Lorne Gale Thesis Lubrication

Biotribological Assessment for
Artificial Synovial Joints: The Role

of Boundary Lubrication

Doctorate of Philosophy
in
Biomedical Engineering

by
LORNE R GALE, BE(Mech, Hons)

Thesis submitted for the degree of Doctor of Philosophy,
School of Engineering Systems,
Institute of Health and Biomedical Innovation (IHBI),
Queensland University of Technology, Brisbane

2007

ii

iii
Abstract
Biotribology, the study of lubrication, wear and friction within the body, has
become a topic of high importance in recent times as we continue to encounter
debilitating diseases and trauma that destroy function of the joints. A highly
successful surgical procedure to replace the joint with an artificial equivalent
alleviates dysfunction and pain. However, the wear of the bearing surfaces in
prosthetic joints is a significant clinical problem and more patients are surviving
longer than the life expectancy of the joint replacement. Revision surgery is
associated with increased morbidity and mortality and has a far less successful
outcome than primary joint replacement. As such, it is essential to ensure that
everything possible is done to limit the rate of revision surgery. Past experience
indicates that the survival rate of the implant will be influenced by many
parameters, of primary importance, the material properties of the implant, the
composition of the synovial fluid and the method of lubrication. In prosthetic
joints, effective boundary lubrication is known to take place. The interaction of the
boundary lubricant and the bearing material is of utmost importance. The identity
of the vital active ingredient within synovial fluid (SF) to which we owe the near
frictionless performance of our articulating joints has been the quest of researchers
for many years. Once identified, tribo tests can determine what materials and more
importantly what surfaces this fraction of SF can function most optimally with.
Surface-Active Phospholipids (SAPL) have been implicated as the bodys natural
load bearing lubricant. Studies in this thesis are the first to fully characterise the
adsorbed SAPL detected on the surface of retrieved prostheses and the first to
verify the presence of SAPL on knee prostheses.

Rinsings from the bearing surfaces of both hip and knee prostheses removed from
revision operations were analysed using High Performance Liquid
Chromatography (HPLC) to determine the presence and profile of SAPL. Several
common prosthetic materials along with a novel biomaterial were investigated to
determine their tribological interaction with various SAPLs. A pin-on-flat
tribometer was used to make comparative friction measurements between the
various tribo-pairs. A novel material, Pyrolytic Carbon (PyC) was screened as a

iv
potential candidate as a load bearing prosthetic material. Friction measurements
were also performed on explanted prostheses.

SAPL was detected on all retrieved implant bearing surfaces. As a result of the
study eight different species of phosphatidylcholines were identified. The relative
concentrations of each species were also determined indicating that the unsaturated
species are dominant. Initial tribo tests employed a saturated phosphatidylcholine
(SPC) and the subsequent tests adopted the addition of the newly identified major
constituents of SAPL, unsaturated phosphatidylcholine (USPC), as the test
lubricant. All tribo tests showed a dramatic reduction in friction when synthetic
SAPL was used as the lubricant under boundary lubrication conditions. Some tribo-
pairs showed more of an affinity to SAPL than others. PyC performed superior to
the other prosthetic materials. Friction measurements with explanted prostheses
verified the presence and performance of SAPL.

SAPL, in particular phosphatidylcholine, plays an essential role in the lubrication
of prosthetic joints. Of particular interest was the ability of SAPLs to reduce
friction and ultimately wear of the bearing materials. The identification and
knowledge of the lubricating constituents of SF is invaluable for not only the future
development of artificial joints but also in developing effective cures for several
disease processes where lubrication may play a role. The tribological interaction of
the various tribo-pairs and SAPL is extremely favourable in the context of reducing
friction at the bearing interface. PyC is highly recommended as a future candidate
material for use in load bearing prosthetic joints considering its impressive
tribological performance.

Keywords
SAPL, orthopaedics, biotribology, boundary lubrication, prosthetics, total joint
replacement, PC, USPC, PyC, Pyrolytic Carbon, surfactant, synovial fluid, SF,
arthritis, joint disease, cartilage, artificial joints, BL.

v
List of Abbreviations
SAPL - Surface Active Phospholipid
TJR - Total Joint Replacement
PC - PhosphatdiylCholine
USPC - Unsaturated Phosphatdiylcholine
SPC - Saturated Phosphatdiylcholine
BL - Boundary Lubrication
HPLC - High Performance Liquid Chromatography
SF - Synovial Fluid
GAG - Glycosaminoglycan
HA - Hyaluronic Acid
DPPC - Dipalmitoyl Phosphatidylcholine
DLPC - Dilinoleoyl Phosphatidylcholine
PLPC - Palmitoyl Linoleoyl Phosphatidylcholine
POPC - Palmitoyl Oleoyl Phosphatidylcholine
DOPC - Dioleoyl Phosphatidylcholine
SLPC - Stearoyl Linoleoyl Phosphatidylcholine
PSPC - Palmitoyl Stearoyl Phosphatidylcholine
OSPC - Oleoyl Stearoyl Phosphatidylcholine
BSA - Bovine Serum Albumin
- Coefficient of friction
OA - Osteo Arthritis
RA - Rheumatoid Arthritis
PyC - Pyrolytic Carbon
LGP - Lubricating Glyco Protein
LTI carbon - Low Temperature Isotropic carbon
GCS - Glucosamine & Chondroitin Sulfate
NSAIDS - Non-Steroidal Anti-Inflammatory Drugs
VS - Visco Supplementation

vi

vii
Glossary of Terms
adsorption: A mechanism of attaching to a surface by chemical or
physical bonding.
amphipathic: One end of a molecule has an affinity for the phase in which
it is in, the other end is repelled by the same phase.
arthrocentesis: The surgical puncture and aspiration of a joint.
arthrology: That part of anatomy which treats of joints.
asperities: roughness of a surface , highest points.
boundary lubrication: Lubrication where there is solid-to-solid contact of the
sliding surfaces.
chondrocyte: The cellular component of the cartilage matrix
contact angle: The angle subtended at the edge of a droplet at the triple
point.
colloid: Microscopic particles suspended in some sort of liquid
medium.
cytokine: Mediator of inflammation.
detritus: A mass of substances worn off from solid bodies by
attrition, and reduced to small portions.
esterified: A chemical reaction in which two chemicals (typically an
alcohol and an acid) form an ester as the reaction product.
glycosaminoglycan: Any of a class of polysaccharides derived from hexosamine
that form mucins when complexed with proteins: formerly
called mucopolysaccharide.
HPLC: HPLC is used to separate components of a mixture by using
a variety of chemical interactions between the substance
being analysed and the chromatography column.
hyaluronic acid: A mucopolysaccharide serving as a viscous medium in the
tissues of the body and as a lubricant in joints: a GAG.
hyaluronidase: An enzyme that catalyses the breakdown of hyaluronic acid
in the body, thereby increasing tissue permeability to fluids.
Hydrodynamic Lubrication where the sliding surfaces are separated by
lubrication: a wedge of fluid.
hydrophilic: Highly compatible with the aqueous phase, or water-
loving.
hydrophobic: Repels the aqueous phase, or water-hating .
lamellar body: Layered structure (A storage form for surfactant).
lipid: Any of a group of organic compounds, including the fats,
oils, waxes, sterols, and triglycerides, that are insoluble in
water but soluble in nonpolar organic solvents, are oily to
the touch, and together with carbohydrates and proteins
constitute the principal structural material of living cells.
lipase: Any of a class of enzymes that break down lipids.
lubricin: A glycoprotein implicated as the boundary lubricant for the
synovial joint.
lyophilic: Characteristic of a material that readily forms a colloidal
suspension.

viii
mucin: Any of a class of glycoproteins found in saliva, gastric juice,
etc., that form viscous solutions and act as lubricants or
protectants on external and internal surfaces of the body.
non-Newtonian: As the rate of shear increases, viscosity decreases; as the
rate of shear decreases, viscosity increases.
osteoarthritis: A degenerative joint disease where the initiating event can
be joint trauma, acute or repetitive.
proteolipid Any of a class of lipid-soluble proteins.
proteoglycan: Any of various mucopolysaccharides that are bound to
protein chains in covalent complexes and occur in the
extracellular matrix of connective tissue.
rheumatoid arthritis: A degenerative disease which is predominantly
inflammatory in nature and origin.
Ringer's solution: An aqueous solution of the chlorides of sodium, potassium,
and calcium that is used topically as a physiological saline.
surfactant: A substance that can effectively modify the surface energy
of an interface.
synoviocyte: The cellular component of the synovial membrane.
synovium: Another term for the synovial membrane.
tribology: The science and technology of surfaces that are in contact
and move in relation to each other.
tribo-pair: The name given to the two materials that slide over each
other in a lubricated system.
triple point: The point where solid, liquid and air all meet.
trypsin: An enzyme capable of breaking down proteins .
turbostratic: A type of crystalline structure where the basal planes have
slipped sideways relative to each other, causing the spacing
between planes to be greater than ideal.
zwitterion: A molecule in which each end carries a different charge
(positive or negative)to produce a charge dipole.

ix
List of Publications and Manuscripts
Refereed Publications
[1] GALE, L.R., R. COLLER, D.J. HARGREAVES, B.A. HILLS, and R.
CRAWFORD (2007): 'The role of SAPL as a boundary lubricant in prosthetic
joints', Tribology International, 40(4), pp. 601-606

[2] GALE, L.R., Y. CHEN, B.A. HILLS, and R.W. CRAWFORD (2006):
'Boundary lubrication of joints: Characterisation of Surface-Active Phospholipids
found on retrieved implants', Acta Orthopaedica, 78(3), pp. 309-314

[3] GALE, L.R., R.W. CRAWFORD, D.J. HARGREAVES and J. KLAWITTER
(2006): 'Boundary lubrication of Pyrolytic Carbon with Surface Active
Phospholipids: Tribological Assessment for Artificial Joints', Acta Orthopaedica,
Under Revision, pp.

[4] L.R. GALE , GUDIMETLA, P., Y. CHEN, R. CRAWFORD and D.J.
HARGREAVES (2007): ' Tribological Testing of Saturated and Unsaturated
Surface Active Phospholipids: Implications for artificial joints', Proceedings of the
Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine,
Submitted, pp.

Refereed Conference Publications
[1] GALE, L.R., Y. CHEN, B.A. HILLS, and R.W. CRAWFORD (2005):
'Boundary Lubrication of Synovial Joints: Characterisation of the Lubricant', 12th
International Conference on Biomedical Engineering ICBME 2005. Singapore,
2005, pp. 3A4-14.

[2] GALE, L.R., B.A. HILLS, R.W. CRAWFORD, and J. KLAWITTER (2005):
'Tribological Evaluation of Pyrolytic Carbon and Surface Active Phospholipid for
Artificial Joints', 12th International Conference on Biomedical Engineering
ICBME 2005. Singapore, 2005, pp. 3A4-12.

x

xi
Table of Contents
Abstract ......................................................................................................................... iii
List of Abbreviations.......................................................................................................v
Glossary of Terms ........................................................................................................ vii
List of Publications and Manuscripts .............................................................................ix
Table of Contents ...........................................................................................................xi
Statement of Original Authorship .................................................................................xv
Acknowledgements ......................................................................................................xvi
List of Figures ............................................................................................................ xvii
List of Tables.................................................................................................................xx
Chapter 1 Introduction 1
1.1 Description of Scientific Problems Investigated.......................................................1
1.2 Overall Objectives of the Study ................................................................................3
1.3 Specific Aims of the study ........................................................................................4
1.4 Account of Scientific Contribution Linking the Scientific Papers............................5
Chapter 2 Literature Review - Biotribology 7
2.1 Tribological studies of the performance of natural synovial joints.........................10
2.1.1 Overview - Lubrication of the Diarthrodial Joint: Search for the
Lubricating Factor.......................................................................................................12
2.2 Tribological aspects of prostheses...........................................................................16
2.3 Summary .................................................................................................................18
Chapter 3 Literature Review Anatomy & Physiology of Diathrodial Joints 19
3.1 Anatomy of the Synovial Joint ................................................................................19
3.2 Articular (Hyaline) Cartilage...................................................................................21
3.2.1 The Articular Surface.........................................................................................22
3.2.2 Articular Cartilage Matrix .................................................................................23
3.2.3 Response to load................................................................................................27
3.3 Synovial Membrane ................................................................................................28
3.3.1 The Synoviocytes...............................................................................................29
3.3.2 Removal of Substances from the Joint Space....................................................30

xii
3.3.3 The Synovium in Disease ................................................................................. 30
3.4 Synovial Fluid......................................................................................................... 30
3.4.1 Synovial Fluid Composition ............................................................................. 31
3.4.2 Production of Synovial Fluid ............................................................................ 34
3.4.3 Functions of Synovial Fluid.............................................................................. 35
3.4.4 Rheology of Synovial Fluid .............................................................................. 36
3.4.5 Synovial Fluid Lipids........................................................................................ 37
3.4.6 Synovial Fluid in Disease ................................................................................. 38
3.5 Osteoarthritis........................................................................................................... 38
3.5.1 Treatments of Osteoarthritis.............................................................................. 41
3.6 Total Artificial Joint Replacement.......................................................................... 43
3.6.1 Artificial joint failure ........................................................................................ 45
3.6.2 Biomaterials ...................................................................................................... 46
3.6.2.1 Pyrolytic Carbon ..................................................................................... 47
3.7 Summary................................................................................................................. 53
Chapter 4 Literature Review Lubrication of Joints 55
4.1 Physical Science of Lubrication, Friction and Wear .............................................. 59
4.1.1 Fluid-Film Lubrication...................................................................................... 65
4.1.2 Boundary Lubrication ....................................................................................... 68
4.1.3 Mixed Lubrication............................................................................................. 71
4.1.4 Wear .................................................................................................................. 71
4.2 Natural Joint Lubrication: A Review...................................................................... 73
4.2.1 Experimental techniques and apparatus used to determine the lubrication
of joints ...................................................................................................................... 75
4.2.2 Fluid-Film Models ............................................................................................ 80
4.2.2.1 Hydrodynamic Lubrication ..................................................................... 80
4.2.2.2 Weeping Lubrication............................................................................... 81
4.2.2.3 Elastohydrodynamic Lubrication............................................................ 83
4.2.3 Mixed Lubrication Models................................................................................ 84
4.2.3.1 Osmotic Lubrication................................................................................ 86
4.2.3.2 Squeeze-Film Lubrication....................................................................... 86
4.2.3.3 Boosted lubrication ................................................................................. 88
4.2.4 Boundary Lubrication Models .......................................................................... 89

xiii
4.2.5 The Search for the Boundary Lubricant ............................................................90
4.2.5.1 Enzyme Studies........................................................................................93
4.2.5.2 Lubricating Glycoprotein.........................................................................94
4.2.5.3 Lipids .......................................................................................................95
4.2.5.4 Surface-active Phospholipid ....................................................................97
4.3 Artificial Joint Lubrication: A Review....................................................................98
4.3.1 Fluid Film Models ...........................................................................................101
4.3.2 Boundary Lubrication Models.........................................................................102
4.3.3 Mixed Lubrication Models ..............................................................................103
4.3.4 Tribological Studies for total joint replacements.............................................104
4.4 Summary ...............................................................................................................107
Chapter 5 Literature Review Boundary Lubrication for Artificial Joints 109
5.1 Surface Chemistry.................................................................................................111
5.1.1 Surfaces and Surface Energy...........................................................................111
5.1.2 Hydrophobic vs Hydrophilic ...........................................................................112
5.1.3 Surface Tension and its Measurement .............................................................113
5.1.4 The Young Equation........................................................................................113
5.1.5 The Contact Angle...........................................................................................115
5.1.6 Surfactants .......................................................................................................115
5.1.7 Electrical Charge..............................................................................................116
5.1.8 Adsorption .......................................................................................................116
5.2 Tribochemistry ......................................................................................................117
5.3 Surfactants for Boundary Lubrication...................................................................118
5.3.1 Lubrication via Surfactants..............................................................................119
5.3.2 Biological Surfactants......................................................................................120
5.3.3 Types of Lipids ................................................................................................122
5.3.4 Phospholipid Analysis .....................................................................................126
5.3.5 Adsorption in Biology .....................................................................................126
5.3.6 Biological Surfactants and Lubrication ...........................................................127
5.4 Boundary Lubrication via SAPL...........................................................................128
5.4.1 SAPL and Wear in artificial joints...................................................................130
5.5 Summary ...............................................................................................................131

xiv
Chapter 6 Scientific Paper I - The Role of SAPL as a Boundary Lubricant in
Prosthetic Joints 133
Chapter 7 Scientific Paper II Boundary lubrication of Pyrolytic Carbon with
Surface Active Phospholipids: Tribological Assessment for
Artificial Joints 141
Chapter 8 Scientific Paper III Boundary lubrication of joints:
Characterisation of Surface-Active Phospholipids found on
retrieved implants 149
Chapter 9 Scientific Paper IV - Tribological Testing of Saturated and
Unsaturated Surface Active Phospholipids: Implications for
artificial joints 157
Chapter 10 General Discussion 165
10.1 Conclusions......................................................................................................... 173
10.2 Future Work........................................................................................................ 174
References 177

xv
Statement of Original Authorship
The work contained in this thesis has not been previously submitted for a
degree or diploma at any other tertiary education institution. To the best of my
knowledge this report contains no material previously published or written by
another person except where due reference is made.

Signed:

Date: 24
th
Sept 2007

xvi
Acknowledgements
I would like to acknowledge my supervisors Professor Doug Hargreaves, Professor
Ross Crawford and Professor Brian Hills. Doug, thank you for your faith and trust
in me. Ross, thank you for your financial support and exposure to the orthopaedic
world. Brian, rest in peace.

A special thank you to my fellow scholars. Your guidance, help and support has
been highly valued.

A big thank you to my friends and family that actually understood and appreciated
what I was doing. Your interest and enthusiasm provided the essential motivation
required along the way.

Most importantly I am indebted to my wife Michelle for her endless patience and
support throughout. Without her love, care and encouragement the journey would
have never been possible.

My daughter Portia has been the best part of this journey, providing me with
continual entertainment and love that only a child knows how.

xvii
List of Figures
Figure 3.1. Diagram of the structure of the knee. Source: (MowHayes)
Figure 3.2: Schematic of a proteoglycan molecule and schematic of a proteoglycan
aggregate (BaderLee)
Figure 3.3: Schematic of the Hypothetical Layers of Articular Cartilage
(BlackHastings)
Figure 3.4. Various forms of PyC indicating surface finish. (Left to Right) As-
deposited, machined and polished.
Figure 4.1. Amontons' Laws of Friction. Source: (Shi)
Figure 4.2. Range of coefficients of kinetic friction reported in the literature for the
mammalian joint are depicted over a physiological range of sliding velocities and
compared with the a modified classical Stribeck diagram. Source: (Hills)
Figure 4.3. Lubrication regimes Source: (Dowson,Wrightet al April)
Figure 4.4. Molecular Structure of Common Solid Lubricants (a) graphite, and (b)
molybdenum disulfide. (Erdemir 2001)
Figure 4.5: Hydrodynamic Lubrication; Diagram showing the formation of the
pressure generated due to a wedge of fluid that separates the moving bearing
surfaces. (Dinnar)
Figure 4.6: Weeping or Hydrostatic Lubrication; a) as load is applied fluid
flows towards the rubbing surfaces at high pressure, carrying the load with minimal
friction. b) when the load is removed the cartilage expands, drawing in synovial
fluid. (Adapted from (McCutchen))
Figure 4.7: Elastohydrodynamic Lubrication; The top surface in this diagram is
deformable, providing a larger fluid film when load is applied.(Adapted from
(Dowson,Wrightet al April))
Figure 4.8: Microelastohydrodynamic Lubrication; Diagram showing the
deformation of the asperities on the surface of the cartilage under load decreasing
the risk of contact and allowing maintenance of a thinner fluid-film. (Adapted from
(DowsonJin))
Figure 4.9. Mixed lubrication showing that one lubrication regime does not answer
the operating conditions in the joint but a combination of mechanisms. Source:
(PanjabiWhite)

xviii
Figure 4.10: Mixed Lubrication; When the fluid film fails, friction is prevented
by boundary lubrication. (Adapted from (Dowson,Wrightet al April))
Figure 4.11: Squeeze Film Lubrication; The arrows indicate the movement of
fluid away from the load-bearing region leaving an enriched film of synovial fluid
between the surfaces. (Adapted from (Hou,Mowet al March))
Figure 4.12: Boosted Lubrication; a) path of the fluid flow into the cartilage
surfaces while loaded. b) schematic diagram of the pools of enriched synovial fluid
formed on the surface of the cartilage. (Adapted from (McCutchen)
Figure 4.13: Boundary Lubrication; The surface layer prevents the articulating
surfaces from coming into contact.(Adapted from (WrightDowson February))
Figure 4.14. Dependence of the efficiency on the friction coefficient in natural and
artificial joints. Source: (Gavrjushenko)
Figure 4.15. Geometric configurations of various tribometers. Source:
(Dumbleton)
Figure 5.1. The triple point in cross section. Depicting the balance of forces at the
edge of a droplet where the liquid, solid and air all meet to subtend a contact angle
().
Figure 5.2. General structure of phosphoglycerides, emphasizing their amphipathic
nature (Schwarz). Various groups for X are given in Figure 5.3.
Figure 5.3. Various polar head groups for the general phosphoglyceride depicted in
Figure 5.2. Source: (http://en.wikipedia.org/wiki/Membrane_lipids)
Figure 5.4. A molecular model for the adsorption of phospholipid zwitterions to a
negatively charged surface in which cations in the plane of the phosphate ions pull
those ions together, thus enhancing close packing of both polar and non polar
moieties and imparting coherency. (Hills)

Figures within Submitted Papers

Chapter 6
Figure 1: Phospholipid model. This model shows the basic mechanism of
phospholipid adsorption to the cartilage surface, rendering it more hydrophobic,
and how interspersed cations pull the phosphate molecules together enabling high
cohesion (adapted from Hills, 2000).

xix
Figure 2: Oligolamellar structure of a common solid lubricant graphite.
Figure 3: Hounsfield test rig set-up showing horizontal position and custom made
attachments.
Figure 4: Schematic of Hounsfield set-up: (1) Hounsfield control panel and drive;
(2) plate head; (3) UHMWPE pin in pin holder; (4) stainless steel plate; (5) heating
resistors and thermocouple; (6) temperature control unit; and (7) force transducer.
Figure 5: Coeff. of friction at ambient UHMW PE/SS.
Figure 6: Coeff. of friction at 371 UHMW PE/SS.
Figure 7: Coeff. of friction at ambient UHMW PE/PyC.

