 8.30% 7.76% 1 1 . 16%
25 30.61% 8.12% 13.73% 13.56%
26 29.90% 12.33% 19.05% 1 7.35%
27 31 .30% 5.60% 7.1 1% 10.55%
28 19.44% 2.28% 10.71% 8.66%
29 36.36% 18.75% 32.29% 26.80%
30 49.37% 1 .63% 2.26% 9.38%
31 61.25% 25.21% 5.71% 25.55%
32 48.72% 13.07% 4.26% 14.89%
33 36.36% 20.96% 5.41 % 16.33%
34 29.82% 3. 39% 0.00% 9.42%
35 36.07% 10.31% 5.41% 12.41%
36 26.92% 22.91% 5. 79% 16.87%
37 34.85% 21 .36% 30.51% 26.24%
38 67. 1 6% 1 6.03% 4.90% 21 .35%
39 27.50% 26.92% 26.09% 26.74%
40 29.10% 6.12% 5.93% 12.27%
Ave rage 38.23% 1 3.1 0% 12.89% 1 8.93%
Std. Dev. 1 2.47% 7.85% 1 0.49% 8.37%

112

Displacement Controlled Experiments

Table 6. Mechanical and weight data for all un-equilibrated explants

 

 

 

 

 

 

 

 

 

 

 

Time
to Peak Peak Initial Dry Percent
Specimen Peak Stress Strain Thickness Weight Weight Water
Is] [MPa] [mm] {91 L9]

1 0.048 60.607 0.608 0.801 0.02581 0.00864 66.52%
2 0.046 30.466 0.618 0.549 0.01753 0.00507 71.08%
3 0.046 49.009 0.567 0.635 0.02017 0.00642 68.17%
4 0.046 40.033 0.626 0.684 0.01807 0.00574 68.23%
5 0.045 27.980 0.624 0.597 0.01803 0.00488 72.93%
6 0.044 41.043 0.621 0.704 0.02300 0.00647 71.87%
7 0.046 28.458 0.573 0.488 0.01556 0.00444 71.47%
8 0.046 31.563 0.540 0.485 0.01509 0.00405 73.16%
9 0.044 33.857 0.644 0.685 0.0221 0.00631 71.45%
10 0.046 34.799 0.626 0.612 0.0217 0.00599 72.40%
11 0.043 33.752 0.584 0.548 0.01892 0.0061 67.76%
12 0.044 33.062 0.600 0.587 0.01871 0.00544 70.92%
13 0.043 37.060 0.605 0.660 0.01952 0.00584 70.08%
14 0.048 36.487 0.645 0.742 0.02462 0.00728 70.43%
15 0.045 23.802 0.613 0.514 0.01275 0.00353 72.31%
16 0.045 23.391 0.552 0.422 0.01195 0.00326 72.72%
17 0.044 66.081 0.535 0.653 0.02118 0.00745 64.83%
18 0.045 72.917 0.629 0.923 0.02784 0.01085 61.03%
19 0.046 73.336 0.578 0.815 0.02370 0.00909 61.65%
20 0.044 66.630 0.590 0.813 0.02357 0.00945 59.91%
21 0.046 61.414 0.548 0.675 0.02118 0.00702 66.86%
22 0.044 50.341 0.602 0.724 0.02001 0.00799 60.07%
23 0.048 57.210 0.568 0.709 0.02328 0.00630 72.94%
24 0.047 42.209 0.589 0.620 0.01876 0.00631 66.36%
25 0.045 34.214 0.563 0.521 0.01529 0.00467 69.46%
26 0.045 35.504 0.566 0.555 0.01676 0.00561 66.53%
27 0.042 38.019 0.537 0.526 0.01561 0.00521 66.62%
28 0.044 33.159 0.571 0.567 0.01764 0.00615 65.14%
29 0.046 44.463 0.570 0.678 0.02060 0.00676 67.18%
30 0.045 45.197 0.568 0.678 0.02261 0.00681 69.88%
31 0.044 38.309 0.607 0.687 0.01809 0.00571 68.44%
32 0.044 33.879 0.556 0.543 0.01785 0.00585 67.23%
33 0.045 39.708 0.567 0.519 0.01539 0.00448 70.89%
34 0.046 47.223 0.594 0.616 0.01918 0.00589 69.29%
35 0.046 50.515 0.633 0.754 0.01930 0.00673 65.13%
36 0.046 41.078 0.624 0.641 0.01897 0.00542 71.43%
37 0.046 26.137 0.627 0.596 0.01586 0.00447 71.82%
38 0.045 23.639 0.574 0.476 0.01358 0.00381 71.94%
39 0.043 50.356 0.572 0.775 0.02290 0.00809 64.67%
40 0.044 33.198 0.572 0.565 0.01960 0.00567 71.07%
41 0.046 32.308 0.595 0.575 0.01845 0.00597 67.64%
42 0.045 44.453 0.579 0.691 0.02380 0.00756 68.24%

113

 

43 0.043
Table 6 (cont)

33.885

0.531

0.508

0.01520

0.00435

7 1 .38%

 

Spechnen

Thus
to
Peak

Peak
Stress

Peak
Strain

Thickness

Weigh;

Dry
VWNght

Percent
VWﬂbr

 

 

 

 

13L

 

[MPa]

 

[mm]

 

191

 

[91

 

 

 

44

45

46

47

43
Ahnuege
Std. Dev.

0.046
0.045
0.043
0.042
0.045
0.045
0.001

36.975
39.134
32.184
37.827
38.276
40.940
12.403

0.570
0.575
0.531
0.569
0.539
0.823
0.031

0.592
0.627
0.505
0.629
0.593
0.828
0.103

0.01820
0.01722
0.01486
0.01743
0.01901
0.01909
0.00348

0.00509
0.0051 1
0.00402
0.00539
0.00607
0.00802
0.001 55

Table 7. Mechanical and weight data for all equilibrated explants

72 . 03%
70.33%
72.95%
69 .08%
68.07%
0.68783
0.03470

 

Sp.