Chapter 7
Figure 1: Schematic of pin-on-flat tribometer: 1 - control panel and drive, 2 - plate
head, 3 - UHMWPE pin in pin holder, 4 - stainless steel plate, 5 - heating resistors
and thermocouple, 6 - temperature control unit, 7 - force transducer.
Figure 2: Coefficient of friction for the material combinations under the three
lubrication conditions. Dark columns represent dry conditions, light grey columns
represent saline lubrication and light columns represent DPPC lubrication.

Chapter 8
Figure 1: Average proportions of PCs (%). Total average PC profile of the 40
implants analysed (all components included). Error bars represent standard
deviation.

Chapter 9
Figure 1: Friction force exhibited by different surfactants (0.2% concentration)
Figure 2: Effect of Concentration on the Friction Force in USPCs
Figure 3: Comparison of the Behaviour of different combinations of surfactant
species.

xx
List of Tables
Table 4.1. Boundary lubricants within SF suitable for tribo tests. Source: (Brown
& Clarke 2006)

Tables within Submitted Papers

Chapter 7
Table 1: Surface roughness of Samples (Ra roughness average)
Table 2: Adsorption of DPPC to PyC

Chapter 8
Table 1: Profile of phosphatidylcholine species detected on retrieved implants: (a)
Polymer components (b) Metallic components

Chapter 1: Introduction
1
Chapter 1
Introduction
This thesis seeks to contribute to solving the problems of inadequate artificial joint
design from a tribological perspective. Artificial joints have a limited life time and
this has been traced to wear related issues. By understanding the methods of
lubrication in joints, in particular boundary lubrication which is the dominant
regime in artificial joints, engineers will be better suited to developing longer
lasting implants.

1.1 Description of Scientific Problems Investigated
The Bone & Joint Decade (2000-2010) has been dedicated to improving the
quality of life for the millions of sufferers of bone and joint disorders. This thesis
represents a contribution to this ongoing effort. The major offender of bone and
joint disorders is arthritis. Osteoarthritis (OA) has emerged to be one of the most
serious and costly health problems encountered in the last century. Simply OA is a
massive problem. OA accounts for more than half of all chronic conditions in the
elderly and statistics show that 85% of the population will suffer from OA in their
lifetime (American Academy of Orthopaedic Surgeons 2002). Any light that can be
shed on this debilitating disease will be most beneficial to the world at large.

Total joint replacement (TJR) offers a partial solution to this problem but has
problems of its own. At best it is a temporary solution to a much larger problem.
Hip and knee replacement surgery has become a common procedure in recent years
as a method of eliminating pain and discomfort and to improve joint functionality
for patients with end stage arthritis in their lower extremities (Scmalzried &
Callaghan 1999). Although a very successful operation, TJR does not offer the
same performance as the natural joint and suffers from a limited lifetime.
Previously, joint replacement was reserved for the elderly. However, due to the
success of the procedure it is increasingly used in younger individuals. This,
2
combined with an ageing population, has resulted in an increase in the incidence of
primary joint replacement. The rate of revision surgery is also increasing. Revision
surgery is associated with increased morbidity and mortality and has a far less
successful outcome than primary joint replacement. As such, it is essential to
ensure that everything possible is done to limit the rate of revision surgery
(Australian Orthopaedic Association 2002). Because we are living to older ages we
are essentially outliving the lifetime of current joint replacement designs. Research
shows that current implants are surviving to the 15 year mark (Charnley 1982;
Donnelly 1997; Kobayashi 1997) but much beyond that is questionable.

The main problem with TJRs is that they fail.
Past experience indicates that the survival rate of the implant will be influenced by
the micro and macro geometry as well as the material properties of the implant
(Donnelly 1997; Kobayashi 1997; Sumner 1998; Simmons, Shaker et al. 2001).
The reasons for failure of current hip and knee joint replacements are, in order of
proportion; loosening, dislocation and wear (Australian Orthopaedic Association
2002). Dislocation is beyond the scope of this thesis but may lie with a better
education of surgeons and the use of computer guided surgery. With dislocation
aside the other two modes of failure, loosening & wear, account for more than half
of the revision procedures and may be related. Loosening may be caused by the
osteolysis of the bone around the implant which is thought to be due to the bodys
response to foreign wear particles produced in the artificial joint. The loss of
material at the bearing surface not only causes the implant to be worn away but it is
this very wear debris which may cause failure in TJR due to loosening. Essentially
many TJRs are failing because of a tribological problem. Any reduction in wear
will be beneficial to the lifetime of the implant. Wear is reduced by lubrication.
There are considerable political, economic, social and technical reasons to improve
the wear performance of biomaterials used in joint replacements.

The tribology of TJRs is not well understood. Boundary lubrication is the last
defence in lubrication engineering. The conditions suited to boundary lubrication
are low relative surface velocities, like reciprocating motion and high loads which
are conditions matched by the human joint. Boundary lubrication can only occur if
3
there is in fact a boundary lubricant present; if not, a dry bearing will exist and
direct contact will occur leading to high wear. The human synovial joint is not a
dry bearing by any means, even in a diseased, arthritic state. In fact the joint is
filled with synovial fluid, a liquid made mostly of water. It is obvious that this is
natures provision for a lubricant, yet it is known in lubrication engineering that
water is a poor lubricant. Even more so, boundary lubrication dictates the
requirement of some surface binding substance that can adsorb to the bearing
surfaces and provide a protecting film. So what else is in synovial fluid that can be
utilised as a lubricant? Synovial fluid also contains small amounts of other
substances. As yet no consolidation has arisen as to the component of synovial
fluid which provides the effective boundary lubrication that is known to exist. The
problem with artificial joints is that they rely upon boundary lubrication; however,
the boundary lubricant has not yet been completely identified. In order to improve
the lubrication of artificial joints the lubricating component of synovial fluid must
first be completely identified and understood.

In summary, TJR is currently not an entirely sufficient solution to OA. Better
artificial joint design should be instituted which means understanding how joints
are lubricated and designing to suit using materials that complement the boundary
lubricant.

1.2 Overall Objectives of the Study
The overall objective of this research included understanding and defining the
tribological aspects of the artificial synovial joint and in particular, the importance
of the role of boundary lubrication. The identification of the boundary lubricant in
synovial joints was essential to this objective. This knowledge may then be used to
increase our understanding of the relationship between the boundary lubricant and
prosthetic materials, both common and novel. Extending our understanding of
artificial synovial joints and their tribological nature may indeed provide an insight
to the crippling disease, osteoarthritis, and, at the very least, provide a better basis
for joint replacement design.
4
1.3 Specific Aims of the study
Specifically, the aims of the thesis were to:

1) Determine the tribological performance of common prosthetic materials
lubricated in vitro by Surface Active Phospholipid (SAPL), which has been
implicated as the boundary lubricant in the joint.

2) Determine the tribological performance of a novel load-bearing prosthetic
material: Pyrolytic Carbon (PyC).

3) Show evidence of SAPL on the surface of retrieved knee implants.

4) Use High Performance Liquid Chromatography (HPLC) to identify the
composition of SAPL found on the surface of retrieved hip and knee
implants.

5) Provide evidence of boundary lubrication in TJR by measuring the
frictional performance of retrieved implants ex vivo.

6) Use the profile of SAPL identified via HPLC to develop a synthetic
boundary lubricant suitable for laboratory testing.

7) Determine the tribological performance of the synthesised test lubricant and
common prosthetic materials.

8) Provide recommendations for the future of artificial joint design.

9) Provide recommendations for the implementation of a standardised test
lubricant suitable for in vitro laboratory testing for the purpose of
evaluating artificial implants.

Aims 2-7 make up the original features of this thesis and to the best of the authors
knowledge have not been investigated else where.
5

1.4 Account of Scientific Contribution Linking the
Scientific Papers
This thesis begins with a literature review of biotribology at large. Chapter 2 is an
overview of the science of biotribology with a focus on the lubrication within the
human body. Subsequent chapters 3-5 give an increasing in detail review to the
area of biotribology of interest. Namely, Chapter 3 reviews the structure and
function of the human diarthrodial joint, its failure due to OA and the current
remedy of Total Joint Replacement and its failure. Chapter 4 is an overview of the
lubrication of human joints. Chapter 5 is an in depth look at boundary lubrication.
Chapters 6-9 are scientific papers that have been written by the thesis author
reporting on research into the lubrication of artificial joint materials.

Chapter 6 is the first scientific paper published and was an initial study to
determine the tribological interaction of a synthetic SAPL, dipamitoyl
phosphatidylcholine (DPPC) and a novel load bearing prosthetic material, PyC.
This study was performed to screen a new candidate material for hip and knee
implants with respect to producing a low friction bearing surface and to further
support the notion that effective boundary lubrication exists between SAPL and
prosthetic materials.

Chapter 7 is the second scientific paper submitted for publication and is a more
advanced study that extends the previous work with PyC given the encouraging
results produced in the preliminary study (Chapter 6). This study was a tribological
assay of several types of PyC lubricated with DPPC. In addition a goniometer was
used to determine the interaction of the SAPL with the PyC surface. Adsorption
tests were performed to establish the tenacity of DPPC to the PyC surface. This
study revealed the frictional performance of many forms of PyC and confirmed the
ability of SAPL to act as a boundary lubricant on artificial surfaces.

Previous work by our research group (Purbach, Hills et al. 2002) established the
presence of SAPL on the surface of retrieved hip implants. Ongoing data collection
6
from the implant retrieval studies generated sufficient data so that a further report
could be published. This report, the third scientific paper of the thesis forms
Chapter 8. This study fully characterised the SAPL found on the surface of
retrieved hip and knee implants by HPLC. This knowledge would allow for a more
accurate definition of the boundary lubricant present on the surface of artificial
joints and support for the boundary lubrication of artificial joints. This study
produced a profile of the constituents of SAPL that revealed that DPPC was not in
fact the dominant portion suggesting that future friction testing of artificial joint
materials should utilise a lubricant similar to the profile of SAPL detected on the
artificial joint surface.

Considering the information obtained in the study described in Chapter 8 it
followed that the next study should test again the common and novel artificial joint
materials for their frictional performance using a synthetic copy of the determined
SAPL profile. The aim of this study was to compare the work done previously
using only DPPC as the lubricant to the results achieved using the more accurate
definition of the boundary lubricant. This paper forms Chapter 9 of the thesis.

The four scientific papers (Chapters 6-9) are the thesis authors original
contribution to this field of research.

The thesis concludes with a general discussion of the outcomes and a summary of
the four papers presented for examination. Nature has provided an effective
boundary lubricant in the form of SAPL as found on retrieved implant surfaces.
Boundary lubrication is instrumental to the reduction of friction between prosthetic
joint materials. Future artificial joint design should incorporate these parameters in
order to improve the currently unsatisfactory lifetime of TJRs. Novel materials,
such as PyC, can interact favourably with SAPL and suggest a tribologically
satisfactory biomaterial suitable for future artificial implants. Better artificial joint
design may be achievable by means of material selection and surface modification
that can capitalise on the nature of the lubricant present. This thesis also promotes
the use of a standardised lubricant for in vitro testing that is similar to what is
found to lubricate artificial joints.

7
Chapter 2
Literature Review - Biotribology
As the thesis author has a background in Mechanical Engineering the following
literature review is broad in order to provide a sound background to the
biotribology field. Biotribology (BT) is a very large field of research indeed. This
thesis will focus on the application of this science, biotribology, to human artificial
synovial joints. This chapter will outline the general topic of BT and the
subsequent literature review chapters will cover further detail of the thesis topic.

The literature review, Chapters 2-5, includes a discussion of the joints themselves,
the fluid within the joint, the disease osteoarthritis (OA) and the current solution of
artificial joint replacement, the bearing materials, a review of the various modes of
lubrication and a focus on boundary lubrication and SAPLs,

BT is a relatively new term introduced in the early 1970s to describe a group of
sciences that converge on one single topic: the study of friction, wear and
lubrication within biology. In consideration that the invention of this term and the
creation of this field of research would not have occurred if it had not been for the
increasing incidence of OA it is essential that a review be made of the natural
synovial joint itself. This will form part of Chapter 3 which will also include a
review of the degenerative joint disease, OA, the current remedy TJR and the
failure of these replacement joints.

Originally BT was applied to natural synovial joints in an effort to understand the
joints mechanical functions and the mechanisms of failure with the hope of
reducing the effects of OA. Physicians, physiologists, biochemists,
rheumatologists, biologists, rheologists and engineers joined forces in an effort to
understand how the natural joint was lubricated and, more importantly, how the
joint was being compromised by OA. This will form part of Chapter 4. OA has not
been cured and the best remedy to alleviate joint dysfunction has been the
invention of TJRs. Artificial joints are prone to failure and the science of BT has
Chapter 2: Literature Review - Biotribology
8
now turned its focus to the artificial joint in an attempt to improve the lifetime of
the prosthesis. The remainder of Chapter 4 discusses the lubrication of artificial
joints.

It is well known that effective boundary lubrication occurs in artificial joints.
Chapter 5 is an in depth look at boundary lubrication.

Biotribo1ogy is a challenging, multidisciplinary field of research, involving
biology, orthopaedics, biomechanics, biomaterials science and tribology.
Tribology, an area of engineering, is the science that studies the lubrication of
interacting surfaces in relative motion. Friction is the resistance to relative sliding
or rolling motion of the surfaces. Overcoming friction dissipates energy and causes
wear of the surfaces. Lubrication provides an effective means of reducing friction
and wear by separating contacting solids with a thin layer of material of low shear
strength. The purpose of tribological research of prosthetic joints is to minimize
friction and wear of the implant, and thereby to increase the lifetime of the joint
(Calonius 2002). So, tribology plays a major role in the effective treatment of one
of the most common medical conditions known in the western world (Unsworth
1991).

Biotribology is natures way of turning a science into an art, so amazing is the
ability of biology to lubricate its mechanical functions. Engineers may be the only
ones that truly appreciate and marvel at natures eloquent yet complex methods of
lubrication. BT applies the principles of lubrication engineering in an attempt to
understand how the body lubricates its articulating bearings: the synovial joints. A
mechanical bearing analysis is not a complete analysis of the human joint, as it is a
biological bearing where biochemistry and surface chemistry play a very important
role, potentially far more important than the mechanical part (Dowson 1990). This
requires engineers to call upon the expertise of other disciplines to understand
human joint lubrication and to design suitable replacements. As will be seen by the
diversity of this literature review, biotribology requires skills from the engineer that
far extend beyond traditional engineering principles.

9
Novel expressions such as biolubrication have been introduced by a research
group (Benz, Chen et al. 2005) to describe the dynamic properties of very thin
aqueous films between two biological surfaces in relative motion or for water
flowing through pores or between two stationary surfaces; but more generally this
expression also covers related phenomena such as the adhesion, friction,
deformations, damage, and wear of the surfaces or the lubricating fluid. These
nanoscale phenomena ultimately determine the way biofluids flow through narrow
pores or effectively lubricate a joint.

BTs large focus is to increase the lifetime of artificial joints. Historically, new
bearing materials were 'tried out' in patients (Charnley 1966; Fisher 2000). The
consequence of incorrect design and material selection and subsequent failure now
means that there are extensive pre-clinical requirements for evaluation of materials
prior to implantation (Fisher 2000). This is by no means a simple problem.
Environmentally (biochemical) conditions in the body are harsh, biomechanical
requirements are complex and variable, and the biological response to wear
particles is largely unknown and dependent on the genetic profile of the recipient
(Fisher 2000). Biotribology is a highly multidisciplinary subject crossing
engineering materials and physical science, biological science and medicine.
Predicting the mechanical and tribological performance of bearing surfaces, and
understanding the biological responses to wear debris and the resulting potential
clinical outcomes, remains a substantial scientific and technological challenge. Key
factors limiting the successful development of improved products are our limited
understanding of biotribological science, the capability and capacity to simulate in
vivo conditions in the laboratory in pre-clinical tests, and a lack of fundamental
understanding of the complex and heterogenous biological reactions and
biocompatibility of wear debris in the body (Fisher 2000).

Tribology is itself an inter-disciplinary subject, being concerned with ". . . the
science and technology of interacting surfaces in relative motion and the practices
related thereto" (Dowson & Wright 1973). It embraces studies of lubrication,
friction, wear, tribochemistry, rheology and surface chemistry each of which calls
10
upon contributions from chemists, physicists, mathematicians, engineers, materials
scientists, and tribologists.

At a 1970 Conference on Rheology in Medicine and Pharmacy, Dr G. W. Scott
Blair, with some justification, referred to "our somewhat precocious sister-science
of tribology" (Ferguson & Nuki 1973). He commented on the interesting link and
that paper was seen as a small attempt to encourage the courtship between the
disciplines.

Dowson & Wright introduced the term "bio-tribology" to mean those aspects of
tribology concerned with biological systems (Dowson & Wright 1973). There is a
growing interest in the relevance of tribology to biological systems, with some of
the main areas of activity being grouped together as follows:
1. The abrasive wear characteristics of human dental tissues.
2. Fluid transport in the body.
3. Locomotion of micro-organisms
4. The motion and lubrication by plasma of red blood cells in narrow
capillaries.
5. The action of saliva.
6. Tribological studies of the performance of natural synovial joints.
7. Tribological aspects of prostheses
This thesis will explore the areas concerned with joints.

2.1 Tribological studies of the performance of natural
synovial joints
It is true that the largest and most successful activity in the field of biotribology has
been the study of human joints.

The human joint is a remarkable bearing. It has a low coefficient of friction (0.002)
(Jones 1934; Charnley 1959; Hills & Crawford 2003) and it is expected to survive
the dynamic loading associated with the normal activities of life for at least 70
years. The bearing material (articular cartilage) is elastic and porous. It has an
11
initial thickness of a few millimeters and, like conventional plain bearing materials,
it is mounted on a hard backing (bone). The lubricant is synovial fluid; a highly
non-Newtonian fluid which is contained within the joint space by the synovial
membrane. It consists of a dialysate of blood plasma with varying amounts of
protein/mucopolysaccharide (hyaluronic acid complex).

Theoretical and experimental studies have suggested that the joint experiences most
of the lubrication modes familiar to tribologists; hydrodynamic,
elastohydrodynamic, mixed, and boundary, together with a unique squeeze film
characteristic. The range of loading experienced by the joint is considerable and
peak loads of more than ten times body weight can be anticipated in some
activities.

In spite of the remarkable characteristics of healthy human joints, many show signs
of wear and general distress during their working life. Osteoarthritis is a process in
which the articular cartilage is roughened and worn away, with consequent
discomfort and loss of mobility. There appear to be certain similarities between the
wear of some engineering bearings and the development of osteoarthritis, and it is
for this reason that physicians, surgeons, biochemists, rheologists and engineers
have joined forces for an attack on the problem.

If an engineering bearing shows signs of distress it can often be cured by
improving the lubricant. This suggests that it might be possible to influence the rate
of development of osteoarthritis by the introduction of synthetic lubricants. The
main problem is that even if the synthetic lubricant could be introduced and
retained within the joint space, we do not have an adequate understanding of the
normal lubrication mechanism to enable us to write a specification for the synthetic
material. In addition, and maybe more importantly, is that consolidation is still
lacking as for the identification of the lubricant in the natural joint.

A large part of BT has focused on the identification of the lubricant within the joint
that facilitates such an engineering feat.

12
2.1.1 Overview - Lubrication of the Diarthrodial Joint: Search for
the Lubricating Factor
The diarthrodial or synovial joint allows relative motion of the bones. The bone
ends meet within a fibrous enclosure termed the joint capsule. The joint cavity is
filled with a pale yellow, viscous fluid known as synovial fluid. The lubricating
factor that allows the relative motion of the bones is believed to be contained
within the synovial fluid. Daily use of the synovial joints of the lower limbs - the
hips, knees and ankles, involves large ranges of relative motion in multiple
directions experiencing loads often as high as six times the body weight during a
normal walking cycle (Mow & Mak 1987). These loads must be sustained by the
biological bearings, the synovial joints, with characteristics of friction, wear and
lubrication that are the envy of modern engineering science. Cartilage rubbing on
cartilage has extremely low coefficients of friction (), in the range of 0.003- 0.024
(Jones 1934; Charnley 1959; Linn 1968) this is much lower than the values attained
using any synthetic bearing materials in equivalent situations, the best of which is
Teflon (PTFE) rubbing Teflon which gives a value for of 0.04.

The mechanics and biochemistry of synovial joint lubrication has been the subject
of detailed investigation since the early 1900s. Interest in the biomechanics of the
joint is widespread, extending into the fields of medicine and veterinary science
due to the high incidence of osteoarthritis (OA) or degenerative joint disease (DJD)
in todays society and the equine industry. Current literature implicates a direct
mechanical cause in the initiation of OA (Lane & Buckwalter 1993; Felson &
Radin 1994), the most commonly affected joints being those that bear load or those
that are likely to be subjected to acute injury (Meachim & Brooke 1984; Wyn-
Jones 1986; Felson 1990; Panush 1990). Deterioration of these joints causes great
pain and loss of mobility for the sufferer. Two of the major aspects in maintaining
joint mobility is lubrication (Cooke, Dowson et al. 1978) and the general
maintenance of a good load bearing surface (Freeman & Meachim 1974). It may be
that the development of OA follows the compromise of the lubricating system of
the synovial joint. What system could act within the joint to enable the exceptional
properties of friction, lubrication and wear seen at the bearing surface?
13

The major theories of joint lubrication are based upon one or a combination of two
main mechanisms: boundary lubrication mechanism, where there is solid-to-solid
contact; or, a fluid-film mechanism, where the two sliding surfaces are separated by
a fluid-film or wedge of liquid which keeps them from touching. Fluid-film
mechanisms typically reach much lower coefficients of friction than boundary
mechanisms (Williams 2005) but require velocities roughly an order of magnitude
higher than typical joint sliding rates in order to maintain the wedge that separates
the two surfaces. Below this velocity, the two surfaces touch and boundary
lubrication is all that remains to facilitate motion.

Initially, the lubrication qualities of synovial fluid were thought to relate to its
characteristic viscosity, a characteristic imparted by the hyaluronic acid component
of the synovial fluid (Ogston & Stanier 1953). It was believed that the greater the
viscosity of the fluid, the better the lubricity. Hyaluronic acid is a high-molecular-
weight polysaccharide which is present in high concentrations in synovial fluid.
Apart from being responsible for the viscosity of synovial fluid, it is also very
slippery. However, it fails to lubricate under any significant load (McCutchen
1967; Linn & Radin 1968; Radin, Swann et al. 1970; Radin, Paul et al. 1970).
Further studies using hyaluronidase (McCutchen 1966) added to synovial fluid
revealed substantially reduced viscosity following hyaluronic acid destruction, but
the lubricating abilities of the synovial fluid were unchanged. Conversely, tryptic
digestion (Wilkins 1968) left viscosity unchanged but severely compromised the
lubricity of synovial fluid. These studies demonstrated two things:
(1) That the lubricating abilities of synovial fluid are independent of its
viscosity and hence, the hyaluronic acid component.
(2) That the lubricating component of the synovial fluid is somehow
associated with a protein.
These experiments were the beginning of the search for the ingredient of synovial
fluid that has the high-load-bearing capabilities necessary for lubrication of the
lower extremity joints.

14
Studies have also been performed to demonstrate the lubricating advantage of
synovial fluid over saline using a system of fresh cartilage rubbing on glass (Jones
1934; Charnley 1959). It was found that synovial fluid had little advantage over
saline in terms of lubricating ability, at least over a short period of time. Over a
longer period, saline ceased to lubricate and the synovial fluid was clearly superior.
The fact that saline lubricated at all was significant. It indicated that there may be a
lubricant attached to the cartilage surface which required replenishing from the
synovial fluid.

Following a rigorous experimental protocol, it was demonstrated that the
lubricating abilities of synovial fluid were completely recovered in the protein
fraction of synovial fluid as opposed to the hyaluronate fraction (Radin, Swann et
al. 1970). Further refinement of the gross protein fraction showed that the
lubricating ability was located in a glycoprotein fraction that could be separated
from the bulk of the synovial proteins. Since serum proteins did not possess similar
lubricating abilities, the glycoprotein was considered unique and termed a
Lubricating Glycoprotein (LGP) or Lubricin (Swann 1978; Swann, Hendren et al.
1981; Swann, Slayter et al. 1981). Numerous characterisation tests were carried out
in an attempt to identify the glycoprotein; however, only 8790.8% has been fully
characterised (Swann, Slayter et al. 1981; Swann, Bloch et al. 1984). Another
interesting fact is that the molecular weight (220,500) varied with the analyses used
to identify the protein (McCutchen 1966). The remaining unidentified 9.213%
has been shown to be lipidic in nature (Schwarz & Hills 1998) and has been
heralded as the active boundary lubricant in SF and labelled Surface Active
Phospholipid (SAPL). Studies demonstrating that LGP could adsorb or otherwise
bind to the articular cartilage surface have also been performed. These showed that
14% of the LGP molecule could actually adsorb to the cartilage surface (Swann,
Hendren et al. 1981). It would seem rather coincidental that these amounts were
very nearly the same suggesting that LGP (Lubricin) may in fact be a carrier for the
active lubricating ingredient (SAPL) , rather than the lubricant per se (Schwarz &
Hills 1998).

15
Other published works, for example (Tsukamoto, Yamamoto et al. 1983), have
shown that the coefficient of friction does not correlate with the concentration of
the common proteins, globulin and albumin, in synovial fluid. Also there is no
correlation between the coefficient of friction and the concentration of hyaluronic
acid. This means that the lubricating properties of synovial fluid depends upon
other substances (Gavrjushenko 1993). Attention has therefore turned to lipids
which are widely distributed in the body. The lubricating ability of lipids is
attractive because lipids have good solubility in SF and the supply is practically
inexhaustible.

Almost all studies concerning lubrication of the joint have ignored the lipids,
despite the oily nature of the cartilage surface and the presence of lipids in the
synovial fluid in concentrations comparable to that of the polysaccharides and
proteins. One exception to this is the study by (Little, Freeman et al. 1969) who
found that rinsing the cartilage surface with a lipid solvent increased friction, i.e.
the value of by 500%. However, the work involving lipids and their role in joint
lubrication appeared to cease at this point. At least a decade later, Hills reignited
interest in the field by suggesting that the surfactant identified in the lung may have
lubricating abilities in many other locations in the body. More recently interest has
grown in the area with several groups researching the role of lipids in lubrication
(Gavrjushenko 1993; Williams III, Powell et al. 1993; Craig & LaBerge 1994;
Higaki, Murakami et al. 1997; Saikko & Ahlroos 1997; Stachowiak & Podsiadlo
1997; Pickard, Fisher et al. 1998; Ethell, Hodgson et al. 1999; Bell, Tipper et al.
2001; Nitzan 2001; Kawano, Miura et al. 2003; Gale, Chen et al. 2006; Gale,
Coller et al. 2007).

Considering the work of Little et al (1969) in light of the work of Radin, Swann
and their co-workers raises the possibility that a form of lipid is involved in the
lubrication of the articular surface and that the glycoprotein is simply the carrier for
the highly insoluble lubricating component. This was confirmed by work done by
(Schwarz & Hills 1998).

16
The equivalent lubrication system used in industry, i.e. solid to solid rubbing, uses
a monolayer of surfactant. Surfactants readily adsorb to surfaces rendering
hydrophilic surfaces more hydrophobic. Interestingly, the articular surface is
hydrophobic, a property readily demonstrated by measuring the contact angle
occurring when a droplet of saline is placed upon the cartilage surface. If a
component of synovial fluid were a surfactant, a surfactant might actually be
responsible for the lubricity under load of synovial fluid and the articular surface.
Indeed, a major portion of the lipid component of synovial fluid is phospholipid
(Rabinowitz, Gregg et al. 1984). Phospholipids are well known for their surface-
activity and have the capability of readily rendering a surface hydrophobic.
Phospholipids also provide values for equivalent to the very low values obtained
by rubbing cartilage on cartilage (Hills 2000). Moreover, phospholipids are also
recognised as possessing substantial load bearing abilities.

2.2 Tribological aspects of prostheses
When prostheses are introduced into the human body and relative motion of the
components are involved, the rheological and tribological features of the material,
the prosthesis and the bodys fluids become important. Examples include the wear
of some heart valves, fretting corrosion of plates and screws, and the friction and
wear characteristics of human joints.

The hip joint has received the most attention and it has been estimated that there
are over one million operations each year the world over (Bowsher & Shelton
2001). A variety of materials and designs have been witnessed in the developments
which have led to the present position. There are three major forms of material
combinations; metal-on-metal, ceramic-on-ceramic and metal-on-plastic. The first
is tribologically undesirable under most circumstances and engineering bearings in
which contact and sliding occur usually consist of a soft bearing material and a
relatively hard metal. However, in the environment of the body, corrosion might
readily occur if dissimilar metals are used. Ceramic-on-ceramic has had varied
success and seem to exhibit similar tribological failings as the other combinations
(Jazrawi, Kummer et al. 1998). The metal-on-plastic arrangement is currently most
17
popular, the favourite materials being stainless steel and ultra high molecular
weight polyethylene. In the early stages of development of this form of prosthesis,
the low friction plastic polytetrafluoroethylene (PTFE) or commonly called Teflon
was employed, but the results were disastrous, owing to the high rate of wear of the
softer material. The present combination of materials provides some confidence in
the long-term future of the prosthesis but beyond the 15 year mark is still in
question. It is worth noting that although wear-rate is probably the most significant
tribological feature of artificial joints, friction is fundamentally important to wear.
In engineering a reduction in friction in a bearing will nearly always reduce wear
and guarantee a longer life for the bearing but there are exceptions to the rule, at
least in the body, as mentioned above in regards to the use of PTFE. The short- and
long- term reactions of body tissue to wear particles is also important.

The knee joint probably suffers more from osteoarthritis than the hip, and yet the
development of the knee joint may be still some way behind the development of
the hip joint. The knee lacks the basic stability associated with the hip, the
geometry and motion is more complicated, and the stress levels generally higher.
Hinge joints have been used successfully, but the motion is in many ways an
unsatisfactory substitute for the natural condition, and there are a number of
medical objections to the arrangement. Present designs are more in the form of
replacement linings of the natural bearing materials. In prosthetics, the combination
of metal and Ultra High Molecular Weight Polyethylene (UHMWPE) seem to be
favoured at the present time.

The number of biological subjects in which the science of tribology has made a
contribution to the overall understanding of the problem is extensive and
expanding. Many of the examples are concerned with the common ground between
the sciences of rheology (Ferguson & Nuki 1973), tribology (Nakano, Momozono
et al. 2000) and surface chemistry (Benz, Chen et al. 2005).

18
2.3 Summary
The main aim of biotribology is to understand natures treatment of tribology and
use this knowledge to design prosthetic joints with the aim to develop joint
couplings that minimise wear and friction in order to improve the long-term
performance of these prostheses. The lubrication of joints is complex as is the role
of the lubricating factors in synovial fluid and both will receive further discussion
in the subsequent chapters. Three substances have been implicated as the
indigenous lubricating portion of synovial fluid: Hyaluronic acid, Lubricin and
Phospholipids. It has been shown that HA fails to lubricate under any significant
load. Lubricin is a protein and never before has a protein been shown to lubricate.
Lipids in particular phospholipids are known for their lubricating abilities.
Interestingly a portion of Lubricin has been identified as phospholipidic in nature
suggesting that phospholipids are indeed the lubricating fraction of synovial fluid.
Hence, the next stage of research into the lubricating component of the joint
environment would seem to be one of testing for the presence of phospholipids at
the bearing surface. Further evidence supporting a role for phospholipid in the
lubrication of the joint would also open a new avenue of research in the
development of an effective artificial synovial fluid for use in both the natural and
artificial joint.

Chapter 3: Literature Review - Anatomy & Physiology of Diathrodial Joints
19
Chapter 3
Literature Review Anatomy &
Physiology of Diathrodial Joints
This chapter includes a discussion of the joints themselves, the fluid within the
joint, the disease osteoarthritis and the current remedy, artificial joint replacement.
It will review the materials used in TJRs and discuss the failure of artificial joints.
It should be noted that in an examination of the literature on synovial joints far
more information is available on the natural joint than its replacement. It is an
essential step in engineering to have a sound understanding of the original and the
reasons for failure of the original before designing a replacement.
3.1 Anatomy of the Synovial Joint
Diarthrodial or synovial joints are found at the articulations of the long bones of the
skeleton (hip, knee, shoulder, fingers etc.). Diarthroses refer to the degree of
movement allowed (function) by the joint: freely movable articulations as opposed
to either amphiarthroses (slightly movable articulations) or synarthroses
(immovable articulations). Synovial refers to the structure of the joint: articular
surfaces covered with hyaline cartilage, connected by ligaments and lined by a
synovial membrane to create a joint cavity filled with synovial fluid (Gray 1918).

Synovial joints allow movement between articulating bones. The loads experienced
by synovial joints are complex and variable, exceeding 100 million cycles within a
lifetime without failure. During walking, for example, joints experience high
loading (five to six times body weight) at low surface velocities during heel strike
and toe off and very low loads at maximum surface velocity during swing phase
(Unsworth 1978). This calls for incredible load bearing capacity combined with an
extremely effective lubrication system.

20
The knee is a diarthrodial or synovial joint and the discussion here will detail the
knee even though the proposed study will also include the hip. Essentially, the
following biotribological review applies equally to the majority of synovial joints
including the hip.

The knee consists of three articulations in one: two condyloid joints (Figure 3.1),
one between each condyle of the femur and the corresponding meniscus and
condyles of the tibia (tibiofemoral joint); and a third between the patella and the
femur (patellofemoral joint). (Figure 3.1)

Figure 3.1. Diagram of the structure of the knee. Source: (Mow & Hayes 1997)

21
The articular capsule surrounds the joint, forming the joint cavity, and is a thin but
strong, fibrous membrane. The inner layer of the capsule is the synovial membrane
(Gray 1918). The menisci or semilunar fibrocartilages are found on the articular
surface of the tibia and improve articulation with the condyles of the femur,
enlarging the joint contact area, hence aiding articular cartilage in load transmission
and distribution. When removed, stress in the subchondral bone can be up to 5.2
times higher than when the menisci are present (Fukuda, Takai et al. 2000). A layer
of articular cartilage 1.5 to 3.5 mm thick (Bader & Lee 2000) lines the femoral,
tibial, and patella articulating surfaces.

3.2 Articular (Hyaline) Cartilage
Articular cartilage covers the articulating bone ends of synovial joints forming a
bearing surface that enables the surfaces to resist compression, transmit and
distribute loads and maintain low frictional resistance and wear. Freeman et al
(Freeman, Swanson et al. 1975) stated that the function of articular cartilage is: to
reduce the stresses present in the articulating bone when load is applied to the joint,
to protect the bones from abrasive wear, and to reduce the friction in the joint.
However, some authors, including Fukuda et al claim that the stress reducing role
of cartilage is only minimal (Fukuda, Takai et al. 2000).

This bearing material has a thickness of between 1 and 5mm, depending upon both
species and location of the joint. It is a porous elastic material with a surface
topography determined largely by the underlying structure of collagen fibres
(Kuettner, Aydelotte et al. 1991). The absence of blood vessels, nerve fibres and
lymphatics as well as basement membranes on either side of the tissue makes adult
articular cartilage unique among connective tissues (Huber, Trattnig et al. 2000). It
is dependent on the diffusion of molecules into the synovial fluid from the well
vascularized synovial membrane for nutrition and the pumping action generated by
the repetitive loading of the joint is essential for sufficient nutrition.

The combination of articular cartilage and synovial fluid provide an almost
friction-free articulation. The biomechanical properties of articular cartilage depend
22
upon the structure of the extracellular matrix, which is composed of collagen fibres
and a well-hydrated ground substance made up of proteoglycans, glycoproteins,
traces of phospholipids and elastin (Kuettner, Aydelotte et al. 1991; Nixon,
Bottomley et al. 1991). The important functional properties of cartilage which
include stiffness, durability and distribution of load, depend on the extracellular
matrix.

3.2.1 The Articular Surface
The articular surface is a thin membrane-like coating of the cartilage matrix. In
electron micrographs it presents as an amorphous, electron-dense layer which at
high magnification shows a particulate and filamentous appearance (Ghadially,
Lalonde et al. 1983) which coats a layer of collagen fibrils. Morphological studies
by Hills have provided visible evidence of oligolamellar phospholipid on the AC
surface (Hills 1989; Hills 1990). Guerra et al reported that the surface of normal
articular cartilage is covered by a discontinuous, mono/multilayered pseudo-
membrane and seems to consist of phospholipids, glycosaminoglycans and proteins
(Guerra, Frizziero et al. 1996). They suggested this membrane-like structure might
have a protecting role in preventing direct contact between the articular cartilage
and toxic agents present in the synovial fluid and/or exert a lubricating effect
within the articular joint. Ballantine also found the same and concluded that a layer
of phospholipids was present on the surface of articular cartilage and that the layer
could clearly be viewed in SEM and OM (Ballantine & Stachowiak 2002). They
suggested that the lipid layer acts as a boundary lubricant and is critically important
to the proper functioning of synovial joints

The surface roughness of the cartilage will be considered briefly from the point of
view of its relevance to lubrication: To the naked eye, the surface appears
glistening, smooth and free from noticeable unevenness, irregularities and
roughness. However, the issue of cartilage surface roughness has been the source of
considerable controversy. Most, if not all observations made with the scanning
electron microscope, whether on cartilage itself (e.g. (Walker, Dowson et al. 1968))
or on cast replicas (e.g. (Dowson, Longfield et al. 1968) show some surface
23
irregularities. Ghadially (1983) explains the presence of surface asperities as
artifacts of tissue preparation; however, there are a number of earlier studies which
describe surface roughness. Davies et al (1962) suggests that the surface is very
smooth, with irregularities in the range of 0.02m. Dowson et al (1968) and Jones
et al (Jones & Walker 1968) found much greater roughness, which increased with
age. The values ranged from about 1m for foetal cartilage to about 2.7 m for
adult cartilage. Further study by Sayles et al reported average roughness between 1
and 6m (Sayles, Thomas et al. 1979). Today, it is accepted that asperities between
2-6m are commonly present (Dowson 1990).

3.2.2 Articular Cartilage Matrix
Cartilage is an anisotropic material consisting of cells (chondrocytes) embedded in
an extracellular matrix consisting of water (60 to 80% by weight), proteoglycans
(PGs), collagen, and some glycoproteins (non-collagenous proteins). The
organisation of these constituents varies with depth from the surface within the
same joint and between joints (Broom 1988). The fact that cartilage is aneural,
avascular, and has slow cell turnover rates means it has minimal reparative abilities
(Caplan, Elyaderani et al. 1997). Even when cartilage appears to repair an
osteochondral defect, the repair tissue often has a significantly lower aggregate
modulus and Poisson's ratio and a higher permeability than the surrounding
cartilage. These changes in properties indicate that cartilage repair tissue may not
be hyaline in nature and, therefore, inadequate for long term function in the joint
(Hale, Rudert et al. 1993).

Chondrocytes
Cartilage is maintained by the chondrocytes, the cellular component of the
cartilage, which account for less than 10% of the total volume of the cartilage.
They are responsible for the production of matrix material, hence for growth and
repair of cartilage tissue. Chondrocytes synthesise all of the basic molecular
components of the extracellular cartilage matrix and maintain the tissue via a
balance between anabolism and catabolism of the appropriate molecules. The
surface layers rely on the synovial fluid and matrix water exchange for chondrocyte
24
metabolism and waste exchange. As the joint ages, the population of chondrocytes
depletes and the mitotic activity of the remaining cells decrease. (Ghadially 1978)

Proteoglycans
PGs are large, electronegative macromolecules found within articular cartilage.
These are embedded within the fibrillar network (Figure 3.2) and give articular
cartilage the ability to undergo reversible deformation. They consist of monomers
formed by a protein core with a large number of glycosaminoglycans (GAGs)
attached in a bottlebrush fashion (Figure 3.2). These GAG side chains are long,
unbranched carbohydrates and are present within the joint mainly as chondroitin
sulphate and keratin sulphate. Proteoglycans are space-filling within the tissue and
show specific interactions with the extracellular cartilage matrix (Comper &
Laurent 1978; Greenwald, Moy et al. 1978; Neame & Barry 1994).

The GAG side chains are linear polymers composed of repeating dimers
(disaccharides). The disaccharide unit contains one or two negatively charged
groups which, when formed into chains of 50-70 dimeric units as is found in a
proteoglycan molecule, represent strongly repelling chains that extend stiffly from
the protein core (Nixon, Bottomley et al. 1991). As highly negatively charged
macromolecules, proteoglycans attract water creating a hydration sheet that gives
the cartilage stiffness and compressibility. Upon loading, the proteoglycan
aggregates are compressed and water is expelled from the tissue thus increasing the
negative charge density which in turn increases the resistance of water flow until
equilibrium between loading forces and swelling pressure is reached. Removal of
the load allows the water to return, with nutrients, and the proteoglycan monomers
swell to their former volume. Swelling is restricted to about 20% of its maximum
by the collagen fibrillar network (Maroudas & Bullough 1968; Maroudas 1976;
Maroudas & Venn 1977; Muller, Pita et al. 1989; Nixon 1991). Damage to the
collagen network restraining the proteoglycan hydration shell allows the cartilage
to swell with water. This is one of the early recognised changes in degenerative
joint disease or osteoarthritis.

25
On one end of the PG is a small hyaluronic acid (HA) binding region. This allows a
large number of PGs to interact with a single HA chain forming PG aggregates
(Figure 3.2). The PG-HA complex is stabilized by link proteins. (Muir 1980)

Figure 3.2: Schematic of a proteoglycan molecule and schematic of a
proteoglycan aggregate (Bader & Lee 2000)

Collagen
Collagen fibres are responsible for the structural element of the matrix, forming a
three dimensional fibrillar network of rope-like molecular aggregates which is
arranged with specific orientation in the various zones related to function of the
bearing surfaces. Trapped in this fibre network are macromolecules formed by
proteoglycan aggregates (fig. 3.2). These consist of proteoglycan sub-units attached
to hyaluronic acid molecules. Collagen provides the cartilage with tensile stiffness
and strength. The property of compressive stiffness is supplied by the
proteoglycans. (Mow, Kuei et al. 1980; Kuettner, Aydelotte et al. 1991; Nixon
1991).

Collagen accounts for approximately half the dry weight of articulate cartilage
(Muir 1980) and about 15 to 20% of the wet weight (Bader & Lee 2000). The
predominant collagen within articular cartilage is type II. Orientation varies
throughout the thickness and over the articular surface.

Collagen type II comprises 90-95% of articular collagen forming cross-banded
fibrils and fibres intertwined throughout the matrix. The remainder is made up of a
combination of at least five other types of collagen (Eyre, Wu et al. 1987; Kuettner,
26
Aydelotte et al. 1991). Collagen fibres show some organisation in the surface
layers of the articular cartilage, tending to be tightly packed and orientated parallel
to the cartilage surface. Those in the deeper layers tend to be roughly perpendicular
but on the whole are reasonably random (Nixon, 1991). The collagen fibrillar
network is essential for maintaining the volume and shape of the tissue. The ability
to provide the cartilage with tensile strength is enhanced by the presence of
crosslinks between the cartilage molecules (Mow, Fithian et al. 1988).

Figure 3.3: Schematic of the Hypothetical Layers of Articular Cartilage
(Black & Hastings 1998)

Structure
Cartilage can be broken down into four hypothetical layers or zones (Glenister
1976) illustrated in Figure 3.3. The superficial or tangential zone (5-20% of full
cartilage thickness) is closest to the joint surface and consists of tightly packed
collagen fibers orientated parallel to the joint surface. The orientation of these
fibers can be determined or mapped using the split-line technique. This entails
pricking the surface of the cartilage with a pin. The cartilage will then split along
the direction of the dominant fiber orientation. The fiber orientations are thought to
be aligned in the direction of the most dominant stress experienced under
physiological loading. (Mow, Fithian et al. 1988) The chondrocytes in the
superficial layer are oval in shape with their long axis parallel to the joint surface.
The superficial layer contains the smallest PG content and the highest water
content. (Meachim & Stockwell 1974)

27
The intermediate or transitional zone makes up about 40-60% of the cartilage
thickness. It contains an interlacing network of collagen fibers which are more
widely spaced and more randomly orientated with respect to the cartilage surface.
The chondrocytes in this zone are spherical and more abundant. This layer has a
higher PG content and lower water content. (Meachim & Stockwell 1974)

The deep or radiate zone is approximately 30% of the cartilage thickness. It has a
tighter meshwork than the intermediate zone with the collagen fibres having a
radial orientation; that is, they run perpendicular to the joint surface (Meachim &
Stockwell 1974). Broom (Broom 1988) has proposed that the fibres are not
straight; rather, each radially aligned fibre has repeating lateral deflections along its
length. This allows neighbouring fibres to overlap or crosslink, forming a three-
dimensional network capable of entrapping the much higher content of PGs. The
chondrocytes in the deep zone are spherical and are arranged in columns. The deep
zone has the lowest water content. (Meachim & Stockwell 1974)

The deep and calcified zone are separated by the tidemark (Fig 3.3). The calcified
zone consists of a few cells imbedded in a matrix containing calcium salt crystals.
(Meachim & Stockwell 1974) The bundles of collagen in this zone are radially
aligned and serve to anchor the cartilage to the underlying subchondral bone.
(Meachim & Stockwell 1974)

3.2.3 Response to load
The cartilage matrix has a sponge-like structure and much of the water within this
structure is free to move when under compression. The effective pore size of the
matrix is only small, in the range of 20A to 65A (McCutchen 1962; Dowson,
Wright et al. 1969), and the permeability is low.