Tune
to
Peak

Peak
Shess

Peak
Shaun

Thick
ness

lnﬂhﬂ
‘Hkﬂght

24hr
“knght

Dry
VWﬂght

Fﬂnal
VWMnr

“ﬁner
(iﬂn

 

 

 

IS]

 

[MPa]

 

 

[mm]

 

19L

 

[ﬂ

 

JHJ

 

 

 

101
102
103
104
105
106
107
108
109
110
111
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129

0.04
0.043
0.038
0.037
0.034
0.042
0.032

0.04
0.045
0.032
0.039
0.032
0.035
0.038
0.027

0.04
0.045
0.044
0.044
0.037
0.045

0.04
0.046
0.039
0.044
0.043
0.046
0.038

29.6132
30.7522
31.3075
31.0925
28.1859
28.1466
26.7588
31.6889
41.6000
31.8526
23.8261
16.0949
32.6873
38.6346
16.2095
56.4879
57.2592
35.7709
41.2968
54.5093
32.7198
46.7933
46.8620
37.4273
44.5050
39.2148
50.2143
18.6666

0.6101
0.6939
0.6260
0.6375
0.6108
0.6342
0.6196
0.6471
0.6163
0.5927
0.7299
0.6218
0.6519
0.6843
0.5877
0.6219
0.6352
0.6195
0.6428
0.5977
0.6766
0.6526
0.6543
0.621 1
0.5816
0.5860
0.6261
0.7040

0.5705
0.8176
0.6205
0.6574
0.6242
0.6062
0.6465
0.6901
0.6585
0.6512
1 .0422
0.7497
0.8105
0.9684
0.5229
0.8006
0.7503
0.5680
0.6864
0.7331
0.7072
0.8123
0.7695
0.6361
0.6154
0.5839
0.7461
0.7376

0.01517
0.02511
0.01856
0.01916
0.01612
0.02028
0.01950
0.02116
0.01762
0.01894
0.02115
0.01876
0.02102
0.02283
0.01617
0.02385
0.02377
0.01570
0.02189
0.02516
0.02037
0.02665
0.02177
0.01787
0.01846
0.01732
0.02168
0.01437

114

0.01666
0.03199
0.02089
0.02135
0.01780
0.02575
0.02800
0.02580
0.01834
0.02166
0.02656
0.02332
0.02447
0.02600
0.02342
0.02556
0.02483
0.01815
0.02356
0.02530
0.02401
0.02895
0.02351
0.02050
0.02042
0.01925
0.02240
0.01661

0.00524
0.00749
0.00585
0.00630
0.00546
0.00665
0.00706
0.00681
0.00619
0.00625
0.00760
0.00525
0.00828
0.00868
0.00581
0.00800
0.00808
0.00492
0.00678
0.00782
0.00668
0.00846
0.00671
0.00563
0.00601
0.00631
0.00601
0.00456

68.55%
76.59%
72.00%
70.49%
69.33%
74.17%
74.79%
73.60%
66.25%
71 . 14%
71 .39%
77.49%
66.16%
66.62%
75.19%
68.70%
67.46%
72.89%
71 .22%
69.09%
72.18%
70.78%
71 .46%
72.54%
70.57%
67.22%
73. 17%
72.55%

9.82%
27.40%
12.55%
1 1.43%
10.42%
26.97%
43.59%
21.93%

4.09%
14.36%
25.58%
24.31%
16.41%
13.89%
44.84%

7.17%

4.46%
15.61%

7.63%

0.56%
17.87%

8.63%

7.99%
14.72%
10.62%
1 1.14%

3.32%
15.59%

 

130 0.045 46.3564 0.6417 0.7545 0.01967 0.02045 0.00664 67.53%
0.046 50.2062 0.5860 0.6815 0.01977 0.02042 0.00633 69.00%
Table 7 (cont.)

131

132
133
134
135
136
137
138
139
140
141
142
143
144
146
147
148
Avg
Std.
Dev.

0.043
0.036
0.044
0.045
0.038
0.045
0.046
0.045
0.033
0.039
0.043
0.037
0.045
0.045
0.039
0.037
0.040

48.4188
41 .7288
58.6704
43.7454
40.9824
44.4755
49. 1 01 3
46.8620
37.6432
33.5299
22.9986
40.5880
24.5931
44.5139
42.4388
49.5821
38.4048

1 0.7872

0.6481
0.6038
0.5824
0.6563
0.6417
0.6555
0.6616
0.5993
0.5961
0.5875
0.5720
0.6197
0.5545
0.5908
0.6245
0.5701
0.8255

0.0371

0.8814
0.6522
0.6777
0.7392
0.7389
0.7573
0.8178
0.641 1
0.7826
0.5814
0.4491
0.7759
0.4342
0.7088
0.8032
0.7717
0.7050

0.1185

0.02499
0.02052
0.02126
0.02153
0.02536
0.02174
0.02147
0.02042
0.02544
0.01783
0.0141 1
0.02290
0.01554
0.02006
0.02509
0.02376
0.02047

0.00320

115

0.02632
0.02181
0.02208
0.02323
0.03001
0.02273
0.02272
0.02107
0.02990
0.01872
0.01505
0.02612
0.01654
0.02129
0.02880
0.02528
0.02299

0.00391

0.00793
0.00647
0.00750
0.00715
0.00847
0.00667
0.00775
0.00655
0.00858
0.00596
0.00427
0.00813
0.00504
0.00735
0.00910
0.00982
0.00883

0.001 25

69.87%
70. 33%
66. 03%
69 22%
71 .78%
70.66%
65. 89%
68.91 %
7 1 .30%
68. 1 6%
71 . 63%
68. 87%
69 . 53%
65.48%
68.40%
61 . 1 6%
70.1 6%

3.1 3%

3.97%
3.29%

5.32%
6.29%
3.86%
7.90%
1 8.34%
4.55%
5.82%
3. 1 8%
1 7.53%
4.99%
6.66%
14. 06%
6.44%
6. 1 3%
14. 79%
6.40%
1 2.44%

9.85%

Appendix B: Chapter 2 Raw Data

Table 1. Mechanical data for all rabbits

 

 

 

 

 

 

 

 

time to
Load peak
Group Rabbit N msec
time zero M54 621.16 3.2
M57 529.94 2.1
BU63 575.8 2.4
BU62 744.43 5.4
BB3 939.64 3.1
BU64 863.48 4.9
ave 712.41 3.52
stdev 164.78 1 .34
4 Days No
Poloxamer M40 738.44 3.2
M72 912.19 4.1
481 BF 550.45 3.1
BU66 525.07 2.3
BU60 595.65 3.8
BU61 600.54 5.1
ave 653.72 3.60
stdev 146.56 0.96
4 Days Poloxamer M74 730 4
M78 641 3
M90 741 4.2
M82 579.1 2.6
48OBF 510.68 3.7
BU65 764.3 2.2
ave 661 .01 3.28
stdev 101 .52 0.81
Total Average 675.7 3.5
Total St Dev 134.4 1 .0

 

116

 

Table 2. Matrix Damage (total surface ﬁssure length) Data for all rabbits

 

 

 

 

 

 

 

 

 

 

117

Matrix Damage (mm)
Un-impacted Impacted

time zero BU62 3.904468085 26.15521277
BU63 0 8.226702128
BU64 0.942659574 2.036914894
BB3 0 8.365744681
Average 1.21 1781915 1 1 .19614362
St Dev 1.84930766 1 0.40022374