Cartilage is poroelastic or porohyperelastic at low rates of loading and viscoelastic
at high rates of loading and under impact load (McCutchen 1982). Under a rapidly
applied load, cartilage shows an initial elastic response followed by a poroelastic,
time-dependent creep response. This creep response is thought to be dependent on
28
the PG content, but unrelated to the collage content as it represents the gradual loss
of fluid which is governed by the strength of the PGs resistance to water loss.
After the removal of the compressive load, cartilage displays a recovery phase.

The ability of articular cartilage to carry large loads during physiological activity is
highly attributed to its unique structure. The superficial layer of cartilage serves a
strain-limiting function. As a compressive load is applied, a tensile force is formed
over the surface of the cartilage and is resisted by the tangentially aligned collagen
fibres. The intermediate and deep zones are responsible for the resistance of
compressive forces. The entrapment of PGs within the network of collagen fibres
causes the formation of an osmotic pressure due to the PGs affinity for water and
their repulsion between each other. The collagen fibres resist this pressure,
resulting in a stiff, highly structured gel system. As load is applied, the initial
response is elastic, with the load fully carried by the fluid component as the
pressure within the cartilage increases. Once the pressure reaches a maximum, a
pressure gradient is formed and there is a continuous flow of fluid out of the
cartilage. The pressure, which was carried by the fluid component, gradually
transfers to the deforming collagen network. This process represents the time-
dependent behaviour of articular cartilage. As the pressure is transferred, a new
equilibrium is reached and no fluid flow occurs. The load is fully supported by the
collagen network, displaying viscoelastic behaviour. When the load is released, the
pressure equilibrium is disturbed and there is a fluid influx until the fluid osmotic
pressure balances the constraining pressure of the collagen network; that is, the
cartilage deformation is fully recovered. (Freeman & Meachim 1974)

What an engineer must consider here is the flow of basically water through a filter
with a pore size smaller than 65A.

3.3 Synovial Membrane
Synovial membrane (or synovium) lines the inner surface of the joint capsule and
covers all intra-articular structures with the exception of the articular cartilage and
menisci. Its functions are: (1) the production of synovial fluid and (2) the removal
29
of synovial fluid and debris from the joint space. On gross inspection the
membrane is smooth and shiny, but under the light microscope microvilli are
apparent. These folds permit the expansion of the synovial membrane in response
to joint movement or to changes in intra-articular pressure (Jayson & Dixon 1970;
Nitzan, Mahler et al. 1992; Lu, Levick et al. 2005). It is a metabolically active
tissue and, as such, requires adequate nutrition and removal of waste products. The
synovial membrane is comprised of: (1) a lining layer adjacent to the joint space
which is called the synovial intima and (2) a supportive or backing layer called the
sub-synovial tissue or subintimal tissue. On its external surface the subintimal
tissue merges with the fibrous capsule of the joint (Ghadially, Lalonde et al. 1983).

3.3.1 The Synoviocytes
The cells of the synovial intima are known as synoviocytes. There are two main
types: Type A and Type B synovial cells. The two main functions of the
synoviocytes are the removal of debris from the joint space and the synthesis of
glycosaminoglycan and hyaluronic acid (3.4.1.) (Henderson & Pettipher 1985).

The Type B cell is well endowed with rough endoplasmic reticulum- an organelle
prominent in cells that produce a protein rich secretion (Ghadially 1983), therefore
it can be argued that the Type B cells must produce a protein rich secretion, the
composition of which has many possibilities as listed below:
(1) the small amount of protein in the synovial fluid which is firmly bound to
hyaluronic acid
(2) procollagen and collagen (in culture Type B cells have been shown to produce
types I and III) (Fox, Lotz et al. 1989)
(3) collagenase and plasminogen activator (Ghadially 1983)
(4) c2 macroglobulin (Glynn 1977)
(5) Lubricating glycoprotein-1 (Lubricin) (Ghadially 1983)
(6) glycosaminoglycans (hyaluronic acid, chondroitin sulfate and dermatan sulfate)
(Fox, Lotz et al. 1989)
(7) Fibronectin (Scott & Walton 1984)

30
3.3.2 Removal of Substances from the Joint Space
Secretion of the synovial fluid cannot continue indefinitely into a cavity of finite
size, like the joint cavity, without a corresponding removal of material. The
presence of micropinocytotic vesicles and pinocytotic vacuoles in the synovial
intimal cells indicates the uptake of fluid from the joint space (Ghadially 1983).
The absence of a basal lamina and the presence of lymphatics and fenestrated
capillaries must all facilitate exchange from the joint cavity. Clearance of
substances injected intra-articularly has also been studied, clearly demonstrating
the exchange between plasma and the joint space (Levick 1987; 1989; Levick
1995). Studies using the electron microscope have demonstrated the uptake of
particulate matter from the synovial fluid by the synovial intimal cells (Ghadially
1983).

3.3.3 The Synovium in Disease
The changes in structure and function of the synoviocytes associated with repeated
trauma, infection, metabolic imbalance or deposition of immune complexes lead to
the production of synovial fluid with altered characteristics. This may contribute to
the development of degenerative and inflammatory joint disease (Tew 1980;
Walker, Boyd et al. 1991; Rorvik & Grondahl 1995). In osteoarthritis and
rheumatoid arthritis, the turnover of the synoviocytes increases and there is both
hypertrophy and hyperplasia of synoviocytes. There may be ulceration of the
synovial surface, proliferating fibroblasts and other histologically-evident changes
(Henderson & Pettipher 1985; Fox, Lotz et al. 1989).

3.4 Synovial Fluid
Synovial fluid is a clear, pale yellow, highly viscous liquid with non-Newtonian
flow properties which occurs in small quantities in normal joints. It occupies a key
position in the function and well-being of the joint and is responsible for the
lubrication and nutrition of the joint tissues. This is natures provision of a
lubricant in the joint and must be well understood by the engineer if he is to
31
provide a suitable solution to the failure of the synovial joint. The specific volume
and composition of synovial fluid varies from joint to joint even within the same
species and sometimes within the same animal (Smith, Habermann et al. 1979;
Currey, Unsworth et al. 1981). The volume of normal SF in the knee joint is
estimated to be 0.5 to 2 ml (Yehia & Duncan 1975).

3.4.1 Synovial Fluid Composition
By understanding the constituents of SF, engineers can utilise this information in
designing suitable implants for the joint and the selection of biomaterials that will
function best. It is essential to consider the environment the implant will be
exposed to and that the material selected will not cause an adverse reaction with the
body and conversely that the body does not have an adverse reaction on the
implant. More so to understand what constituents will work with the biomaterial
and what constituents wont and vice versa.

Synovial fluid is essentially a dialysate of plasma with the addition of hyaluronic
acid which is locally synthesised by the synovial cells. The total protein
concentration of synovial fluid is lower than that in plasma, although some
individual proteins are present in higher quantities than in plasma (Weinberger &
Simkin 1989; Simkin 1991; Prete, Gurakar-Osborne et al. 1993). The protein
content of SF is about 20 mg/ml (2 %) and the proteins of SF are identical with the
proteins of blood plasma. Several of the synovial fluid constituents are unique to
the synovial fluid and are either synthesised in the synovial membrane or are a
result of catabolism within the articular cartilage and other structures associated
with the joint. The constituents of synovial fluid will be considered as groups of
substances based on their origin:
I. Constituents derived from the blood
II. Substances secreted by the synovial membrane
III. Products derived from the catabolism of the joint.

I. Constituents Derived from the Blood
32
The major constituents of the synovial fluid are derived directly from the blood.
The path for direct fluid exchange between plasma and the synovial fluid comprises
the capillary endothelium and interstitial spaces between synoviocytes (Knight &
Levick 1983). The degree of permeability of the capillary walls is believed to limit
the molecular size of components reaching the synovial fluid. Small molecules
such as ions and non-electrolytes diffuse freely between plasma and synovial fluid
via the large vascular network in the synovium. As a result of this direct exchange,
the concentrations of these solutes in both fluids is almost identical.

Glucose is one of the most important nutritional requirements of the chondrocytes
(Simkin 1993; Levick 1995). The concentration of glucose in synovial fluid is
usually close to that of plasma. Between 3 to 4 hours following a meal, levels have
regularly been found to be higher in the joint space than in the perfusing blood
(Ropes, Muller et al. 1960), leading investigators to suggest that there might be a
specific glucose transfer system. Most small molecules move freely in both
directions between plasma and synovial fluid in accord with simple diffusion
kinetics. Glucose, however, enters the joint more rapidly than would be expected
from its size, but its rate of return to the plasma is unaffected (Simkin 1993).

Proteins do not equilibrate between plasma and synovial fluid (Ropes, Rossmeisl et
al. 1939). Their distribution in synovial fluid is primarily due to capillary selective
permeability and the size of the molecules that can permeate the synovial
membrane. The concentration of total proteins in normal synovial fluid varies from
joint to joint and is approximately one-third of the protein concentration in serum
(Weinberger & Simkin 1989). Synovial fluid contains many of the lower molecular
weight proteins such as albumin, globulin, and seromucin (Currey, Unsworth et al.
1981). Proteins with molecular weights exceeding 160 000, such as fibrinogen,
macroglobulins and lipoproteins, are found in trace amounts only (de Medicis,
Reboux et al. 1976).

Fat soluble solutes can diffuse easily both through and between cell membranes, so
they do not face the same restrictions as the hydrophilic molecules. The entire
surface area of the synovium is available for the diffusion of these molecules.
33
Physiologically, the most important of these are the respiratory gases: oxygen and
carbon dioxide (Simkin 1993). Lipophilic molecules tend to accumulate in fatty
tissues and are eluted slowly by the surrounding interstitial water (Prete, Gurakar-
Osborne et al. 1993).

Normal synovial fluid contains few cells and there are no erythrocytes or platelets
present. Mononuclear cells such as modified macrophages and synovial cells are
the predominant cell type. The average number of cells per mm
3
is reported to be
63 (Currey, Unsworth et al. 1981). The mononuclear phagocytes are probably
present in order to remove wear debris and particulate matter from the synovial
fluid.

II Constituents derived from the Synovial Membrane
Although the source of these constituents has already been discussed, the structure
and role of hyaluronic acid will be covered further:

Hyaluronic Acid
Hyaluronic acid (HA) is the major secretory product from the synovial membrane.
It is a glycosaminoglycan of high molecular weight which confers the
viscoelasticity and shear thinning behaviour of synovial fluid. The viscosity of the
synovial fluid is directly related to the concentration of HA (Ogston & Stanier
1953) which is about 2 to 3 mg/ml in the normal human living knee (Balazs 1968).
In some inflammatory disease states where the volume of synovial fluid is greatly
increased, the viscosity is correspondingly decreased due to the reduction in HA
concentration. It is suggested that the concentration of HA in synovial fluid is
passively equilibrated with its concentrations in the synovial tissue (Henderson,
Revell et al. 1988). HA is rapidly removed from the joint cavity through the
lymphatic pathways. It is then taken up from circulation and metabolised by
endothelial cells in the liver sinusoids and the lymphatic tissue (Fraser & Laurent
1989). The factors controlling HA synthesis and turnover in the synovial space are
still the subject of investigation (Fraser & Laurent 1989; Momberger, Levick et al.
2005).

34
III Constituents derived from the Catabolism of Joint Tissues
The articular cartilage matrix is basically composed of collagen and proteoglycan.
These two components are constantly being turned over throughout the life of the
cartilage (Mankin & Lippiello 1969; Lohmander 1988). The protein core of the
proteoglycans may be cleaved by proteases and as a result, proteoglycan fragments
may be released into the synovial fluid (Sandy, Brown et al. 1978). These are taken
up by the afore-mentioned phagocytic cells. Chondroitin sulphate, a
glycosaminoglycan, is present in small amounts in normal synovial fluid due to the
turnover of the matrix proteoglycans (Swann 1978).

3.4.2 Production of Synovial Fluid
It has already been noted that most of the components of synovial fluid are derived
from the blood, presumably by a process of simple diffusion through the synovial
membrane, without intervention of cellular activity or active transport mechanisms.
However, the presence of hyaluronic acid and proteins, other than those found in
plasma, (and their degradation products) cannot be explained by the above scheme.

Hyaluronic Acid
Studies including light microscope and histochemical work have been used to try to
identify the source of these extra components of the synovial fluid. It would
appear that conclusive evidence supporting the role of Type B cells in the synthesis
of hyaluronic acid is still lacking, although it has been clearly illustrated that the
intact synovium synthesises hyaluronic acid (Yielding, Tomkins et al. 1957; Myers
& Christine 1983).

Protein
Most proteins in synovial fluid are filtered across the synovial membrane directly
from the circulating blood. Normal synovial fluid from the human knee contains
13g of total protein /l00ml (2.9% of wet weight), a large quantity of the protein
being albumin (Rabinowitz, Gregg et al. 1984; Simkin 1993). Between 10% and
30% of the proteins in synovial fluid are loosely associated with hyaluronic acid
(Balazs 1968), these are believed to be derived from the blood. However, 2% of the
35
protein is firmly bound to the hyaluronic acid, is different from plasma protein and
is probably of synovial origin (How, Long et al. 1969).

Another protein apparently unique to the synovial fluid is the lubricating
glycoprotein, proposed to be the boundary lubricant for the cartilage, which
constitutes around 0.5% of the total protein in synovial fluid and is absent in serum
(Swann & Radin 1972). Little is known of the synthesis of this protein, although it
is clear that it is not produced in the articular cartilage. In vitro cultures of cells
from the synovial lining have produced a molecule resembling this glycoprotein
(Ghadially 1983).
3.4.3 Functions of Synovial Fluid
The functions of synovial fluid lie largely in two areas:

I. Nutrition to the avascular cartilage.
II. A role in lubrication of the synovial joint, including load bearing and
shock absorption.

I Nutrition
As previously noted, the articular cartilage is avascular and it must depend largely,
if not entirely, on diffusion from the synovial fluid for nutrition (Maroudas 1976;
O'Hara, Urban et al. 1990). In immature animals, before calcification is complete
and there is closure of the subchondral plate, the deeper parts of the cartilage may
receive some nutrition from blood vessels in the marrow space (McKibbin &
Maroudas 1979; Mankin & Brandt 1984). The transport of nutrients and the
removal of the by-products of metabolism is believed to be assisted by the
movement of fluid in and out of cartilage in response to cyclic loading of the joint
tissues known as pumping (Mankin & Brandt 1984; O'Hara, Urban et al. 1990).

Joint motion is not considered to be important for the supply of nutrients of low
molecular weight, such as oxygen and glucose to the cartilage. However, as the
molecular weight of solutes increases, simple diffusion of the solute becomes more
difficult and the influence of convection becomes more important. Pumping may
36
assist the transport of substances such as enzyme inhibitors or hormone carrier-
proteins through the matrix. It has also been thought that the value of joint motion
is to stir the bulk liquid and reduce the stagnant boundary layer over the cartilage
surface (Maroudas & Bullough 1968).

II. Lubrication
The synovial fluid has a role in maintaining good lubrication between the opposing
cartilage surfaces and also between the soft tissue structures within the joint cavity.
Lubrication of the cartilage surfaces will be dealt with thoroughly in chapter 4 and
will receive no further discussion here. However, soft tissue lubrication warrants
some discussion here. It is generally believed that the hyaluronic acid component
of synovial fluid is responsible for soft tissue lubrication (Radin, Paul et al. 1971;
Swann 1978). In this location, the hyaluronate experiences considerably lower
loads than at the cartilage surface. This enables the polysaccharide to lubricate the
soft tissues, such as the synovium and the tendons, most effectively. Lubrication of
these soft tissues is probably via the boundary mechanism (Radin, Paul et al. 1971;
Swann, Radin et al. 1974). Alternatively Wright et al present evidence for a
hydrodynamic mechanism in the lubrication of the synovial membrane and stress
the importance of the viscous properties of the synovial fluid (Wright & Dowson
1976).

3.4.4 Rheology of Synovial Fluid
When the viscosity of a fluid is independent of the rate of shear to which it is
subjected it is said to be Newtonian. Synovial fluid (SF) is highly non-Newtonian
(McCutchen 1966), as the rate of shear increases, the viscosity decreases greatly
(shear thinning) and, as the rate of shear decreases, the viscosity increases.
Hyaluronic acid is responsible for the viscosity of the SF. More precisely, SF is
thixotropic and the viscosity is time-dependent, resulting in different maximum and
steady-state values. Synovial fluid also exhibits an elastic response, producing non-
zero normal force differences under load (Caygill & West 1969). The rheological
properties of both the cartilage, as mentioned earlier, and the synovial fluid are
responsible for optimal lubrication. The viscoelastic properties of the synovial fluid
37
also play an important role in maintaining continual effective lubrication (Schurz &
Ribitsch 1987). These properties facilitate a high rate of motion, but also lead to a
thin film of fluid at high speeds (Hamerman, Rosenberg et al. 1970) and, as
discussed in section 4.2.1., this presents a number of problems in the development
of lubrication theories for the joint, especially with respect to the size of cartilage
asperities.

The viscosity is little more than twice that of water under the same conditions. The
lubricant is more akin to water than to mineral oil or silicone fluid as far as
viscosity is concerned (Dowson 1990).

3.4.5 Synovial Fluid Lipids
This component of synovial fluid will receive specific attention because (1) its
source in the normal healthy joint is not clear and discussion of this component is
frequently omitted from reviews on synovial fluid composition; and (2) lipids,
specifically the phospholipids of the synovial fluid, have been implicated in the
boundary lubrication of the joint (Hills 1988; 1989; Hills 1990; Gavrjushenko
1993; Williams III, Powell et al. 1993; Craig & LaBerge 1994; Higaki, Murakami
et al. 1997; Saikko & Ahlroos 1997; Pickard, Fisher et al. 1998). As discussed,
synovial fluid is considered to be a filtrate of plasma combined with HA which is
synthesised locally and most biochemical investigations of synovial fluid have
focused on the protein and polysaccharide constituents. Because of the
comparatively low concentrations of lipid components in normal synovial fluid and
the general lack of interest in their clinical significance, (Wise, White et al. 1987)
little attention has been devoted to their discussion.

Lipids constitute 2 mg/ml of the wet weight of synovial fluid of a healthy human
knee joint (Rabinowitz, Gregg et al. 1979). Neutral lipids make up 67.5% of the
total synovial lipids and the remainder is phospholipid (Rabinowitz, Gregg et al.
1984). Studies on the lipid content of synovial fluid (Bole & Peltier 1962; Chung,
Shanahan et al. 1962; Rabinowitz, Gregg et al. 1984) clearly show
dipalmitoylphosphatidylcholine (DPPC) to be the predominant phospholipid
38
present (14.8% of total lipid). Phosphatidylethanolamine, phosphatidylserine,
phosphatidylinositol and sphingomyelin are also present in synovial fluid (Hills &
Butler 1984; Rabinowitz, Gregg et al. 1984). Lipoprotein levels in synovial fluid
average 40% of those found in serum (Small, Cohen et al. 1964).

The source of these synovial lipids in the normal healthy joint is not clear (Prete,
Gurakar-Osborne et al. 1993) as lipids (lipoproteins) appear to be prevented from
freely entering the synovial cavity (Bole & Peltier 1962). Following trauma, such
as a fracture or soft tissue damage, the level of lipid droplets (largely triglycerides)
in the synovial cavity increase dramatically (Rabinowitz, Gregg et al. 1984; Wise,
White et al. 1987). Conversely, the levels of phospholipids decrease considerably
(Rabinowitz, Gregg et al. 1984). If phospholipid is the boundary lubricant (4.2.5.3
& 5.6) for the joint this may have implications for the development of osteoarthritis
or degenerative joint disease following trauma.
3.4.6 Synovial Fluid in Disease
As already mentioned, SF shows a reduction in viscosity with disease and the lipid
concentration is also affected. Some diseases reduce the viscosity of the SF and
eventually it becomes independent of the shear rate. This is caused by the reduced
concentration of the hyaluronic acid in the pathological SF (Yehia & Duncan
1975). However, the lubricating ability does not seem to be greatly affected
(Swann, Bloch et al. 1984). Only 20 out of 180 diseased joints showed a reduction
in lubricating ability. SF from OA joints is far more Newtonian in its
characteristics (Dumbleton 1981). The question is raised in respect to joint fluid
lubrication: in what way is the function and nature of the joint fluid altered in the
presence of disease and is the change a cause or effect of the pathological
condition? Essentially, little work has been carried out on the nature of joint fluid
around artificial joints and the role of the synovial fluid in artificial joints.

3.5 Osteoarthritis
There are more than 100 different types of arthritis, and the cause of most is
unknown. Arthritis causes pain, stiffness, and sometimes swelling in or around
39
joints. Arthritis is a complex collection of diseases that may have different causes
but result in the same outcome. Arthritis is a common disorder in all populations
and often results in disability or activity limitation and eventual loss of mobility
(Felson 1998).

Osteoarthritis, or degenerative joint disease (DJD), is one of the oldest and most
common types of arthritis. Osteoarthritis affects the quality of life of hundreds of
millions of people the world over. Osteoarthritis is a disease of the synovial joint.
The term osteoarthritis is the name this disease is well known by, but it is
somewhat technically and pathologically incorrect. Osteoarthritis refers to a
primarily inflammatory problem, which is not the case. The pathogenesis described
below is the end result of multiple injuries to the joint surface with minor
inflammation to the joint capsule also occurring. The more accurate term for this
disease is osteoarthrosis (OA), implying primarily a degeneration of the joint. This
thesis, however, will continue to use the common term of osteoarthritis or OA.