4 Days No

Poloxamer 481BF 0.796702128 6.747659574
BU66 0 4.297765957
BU60 0 2.558617021
BU61 5.112978723 16.67085106
Ave rage 1 .47742021 3 7.568723404
St Dev 2.452631496 6.306694917

4 Days Poloxamer M74 13.95170213 26.654361?
M90 11.07319149 19.24531915
4808F 0.620212766 5.912234043
BU65 2.878085106 18.76308511
Ave rage 7.1 30797872 1 7.64375

St Dev 8.391 152389 8.814868203

 

Table 3. Time Zero group cell viability counts

%

°/o

 

 

 

 

 

 

 

Controls Dead test Dead
Super Middle Deep Total Super Middle Deep Total
6jc1a 7.67 9.63 8.20 8.54 6jt1a 23.71 33.82 44.32 31.70
6jc2a 19.77 21.71 11.00 17.70 6jt25 21.35 22.35 29.58 23.60
6jc33 8.37 12.54 0.00 9.45 6jt3a 58.79 41.52 16.67 41.29
6jo4a 8.66 6.21 14.94 8.78 6jt4a 22.99 29.91 32.37 28.97
6jc5a 2.94 5.47 9.34 5.51 6jt5a 32.96 23.40 31.32 28.01
6j06a 16.32 18.90 8.33 15.39 6jt6a 8.40 9.29 19.12 11.43
6jc7a 8.31 11.51 11.06 10.27 6jt7a 25.48 16.92 19.97 19.84
' 6jt8a 7.72 13.10 25.00 13.29
avg 8.71 12.28 10.48 9.66 20.37 23.79 24.86 24.77
std dev 4.30 6.11 2.51 3.24 9.19 10.86 6.38 9.89
6jc1a 12.00 6.83 26.87 11.88 6jt1b 34.02 25.20 33.33 29.68
6jc3a 14.83 16.61 37.35 18.81 6jt2b 13.14 22.18 14.57 16.99
6jC48 16.78 15.50 13.99 15.36 6jt3a 41.10 34.64 22.02 31.01
6j05b 10.94 8.53 6.96 8.68 6jt4b 21.66 16.55 24.58 19.69
6jt5a 9.65 20.26 28.54 20.83
6jt6a 55.84 37.40 35.92 40.89
avg 13.64 11.87 15.94 13.68 29.24 26.04 28.88 23.64
std dev 2.66 4.91 10.10 4.37 17.73 8.27 5.81 6.30
BU62La 1.32 16.32 9.23 12.13 BU62Ra 41.10 48.65 28.10 38.63
BU62Lb 6.02 3.15 14.51 6.96 BU62Rb 36.29 58.79 37.75 46.20
BU62Lc 4.84 15.34 21.64 12.89 BU62Rc 4.70 14.25 20.22 13.84
BU62Ld 17.72 10.21 16.07 13.51 BU62Rd 22.42 2.54 20.39 11.53
BUSZLe 4.12 6.49 19.39 8.88 BU62Re 57.55 61.76 44.90 55.36
BU62Lf 7.09 4.03 34.72 11.81 BU62Rf 16.67 10.24 14.17 11.95
BU62Lg 3.80 3.98 12.40 5.63 BUGZRg 44.44 31.17 21.20 29.42
BU62Lb 13.33 4.88 7.39 7.66
Average 5.79 8.05 14.37 9.93 31.88 32.49 23.64 29.56
St Dev 3.79 5.28 5.16 3.01 18.17 24.28 8.21 17.83
BU63Le 14.47 18.90 8.81 15.45 BU63Ra 28.46 12.48 5.87 12.19
BU63Lb 2.53 9.66 5.71 7.62 BU63Rb 49.66 9.40 19.70 18.40
BU63Le 2.17 2.29 4.56 3.14 BU63Rc 59.47 39.85 25.45 41.66
BUG3Ld 3.87 0.88 3.57 2.33 BU63Rd 84.31 36.67 18.23 34.37
BU63Le 14.81 7.59 8.83 8.74 BU63Ra 75.29 35.49 33.91 45.41
BU63Lf 1.69 6.63 9.79 6.91 BU63Rf 89.16 46.12 47.95 56.35
BU63Lg 13.16 6.19 29.13 14.32 BU63Rg 58.82 15.29 27.66 29.62

118

 

 

 

 

 

BU63Lh 6.81 5.87 3.09 5.39 BU63Rb 55.94 25.59 11.30 30.60
Average 7.44 5.59 6.34 7.99 67.52 27.61 20.30 33.57
St Dev 5.79 3.03 2.77 4.78 15.30 13.90 9.66 14.33
BU64La 5.93 3.92 11.41 6.72 BU64Ra 37.86 41.87 27.59 38.10
BU64Lb 16.33 3.00 1.65 4.04 BU64Rb 27.35 36.60 19.58 31.71
BU64Lc 5.13 7.65 7.35 7.38 BU64Rc 58.33 25.52 15.91 30.96
BU64Ld 22.22 5.28 14.02 9.96 BU64Rd 65.27 38.10 23.36 41.02
BU64Le 17.65 7.01 7.28 8.43 BU64Re 48.98 35.61 17.74 31.48
BU64Lf 3.96 9.31 12.23 9.32 BU64Rf 4.17 27.06 16.38 21.65
BU64Lg 14.68 5.43 7.33 7.42
BU64Lb 11.57 11.79 12.20 11.83
Average 12.18 5.94 10.26 8.72 47.56 34.13 18.59 34.65
St Dev 6.66 2.19 2.86 1.79 15.28 6.45 3.02 4.60
BB3La 24.01 11.99 25.76 18.05 BB3Ra 58.18 33.05 26.92 40.00
BB3Lb 31.95 25.74 10.77 23.20 BB3Rb 96.94 79.41 21.43 75.51
BB3Lc 20.69 1 1.48 26.86 17.52 BB3Rc 70.65 35.94 14.61 33.33
BB3Ld 15.63 24.79 7.50 17.74 BB3Re 69.00 47.97 15.25 45.10
BB3Rf 76.50 42.97 6.83 38.38
BB3Rg 48.00 39.52 31.30 38.13
BBBRh 87.39 48.33 15.23 41.73
Average 20.11 18.50 17.72 17.77 72.38 41.30 18.80 39.45
St Dev 4.22 7.82 10.01 0.27 16.61 6.27 8.32 3.94

119

Table 4. 4 Day No Poloxamer cell viability counts
%

 

 

 

 

 

 

 

 

 

Controls % Dead test Dead
Super Middle Deep Total Super Middle Deep Total

M406jc1a 14.80 15.09 12.08 14.43 M406jt1e 15.15 21.50 14.29 18.02
M406jc2a 8.45 8.89 7.95 8.60 M406jt4a 27.88 29.29 20.43 27.23
M406j03a 12.28 20.00 13.66 15.51 M406jt5a 18.87 20.11 20.14 19.81
M406jo4a 36.67 22.08 8.68 18.37 M406jt63 34.53 27.97 17.34 25.63
M406'L05a 16.37 11.52 16.04 14.23