OA can affect all synovial joints of the body either singly or in combination but
most commonly affects the load bearing joints of hips and knees. It is
characterized by the breakdown of the articular cartilage and the eventual loss of
articular cartilage from all or a portion of the joint surface. The disease not only
affects the cartilage, but also has ruinous influence on the subchondral bone, joint
capsule, synovial membrane, and even surrounding ligaments and muscle. (Flores
& Hochberg 1998). The cartilage breakdown causes bones to rub against each
other, causing pain and loss of movement. As the cartilage continues to degenerate,
changes occur in the underlying bone, which becomes thickened with the formation
of bony growths from the bone surface called bone spurs. Fluid-filled cysts may
form in the bone near the joint. Bits of bone or cartilage can break off into the joint
space and irritate soft tissues, such as muscles, and cause problems with movement.
Much of the pain is a result of the increased load on muscles and the other tissues
that help joints move, which is a result of the damage to the cartilage. Cartilage
itself does not have nerve cells, and therefore cannot sense pain (inflammatory), but
the muscles, tendons, ligaments and bones do. The synovium becomes inflamed as
a result of the arthritic cartilage. This inflammation leads to the production of
40
cytokines (inflammatory proteins) and enzymes that may damage the cartilage
further.

While it is widely accepted that end stage OA is represented by full thickness loss
of articular cartilage, early OA is much more ambiguous. Osteoarthritic
degeneration progresses through four phases as the cartilage damage becomes more
severe; these grades of degradation are explained by Marcinko (Marcinko &
Dollard 1986). In Grade I, surface and subsurface damage are minor, and limited to
small fissures and pits. The damage can be observed only at points of highest stress
and the rest of the joint functions normally. In Grade II, more severe cartilage
damage can be seen, though the damage is still confined to the areas of greatest
loading. Some cartilage loss can occur in this stage. Grade III of the degradation
marks the complete loss of cartilage in heavily loaded areas and possibly the
formation of bony growths. Pain in the joint would typically begin during this
stage. Grade IV is the most severe level of degradation; in this end stage, large
areas of bone may be completely exposed. The surfaces of the bone can become
misshapen and the articular surfaces become irregular (Marcinko & Dollard 1986).

The causes of osteoarthritis have been subject to a number of theories. Several
theories are related to trauma in the joint resulting from injury (Panush 1990;
Buckwalter & Lane 1997; Jay, Elsaid et al. 2004). Other theories suggest that OA
is just a wear and tear form of arthritis. Yet other theories point to genetic factors
involved in OA that can predispose the patient to or produce osteoarthritic lesions
(Neame, Muir et al. 2004). Risk factors for OA are many and varied, single or
multiple, and include: age, sex, obesity, height, genetic influences, socio-cultural
influences (nutrition, sitting habits etc.), exercise levels, joint injury, existing
diseases (for example, osteoporosis), congenital factors etc. (Felson 1998; Wildner
& Sangha 2000)

It has not yet been determined exactly which mechanism causes OA or what is the
initial trigger. It is likely that there are multiple pathways. Continuing efforts of
research are being made to gather more information about the early stages of OA
and the chain of events that occur during the progression of the disease.
41

From an engineering perspective, when considering the failure of bearings, with
failure due to overload (i.e. trauma) aside, nearly all bearing failures can be traced
to a lubrication deficiency. Indeed, one of the models proposed for the pathogenesis
of OA is that the joint fails because of a lubrication related issue (Davis, Lee et al.
1978; Marcinko & Dollard 1986; Batchelor & Stachowiak 1996).

3.5.1 Treatments of Osteoarthritis
Few treatments exist for osteoarthritis patients. Part of the difficulty in treating
OA results from the difficulty in diagnosing the disease early; the pain and stiffness
associated with later stages (Grades III and IV) of the disease are not generally
evident until much cartilage damage has occurred. Attempts to slow the
progression of osteoarthritis have been made using nutritional and pharmaceutical
supplements, physical therapy, and other approaches. Some common treatments are
weight control, pain-control medicine, and hot/cold therapy. These treatments have
been shown to reduce pain, but do little to stop the progression of the disease and
cannot reverse its effects (Furey & Burkhardt 1997).

Weight loss and physical therapy are generally prescribed to treat mild OA.
Traditionally more advanced OA is treated with non-steroidal anti-inflammatory
drugs (NSAIDS) until the drugs no longer provide relief from pain at which point
the treatment usually progresses to surgery. While NSAIDS are the most
commonly prescribed medications for OA, they do not seem to alter the natural
course of the disease and they produce significant gastrointestinal side effects.

There are a growing number of non-traditional approaches being used throughout
the world. In Europe and Asia, glucosamine and chondroitin sulfate (GCS) have
long been used to treat arthritis. Glucosamine is a constituent of
glycosaminoglycans (GAGs) and proteoglycans that are naturally found in cartilage
and synovium. It is also a building block for hyaluronic acid, which forms the
backbone of the proteoglycans embedded in the articular cartilage. Chondroitin 4-
and 6-sulfates are the main GAGs of articular cartilage (Buckwalter 1983). In
42
theory, by providing the building blocks for healthy articular cartilage, the joint
will be able to repair and regenerate cartilage, but this ability of chondroprotection
is unproven. It is worth noting that there is no single treatment that works for the
entire affected population but that certain treatments do seem to help some
individuals.

Another non-surgical treatment for OA is intraarticular injections of hyaluronic
acid (HA) known as viscosupplementation (VS). Balazs (Balazs 1968) has
advocated the use of artificial synovial fluid since the 1960s and is truly a pioneer
in the field. VS is a procedure in which a hyaluronic acid derivative is injected
directly into the affected joint. The goal of viscosupplementation is to recreate the
environment of a healthy joint in an effort to encourage the joint to escape the
negative feedback cycle of OA progression. Once normal viscoelastic properties
are returned to the joint fluid, the joint will increase production of healthy synovial
fluid. Initially the mode of action is mainly mechanical, but it is suspected that
additional biologic benefits are eventually levied. The half-life of the injected fluid
is only a few days. Therefore after one week there is no longer a significant amount
of artificial fluid in the joint. However, clinical studies have shown that the benefits
of viscosupplementation can last from 6 months up to a year after the last injection
(Wobig, Dickhut et al. 1998). It is believed that because of the transient nature of
the HA within the joint, it is not actually the restoration of joint viscosity that is the
cause of the noted changes (Brandt, Smith et al. 2000). More likely, the benefits
observed in VS are due to interactions on the molecular level between the HA and
pain receptors in the joint. However, since the cause for the initial onset of arthritis
is not addressed (subchondral sclerosis, varus/valgus deformity, obesity etc.) the
benefits of viscosupplementation are not and may never be long term. From an
engineering perspective the idea is good; renewal of the lubricant will increase the
lifetime of the component. However without fully understanding the nature of the
bearing and how it is lubricated it is foolhardy to suggest a suitable replacement
lubricant.

When these treatments do not work or the situation is continually getting worse,
surgery is needed to relieve chronic pain (Australian Orthopaedic Association
43
2002). Joint replacement is a treatment typically reserved for patients over 55 years
of age, and is undertaken when no other measures can be used to save or prolong
the life of the joint.

Although new research continues to provide more information about the causes and
mechanisms of osteoarthritis, the roles of tribology and biochemistry in the process
of joint degradation must be more clearly understood before the disease can be
adequately prevented or cured. Ultimately, preventative measures could become
available that may force joint replacements and other such radical procedures into
obsolescence.

3.6 Total Artificial Joint Replacement
Total joint replacement (TJR), also called arthroplasty and referred to commonly as
hip/knee surgery or a hip/knee replacement, is a surgical procedure that removes
the diseased components of the natural joint and replaces them with an artificial
equivalent.

In a partial joint replacement, one of the articulating surfaces of the joint is
surgically removed and replaced with an approximate artificial replica. In total joint
replacements, both articulating surfaces are removed. The prosthesis design is
usually chosen from a library of geometries, then modified by a varying selection
of ball sizes, in the case of the hip, to fit the specific patient. When all or a portion
of a joint is replaced, a portion of the patients bone must be cut away and
removed; the prosthesis is then mounted or cemented in place on the truncated end
of the bone.

The earliest successful implants were bone plates, introduced in the early 1900s to
stabilize bone fractures and accelerate their healing. In 1925, a surgeon in Boston,
Massachusetts, M.N. Smith-Petersen, M.D., moulded a piece of glass into the
shape of a hollow hemisphere which could fit over the ball of the hip joint and
provide a new smooth surface for movement. In the 1930s PhiIlip Wiles from the
Middlesex Hospital, London, UK designed and inserted the first total hip
44
replacements. G.K.McKee, who was following Wiles, introduced metal-on-metal
hip prostheses. He developed various uncemented prototype total hip replacements
in the 1940's and 1950's. Dental cement brought about a new era in fixation
techniques and the McKee-Farrar Total Hip Replacement (THR) was the first
widely used and successful THR. In the late 1950s, the first total hip replacement
was introduced using PTFE as the cup-bearing surface (Blanchard 1995). However,
since the PTFE undergoes aseptic loosening, PTFE did not proved to be an
appropriate material to use as a load-bearing surface in the body. In the late 1960s
Sir John Charnley, a British Orthopaedic surgeon, developed the fundamental
principles of the artificial hip that still sees widespread use today. Frank Gunston
developed one of the first artificial knee joints in 1969. Since then, joint
replacement surgery has become one of the most successful orthopaedic treatments
(http://www.utahhipandknee.com/history.htm 2004).

It is now accepted that a Charnley-type total hip replacement can give satisfactory
results in an elderly sub-active population. Several metals are used including
stainless steel, alloys of cobalt and chrome, and titanium, and the plastic material
used is the durable and wear resistant UHMWPE. Recent research on hip and knee
replacement prostheses covers areas such as tribology of metal-on-metal hip joints,
coatings for enhanced tribological performance, soft layer and hard-bearing
surfaces in hip replacements, generation and biological activity of wear debris,
wear measurement and component geometry, wear of ceramic-on-ceramic joint
replacements, validation of knee simulator wear, wear measurement on knee
prostheses, lubricating film thickness in hip prostheses, and advances in simulator
testing (Hutchings 2002).

Current artificial joints made with UHMWPE and metals undergo degradation after
10- 15 years (Spector 1992; Shi et al. 2001). The research on how to improve the
design and materials to improve the durability of the artificial joints is the key for
the most recent research.

Total artificial joints must not only survive the rigorous loading pattern of the
human synovial joint, but must do so under extreme biological conditions. Low
45
coefficients of friction are essential in order to minimise the torque transmitted to
the surrounding bone. They must be long lasting, requiring an implantation life of
up to, and sometimes over 20 years depending on the age of the patient. They must
also exhibit very low wear rates, with the wear particles not causing adverse
cellular reactions that lead to loosening and failure. (Fisher & Dowson 1991)

For forty years, the basic tribological design of hip joints has not changed. There
have been many innovations regarding the strength, shape, function and behaviour
of the femoral stem which holds the prosthesis into the bone, but the surfaces have
changed very little indeed since Charnley first moved from PTFE to UHMWPE
(Maroudas 1967).

3.6.1 Artificial joint failure
The reasons for failure of current hip and knee joint replacements are, in order of
proportion, loosening, dislocation and wear (Australian Orthopaedic Association
2002). Failure due to dislocation is beyond the scope of this thesis but lies with a
better education of surgeons and the use of computer guided surgery. The other two
main modes of failure for implants are loosening and wear, and account for more
than half of revision procedures. Considering the nature of two articulating surfaces
and with wear (directly and indirectly through loosening) being the biggest mode
of failure of joint replacements, it is clear that the failure of many artificial joints is
a tribological issue. Hence tribology plays a major role in the effective treatment of
one of the most common medical conditions known in the western world
(Unsworth 1991).

Wear particles of UHMWPE are known to cause adverse cellular reactions
resulting in loosening and eventual failure of the prosthesis. The wear particles are
released into surrounding tissues and fluids where they initiate macrophage
activity. Since the wear particles cannot be broken down, the macrophages secrete
cytokines and other inflammation mediators that cause necrosis (death) and
resorption of the bone near the joint replacement. This reaction is a major cause of
prosthetic loosening and failure (Fisher & Dowson 1991) and (Ingham & Fisher
46
2000). Other causes of loosening include the response of the bone to stress
shielding and micromotion between the bone and prosthesis (Ingham & Fisher
2000).

High friction in the artificial joint causes high torque loads to be transmitted to the
bone-implant interface which may accelerate the loosening of the implant in light
of loosening being the primary reason for failure. Combine this with the immune
reaction to wear debris and friction in the joint is very important. It is therefore
important to improve lubrication mechanisms in order to reduce friction and hence,
wear.

It is obvious there is still a need for a better solution for TJR and one way to offer
this is to consider the tribological aspects of the joint and provide materials that
better suit the indigenous lubricant.

3.6.2 Biomaterials
Well over a century ago a great search was already under way for some material
which could be utilised to resurface or even replace diseased hips. Several
proposals and trials were made including the use of muscles, fat, chromatized pig
bladder, gold, magnesium and zinc. All met with failure. Surgeons and scientists
were unable to find a material which was biocompatible with the body, and yet
strong enough to withstand the tremendous forces placed on the hip joint. Stainless
steel was trialled in the late 1920s but the poor steel processing techniques of the
day led to corrosion in the body. A break through occurred in the 1930s when
scientists manufactured a cobalt-chromium alloy that was almost immediately
applied to the orthopaedic problem. This new alloy was both very strong and
resistant to corrosion, and has continued to be employed in various prostheses since
that time. The next hurdle in the treatment of the arthritic hip was to replace the
pelvic component, the acetabulum cup, with a suitable biomaterial. An attempt was
made in the late 1950s by John Charnley when he replaced the arthritic socket with
a polytetrafluoroethylene (PTFE or Teflon) implant. Although PTFE is one of the
most slippery materials know to man in the engineering world, it had disastrous
47
results in the body. Undeterred, Charnley went on to try polyethylene (PE) with
great success. Over a very short time, this became the 'gold standard' for joint
replacement, starting with total hip replacement and quickly spilling over to total
knee replacement.

The most common combination of materials used for artificial joints today is ultra
high molecular weight polyethylene (UHMWPE) cups sliding on hard metallic or
ceramic heads (Fisher & Dowson 1991), (Ingham & Fisher 2000). Typical
materials used for the heads include: stainless steel (316L), titanium alloy (Ti 6Al
4V), cobalt chrome molybdenum alloy (Co 28Cr 6Mo), and aluminium oxide
ceramic (alumina Al
2
O
3
). This combination of UHMWPE on metallic or ceramic
heads has been found to be quite successful in providing both low coefficients of
friction and wear rates.

However, considering the previous section, it is clearly evident that the optimal
material has not yet been found for use in prosthetics. Research is ongoing in
search of materials that will allow for a longer lasting implant.

3.6.2.1 Pyrolytic Carbon
A search is underway for new implant materials that have low friction and are
highly wear-resistant and biocompatible. The friction and wear properties are only
pertinent to the bearing surfaces themselves rather than to the bulk material. In
tribology, various methods are used to modify the friction and wear behaviour of
the contacting surfaces. A common method for modifying surface properties for
tribological applications is through deposition of a distinctive surface layer in the
form of a coating. One such coating is Pyrolytic carbon, a very successful
biomaterial (More, Haubold et al. 2000) already used for prostheses in the heart
and the upper limbs. This low friction and wear resistant thin film coating may be a
candidate material for articulating surfaces of total joint prostheses.

While many engineering materials and biomaterials are based on carbon or contain
carbon in some form, elemental carbon itself is also an important and very
successful biomaterial. Elemental carbon is found in nature as two crystalline
48
allotrophic forms: graphite and diamond. Elemental carbon also occurs as a
spectrum of imperfect, turbostratic crystalline forms that range in degree of
crystallinity from amorphous to the perfectly crystalline allotropes (More, Haubold
et al. 2000). Recently a third crystalline form of elemental carbon, the fullerene
structure, has been discovered. There exist many possible forms of elemental
carbon that are intermediate in structure and properties between those of the
allotropes diamond and graphite. Such 'turbostratic' carbons occur as a spectrum of
amorphous through mixed amorphous, graphite-like and diamond-like to the
perfectly crystalline allotropes. Because of the dependence of properties upon
structure, there can be considerable variability in properties for the turbostratic
carbons. Properties found in one type of carbon structure can be totally different in
another type of structure.

The biomaterial known as Pyrolytic Carbon (PyC) is not found in natureit is
manmade. Pure carbon was the original objective of the development of PyC
because of the potential for superior biocompatibility (LaGrange, Gott et al. 1969).
The successful pyrolytic carbon biomaterial was developed at General Atomic
during the late 1960s using a fluidized bed reactor (Bokros 1969). In the original
terminology, this material was considered a low temperature isotropic carbon (LTI
carbon). Pyrolytic carbon components have been used in more than 25 different
prosthetic heart valve designs since the late 1960's and have accumulated a clinical
experience on the order of 16 million patient-years (More, Haubold et al. 2000).
Clearly, pyrolytic carbon is one of the most successful and critical biomaterials
both in function and application. Among the materials available for mechanical
heart valve prostheses, pyrolytic carbon has the best combination of blood
compatibility, physical and mechanical properties and durability.

The term pyrolytic is derived from pyrolysis which means thermal
decomposition. PyC is formed from the thermal decomposition of hydrocarbons
such as propane, propylene, acetylene and methane, in the absence of oxygen.
Without oxygen, the typical decomposition of the hydrocarbon to carbon dioxide
and water cannot take place; instead, a more complex cascade of decomposition
products occurs that ultimately results in a polymerization of the individual
49
carbon atoms into large macroatomic arrays (More, Haubold et al. 2000).
Decomposition products, under the appropriate conditions, can form gas-phase
nucleated droplets of carbon/hydrogen, which condense and deposit on the surfaces
of items within the reactor (Bokros 1969). This pyrolytic carbon coating can be
produced in a variety of structures such as laminar or isotropic, granular or
columnar. The structure of the coating is controlled by the gas flow rate (residence
time in the bed), hydrocarbon species, temperature and bed surface area. This
control allows for a tailor made material, inasmuch that the structure can be
controlled to provide required mechanical properties and the surface can be
modified to give the desired surface properties.

Since pyrolytic carbon is a coating, it must be deposited on an appropriately
shaped, pre-formed substrate (preform). Because the pyrolysis process takes place
at high temperatures, the choice of substrates is severely limited. Only a few of the
refractory materials such as tantalum or molybdenum/rhenium alloys and graphite
can withstand the conditions at which the pyrolytic carbon coating is produced. It
is important for the thermal expansion characteristics of the substrate to closely
match those of the applied coating: otherwise, upon cooling of the coated part to
room temperature, the coating will be highly stressed and can spontaneously crack.
For contemporary heart valve applications, fine-grained isotropic graphite is the
most commonly used substrate. The graphite substrate does not impart structural
strength; rather, it provides a dimensionally stable platform for the pyrolytic carbon
coating at both the reaction temperature and at room temperature.

PyC was found to have not only remarkable blood compatibility but also the
structural properties needed for long-term use in artificial heart valves (LaGrange,
Gott et al. 1969). PyC flexural strength is high enough to provide the necessary
structural stability for a variety of implant applications, and the density is low
enough to allow for components to move easily under the applied forces of
circulating blood. With respect to orthopaedic applications, Youngs modulus is in
the range reported for bone (Reilly & Burstein 1974; Reilly, Burstein et al. 1974),
which allows for compliance matching. Relative to metals and polymers, the
50
pyrolytic carbon strain-to-failure is low; it is a nearly ideal linear elastic material
and requires consideration of brittle material principles in component design.

Wear resistance of the pyrolytic carbon is excellent. The strength, stability and
durability of pyrolytic carbon are responsible for the extension of mechanical valve
lifetimes from less than 20 years to more than the recipients expected lifetime
(More, Haubold et al. 2000). This property alone makes PyC attractive for use in
the body where wear is known to be an issue. It is interesting to note that some PyC
heart valves have PyC/PyC hinge joints that allow the valve leaflets to open and
close about 40 million times a year and have shown very little wear after 10 20
years of service. Contact pressures at the PyC/PyC hinge joint are high due to the
water hammer effect at valve closure.

The surface of as-deposited, machined and polished components is shown in Figure
3.4. It was found early on in experiments (LaGrange, Gott et al. 1969) that clean
polished pyrolytic carbon surfaces of tubes, when placed within the vasculature of
experimental animals, accumulated minimal if any thrombus and certainly less than
pyrolytic carbon tubes with the as-deposited surface. Consequently, the surfaces of
pyrolytic carbon have historically been polished, either manually or mechanically,
using fine diamond or aluminium oxide pastes and slurries. The surface finish
achieved has roughness measured on the scale of nanometers. The surfaces of
polished pyrolytic carbon (30-50nm) are an order of magnitude smoother than the
as-deposited surfaces (300-500nm).

51

Figure 3.4. Various forms of PyC indicating surface finish. (Left to Right) As-
deposited, machined and polished.

Pyrolytic carbon surface chemistry is important because the manufacturing and
cleaning operations to which a component is subjected can change and redefine the
surface that is presented to the bodys fluids. For example, oxidation of carbon
surfaces can produce surface contamination that detracts from blood compatibility
(LaGrange, Gott et al. 1969). The effect of modified surface chemistry on blood
compatibility is not well characterised.

It is believed that pyrolytic carbon owes its demonstrated blood compatibility to its
inertness, and to its ability to quickly absorb proteins from blood without triggering
a protein denaturing reaction. Ultimately, the blood compatibility is thought to be a
result of the protein layer formed upon the carbon surface. Baier observed that
pyrolytic carbon surfaces have a relatively high critical surface tension of 50
dyne/cm, which immediately drops to 28 to 30 dyne/cm following exposure to
blood (Baier, Gott et al. 1970). A more contemporary version of the mechanism of
pyrolytic carbon blood compatibility might be to reject the assumption that the
surface is inert, as it is now thought by some that no material is totally inert in the
body (Williams 1998) and accept that the blood-material interaction is preceded by
52
a complex, interdependent and time-dependent series of interactions between the
plasma proteins and the surface (Hanson 1998) that is as yet poorly understood.

The suitability of a material for use in an implant is a complex issue. The merit of
success of PyC in heart valves can not be simply passed onto other potential
applications for PyC in the body. The suitability of carbon materials for long-term
implants is not assured simply because the material is carbon. Elemental carbon
encompasses a broad spectrum of possible structures, mechanical properties and
surface characteristics. Each new candidate carbon material requires a specific
assessment of biocompatibility.