Average 12.98 15.52 11.68 15.63 20.63 24.72 19.30 22.67
St Dev 3.45 5.55 3.39 1.91 6.55 4.59 1.70 4.45
M72La2 58.06 40.61 27.82 43.68 M72Ra 13.04 5.24 11.97 9.00
M72Lb 16.33 11.89 25.24 16.15 M72Rb 25.89 27.01 22.52 25.34
M72Lc 18.37 10.48 28.22 17.68 M72Rc 33.03 51.69 49.67 46.05
M72Ld 7.94 11.53 46.89 20.07 M72Rd 84.73 64.17 31.97 60.68
M72Lc 2.47 2.77 7.04 3.93 M72Re 53.39 43.21 49.74 48.01
M72Lf 16.67 8.52 10.69 10.90 M72Rf 59.52 50.89 1.96 48.43
M72Lg 6.74 13.71 12.78 11.05 M72Rg 18.13 5.76 6.92 8.70
M72Lh 10.66 11.21 29.57 15.91 M72Rh 40.00 25.40 23.32 27.72
M72Li 16.51 4.51 6.32 7.77 M72Ri 6.03 15.04 41.07 16.77
Average 11.96 9.33 18.46 12.93 31.13 32.04 26.57 32.30
St Dev 5.83 3.82 10.16 5.45 19.04 21.40 17.82 19.08
481BFLa 5.29 6.01 2.72 5.00 481BFRa 94.00 47.33 5.80 41.77
481BFLb 12.93 0.99 6.40 4.79 481 BF Rb 42.16 59.00 37.93 49.86
481BFLc 6.45 8.00 11.34 8.73 481BFRc 61.45 14.51 8.21 22.69
481BFLd 7.92 14.04 13.77 12.50 4813FRd 64.13 25.09 29.30 37.05
481 BF Le 5.02 4.13 9.63 5.52 481BFRe 40.54 15.69 28.02 22.50
481BFLf 12.66 7.69 12.07 9.84 481BFRf 85.09 14.79 10.34 26.36
Average 8.38 5.36 10.64 6.78 64.56 23.48 19.93 33.37
St Dev 3.57 2.89 2.80 2.34 21.80 14.03 13.46 11.29
BU66La 17.13 11.11 7.63 12.01 BU66Ra 28.38 34.76 25.00 31.16
BU66Lb 5.00 1.55 6.32 3.65 BU66Rb 29.24 22.86 27.98 25.94
BU66Lc 3.74 3.39 2.71 3.32 BU66Rc 63.25 16.82 33.17 33.04
BU66Ld 3.80 2.47 8.20 3.89 BU66Rd 32.20 18.87 9.94 20.30
BU66La 9.20 2.83 6.02 4.83 BU66Re 52.30 28.27 1 1.99 28.08
BU66Lf 2.70 6.97 1.55 4.98 BU66Rf 42.11 29.97 46.75 36.35
BU66Lg 9.77 6.87 17.82 9.72 BU66Rg 48.47 27.05 19.62 32.37
BU66Lb 25.29 15.75 10.53 15.26 BU66Rh 43.51 15.21 14.84 21.27
Average 7.33 5.03 6.14 6.06 39.46 24.23 20.36 28.56
St Dev 5.13 3.42 3.12 3.40 9.57 6.92 8.69 5.74
BU60La 60.00 20.30 13.27 27.85 BU60Ra 57.24 29. 10 8.70 31.60
BUSOLb 21.90 9.72 9.68 13.74 BU60Rb 76.83 25.24 5.30 28.12
BU60Lc 6.25 4.17 15.73 7.96 BUGORc 40.37 16.90 14.38 19.84
BUGOLd 37.38 23.81 21.65 27.52 BU60Rd 66.15 27.32 9.09 32.82
BUBOLe 20.75 8.85 9.28 11.41 BU60Re 47.97 26.25 3.29 28.22

120

 

 

 

BU60Lf 14.29 3.66 4.55 5.54 BU60Rf 38.81 29.92 10.19 26.49

BU60Rg 88.74 30.80 27.68 44.93

BU60Rb 65.56 43.04 80.91 56.50
Average 20.1 1 1 1.75 10.50 15.67 60.21 26.50 1 1.23 30.29
St Dev 11.49 8.42 4.26 9.72 17.59 4.68 8.07 7.69
BU61La 3.85 9.19 12.02 9.35 BU61 Ra 8.57 18.09 3.74 11.87
BU61Lb 12.50 18.32 27.54 20.26 BU61 Rb 5.26 8.33 14.17 8.94
BU61Lc 0.00 10.47 15.89 11.51 BU61Rc 0.00 0.99 12.77 3.56
BU61Ld 5.41 11.39 29.67 15.66 BU61 Rd 18.53 10.63 21.05 14.86
BU61 Le 28.89 17.96 31.67 23.18 BU61Re 16.77 17.13 7.89 13.58
BU61Lf 2.78 7.66 6.11 6.72 BU61Rf 6.45 9.68 7.89 8.85
BU61Lg 4.65 7.64 5.33 6.62 BU61Rg 9.21 7.99 23.53 13.42

BU61 Rh 10.90 17.63 33.19 20.87
Average 4.86 11.80 18.32 13.33 9.46 12.78 13.01 10.73
St Dev 4.19 4.54 11.22 6.56 6.03 i 4.61 7.25 3.91

121

Table 5. 4 Day Poloxamer Group cell viability counts

 

 

 

 

 

 

 

 

 