Cook et al have researched the application of PyC in a hip implant as trialled in
canines (Cook, Thomas et al. 1989). They compared the effect of common
prosthetic metallic surfaces versus the PyC articulating against the cartilaginous
acetabulum. They found that PyC resulted in significantly less damage to the
cartilage surface than with either the CoCr or Ti alloys. Survivorship analysis
indicated a 92% probability of survival for cartilage articulating with PyC as
compared to only a 20% probability of survival for cartilage articulating with either
of the metallic alloys. The reason for this far superior performance was unknown.

It is clear that PyC has some very interesting surface properties. The research to
date seems to indicate that the PyC surface is active and binds biologic molecules
that result in blood compatibility and possibly a very slippery, wear resistant
surface. The tribology, and more accurately the tribochemistry, of this coating is of
keen interest as a material that may be suitable for load bearing implants such as
the hip and knee where wear is known the be a problem. Modification of the
activity of the PyC surface by both chemical and thermal treatments is possible to
enhance the attachment of bio macro-molecules, for example SAPL. Knowledge
gained from tribological studies of PyC and the bodys fluids will be invaluable to
the future use of PyC in the human body.

53
This chapter describes the natural bearing - the synovial joint and its failure due to
OA and the replacement bearing - the artificial joint and its failure due to
tribological reasons, establishing the need to provide a more satisfactory solution to
the problem of joint failure.
3.7 Summary
The synovial joint is an amazing machine element from the engineers perspective.
The ability to provide such effortless motion over a large range of loads for the
lifetime of the individual is astounding. A large amount of this feat is thought to be
due to the synergy between the unique properties of both the articular cartilage and
the synovial fluid. Surface active phospholipids have been identified in both the
synovial fluid and the articular cartilage and are thought to play an important role
in the lubrication of the joint. DPPC has been identified in the natural joint. DPPC
has been identified in the artificial joint. DPPC has been shown to reduce friction
and wear between prosthetic materials (Sect 4.4.). The presence of not only DPPC
but all PCs (SAPLs), essentially all surface active molecules (that may be
beneficial in lubrication) need to be qualified in the joint environment. Part of this
dissertation seeks to identify the surfactants present in the artificial joint.

As amazing as this natural bearing is it still fails in a large number of cases due to
primarily osteoarthritis. Few treatments exist for these failures, none of which offer
a long term solution. Viscosupplementation receives some credit from the
engineers standpoint as the idea of renewing the lubricant in the bearing has merit.
However until the nature of the joint and its lubrication is fully understood it
would be unwise to suggest such a replacement lubricant. The long term solution to
OA is the removal of the diseased joint and replacement with an artificial
equivalent in as procedure called a total joint replacement. This equivalent is not a
true equivalent and suffers from a limited lifetime. There is evidence that suggests
that many artificial joints fail due to a tribological situation in vivo.

It has been shown that the lubricating ability of pathologic synovial fluid remains
largely unaffected therefore it is the radical replacement of the natural bearing
surface with man-made materials that may be the source of the problem in TJR
54
failures. Most of the research to date attempting to solve this problem of wear-
related failure in TJRs has been divided into two areas; developing materials that
mimic the properties of AC and the development of more wear resistant
biocompatible material pairs. The thesis considered the latter and investigated the
tribological performance of pyrolytic carbon a highly successful biomaterial used
in non-load bearing locations in the body.

Chapter 4: Literature Review Lubrication of Joints
55
Chapter 4
Literature Review Lubrication of Joints
This chapter includes a discussion of tribology in general with a focus on the
lubrication of synovial joints, both natural and artificial. It should be noted that in
an examination of the literature on synovial joint lubrication far more research has
occurred and hence considerably more information is available for the natural joint
as compared to its replacement. As previously stated it is a crucial step in
engineering to have a sound understanding of the function of the original
component before designing a replacement component.

The major load bearing joints, the joints of the lower extremities such as the hip,
are essentially ball and socket joints. Human joints are required to maintain
efficient and painless function for at least 7 to 8 decades under normal conditions.
In order for two opposing surfaces to be continually rubbing against one another
for such a period without major functional deterioration, i.e. wear and tear, a
remarkable system of regeneration and/or lubrication is required within the joint
capsule. It is accepted that the joint itself has a relatively low level of metabolic
turnover (Mankin & Lippiello 1969; Mankin & Brandt 1984; McIlwraith &
Vachon 1988) and, as a result, has a slow rate of repair following injury. This
would tend to negate the argument that the joints maintain their function through
regeneration. Hence, it would seem clear that it is the role of a lubricant present in
the joint capsule to keep the joint mobile through a lifetime of normal wear and
tear. Over the last century, lubrication mechanisms ranging from the simple to the
complex have been examined with reference to their possible function at the
synovial joint surface. There is still no clear agreement on which mechanism acts
within the joint. However, it is clear that either of the two fundamental mechanisms
of lubrication, as described in the physical sciences; fluid-film lubrication or
boundary lubrication, are responsible. Further, it is probable that combinations or
modifications of these mechanisms play a role in the lubrication of the joint.

56
Today our knowledge of diarthrodial joint lubrication is based not only upon these
basic lubrication mechanisms, but also upon our knowledge of the structural and
deformational characteristics of the articular cartilage; the biochemical and
biorheological (flow) properties of synovial fluid; the roughness of the articular
surface; the kinematics associated with each particular joint and the nature of the
loads applied to the articular surface (Mow & Mak 1987). It may be stated at this
stage that the unique lubricating properties of synovial joints are not solely due to
either synovial fluid or to the articular cartilage but to some complex synergistic
interaction of both (Dintenfass 1963). Understandably this situation/system is
radically changed when the cartilaginous bearing surfaces are replaced with an
artificial/man-made material.

It is interesting to consider that the development of ideas on human joint
lubrication has followed a very similar pattern to the general development of the
understanding of highly loaded lubricated components in engineering (Dowson,
Wright et al. 1969). It may be oversimplified to consider such a complex creation
with engineering principles but it is the engineer that has been called upon to help
elucidate the mystery of synovial joint lubrication.

Essentially the synovial joint can be likened to a mechanical bearing, by
description, a plain bearing. A bearing is made up of two bearing surfaces
(cartilage on cartilage for the original joint and several different man made
materials for the replacement joint), a load and/or motion (ambulation) and a
lubricant (synovial fluid). It must be noted here that in the case of a replacement
joint the only change to the system is the bearing surfaces and consideration must
be given to the interaction of the lubricant (which may also be affected by the
condition that brought about the failure of the original bearing surfaces) with the
new bearing surfaces. The original bearing surfaces and the lubrication of cartilage
are considerably different to the replacement bearing surfaces and their lubrication.
Articular cartilage, which covers the ends of the bones in the joint as a thin layer, is
not a rigid structure like the replacement material; it is flexible and compliant,
almost a gelatinous substance which shares with synovial fluid the fact that in its
composition there is more water than solid substance (Charnley 1959).
57

The natural joint is a remarkable bearing. It is expected to operate in the human
machine for a lifetime while transmitting varying loads and yet accommodating a
wide range of movement. It normally performs these functions with high efficiency
and coefficients of friction as low as 0.002 have been recorded. In engineering
situations bearing life is sometimes measured in minutes, often in hours and
sometimes in years, but it is rare indeed for the bearing designer to be asked to
provide a bearing with an estimated life of 70 years or more! It is fair to say that the
engineer would find it extremely difficult to provide a bearing which would operate
within the environment of the body, in the same space as the natural bearing, under
the same loads, with the same degree of movement, with similar friction and
comparable mean life to the natural bearing (Dowson 1973).

The influence of lubrication, friction, wear and good bearing design upon the
efficiency and general economic performance of the machinery in our
technological society was recognised in 1966 by the introduction of a new word,
'tribology', meaning 'the science and technology of interacting surfaces in relative
motion and practices related thereto'. In recent years tribologists have worked with
rheumatologists, orthopaedic surgeons, and biochemists in an effort to understand
the remarkable characteristics of healthy joints and, perhaps more importantly, the
reasons why some human bearings wear out more rapidly than others. Engineers
tend to classify forms of lubrication into a few well known physical and chemical
mechanisms, and a good deal of progress has now been made towards an
understanding of the performance of the natural joint. Attempts have also been
made to develop synthetic lubricants which might be introduced into the joints
which start to show signs of failure in the same way that an engineer might prolong
the life of a bearing in a machine by introducing an improved lubricant (Dowson
1973). However it would seem pointless to design such pseudo-synovial fluids
until the lubrication of the human bearing is clearly understood and the indigenous
lubricant is fully defined.

Relative motion between two surfaces is characterised by frictional forces and wear
of the rubbing surfaces. In general, the purpose of a lubricant is to reduce friction
58
and prevent the wear of the surfaces. The ability of synovial joints to function with
almost negligible frictional resistance with a coefficient of kinetic friction ()
ranging from 0.001- 0.03 (Jones 1934; Charnley 1959; Dintenfass 1963; Linn
1968), and to do so under appreciable stress, up to l3kg/cm
2
or 1.3MPa indicates a
superior system of lubrication which has not been matched in non-biological
systems (Hills 1989).

The lubrication of synovial joints has received comparatively little attention from
physiologists from a point of view which might seem rather obvious to engineers.
The possibility that the lubrication of synovial joints might be a more subtle
problem than had been suspected presented itself when investigating certain
disappointing results (Hall, Unsworth et al. 1997) encountered after attempts to
construct artificial joints by existing surgical techniques. The hip joint, which in
form is an almost perfect ball and socket, is frequently affected by types of arthritis
which destroy the polished rubbing surfaces and which distort the geometry of the
concentric spherical surfaces, with the result that motion is restricted and pain
develops. Early artificial joints designed on accepted ideas of joint lubrication at
the time seemed adequate to replace the distorted and destroyed ball (the head of
the femur) by a sphere of any substance which could be tolerated without adverse
reaction of the body tissues, such as stainless steel, cobalt-chromium alloy or
plastic. Provided that the surface of the sphere was highly polished it has been
assumed without question that its 'slipperiness' when bathed in tissue fluids would
be adequate to function as a weight-bearing joint (Charnley 1959). Even if the
coefficient of friction of this artificial joint was not quite as favourable as the
normal joint, it has been accepted that from a mechanical standpoint a polished
steel sphere would be an improvement on the diseased and distorted head which it
was used to replace. The early results of many of these operations, often quite
dramatically successful in the early months after operation, suggest that surgical
research is very close to success; but, the later results, with return of pain and
disability in many cases, showed that important details still eluded discovery. The
suspicion was aroused that accepted ideas of joint lubrication at that time might be
too simple, and it was necessary re- investigate the subject of normal joint
59
lubrication before investigating the frictional properties of substances suitable for
making an artificial joint.

4.1 Physical Science of Lubrication, Friction and Wear
The word Tribology comes from the Greek word T or tribos, meaning to
rub. It is concerned with surface interactions between two bodies moving relative
to one another. Tribology is generally defined as the study of three areas: friction,
wear, and lubrication of interacting surfaces. The three are highly related, although
in the case of friction and wear their relationship is not well understood. Generally
speaking, wear is friction-related, but there are exceptions to the rule. These
exceptions relate to when wear is regarded as phenomena. Basically, friction is
produced when two sliding surfaces come into contact and inevitably wear will
occur. The defence against wear is achieved by lubrication; the separation of the
two surfaces by a lubricant that will result in a reduction of friction.

Friction
Friction can be defined as the force that acts at the surfaces of two articulating solid
bodies so as to resist sliding on one another. Simply, it is the resistance to motion
of surfaces in relative motion. This force which tends to prevent one surface sliding
over another is caused by shearing of local adhesions that occur at the regions of
real contact between asperities and is very conveniently quantified by a simple
index - the coefficient of friction (). The first published study of friction was by
Amonton in 1699 (Shi 2004) who revealed certain laws of friction which still
form the basis for the analysis of many sliding contacts today. For dry friction, 1)
the frictional force is directly proportional to the applied load (Amontons 1
st
Law);
2) the force of friction is independent of the apparent area (A) of contact
(Amontons 2
nd
Law); and 3) kinetic friction is independent of the sliding velocity
(Coulombs Law). When a block is placed on a dry surface as illustrated in Figure
4.1, the force (F) needed to be applied in the plane of the interface in order to cause
motion is found to be proportional to the weight (W) or other force pressing the
two surfaces together (Adamson, 1967).
60

Figure 4.1. Amontons' Laws of Friction. Source: (Shi 2004)
The proportionality constant () is termed the coefficient of friction and tends to be
lower after motion is established (Stachowiak & Batchelor 2005), so it is more
appropriate to use the coefficient of kinetic friction in the physiological context. In
order to determine the coefficient of friction, two measurements are needed: the
force (F) required to initiate and/or sustain sliding and the normal force (W)
holding the two surfaces together. The coefficient of friction is then equal to the
initiation/sustaining force divided by the normal force. Since a lubricant reduces
friction, provides a very convenient index for quantifying the lubricating ability
of any system. It should be noted that must be considered with care as it is
subjective to the lubricated conditions and type of tribometer used. Comparisons
between values generated by the apparatus under the same conditions is
acceptable but the comparison of friction coefficient values from differing
apparatus and dissimilar conditions need to be treated with care. This is especially
true for boundary lubrication where many variables play a role in lubrication
(4.1.2).

Lubrication
'Lubrication may be defined as any means capable of controlling friction and wear
of interacting surfaces in relative motion.' (Hsu & Gates 2001). From this definition
it is clear that an effective lubricant is one which minimises both the coefficient of
friction and the amount of wear particles produced. Essentially, effective
lubrication can almost entirely eliminate the friction force by keeping the two
interacting surfaces apart. This is made possible by some substance that keeps the
bearing surfaces from touching. This substance is referred to as the lubricant
and can be a liquid, a solid, a mixture of the two or even a gas.
61

There are basically two types of lubrication - one in which there is solid-to-solid
contact of the sliding surfaces known as boundary lubrication and the other where a
thin layer of fluid is either pumped into the intervening space (hydrostatic
lubrication) or the motion itself produces a wedge of fluid on which the moving
surface planes over the counterface (hydrodynamic lubrication). In most
engineering applications, hydrodynamic lubrication is much more effective than
boundary lubrication, but requires high velocities for the wedge of fluid to generate
enough pressure to overcome the load and keep the surfaces separated. In starting
up machinery, the friction and wear are much higher as boundary lubrication gives
way to hydrodynamic lubrication as the fluid film is established and the same
applies in the opposite sense at shutdown. The transition between these modes is
very important, as originally recognised by Stribeck in his classical diagram
(similar to Figure 4.2) demonstrating the change in friction as one mode supplants
the other.

Figure 4.2. Range of coefficients of kinetic friction reported in the literature
for the mammalian joint are depicted over a physiological range of sliding
velocities and compared with the a modified classical Stribeck diagram.
Source: (Hills 2000)

62
The Stribeck diagram clearly indicates the drop in the coefficient of friction as the
velocity increases and the transition occurs from solid-to-solid contact to the
establishment of a fluid-film. The friction of fluid film lubrication is proportional to
both the sliding velocity and the bulk fluid viscosity and inversely proportional to
the film lubricant thickness. Therefore as the velocity reduces the fluid film
thickness diminishes and the friction increases. There are three regions in the
Stribeck-Curve: boundary lubrication, a transition region which refers to mixed
lubrication and hydrodynamic lubrication (which includes elasto-hydrodynamic
lubrication (EHL)). Boundary lubrication is characterised by the absence of
hydrodynamic pressure or fluid pressure. One hundred percent of the loading is
carried by the asperities in the contact area, protected by adsorbed molecules of the
lubricant. Mixed lubrication is the intermediate region between boundary
lubrication and hydrodynamic lubrication. Hydrodynamic lubrication is
characterised by the presence of a pressurised fluid that keeps the surfaces apart.
The lubrication regimes are shown schematically below in Figure 4.3. where h is
referred to as the film thickness that separates the bearing surfaces.

63

Figure 4.3. Lubrication regimes Source: (Dowson, Wright et al. 1969)

In systems where both static and kinetic friction occur at some stage (eg most
engines) a force known as the break-out or start-up force must be overcome.
The coefficient of friction associated with static friction is generally much higher
than that associated with kinetic friction and it is a known fact that the majority of
wear in engines occurs at these times of start-up and shut-down. It is well
established in the physical sciences that, to overcome the problem of the break-out
force, it is necessary to use additives in oils designed for use in systems where
64
there will be components of both static and dynamic friction. The role of the
tribologist is to define a lubricant that can cater to all operating conditions.

When an engineer investigates the performance of bearings he tries to determine
the mode of lubrication and to classify the system as "fluid-film" or "boundary" or
some combination of the two. As mentioned, in fluid-film lubrication the bearing
surfaces are completely separated by a film of fluid for most of their working life
and the frictional characteristics of the bearing can be explained in terms of the
rheological properties of the bulk fluid. In boundary lubrication the bearing
surfaces come much closer together and the frictional characteristics of the bearing
are determined by the properties of the surface materials. The predominant surface
properties are rarely related to the bulk properties of the fluid or the solid and are
usually determined by thin layers of molecular proportions which might take the
form of material arising from chemical reaction between the bearing material and
its immediate environment of thin layers formed by physical or chemical
adsorption (Dowson, Wright et al. 1969).

Wear
Wear can be defined as the progressive loss of substance from articulating surfaces
in relative motion. Mechanisms of wear are many and varied and include abrasion,
adhesion, surface fatigue, corrosion etc. These mechanisms may occur singly or at
the same time. They may also operate independently or interact (Axn, Hogmark et
al. 2001). Friction is caused by the two surfaces coming into contact and this
contact eventually causes wear. It is a simple conclusion to presume that this would
logically mean the lower the friction the lower the wear, however this is not always
the case. Low friction does not always result in low wear and high friction does not
always result in high wear. But generally in bearing design the lower the friction
the lower the wear (Bhushan 2001). There is also evidence (Williams 2005) where
a small increase in friction has led to a large increase in wear and conversely a
decrease in friction has lead to a reduction in wear. The engineering tribologists
must consider the laws of friction and the exceptions to the laws when designing
bearings.

65
4.1.1 Fluid-Film Lubrication
Fluid-film lubrication is a class of lubrication mechanisms having in common a
film of fluid that completely separates the opposing surfaces. This layer of fluid is
thicker than the sum of the surface roughness of the two surfaces. The development
of a film between the opposing surfaces may be achieved in various ways
depending on the properties of the fluid and the surface motion. The load on the
surface is supported by the pressure in the fluid film. The fluid film thickness is
usually less than 10m according to (Dowson, Wright et al. 1969). The resistance
to motion arises entirely from the shearing of the viscous layer, and in general the
friction force can be represented by an equation of the form:
Equation 4.1. A
h
V
F =
where F = friction force;
V = relative sliding velocity
h = lubricant film thickness
A = effective area of bearing
= coefficient of viscosity

The coefficient of viscosity is the most important property of a lubricant in fluid-
film bearings since it governs not only the resistance to motion but also the ability
of the bearing to develop adequate load-bearing pressures in the fluid.

The lubricating films in fluid-film bearings are themselves quite thin in physical
terms, normally between 1m and 25m. (Dowson, Wright et al. 1969) This
magnitude is normally adequate to separate the opposing bearing surfaces which
are, of course, manufactured with a high degree of precision. Surface roughness of
typical engineering bearings usually range from 0.1m to 1m Ra.

If the opposing bearing surfaces can be prevented from touching each other by
interposing a layer of fluid, wear is almost totally prevented and the resistance to
sliding is low. It is for this reason that 'fluid film' is often described as the ideal
mode of lubrication. If the load-carrying pressures are generated by the sliding
66
motion of the surfaces the mechanism is known as 'hydrodynamic lubrication', but
if the pressure is generated in a pump outside the bearing the action is known as
'hydrostatic lubrication'.

Hydrodynamic Lubrication
In the last decades of the 19th century engineers began to understand the manner in
which industrial bearings worked. The fact that they worked had long been
accepted but no-one could explain why oils and greases, which were obviously
essential to the continued running of bearings, could make things slippery. In 1883,
Beauchamp Tower (cited by (McCutchen 1978)) was experimenting with a bearing
shell that sat on an oiled rotating shaft when he noticed that oil flowed out of a hole
that was in the centre of the shell. A wooden plug was driven into the hole but was
slowly forced out by the oil. By mapping the pressure in the lubricating oil, Tower
found that the total force it exerted equalled the bearing load. The shell was
supported by liquid and did not touch the shaft at all so long as the shaft turned fast
enough. Lord Rayleigh (1884) and Sir Osborne Reynolds (1886) (both cited by
(McCutchen 1978)) independently realised that the bearing shell sets itself slightly
off centre horizontally relative to the shaft, forming a thin, curving wedge-shaped
space between itself and the shaft that converges in the direction of shaft motion.
Because the oil is viscous, the moving shaft drags it towards the thin edge of the
wedge where there is less and less room for it. A high pressure results which keeps
the surfaces apart. This was probably the first description of hydrodynamic
lubrication.

Hydrodynamic lubrication is essentially the classical model of fluid-film
lubrication. It occurs when the relative motion of two bearing surfaces draws fluid
into the space between them, keeping them apart, the height of the fluid decreasing
in the direction of motion (wedge action). This mechanism requires uninterrupted
motion in the same direction to maintain the integrity of the fluid wedge and is
especially efficient in high speed bearings.

67
The thickness (h) of the fluid layer in a hydrodynamic bearing in which the
viscosity of the fluid remains constant increases as the sliding speed (U) increases
and decreases as the load (W) increases, or
Equation 4.2.
W
U
h
Another important feature of hydrodynamic bearings is that the film thickness
decreases in the direction of sliding (Figure 4.3b).

Elastohydrodynamic Lubrication
Elastohydrodynamic lubrication (EHL) is defined as hydrodynamic lubrication
when applied to solid surfaces of low geometric conformity that deform elastically
(Szeri 2001). In bearings utilizing this mode of lubrication, the pressure and film
thickness are of order 1 GPa and 1 m, respectively under such conditions,
conventional lubricants exhibit behaviour distinctly different from their bulk
properties at normal pressure. In fact, without taking into account the viscosity
pressure characteristics of the liquid lubricant and the elastic deformation of the
bounding solids, hydrodynamic theory is incapable of explaining the existence of
continuous lubricant films in highly loaded gears and rolling contact bearings. The
principal feature of EHL contacts is the film thickness is nearly uniform over the
contact zone, but displays a sudden decrease just upstream of the trailing edge
(Figure 4.3). EHL is a complex regime whereby large pressures are generated in
the fluid film separating the surfaces, large enough that one or both of the bearing
surfaces will elastically deform actually increasing the fluid film thickness between
the surfaces.

Hydrostatic Lubrication
This mechanism also falls under the general heading of fluid lubrication. Bearing
surfaces are held apart by a film of lubricant which is maintained under pressure
and usually supplied by an external pump. The load is completely supported by the
pressure supplied from an external source. This mechanism is especially suited to
oscillating bearings in low-speed applications when loads are high.

68
4.1.2 Boundary Lubrication
Boundary lubrication is a science on its own and it will receive further discussion
in Chapter 5.

Boundary lubrication is perhaps the most confusing and complex aspect of the
subject of friction and wear prevention (Ling, Klaus et al. 1969). Confusing in
part by the fact that the boundary is not always well defined and complex in part by
the large number of variables involved. In reference (Larsen & Perry 1950), twenty
nine variables are listed and the list is not by any means complete.

It is clear from Equation 4.1 that the thickness of the lubricant layer will decrease at
low speeds and high loads and in due course the opposing bearing surfaces will
come into contact at high spots or asperities. As the severity of this condition
increases, the effectiveness of the lubricant layer diminishes, friction rises, and the
bearing surfaces start to wear.

It is found that friction and resistance to wear are no longer governed by the
viscosity of the bulk fluid but by the properties of thin surface layers formed by
physical and chemical action on the surfaces of the solids; hence the term
'boundary lubrication' (Dowson, Wright et al. 1969).

Boundary lubrication has developed from experience from very early times. The
first crude wheel could not have been supported by an elegant hydrodynamic
bearing, so some system of boundary lubrication must have been used (Ling, Klaus
et al. 1969).

In engineering, boundary lubrication is known to occur during start, stop and under
severe operating conditions of mechanical machinery. During these conditions,
high loads and/or low speed cause the breakdown of the fluid film causing the
bearing surfaces to come into contact at their asperities. Effective boundary
lubrication depends on the properties of an anti-wear film which must be adsorbed
to the surfaces of the bearing materials and have high cohesion. This film may
operate to prevent wear in many ways. A sacrificial layer prevents wear by being
69
removed instead of the asperities of the bearing surfaces. In order for the surfaces
to remain protected, the rate of formation of the sacrificial layer must be greater
than the rate at which it is being removed. Many solid lubricants work using this
mechanism. Other mechanisms of boundary lubrication include shear resistant
layer, low shear interlayer, friction modifying layer, and load bearing glasses (Hsu
& Gates 2001).

Solid lubricants are a thin film of solid interposed between two rubbing surfaces.
They provide effective lubrication by shearing easily to prevent wear and
maintaining low coefficients of friction. Most solid lubricants can shear easily due
to their lamellar crystaline structure allowing ease of sliding between layers. They
contain close, strongly bonded layers connected by only weak forces between each
layer. It is thought that the strong bonding within the layer helps reduce wear
damage. Common examples of lamellar solid lubricants are graphite and
molybdenum disulfide (Erdemir 2001). Figure 4.4 shows the structure of these
solid lubricants. Evidence has been presented which suggests that surfactant found
in synovial fluid acts in the same way as lamellar solid lubricants, such as graphite,
in joint prosthesis. (Purbach, Hills et al. 2002)

Figure 4.4. Molecular Structure of Common Solid Lubricants (a) graphite,
and (b) molybdenum disulfide. (Erdemir 2001)

70
In commercial situations where surface contact between mating parts is known to
occur, substances are often added to the lubricant (for example engine oils) to
enhance its efficiency. Additives that improve lubricity are often commercial
surfactants. (Larson & Larson 1969) Surfactants adsorb to the bearing surfaces to
form a protective coating that prevents wear and decreases static friction when
contact occurs. They are also an effective anti-stick agent. Although these
surfactants are capable of producing extremely low levels of friction they are too
toxic to be used in clinical situations (Hills 2002).

When two surfaces touch, asperities from one surface make direct physical contact
with the other thus generating friction and possibly adhesion and abrasion if sliding
is attempted. On bearing surfaces these asperities are minimised by making them
macroscopically smooth, but friction can be further reduced if the material at the
interface shears. This could be done within the substance of the bearing, e.g.
Teflon, or within a layer of foreign substance physically adsorbed to the surface
(Figure 4.3.). Substances which are ideal for adsorption to the surface are
surfactants where the polar group bonds the molecule by strong chemisorption to
form an adsorbed monolayer. An example in everyday life is experienced in
touching a bar of wet soap when ones fingers remain slippery, even after
squeezing out the water. Boundary lubricants are described in terms of their
chemistry and the chemistry of the surfaces they are to protect (Ling, Klaus et al.
1969).

Boundary lubrication is then highly dependent on maintaining strong cohesion of
the adsorbed monolayer in order to avoid its penetration by asperities on the
counterface. The basic molecular requirements for surfactant monolayers to act as
efficient boundary lubricants are described later (5.4.4.). An interesting sub-section
is lamellated-solid lubricants (Erdemir 2001) where sliding surfaces are separated
by multiple layers of the lubricant which are readily sheared, leaving the material
on both surfaces as determined by the cleavage plane. An everyday example is the
graphite pencil, which has the ability to withstand a perpendicularly applied load
while shearing. Thus graphite and molybdenum disulphide (MoS
2
) are much used
lubricants on their own or as additives in oils (Ling, Klaus et al. 1969).
71

Boundary lubrication is essentially independent of the physical properties of either
the lubricant (e.g. viscosity) or the contacting bodies (e.g. stiffness). The best
values for the coefficient of friction obtained in the non-biological field using
boundary mechanisms, is that of Teflon (polytetrafluroethylene PTFE) rubbing
upon Teflon providing a value for of 0.04. Boundary lubrication is generally
implicated under conditions where the load is very high, or the surface motion is
very small and continuous motion is not possible. It must act when the surface
asperities come into contact or when, under prolonged and possibly severe load, the
fluid film is depleted.

4.1.3 Mixed Lubrication
In many bearings there is a combination of both boundary and fluid film regimes, a
situation broadly termed as mixed lubrication. This lubrication condition is of
importance especially in systems where movement is intermittent; hence, there is
continual changing from boundary to fluid-film regimes.

4.1.4 Wear
Wear is defined as the damage to a surface that generally involves progressive loss
of material and is due to relative motion between that surface and a contacting
surface or substance(s) (Shi 2004).

Although practically everything which an engineer makes will ultimately wear out,
this aspect of tribology may be the least explored. The physical and chemical
actions involved in the wear process are complex and much current research is
devoted to their elucidation.

Well-known forms of wear include abrasion, adhesion, fatigue, erosion, fretting
and corrosive wear. Abrasive wear occurs when asperities of a rough, hard surface
slide over a softer one (e.g. steel over plastic) and damage the interface by plastic
deformation of fracture. Abrasive wear may also occur when hard particles are
72
trapped between rubbing surfaces. In the latter case, the abrasive material may
enter the system from the environment or it may be debris resulting from the wear
process (Bhushan 2001).

Adhesive wear, which occurs as a result of local welding at asperity junctions, is by
far the most common wear mechanism (Stachowiak 2005). In many cases the
volume of material removed in the adhesive wear process satisfies the following
relationship:

Equation 4.3.
m
p
Wx
k V
3
=
where
V = volume of material worn away
W = applied load
x = total sliding distance
p
m
= hardness of softer material
k = wear coefficient

This equation enables the bearing designer to estimate the useful life of a bearing or
prosthesis under given operating conditions. It also enables the wear resistance of
different materials to be evaluated if the hardness can be measured and the
dimensionless wear coefficient evaluated (Dowson, Wright et al. 1969).

Fatigue wear occurs when repeated loading and unloading cycles are applied to the
materials and induce the formation of sub-face or surface cracks, which finally will
result in the break-up of the surface (Axn, Hogmark et al. 2001). Chemical and
corrosive wear occurs when sliding takes place in a corrosive environment. Impact
wear by erosion occurs due to jets and streams of solid particles, liquid droplets,
and implosion of bubbles formed in the liquid and by percussion from repetitive
solid body impacts. Fretting wear and fretting corrosion occur where low-
amplitude oscillatory motion in the tangential direction takes place between
contacting surfaces which are nominally at rest (Bhushan 2001; Ludema 2001).
73
Combinations of these mechanisms can be involved in the wear of biological
bearings and their replacements (Furey 2000).

Wear is related to friction, although in some cases the extent of this relationship is
not well understood. Effectively, high friction will lead to wear in bearings and low
friction is regarded as high wear resistance in most practical situations (Axn,
Hogmark et al. 2001).

4.2 Natural Joint Lubrication: A Review
The human joint is an extraordinary bearing and the engineer would find it difficult
to produce a simple bearing system of the same efficiency for similar operating
conditions. The friction forces in healthy joints are extremely low and coefficients
of friction as low as 0.002 have been recorded. However, in spite of the
extraordinarily low friction, weight-bearing human joints sometimes fail in the
mechanical sense and it is important to establish an understanding of the
lubrication mechanism in healthy joints if the reasons for these failures are to be
examined and understood (Dowson, Wright et al. 1969).

Under normal conditions the combined modifications of the various lubrication
mechanisms are more than adequate to keep the cartilage lubricating effectively
and keep wear minimal for what is essentially a lifetime. However, the failure of
lubrication would be expected to lead to increased cartilage wear and the process of
degeneration could begin (Cooke, Dowson et al. 1978). Indeed, this is one of the
models presented for the pathogenesis of osteoarthritis and other degenerative joint
diseases.

How extraordinarily low is a coefficient of friction of even 0.01 can be appreciated
only if it is realised that it is approximately three times better than the coefficient
for ice sliding on ice (which is approximately 0.03) and between ten and twenty
times better than that for a polished steel surface moving on a lubricated brass
bearing (which is within the range of 0.1 to 0.2). The extraordinary fact of this very
low coefficient of friction in a synovial joint may not be appreciated by all
74
scientists. A coefficient of 0.01 means that a load of 50kg (the weight of a small
adult) could be made to slide by applying a force of only 500g. If articular cartilage
were to be no more slippery than a plain engine bearing the same load would need
a force of 5-10kg to make it slide. If we consider that in the lower extremity of a
human being four sliding surfaces are used simultaneously in each limb when
walking or running (hip, knee, knee-cap, and ankle) we see that for an adult
weighing 50kg, 2kg would be needed to move all these joints, but, if in place of
articular cartilage lubricated metal bearings were substituted, a total force of 20-
40kg would be needed to overcome frictional resistance. As Charnley said it best,
The remarkable way in which nature has solved this friction problem will be
appreciated when it is realised that to equal it an engineer would have to discard
plain bearings and employ a so called 'frictionless' bearing where the rolling
mechanism of a ball-race takes the place of sliding friction (Charnley 1959).

In the efforts to elucidate how the synovial joint maintains its remarkable
capabilities, showing almost negligible frictional resistance and wear rates, the
engineering analyses of lubrication were soon applied to the joint. This, however,
created an apparent conflict between support for the role of fluid-film lubrication
and support for the role of boundary lubrication. The very low values of 0.003-0.02
that were obtained with the joint would seem to indicate that lubrication had to be
hydrodynamic, as values obtained for solid- to-solid contact (boundary lubrication
mechanism) were typically at least an order of magnitude higher than these.
However, as clearly demonstrated by the Stribeck diagram (Figure 4.2), it is
necessary to reach a certain velocity (of at least 5cm/s) to attain and maintain a
fluid-film, yet the articular joint frequently performs many actions at extremely low
velocity, insufficient to maintain a system of hydrodynamic (fluid-film) lubrication.
Also, although joints are often stationary under load, there is no evidence of a
break-out (start-up) force when movement is resumed i.e., there is no obvious
transition from static to kinetic friction. Because of these complications, engineers
have been postulating models of joint lubrication that would enable fluid-film
lubrication to function at low velocities (certainly lower than that predicted by
Stribeck).

75
In hydrodynamic, or full-film lubrication, the nature of the substances composing
the sliding surfaces, and the nature of the fluid used as a lubricant, are theoretically
unimportant; the essentials are the geometry of the surfaces and the viscosity of the
fluid. In this type of lubrication it is not necessary for the fluid to possess the
property of 'oiliness'. If the hydrodynamic theory were to be the basis of joint
lubrication as it was thought to be in the pioneering years of the development of
TJRs it would be quite logical to expect success by replacing one, or even both, of
the spherical joint surfaces with polished surfaces of stainless steel or plastic of the
correct dimensions. Surgical experience at that time showed that this was not borne
out in practice and it is was the scientists of the time that deemed it necessary to
examine the other forms of lubrication that might be have present (Charnley 1959).

Since that time some thirty or more theories have been proposed for joint
lubrication (Furey & Burkhardt 1997) and several more have been hypothesised
since 1997 (Murakami, Higaki et al. 1998; Skotheim & Mahadevan 2004). The
sheer number of models shows the lack of complete understanding of joint
lubrication. It is difficult for those working on replacement bearings to design
suitable replacements when the lubrication of the original is not fully understood.

4.2.1 Experimental techniques and apparatus used to determine
the lubrication of joints
Generally, two types of frictional measurements have been conducted on joints. In
the first type, the frictional properties of entire joints have been tested using joint
testers or arthrotripsometers. All of these studies are of the pendulum or
reciprocating motion type, where one bone of the joint swings relative to the other,
either freely or through forced motion. In the second type of experiments, cartilage
explants have been rubbed against each other or against other materials.

Pendulums
Pendulums have been used by many research groups to attempt to identify the
mode of lubrication acting at a surface. In boundary lubrication, friction is
independent of speed so the amplitude of pendulum oscillation drops by an equal
76
amount with successive swings. With hydrodynamic lubrication, however, the loss
between swings falls as the amplitude falls, so long as the fluid film is complete. A
pendulum used in this way to assess friction by measuring the energy loss per cycle
is called a Stanton pendulum after T.E. Stanton (1923, cited by (McCutchen 1978))

In 1934, Jones measured the friction coefficient of cartilage against cartilage (using
a horse stifle joint) at very low rubbing speed and found a value of 0.02 regardless
of whether he used synovial fluid or Ringers solution as the lubricant. Later (Jones
1936), he measured friction using a pendulum with a human interphalangeal
(finger) joint as the pivot and observed how the oscillations decreased over time.
The behaviour of Joness pendulum fell between the expectations for boundary and
for hydrodynamic lubrication and, from this, Jones concluded that hydrodynamic
lubrication acted at high rubbing speeds, switching to boundary lubrication as the
rubbing-speed fell. This was the first suggestion of a mixed lubrication mechanism.

In 1959, Charnley repeated Jones experiments with a somewhat different
apparatus. He also removed the tendons and ligaments from the joint, but used the
same principle. He found very low coefficients of friction (0.005 - 0.024) which
varied little with speed. This prompted Charnley to suggest that the mechanism of
joint lubrication was purely boundary lubrication (Charnley 1959). Probably the
largest obstacle preventing widespread acceptance of this theory was that boundary
lubrication typically yields coefficients of friction of about 0.1 in engineering
applications, and the values reported by Charnley were considered to be attainable
only in situations which lubricated via fluid-film mechanisms. Once the possibility
was raised that the lubrication mechanism may be either fluid-film or boundary in
nature, an increasing number and variety of experiments followed.

Barnett and Cobold (1962) removed the skin, tendons and ligaments from their test
joint in stages. This enabled them to show that the frictional force between the
cartilage surfaces was independent of the amplitude and hence of the speed of the
oscillations, a characteristic of boundary lubrication. (Barnett & Cobbold 1962)

77
In 1967, (Faber, Williamson et al. 1967) used an apparatus other than a pendulum
to assess friction at the full physiological range of rubbing speeds within the living
joint. Using a spring-loaded vibrating device, they noted that the decay of
oscillations was neither purely linear nor purely exponential. This implicated the
presence of both a boundary mechanism and a fluid-film mechanism, the
contribution of each varying with the load applied to the joint. (Little, Freeman et
al. 1969) used the femoral head and acetabulum of human hip joints as the pivot on
a Stanton pendulum. The hip was oscillated at various loads, with either synovial
fluid or Ringers solution as the lubricant. Under all conditions, the decay of free
oscillation appeared to be independent of load and rate of movement. This led the
group to conclude that the dominant frictional forces were boundary in nature.

Alternate Testing Apparatus
Another friction testing apparatus frequently employed to assist in solving the
riddle of the lubrication mechanism acting within the joint is to slide small pieces
of cartilage over another surface, usually cartilage or an artificial material such as
glass. This system offers the means to separate purely surface friction effects from
ploughing effects (Mow & Mak 1987). A slight problem with this apparatus is
that, unlike the in vivo situation, the entire surface of the cartilage is under load.

It was while using this type of apparatus (cartilage on glass) that McCutchen
(McCutchen 1962) was able to demonstrate that as good a bearing as cartilage is
when lubricated by distilled water or saline, it is better with synovial fluid.
However, it was also demonstrated that over short periods of time, synovial fluid
showed little advantage over water or saline in the lubrication of this set-up.
However, as time progressed the synovial fluid was obviously superior. The
significant point here would seem to be not that synovial fluid was superior to
saline over extended periods of time, but the fact that there was any lubrication
with saline at all. This would strongly support an argument for a lubricant adsorbed
to the cartilage surface. As this lubricant was worn away without replacement, as
would be the situation when using saline and water as the fluid lubricant, the
coefficient of friction would rise. With synovial fluid as the lubricant, the adsorbed
layer could be continually replaced and should remain constant. McCutchen went
78
on to propose a model of lubrication that utilised boundary lubrication and could
combine with his theory of weeping lubrication.

Malcolm (Malcolm 1976) tested cartilage against cartilage in a continuously
rotating articulation and found that the interfacial coefficient of friction; (1)
increased with time under load, (2) increased with magnitude of load, (3) was
lower when using synovial fluid as a lubricant rather than saline and (4) was very
sensitive to small vertical oscillations imposed on the rotating cartilage. These
dynamic loads dramatically decreased the coefficient of friction between the
rotating surfaces. This effect was attributed to the expression of interstitial fluid
caused by the imposed oscillation. This frictional experiment demonstrated that
when the interstitial fluid was forcibly expressed into the gap by the oscillation, a
fresh fluid film was created and thus frictional resistance was reduced: i.e. the
lubrication mechanism changes from boundary to fluid-film. Other experiments
that explored a wide range of speeds utilising this type of apparatus (Faber,
Williamson et al. 1967; Linn 1967) produced observations that also indicated the
simultaneous presence of both boundary and fluid film lubrication.

The diversity of possible lubrication mechanisms has created considerable interest
and controversy over the last century. Evidence cited to support the various
hypotheses has often been indirect and conjectural; for example, the two theories,
weeping and boosted lubrication, are apparently in complete contradiction of each
other indicating that there still remains a basic lack of knowledge of the
biomechanics involved in the load bearing of the articular cartilage. Conversely,
theories derived from a purely engineering point of view often demonstrate a lack
of understanding of the nature of biological systems.

More recent biomechanical evidence, first described by (Mansour & Mow 1977),
(Lai, Kuei et al. 1978) and reviewed by (Mow & Mak 1987), would seem to
suggest the simultaneous existence of both fluid exudation and imbibition under a
moving load in a system of self-lubrication. Hence, both the weeping and boosted
lubrication models are probably only partially correct.

79
It would appear that the attempts to isolate a single mode of lubrication with
respect to the synovial joint may have been misguided and somewhat fruitless as
there is considerable evidence to support a mixed lubrication regime. The joints
appear to enjoy all of the common physical modes of lubrication plus some novel
ones which are, perhaps, unique to the synovial joint. Many of the studies
completed regarding the modes of lubrication acting in the joint demonstrate the
futility of trying to ascribe a single mode of lubrication to the synovial joint. Few
engineering bearings operate under a single mode of lubrication so it would seem
ludicrous to expect such a complex structure as the joint to be able to behave
differently to some of the comparatively simple bearings that are utilised in
industry.

Engineers have been tackling the issue of joint lubrication in a slightly different
manner (Mow, Holmes et al. 1984; Hou, Holmes et al. 1989; Schreppers, Sauren et
al. 1991; Jin, Dowson et al. 1993), developing mathematical laws regarding the
characteristics of both the articular cartilage and the synovial fluid before going on
to determine how applicable these laws would be in vivo. Using known anatomic,
biochemical, kinematic and loading characteristics for a specific joint, one could, in
theory, calculate which lubrication system is likely to be acting. To date, there is
still no clear understanding of the lubrication mechanisms except that it does
involve both boundary lubrication and fluid-film lubrication acting together, the
importance of each mechanism depending on the type of loading to which the joint
is subjected.

Although evidence supporting the role of fluid-film lubrication is strong, cartilage
is undoubtedly better lubricated by synovial fluid than by saline (McCutchen
1966). If the advantage is not provided by the behaviour of the liquid between the
surfaces, it must result from the behaviour of the liquid at the surfaces therefore it
must be boundary lubrication.

The common theories of joint lubrication will now be briefly reviewed.

80
4.2.2 Fluid-Film Models
Under initial consideration it is easy to see how fluid-film lubrication theory was
first selected for the method of joint lubrication. Simply, it is the only method of
lubrication known in engineering that can achieve such low values of friction as
reported in the human joint. Several different models based on fluid-film
lubrication have been proposed and will be discussed briefly in the following
sections.

4.2.2.1 Hydrodynamic Lubrication
In 1932 MacConaill proposed that joint lubrication was hydrodynamic (Fig. 4.5)
(MacConaill 1932). He theorised that a wedge-shaped film would be formed
between the articulating joint surfaces by the synovial fluid, hence bearing the load
and preventing friction and wear of the joint surfaces. However, there was no
explanation for the low friction of joints starting from rest and he hadnt taken into
account the incongruent nature of the opposing surfaces. Also, it seemed unlikely
since hydrodynamic lubrication requires low loads and high surface velocities
which are not found in the human synovial joint.