Controls % Dead test % Dead

Super Middle Deep Total Super Middle Deep Total
M74LFT1a 3.82 5.19 6.29 5.10 M74RT1a 25.97 14.63 9.22 15.54
M74LFT2a 14.63 30.15 6.08 21.93 M74RT2a 38.76 8.10 26.28 20.09
M74LFT3a 2.04 1.48 0.00 1.46 M74RT3a 9.06 9.76 38.00 15.61
R74LFT4a 9.22 12.70 9.38 11.08
Average 5.03 6.46 7.25 5.88 24.60 10.83 24.50 17.08
St Dev 3.74 5.72 1.84 4.86 14.90 3.39 14.47 2.61
M82L1 13.97 9.66 10.71 11.28 M82R1 7.27 3.43 8.65 5.51
M82L2 65.85 45.96 32.84 45.68 M82R2 13.66 8.74 20.71 13.33
M82L3 2.31 3.00 15.55 6.85 M82R3 7.77 5.41 13.16 7.54
M82L4 4.62 5.21 19.64 8.07 M82R4 32.57 20.05 19.05 23.29
M82L5 6.43 7.47 4.73 6.54 M82R5 30.52 11.20 5.50 17.41
M82L6 1.05 0.79 10.64 2.61 M82R6 70.18 31.17 6.41 35.59
M82L7 9.78 2.95 13.86 7.68
Average 6.36 4.85 12.52 7.17 18.36 9.77 12.25 13.42
St Dev 4.84 3.27 5.08 2.80 12.32 6.48 6.50 7.26
M78LFT1a 1.60 0.24 0.00 0.40 M78RT1a 2.34 0.77 1.69 1.47
M78LFT2a 2.78 3.07 7.54 4.26 M78RT4a 21.52 13.89 10.63 15.57
M78LFT4a 17.91 12.88 16.75 15.05 M78RT5a 27.61 7.86 4.17 10.89
M78LFT5a 2.41 4.99 1.38 3.57 M78RT6a 1.33 5.33 6.60 4.83
M78LFT6a 4.39 3.93 5.59 4.46
Average 2.79 3.06 3.63 3.17 13.20 4.65 4.15 8.19
St Dev 1.17 2.03 3.53 1.89 13.36 3.59 2.46 6.27
M90L1 31.15 15.73 ' 7.84 16.06 M90R1 55.38 21.48 25.56 27.15
M90L2 0.00 0.22 0.00 0.17 M90R2 89.33 95.80 77.27 91.29
M90L3 82.72 38.03 24.22 44.06 M90R3 35.85 32.25 21 .27 29.66
M90L5 80.79 54.98 28.22 55.23 M90R4 14.29 12.75 7.78 11.95
M90L6 11.68 4.14 3.77 6.70 M90R5 50.98 61.27 12.93 40.70
Average 41.27 14.53 12.81 24.44 39.12 31.94 16.88 27.36
St Dev 38.60 17.00 12.63 24.02 18.55 21.12 8.02 11.84
4808FLa 19.13 9.56 2.52 10.18 4808FRa 35.09 15.24 23.62 24.69
4808FLb 17.47 4.72 3.61 7.90 48OBF Rb 27.59 6.09 8.87 1 1.59
4808FLc 10.39 2.16 5.26 3.79 4808FRc 8.49 3.47 5.07 5.16
48OBFLd 29.79 13.03 7.10 12.71 4803FRd 15.38 11.68 9.68 11.62
4808FLe 31.46 2.47 3.53 6.27 4808FRe 16.59 9.45 4.72 9.99
4808FLf 16.07 13.66 7.53 13.09 4808FRf 4.55 2.19 5.88 3.99
4808FLg 13.04 6.79 3.49 6.41 4803FRg 27.19 3.98 9.45 10.77

122

 

 

 

4803FLh 3.67 1.18 4.43 2.61 480BFRh 15.79 9.72 7.58 11.22
Average 17.63 6.70 4.68 7.87 18.83 7.73 7.32 9.19
St Dev 9.35 4.92 1.81 3.87 10.34 4.54 2.10 3.22
BU65La 5.94 8.04 2.53 5.91 BU65Ra 40.28 34.47 5.47 27.05
BU65Lb 3.16 4.35 6.14 4.73 BU65Rb 44.67 24.54 11.41 24.08
BU65Lc 21.52 9.50 7.33 9.96 BU65RC 64.90 29.22 4.47 24.07
BU65Ld 25.00 7.35 15.18 13.19 BU65Rd 60.30 44.71 12.90 38.97
BU65Le 1.35 0.93 7.64 3.34 BU65Re 48.78 23.61 3.16 20.00
BU65Lf 13.51 4.55 8.66 7.06 BU65Rf 12.12 4.56 3.79 5.88
BU65Lg 7.98 2.78 10.45 6.86 BU65Rg 40.33 23.71 1.99 22.03
BU65Lh 2.35 3.50 26.82 10.70 BU65Rb 30.07 23.65 5.30 19.35
Average 10.10 5.13 8.28 7.72 47.05 29.13 5.08 22.76
St Dev 9.03 2.91 - 3.90 3.31 12.13 7.98 3.04 2.88

123

Appendix C: Chapter 3 Raw Data

Table 1. Mechanical data for the high rate of loading “with GlcN” (1-28) and “no GlcN”
(29- 56) explants

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Time to Peak Peak
Specimen Peak Stress Strain Thickness
[8] [MP8] {mm}

1 0.034 28.954 0.374 0.615
2 0.032 28.479 0.397 0.638
3 0.031 28.529 0.606 0.593
4 0.033 28.830 0.539 0.550
5 0.04 28.41 1 0.526 0.545
6 0.036 28.404 0.573 0.556
7 0.032 28.299 0.481 0.634
8 0.04 28.319 0.539 0.590
9 0.028 28.918 0.532 0.597
10 0.037 28.784 0.681 0.793
11 0.031 28.760 0.515 0.814
12 0.034 28.852 0.568 0.663
13 0.029 28.692 0.483 0.659
14 0.038 28.670 0.479 0.617
15 0.034 28.705 0.528 0.735
16 0.033 28.749 0.479 0.856
17 0.029 28.952 0.529 0.738
18 0.033 28.620 0.625 0.742
19 0.026 29.024 0.571 0.789
20 0.036 28.462 0.445 0.678
21 0.027 28.010 0.525 0.634
22 0.039 27.408 0.491 0.687
23 0.039 27.454 0.468 0.777
24 0.036 28.061 0.508 0.906
25 0.039 27.346 0.442 0.725
26 0.04 27.261 0.469 0.512
27 0.034 28.090 0.605 0.931
28 0.037 27.379 0.523 0.690
Avergge 0.034 28.31 5 0.517 0.681
Std. Dev. 0.004 0.575 0.065 0.1 1 2

 

124

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Time to Peak Peak
Specimen Peak Stress Strain Thickness
[8} [MP8] {mm}
29 0.039 29.128 0.676 0.530
30 0.031 28.849 0.559 0.566
31 0.034 28.840 0.457 0.603
32 0.036 28.777 0.504 0.338
33 0.030 28.921 0.617 0.548
34 0.031 29.183 0.536 0.609
35 0.031 28.630 0.483 0.657
36 0.032 28.764 0.471 0.715
37 0.029 28.858 0.610 0.546
38 0.025 28.620 0.569 0.685
39 0.034 28.986 0.475 0.698
40 0.034 28.960 0.563 0.722
41 0.029 29.060 0.529 0.677
42 0.033 29.032 0.530 0.597
43 0.037 28.579 0.494 0.695
44 0.032 28.865 0.509 0.587
45 0.025 29.339 0.537 0.787
46 0.029 29.550 0.676 0.732
47 0.032 28.918 0.664 0.601
48 0.026 28.192 0.647 0.871
49 0.034 28.757 0.514 0.688
50 0.033 28.405 0.524 0.696
51 0.038 28.069 0.582 0.604
52 0.033 28.444 0.608 0.987
53 0.036 28.514 0.543 0.763
54 0.037 28.011 0.480 0.725
55 0.039 27.789 0.539 0.612
56 0.040 27.808 0.613 0.549
Average 0.033 28.641 0.551 0.661
Std. Dev. 0.004 0.473 0.062 0.11 6

 

 

 

 

 

125

 

Table 2. Mechanical data for the low rate of loading “with GlcN" (57-84) and “no

GlcN” (85-112) explants

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Time to Peak Peak
Specimen Peak Stress Strain Thickness
[8} [MP8] {mm}
57 0.901 29.372 0.576 0.437
58 0.879 29.370 0.637 0.425
59 0.797 29.452 0.588 0.648
60 0.800 29.471 0.686 0.687
61 0.875 29.367 0.569 0.534
62 0.858 29.390 0.664 0.548
63 0.878 29.391 0.697 0.585
64 0.833 29.449 0.716 0.744
65 0.830 29.396 0.722 0.696
66 0.883 29.413 0.639 0.638
67 0.954 28.931 0.525 0.606
68 0.954 28.945 0.628 0.561
69 0.948 28.918 0.651 0.571
70 0.948 28.893 0.622 0.383
71 0.941 28.934 0.687 0.585
72 0.955 28.905 0.499 0.642
73 0.936 28.914 0.571 0.616
74 0.948 28.939 0.546 0.553
75 0.941 28.939 0.608 0.802
76 0.947 28.941 0.595 0.723
77 0.878 28.075 0.536 0.776
78 0.866 28.247 0.659 0.973
79 0.840 28.162 0.579 0.818
80 0.912 28.038 0.570 0.713
81 0.823 28.169 0.692 0.914
82 0.895 28.223 0.521 0.721
83 0.837 28.087 0.582 0.590
84 0.903 28.205 0.564 0.646
Avergge 0.890 28.833 0.61 5 0.653
Std. Dev. 0.049 0.521 0.062 0.135

 

 

 

 

 

 

126

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Time to Peak Peak
Specimen Peak Stress Strain Thickness
[8] [MP8] 1mm}

85 0.823 29.467 0.619 0.574
86 0.828 29.529 0.715 0.747
87 0.874 29.432 0.625 0.583
88 0.886 29.452 0.578 0.552
89 0.880 29.409 0.556 0.523
90 0.861 29.452 0.636 0.681
91 0.886 29.411 0.540 0.452
92 0.840 29.452 0.605 0.708
93 0.826 29.524 0.655 0.949
94 0.852 29.426 0.645 0.649
95 0.941 28.882 0.413 0.569
96 0.945 28.916 0.536 0.700
97 0.945 28.868 0.612 0.550
98 0.946 28.882 0.550 0.468
99 0.943 28.880 0.481 0.411
100 0.983 28.927 0.569 0.592
101 0.946 28.895 0.491 0.627
102 0.942 28.939 0.584 0.653
103 0.934 28.921 0.631 0.675
104 0.910 28.973 0.712 0.889
105 0.870 28.306 0.572 0.839
106 0.867 28.260 0.644 0.908
107 0.857 28.156 0.622 0.720
108 0.925 28.208 0.600 0.713
109 0.866 28.188 0.685 0.875
110 0.854 28.151 0.628 0.597
111 0.916 28.113 0.485 0.642
112 0.912 28.139 0.556 0.623
Average 0.895 28.851 0.590 0.662
Std. Dev. 0.043 0.529 0.068 0.133

 

 

 

 

 

127

 

Table 3. Total length of surface ﬁssures for the high rate of loading “with GlcN” (1-28)
and “no GlcN” (29-56) explants

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Specimen Surface Fissure Lengtlu
[ml

1 11.574
2 12.575
3 47.798
4 21.272
5 32.920
6 37.343
7 36.256
8 45.666
9 30.735
10 45.101
1 1 40.625
12 27.741
13 29.936
14 13.609
15 26.345
16 49.226
17 35.766
18 48.512
19 48.906
20 11.126
21 17.861
22 16.551
23 29.659
24 30.885
25 7.417
26 32.238
27 47.488
28 37.865
Average 31 .480
Std. Dev. 1 3.31 5

 

 

 

 

128

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Specimen Surface F issure Lenﬂ
{mm}

29 40.615
30 53.915
31 35.265
32 14.622
33 56.803
34 40.966
35 50.718
36 25.471
37 48.885
38 54.458
39 54.458
40 29.659
41 39.005
42 1 5.741
43 38.888
44 72. 107
45 46.870
46 59.243
47 56.750
48 13.460
49 17.542
50 63.176
51 37.716
52 47.989
53 53.339
54 53.670
55 55.247
56 48. 107

Average 42.787

Std. Dev. 1 5.273

 

 

 

129

Table 4. Number of dead cells per ﬁssure, per millimeter of ﬁssure depth, and
percentage of dead cells in each zone for the high rate of loading “with GlcN” (1-28) and
“no Glc ” (29-56) explants

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dead Cells Dead Cells
Spec per Depth per ﬁssure Sugrficial Middle Deep
%
% Dead Dead % Dead
cells - cells cells
1 271.244 45.5 47.57% 9.30% 9.23%
2 589.093 79 74.68% 14.90% 17.29%
3 395.119 69 25.61% 21.90% 4.76%
4 576. 166 54 38.33% 9.45% 12.50%
5 740.307 67 28.09% 1 3. 1 3% 3.88%
6 165.768 72 44.64% 23.41% 19.35%
7 203.675 23 28.99% 3.67% 7.84%
8 103.434 10.5 20.00% 14.47% 3.45%
9 344.611 59 37.50% 15.19% 9.03%
10 167.447 27 32.99% 8.66% 4.61%
11 492.672 61 57.07% 14.10% 14.77%
12 656.843 28 15.79% 21.22% 2.60%
13 763.862 121 52.19% 7.65% 5.56%
14 867.998 187 35.00% 76.66% 30.51%
15 621 .828 45 53.49% 49.02% 58.87%
16 252.752 32 14.25% 1.44% 0.64%
17 494.330 43 38.52% 30. 18% 5.59%
18 50.331 10 6.56% 8.39% 30.26%
19 52.011 11 8.21% 27.74% 31.71%
20 195.570 12 4.72% 4.39% 3.94%
21 485.298 76 23.08% 37.58% 18.37%
22 41 1 .184 50 28.28% 13.40% 94.95%
23 76.046 6.5 13.85% 9.18% 23.08%
24 364.015 70 42.86% 12.59% 7.74%
25 877.514 144 51.85% 15.64% 23.53%
26 295.421 18.5 34.88% 12.56% 8.96%
27 288.147 48 36.59% 16.04% 18.26%
28 92.807 25.5 31 .31 % 6.78% 3.64%
Avergge 384.036 52.083 32.81% 17.57% 16.75%
Std.
Dev. 242.015 40.722 16.46% 15.34% 19.51%

 

 

130

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dead Cells Dead Cells
Spec per Dean per fissure Superficial Middle Deep
% Dead % Dead % Dead
cells cells cells

29 1343.968 39.5 87.64% 30.51% 4.30%
30 1 1 11.769 30 76.83% 18.09% 1.89%
31 506.197 38 60.40% 42.13% 10.23%
32 715.533 62 41.67% 64.03% 16.83%
33 229.293 21 88.64% 23.23% 6.50%
34 778.702 81 56.90% 19.49% 0.82%
35 640.310 67 64.20% 19.65% 21.24%
36 849.974 89 36.36% 1 1.50% 21 .97%
37 625.101 43.5 53.70% 53.52% 38.10%
38 1294. 306 138 23.33% 27.37% 39.51%
39 1000.114 90 44.93% 49.46% 21.78%
40 403.560 98 87.64% 83.08% 20.00%
41 631.756 153 26.76% 17.33% 2.13%
42 384.203 58 77.08% 86.73% 32.26%
43 1075.168 120 74.53% 59.23% 8.00%
44 1451.708 254 63.30% 56.44% 11.71%
45 1036.651 199 35.65% 11.88% 0.00%
46 910.239 89 55.56% 72.88% 62.64%
47 850.355 77.50% 75.59% 43.1 7%
48 1965.854 205 48.59% 43.50% 17.50%
49 766.297 84 44.87% 7.54% 16.89%
50 243.619 17.5 48.62% 26.36% 14.04%
51 823.129 121 60.38% 29.33% 24.00%
52 431.722 92 32.89% 17.91% 22.81%
53 568.862 88.5 63.77% 14.93% 10.26%
54 615.385 36 75.31% 21.01% 26.62%
55 1919.585 82 49.69% 8.77% 17.46%
56 994.941 61 40.16% 25.00% 15.15%

Average 831.779 90.150 56.30% 35.08% 18.31%

Std.

Dev. 438.445 55.038 18.50% 23.85% 14.09%

 

131

 

Table 5. Number of dead cells per ﬁssure, per millimeter of ﬁssure depth, and
percentage of dead cells in each zone for the low rate of loading “with GlcN” (57-84) and
“no Glc ” (85-112) explants

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dead Cells Dead Cells
Spec per Depth per fissure Superficial Middle Deep
% Dead % Dead % Dead
cells cells cells

57 729.990 44.5 59.76% 21.51% 12.61%
58 1285.474 98 74.32% 26.29% 16.42%
59 276.380 28.5 52.17% 18.11% 6.33%
60 502.623 47 68.09% 9.31% 9.68%
61 406.038 31 .5 30.83% 4.92% 4.27%
62 244.451 42 26.67% 10.34% 5.26%
63 338.911 47 50.65% 12.05% 2.86%
64 101.608 11 39.13% 3.80% 1.67%
65 250.448 29 46.91% 22.73% 13.95%
66 18.03% 13.28% 15.71%
67 712.474 158 58.46% 30.19% 6.73%
68 1849.571 127 47.44% 25.29% 97.70%
69 1238.857 102 43.30% 12.50% 5.13%
70 1602.702 214 66.09% 54.25% 8.64%
71 305.938 73 41.38% 21.70% 12.24%
72 917.397 72 37.69% 7.42% 2.78%
73 667.539 113 27.71% 63.56% 27.56%
74 1003.215 160 36.73% 41.48% 23.39%
75 729.990 44.5 57.80% 22 .03% 16.82%
76 1285.474 98 46.58% 23.74% 6. 77%
77 276. 380 28.5 64.00% 14.46% 17.86%
78 502.623 47 72.73% 12.82% 7.96%
79 406.038 31.5 58.76% 35.03% 17.91%
80 244.451 42 25.00% 6.25% 23.21%
81 1096.694 121 41.18% 11.01% 14.00%
82 338.911 47 57.35% 23.67% 8.18%
83 101.608 11 66.67% 14.58% 8.16%
84 250.448 29 53.45% 26.67% 22.62%

Avegge 664.740 72.1 03 48.89% 21 .04% 1 4.87%

Std.

Dev. 467.848 50.385 15.20% 14.13% 17.68%

 

132

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dead Cells Dead Cells
Spec per Depth per fissure Superficial Middle Deep
' % Dead % Dead % Dead
cells cells cells

85 1164.815 160 61.76% 59.04% 19.80%
86 621.316 70 78.38% 64.00% 37.04%
87 1065.220 182 81 82% 56.68% 29.71%
88 1488.102 67.5 77.06% 25.96% 14.71%
89 708. 396 89 83.56% 32.38% 12.50%
90 809.758 126 64.08% 16.67% 17.48%
91 930.205 108 58.33% 33.74% 26.27%
92 990. 388 84 72.34% 49.22% 1 6.16%
93 1209.386 87 38.54% 20.91% 2.60%
94 1298.088 274 64.08% 46.41% 34.33%
95 652.543 94 43.48% 25.40% 4.23%
96 946.971 72.5 67.95% 1 1 .58% 10.62%
97 931.217 62.5 44.26% 19.67% 6.58%
98 896.670 142 66.00% 34.29% 21.71%
99 808. 595 109.5 64.84% 22.86% 12.33%
1 00 1286. 551 244 80.56% 54.01% 9.85%
101 703.538 149 44.23% 69.09% 23.31%
102 662.031 216 68.18% 53.42% 22.40%
103 398.012 77 50.00% 9.17% 13.75%
104 646.105 62 87.50% 28.24% 11.96%
105 522 .482 54 66.67% 29.02% 9.62%
106 1360.639 131 72.15% 30.57% 16.67%
107 611.612 72 41.94% 15.61% 8.70%
108 645.588 50.5 57.14% 6.09% 3.15%
109 907.029 80 51.30% 25.48% 28.21%
1 10 479.660 35 37.96% 2.27% 0.88%
111 1151.738 92 64.37% 34.66% 19.67%
112 925.558 162 61.36% 54.12% 16.85%

Avme 898.920 1 10.794 62.49% 33.23% 1 6.1 1 %

Std.

Dev. 290.508 58.817 1 4.25% 18.41 % 9.41 %

 

133

 

Table 6. Material property values for the theoretical model

 

 

 

 

 

 

“no GlcN” “with GlcN”
E1 (MPa) 7.835 7.835
E3(MPa) 1.53 1.53
k1(*10'” m2) 5.95 3.149

k3 («10'13 mT) 1.222 0.858

 

 

 

 

134

Appendix D: Rabbit impact standard operating procedure

RABBIT PATELLOFEMORAL IMPACT SOP

 

What to bring with you:
1.) Portable computer w/ A2D board, 2.) LPS lubricant, ethanol, 3.) meter stick, 4.) rabbit

data sheets (a copy is attached to the end of this SOP), 5.) blank IBM formatted disks

Pre-test set up: (* Leave portable computer off until all cables are connected)

1.

Plug in Valadyne strain gauge ampliﬁer (SGA) to the wall. Turn on the SGA and
insure that the trigger release switch (see ﬁgure 1) is turned off (down position) to
protect from accidental triggering.

The SGA will need to be on for 15 minutes in order to allow the electronics to
stabilize.

Assure that all electronic connections are in place. Figures 2 through 4 carefully
detail where all connections should be made.

Once you have made all the necessary connections, turn on and start up computer.
Computer will give a list of possible conﬁgurations, choose “Ethernet
conﬁguration”.

Spray some LPS greaseless lubrication on a rag and wipe down the steel rod and
gray rails of the impact cart (see ﬁgure 5). Do this very sparingly.

Use alchohol to clean the sides of the cart (see ﬁgure 5) i.e., the portion of the cart
where the brakes act. Keep rabbit hair off the rubber brake pads.

After the 15 minutes are up, run the program “rabinsur.vi” located on the portable
computers desktop. If the program is running (hit small arrow in the top left of
screen in LabView) you will see a readout for the load.

Using the small screwdriver, zero the load cell on the SGA. Use the opening to
the top left of the black dial (see ﬁgure 1). Use the readout in LabView.

Calibrate the load cell by depressing and holding the shunt cal switch on the
Valadyne strain gauge ampliﬁer. If needed readjust the set point to 3349 N on
LabView (or 7.53 Volts on the voltmeter) by using the Gain knob (See ﬁgure 1).
**** The values in step 9 are susceptible to change whenever the load cell is

changed. Always double check these values with the load cell speciﬁcations.

135

10. Double check to make sure the load cell is still zeroed, adjust if necessary.

Rabbit preparation:

1. On the Data Sheet, record the Rabbit name, weight (kg), and sex.

2. Once the rabbit is fully sedated, pull the left hind foot through the very bottom
hole of the leather strap of the holding chair and tuck the rest of the strap under
the rabbit so it can be ﬁxated to the underside of the chair.

3. Position the right leg so that the femur is pointing vertical and the impactor is
directed to hit the middle of the patella.

4. Place the black strap around the right hind foot and pull tight to insure the foot is
fully constrained. It is important that the femur remain vertical, but the tibia can
be horizontal in order to ﬂatten the patella. Fix the end of the strap, as well as the
leather strap, to the Velcro pad on the underside of the chair.

5. Move the clamping bar into position and attach the free end to the distal clamp.

6. Slowly apply even pressure to both clamps until the clamps lock in place.

7. Slide the chair into position so the patella is directly under the head of the
impacting cart.

8. Lower the cart so the head of the impacting cart is just above the patella, checking
to insure the patella is centered under the head.

9. Raise the impact cart to the desired height. Measure from top of patella to the
bottom of the impactor head using the meter stick (see ﬁgure 6)

10. The position of the beam holding the upper brake may need repositioning to
accommodate the desired height. Do this by removing the screws of the beam
and moving the beam to the desired height above the patella.

1 l. The height and weight of the sled should be:

Energy Height Mass
6 J 0.46meters 1.33kg
10 J 1 meter 1 kg

0 ﬁgure 5 shows 6 J set-up I
0 6 J impacts should see between 500 and 600 N as to where 10 J impacts
should see from 900 to 1200 N

136

Impacting the rabbit:

1.
2.

99:59“

10.
11.

Turn on the trigger enable switch on the Valadyne strain gauge ampliﬁer.

Click the arrow at the top left of the LabView interface (the “run” arrow) (see
ﬁgure 7). This should enable the load readout to be constantly changing (if load
readout is already constantly changing then the arrow has already been hit). Then
click on the large green START button below the graph area on LabView (button
that reads “disabled” in ﬁgure 7). WAIT AT LEAST 5 SECONDS BEFORE
PROCEEDING.

Press the red trigger button on the Valadyne SGA to drop the impact cart.
Impact cart will drop and the A/D board will trigger and save the data

LabView will prompt user for a ﬁlename and location.

Switch the trigger enable switch to disable (down position) before removing the
rabbit.

Excel will automatically run and a macro will plot the data Note the peak load
and time to peak on the data sheet. Also sketch the graph and note any comments.
Choose “save as” from the ﬁle menu. Save the ﬁle as an Excel Worksheet i.e.,
“.xls” format. Back up all data ﬁles to a ﬂoppy disk.

FOLLOWING IMPA CT ION OF ALL ANIMALS, REMO VE DISK FROM
DRIVE ((1:12 AND COPY ALL FILES TO THE (g:luserlbimﬂadz
DIRECTORY FOR PERMANENT DOCUMEN TA TION.’

When testing is done shut down computer before disconnecting cables.

Turn off voltmeter and leave on SGA if more testing is going to be done in the

next week.

137

Figures

    
  
     
  
  
  

Load Cell
Balance Screw

 

Trigger Enable
Switch

 

   
  
  
 
  
    
 
    
    
 
    
   

Load Cell
Gain Knob

Trigger
Button

   

Load Out: to
Interface Box
see Fig 3

 

To Catch Release

(yellow chord on

drop ﬁxture 120
VAC)

Load In:
from load cell

  

.3 a
Trigger Out: to
Intterface Box
see Fig 3

Power
from Wall

 

 

.
4

Figure 2. Rear view of Validyne SGA

 

138

 

 
  
 

 
     
 

Load Out from
SGA (see ﬁgure 2)

 
   
 
   
 
   
    
  

Voltmeter set to
20 VDC

 

 

 
 

Trigger Out from
SGA (see ﬁgure 2)

 

Interface Box Out to A/D
Board (see ﬁgure 4)

Portable
Computer with
A2D Card

Out from
Interface Box

 

n.»nnnn»» .
”nun/2”

"Mme.” .

    

Figure 4. Connection from interface box to A2D card

139

 

 
 
  
  
 
   

Removable Weights (nuts,
screws included).

Cart Sides

Gray Plastic Rails

Steel Rod

Figure 5. Cart with 1.33 Kg (6 Joule) set up

140

 

Load Cell

 

Figure 6. Lower view of impact drop ﬁxture w/ load cell and impactor head

141

L ad Cell Reado t

 

 

Figure 7. Screen shot of labview layout

 

 

"Trigger Value has been changed to
50 in order to prevent accidental
triggering

 

 

142

sl ‘1] I11 1.

uljt‘jjjjjjjjj1111111