Figure 4.5: Hydrodynamic Lubrication; Diagram showing the
formation of the pressure generated due to a wedge of fluid that
separates the moving bearing surfaces. (Dinnar 1975)

It had long been recognised that cartilage was permeable, fluid soaked
(Benninghoff, 1925 cited by (McCutchen 1983)) and not rigid. Despite the
knowledge presented by Beninghoff, most engineers treated the cartilage as a
standard, non-deformable, non-porous and smooth material and it wasnt until the
late 1950s that lubrication theories allowed for the incongruent surfaces and the
porous and elastic nature of the articular cartilage began to emerge.
81

Fluid-film lubrication may function in a number of different situations and in a
number of different modes: hydrodynamic (light loads at high speed),
elastohydrodynamic (moderate loads and speeds), and squeeze-film mechanisms
(impact loads) including boosted lubrication and weeping lubrication (a
modification of both hydrostatic and squeeze film mechanisms). The source of the
fluid film in the joint capsule depends upon the joint kinematics and the magnitude
of joint loading. Both synovial fluid and cartilage exudate are viable sources of the
fluid film and the various fluid-film models implicate both possible sources
(Dowson 1990).

Hydrodynamic lubrication, as already described, was one of the earliest modes of
lubrication implicated within the joint capsule. This mechanism requires
uninterrupted motion in the same direction to maintain the integrity of the wedge
(Figure 4.5.) and is especially efficient in high speed bearings. These requirements
indicate that pure hydrodynamic lubrication is unlikely to be solely applicable to
the synovial joint which requires frequent changes of direction or efficient
functioning at slow speeds, up to 0.1m/s (Dowson 1973; 1990) or at rest. Further
problems with this mechanism are highlighted when one examines the fluid film
thickness: minimum fluid thickness is in the order of 10
-7
-10
-3
m during the peak
loading period of a normal walking cycle (Mow & Mak 1987). When compared to
the average height of the healthy cartilage asperities (approximately 2-6m)
(Dowson 1990) it is obvious that hydrodynamic lubrication alone would rapidly
result in considerable wear of the cartilage and is inadequate under the conditions
set within the synovial joint.

4.2.2.2 Weeping Lubrication
In 1959 McCutchen proposed that a synovial joint could 'be thought of as a bearing
with a thick film of lubricant, where 'weeping' through the porous wall supplies
enough liquid to maintain the (fluid) film'. (McCutchen 1959) He surmised that
application of a load to the cartilage must pressurise the liquid within it and, if the
cartilages were permeable enough, the liquid would flow out to the rubbing
surfaces, thus carrying the load (Figure 4.6.). Using this information he tried a
82
system using Rubazote (closed cell rubber foam which has a pocketed surface,
lubricated with soapy water) and obtained coefficients of friction of less than 0.003
when the load was first applied. This slowly rose as the lubricant leaked away. The
lubrication system with which he worked was a form of hydrostatic lubrication in
which the interstitial fluid of hydrated articular cartilage would flow out onto the
bearing surface when a load was applied to it. The cartilage would act as a self
pressurising sponge. When the pressure was released the fluid would flow back
into the cartilage, i.e. self-pressurised hydrostatic lubrication. Because the cartilage
would appear to be weeping under load, this mechanism was termed weeping
lubrication. In 1967, McCutchen coupled this basic description of weeping
lubrication with a unique form of boundary lubrication known as osmotic
lubrication (4.2.3.1.).

Figure 4.6: Weeping or Hydrostatic Lubrication; a) as load is applied
fluid flows towards the rubbing surfaces at high pressure, carrying the
load with minimal friction. b) when the load is removed the cartilage
expands, drawing in synovial fluid. (Adapted from (McCutchen 1978))

Later, with Lewis, (Lewis & McCutchen 1959) McCutchen tested his weeping
lubrication theory using cartilage. They showed that cartilage does exude fluid as
load is applied and, under typical joint loads, exudes enough fluid to form an
effective lubricating surface layer. Later, in 1970, Radin et al concluded that the
'hydrostatic or "weeping" mechanism is a significant functioning process in joints'.
(Radin, Paul et al. 1970)

83
4.2.2.3 Elastohydrodynamic Lubrication
A variation of the hydrodynamic and squeeze-film modes of fluid-film lubrication
occurs when bearing surfaces are elastic enough for the lubricant pressure
generated by motion under a given load to cause elastic deformation of one or both
of the opposing surfaces i.e. both the resistance due to the viscosity of the lubricant
(synovial fluid) and the elastic deformation of the bearing surface (articular
cartilage), playing a prominent role in the lubrication process (Mow & Mak 1987).
Dintenfass, in 1963, was probably the first in the biological field to highlight the
significance of the deformability of the articular cartilage in fluid-film mechanisms.
He postulated that the elastic deformation of the cartilage could
spread the load to a larger bearing area, thus reducing the velocity necessary to
maintain a fluid film between the bearing surfaces (Figure 4.7.). His theory
depended on 'the existence of a thixotropic and elastic fluid (synovial fluid)
between the articular surfaces, that the area of the load-carrying film depends on
the elasticity of the cartilage, and that the velocity gradient existing in the gap
between the articular surfaces depends also on the lateral deformation of these
surfaces' (Dintenfass 1963). Dintenfass found support through Medley, Dowson
and Wright (Medley, Dowson et al. 1984).

Figure 4.7: Elastohydrodynamic Lubrication; The top surface in this
diagram is deformable, providing a larger fluid film when load is
applied.(Adapted from (Dowson, Wright et al. 1969))

As with hydrodynamic lubrication, the thickness of the film developed from
elastohydrodynamic lubrication would not always be adequate to avoid wear and
tear of the cartilage asperities (Tanner 1966; Dowson 1966-67; Higginson 1978;
Dowson, Unsworth et al. 1981; Mow & Mak 1987). One model which was claimed
84
to overcome the problems of fluid thickness was a model derived from EHL called
microelastohydrodynamic lubrication (mEHL). Microelastohydrodynamic
lubrication refers to the elastic deformation of the asperities (Figure 4.8.) of the
articulating surfaces under load, decreasing the risk of surface contact and allowing
a thinner fluid film to be maintained. In 1970 Bennett and Higginson suggested that
microelastohydrodynamic lubrication is present in the human synovial joint
(Bennett & Higginson 1970). They were later supported by Dowson and Jin who
found that the microelastohydrodynamic action effectively smooths the rough
cartilage surfaces allowing a sufficiently thick fluid film to separate the surfaces of
the articulating bone ends (Dowson & Jin 1986).

Figure 4.8: Microelastohydrodynamic Lubrication; Diagram showing
the deformation of the asperities on the surface of the cartilage under
load decreasing the risk of contact and allowing maintenance of a thinner
fluid-film. (Adapted from (Dowson & Jin 1986))

4.2.3 Mixed Lubrication Models
In fact, problems with the theoretical predictions of film thickness in synovial
joints has been a major stumbling block for many of the proposed lubrication
theories and is probably the strongest argument for a mixed lubrication regime
within the joint.

It has been said that either hydrodynamic lubrication or boundary lubrication by
themselves are far too crude to represent the complex lubrication mechanism in
joints (Dintenfass 1963). This has led to several mixed lubrication models, more
recently called Adaptive Multi-mode Lubrication (Murakami, Higaki et al. 1998)
which is a combination of many different lubrication regimes occurring (Fig. 4.9)
at the one time or individual regimes occurring throughout the loading cycle.

85

Figure 4.9. Mixed lubrication showing that one lubrication regime does not
answer the operating conditions in the joint but a combination of mechanisms.
Source: (Panjabi & White 2001)

As discussed in 4.2.1. Jones used a finger joint as the pivot of a pendulum to see
how the friction varies and whether or not it is proportional to the fixed load. The
results suggested mixed lubrication, with Jones concluding that human joints are
usually lubricated by fluid film lubrication but, when speed and/or eccentricity are
not enough to maintain a fluid film, a form of solid friction must occur, that is, the
surfaces come into contact. (Jones 1936) Jones' work on mixed lubrication has
received much support including Linn (Linn 1968), Dowson (Dowson 1973)) and,
Murakami et al (Murakami, Higaki et al. 1998).
86

Figure 4.10: Mixed Lubrication; When the fluid film fails, friction is
prevented by boundary lubrication. (Adapted from (Dowson, Wright et
al. 1969))

4.2.3.1 Osmotic Lubrication
In 1966 McCutchen suggested that mucin molecules might adsorb to the articular
surfaces at several particular points along the molecule leaving loops of mucin
chain still in solution. This would leave a higher charge density at the cartilage
surface compared to that in the bulk fluid, creating the effect of an osmotic
gradient. When a compressive load is applied to this layer of dissolved mucin
chains, it is resisted by a stress with the same origin as osmotic pressure; that is, the
tendency of solutes to distribute themselves evenly in a solvent (Davis, Lee et al.
1979). For this reason, McCutchen called this theory osmotic lubrication
(McCutchen 1966). It was McCutchens investigation into his osmotic theory that
led to the discovery that synovial fluid will only lubricate effectively after
sufficient soaking and that the friction will increase if resoaking is not allowed.
Both of these discoveries provide strong support for boundary lubrication.

4.2.3.2 Squeeze-Film Lubrication
Fein (1967) was the first to demonstrate the importance of such a mechanism in
synovial joint lubrication. He concluded that 'synovial joints are probably squeeze-
film lubricated with the squeeze film being replenished by hydrodynamic action
(entrainment of fluid when the joint is moved)'. The approaching surfaces generate
87
pressure (Figure 4.10.) in the lubricant as they squeeze it out of the area of
impending impact between them while the resulting pressure keeps the two
surfaces apart. The fluid must be squeezed out before the surfaces can come in
contact. This mechanism would fail if the maximal load were applied continuously.
However, if the load is only for short periods with the load being reduced while the
film remains reasonably thick, the film thickness can recover. In cartilage-on-
cartilage systems the film that forms in the transient area of impending contact has
been referred to as the squeeze film. Unfortunately, the work of Fein (1967)
ignored the permeability of the cartilage to water and small solutes. The effect of
cartilage permeability in a squeeze-film bearing would usually be to increase the
rate of leakage, thus reducing the time that such a bearing could support load
(Swanson 1979). However, in the articular joint, where synovial fluid fills the joint
cavity, the loss of water and small solutes would leave the hyaluronate and protein
between the bearing surfaces to potentially assume some role in lubrication. Such
an effect is covered in detail in the next section.

Hou et al and Hlavcek have both performed mathematical analyses of the squeeze
film theory, concluding that squeeze film lubrication is likely to occur but is most
probably supplemented by either boosted or boundary lubrication (Hou, Mow et al.
1992; Hlavacek 1995). Squeeze film lubrication occurs when a film of viscous
fluid is caught between a pair of normally approaching surfaces. The pressure due
to the applied normal force resists the tendency for the surfaces to approach each
other. (Archibald 1969)

Figure 4.11: Squeeze Film Lubrication; The arrows indicate the
movement of fluid away from the load-bearing region leaving an
enriched film of synovial fluid between the surfaces. (Adapted from
(Hou, Mow et al. 1992))
88

4.2.3.3 Boosted lubrication
Walker et al (1970) suggested that as load increases the low molecular weight
fraction of the synovial fluid is forced into the cartilage pores leaving trapped pools
of enriched synovial fluid in the rough surfaces of the cartilage. This theory was
called 'boosted' lubrication. This is a combination of both boundary and fluid
conditions, proposed independently by Maroudas (Maroudas 1967) and Walker
(Walker, Dowson et al. 1968). As the articulating surfaces approach each other, the
water and small solutes (<5m diameter) in the synovial fluid are uniformly
squeezed out from between the region of contact and taken up into the cartilage
matrix. The closing gap, as load is applied, between the bearing surfaces increases
the resistance to sideways movement of the fluid from the gap tangent to a point
beyond the resistance of flow into the cartilaginous bearing material normal to the
articulating surface (Figure 4.11.). The effective pore size of healthy, non-
pathogenic cartilage (20A to 65A) (McCutchen 1962; Dowson, Wright et al. 1969)
does not allow passage of the macromolecular components of the synovial fluid,
including hyaluronic acid (HA) or the HA-protein complex (HAP), into the
cartilage matrix. The articular surface acts as an ultrafiltration membrane so that a
highly concentrated and viscous layer is left at the cartilage surface. Walker et al
postulated that this gel layer was capable of carrying much greater loads for a
much longer time than normal synovial fluid with the diluted concentration of
HAP. It may be that the micro asperities at the articular surface provide traps for
the gel, providing enhancement for the formation of gel pockets at the joint surface
(Walker, Dowson et al. 1969; Walker, Sikorski et al. 1970). More recently, Tandon
et al have proposed a mathematical model that supports Walker's theory of boosted
lubrication (Tandon, Bong et al. 1994).
89

Figure 4.12: Boosted Lubrication; a) path of the fluid flow into the
cartilage surfaces while loaded. b) schematic diagram of the pools of
enriched synovial fluid formed on the surface of the cartilage. (Adapted
from (McCutchen 1978)

4.2.4 Boundary Lubrication Models
Charnley first suggested human joints acted under pure boundary lubrication in
1959 (Charnley 1959). Even before examining the problem experimentally there
were several theoretical criticisms which make the hydrodynamic theory unlikely.
Firstly, the articular cartilage, which is present as a layer about 3.25mm thick over
the sliding surfaces of the joint, is very resilient, being easily indented by the
pressure of the thumb-nail. An ankle joint of an adult male has a projected surface'
area of less than 13cm
2
, so that a man weighing 75kg carrying a 45kg load on his
shoulders will expose an ankle joint to pressures of about 10kg/cm
2
. There can be
little doubt that these resilient surfaces are intimately applied to each other over the
whole area even when not carrying loads. Secondly, there is the well-known fact
that hydrodynamic lubrication is not suited to conditions where the motion is
reciprocating, because no sooner has a fluid wedge been established for motion in
one direction than it is destroyed as motion starts in the other. Thirdly,
hydrodynamic lubrication is not easy to achieve with slow-moving surfaces under
heavy loads. In addition the in-viscous nature of synovial fluid and its shear
thinning behaviour would increase the difficulty of forming a fluid film. In further
support of boundary lubrication, Dowson showed that the fluid film thickness can
be maximised by increasing the femoral head diameter (Dowson, Fisher et al.
1991) but in considering the evolution from large femoral heads to smaller femoral
90
heads in prosthetic design this further reduces the chances of a suitably thick fluid
film ever forming (Hutchings 2003). Wang et al noted that the larger the femoral
head the worse the fluid-film lubrication condition (Wang, Essner et al. 1998).

Charnley came to this conclusion of pure boundary lubrication when he discovered
that there was little variation in friction with speed in pendulum experiments
similar to Jones (Unsworth 1991). Caygill et al came to the same conclusion in
1969 and suggested that hydrodynamic lubrication be discarded in favour of
alternative mechanisms (Caygill & West 1969). Boundary lubrication requires both
strong adsorption to the bearing surfaces and strong cohesion between the adsorbed
polymers. Boundary lubrication is said to occur when the surfaces of the bearing
solids are separated by a film of molecular proportion which are bonded in some
way to them. This type of action is more suited to slow reciprocating motion under
heavy loads.

Figure 4.13: Boundary Lubrication; The surface layer prevents the
articulating surfaces from coming into contact.(Adapted from (Wright &
Dowson 1976))

The essential requirement for boundary lubrication is the provision of a boundary
lubricant. Although the operating conditions of the joints indicate boundary
lubrication the existence of a boundary lubricant is necessary to validate the theory.

4.2.5 The Search for the Boundary Lubricant
It was obvious to everyone working in the field of joint lubrication that the
lubricant for fluid film lubrication had to be synovial fluid. However, the identity
of the boundary lubricant was not nearly so obvious. The search for an actual
91
lubricating factor, as required under boundary conditions, began well after the
start of the search for the lubrication mechanism itself. During the 1960s, research
groups were starting to analyse the synovial fluid in greater detail as it was
becoming more widely accepted that a boundary mechanism had a role in the
lubrication of the joint. The lubricant had to be contained within the synovial fluid
and also be located at the articulating surface. Initially there was little difference in
the fractions of synovial fluid that were identified but slowly it became easier to
characterise these fractions and to identify the role of each fraction. This led to the
discovery of fractions within fractions, i.e. sub-fractions. Gradually the identity of
the indigenous lubricant has emerged.

The role in lubrication of the macromolecular components of synovial fluid was
investigated because these constituents, collectively known as synovial mucin, are
one of the distinguishing features of synovial fluid when compared with blood
plasma (McCutchen 1962). Linn and Sokoloff analysed the fluid exuded from the
cartilage and found that it was essentially water and electrolytes (Linn & Sokoloff
1965). If the superior lubricating abilities of synovial fluid as compared to water
are supplied by the mucin, it is obvious that their supply is not from within the
cartilage. Thus, this work confirmed the general opinion that the source of the
lubricating factor is the synovial fluid.

Experiments to test synovial fluid lubrication have, as described, often used
systems involving cartilage rubbing on glass or cartilage rubbing on cartilage using
either saline or synovial fluid as the lubricant. Neither of these systems eliminates
the role of the interstitial fluid and the possibility of boundary lubrication from
substances adsorbed to the cartilage surface and so are not conclusive in identifying
the active lubricating mechanism. However, they are good systems in which to
compare the lubricity of different solutions and surfaces. Several experiments
(Ropes, Robertson et al. 1947; McCutchen 1966; Tanner 1966; Wilkins 1968) have
used alternative systems; for example, rubber rubbing on glass with either synovial
fluid or saline as the lubricant. The trial of McCutchen (1966), using rubber on
glass, demonstrated that synovial fluid was a superior lubricant, compared to 0.9%
saline, in the absence of the cartilage. Some of the other experiments which used
92
articular cartilage for at least one of the rubbing surfaces did not show a clear
difference between synovial fluid and saline, at least not over a short period of time
(Little, Freeman et al. 1969; Radin, Swann et al. 1970). Unsworth and Dowson et al
wiped the cartilage surface dry before testing. Under a load of 800N or greater, the
dry cartilage performed equivalent to that lubricated with synovial fluid
(Unsworth, Dowson et al. 1975). Hence, the low coefficient of friction measured at
the joint surface would seem, to some extent, to depend on substances left on the
cartilage surface.

As just mentioned, McCutchen performed lubrication tests on the synovial fluid in
a system using surgical rubber against glass microscope slides. He found that
lubrication was poor until the rubber and glass had been in contact with the
synovial fluid for some minutes, indicating that a chemical reaction must occur
between the rubbing surfaces and the lubricant, as might be expected if a boundary
lubrication mechanism were acting. McCutchen further analysed the synovial fluid
using ultrafiltration of the synovial fluid with filters of varying porosity
(McCutchen 1966). He demonstrated that use of a 0.22m filter retained the
lubricating ability in the residue, a larger porosity (0.65m) allowed all of the
lubricating ability to be completely filtered. With a 0.45m filter, the lubricating
ability was split between the filtrate and the residue. When using hyaluronidase
treated synovial fluid, a 0.1m filter system divided the lubricating ability between
filtrate and residue. This suggested that the lubricating ability of the synovial fluid
was related to the hyaluronic acid, the enzymatic digestion breaking up the
lubricating molecules.

This implication of the hyaluronic acid, the component of synovial fluid that
imparts viscosity to the synovial fluid, in joint lubrication was supported in general
by most other studies preceding this work (e.g. (Ogston & Stanier 1953)), and
several more which followed McCutchens study (e.g. (Chikama 1985; Laurent,
Laurent et al. 1996; Marshall 2000; Mori, Naito et al. 2002)). But as discussed in
the overview (2.1.1.) HA does little more than effect the viscosity of SF in terms of
lubrication as it has been shown to fail to lubricate under any significant load.

93
4.2.5.1 Enzyme Studies
Research following on from this work of McCutchen also treated synovial fluid
with various enzymes to selectively destroy first the polysaccharide and then the
protein (Linn 1968; Linn & Radin 1968; Wilkins 1968), in order to determine
which part of the mucin component was essential for boundary lubrication.
Cleavage of the polysaccharide, using a 10 minute digestion protocol, with bovine
testicular hyaluronidase left a watery product which lubricated as well as whole
synovial fluid, something already noted by McCutchen, (1966). This was further
evidence that the viscosity or rheology of synovial fluid was not the vital
component in the joint lubrication system. However, extended digestion with the
hyaluronidase (43 hours at 38C, pH 5.2) did affect the lubrication abilities under
boundary lubrication conditions in the study by Wilkins (1968), suggesting that, to
some degree, hyaluronic acid is required. However, these results must be
considered in light of the experimental conditions. The 10 minute digestion was
performed at 25C and at neutral pH. The conditions for the extended digestion
were vastly different and could have had more widespread effects on the synovial
fluid than the short digestion. Linn and Radin found no deterioration in lubricating
ability following extended digestion (24 hours) under the same conditions (Linn &
Radin 1968).

Wilkins found that the separation of mucin from the synovial fluid by ultrafiltration
followed by treatment with proteolytic enzymes destroyed the lubricating ability
under conditions of boundary lubrication, even though the viscosity remained
completely intact (Wilkins 1968). The ultrafiltration was necessary as whole
synovial fluid contains protease inhibitors (Davis, Lee et al. 1979). Treatment with
papain also destroyed the lubricating ability of the synovial fluid. This strongly
indicated that the protein part of the synovial mucin was essential to its lubricating
ability, even though it accounts for only a small part of the total molecular weight
of the synovial mucin. Wilkins went on to suggest that the protein must act in a
relatively small domain, possibly serving as the anchor by which mucin clings to
the surface or as a link to some small, undiscovered anchoring group. Hence, the
intact protein component of the synovial fluid is essential for boundary lubrication.
While the hyaluronic acid of the mucin complex also seems to be necessary, its
94
natural length can be considerably shortened before any detrimental effects on
boundary lubrication are obvious.

Jay (Jay & Cha 1999) was able to show that phospholipase digestion didnt cause
the friction to increase by much whereas trypsin digestion caused the friction to
increase dramatically in contradiction to Hills (Hills & Monds 1998) studies which
showed the opposite. This has led to some debate (Hills & Jay 2002) and is still a
matter of current research.

4.2.5.2 Lubricating Glycoprotein
Radin et al pursued the implication that protein was essential for lubrication of the
cartilage (Radin, Swann et al. 1970). To test whether or not a protein fraction was
the active lubricant, the group had to separate the hyaluronate component of the
synovial fluid from the protein component without disrupting the physiochemical
properties of either component. This was achieved through density gradient
sedimentation equilibration. All of the hyaluronate banded in the middle of the
gradient, while the gross protein content was present in the top layer of the
gradient. The lubrication abilities of the different fractions were tested using a
modified Linn arthrotripsometer, which enabled the measurement of in
continually oscillating joints (Linn 1967; Radin, Paul et al. 1970). The hyaluronate
fraction showed no lubricating advantage over buffer, but the protein fraction had
lubricating abilities equivalent to that of whole synovial fluid. Further
characterisation of the protein fraction continued and a rigorous purification and
separation protocol was developed for the synovial fluid. The major glycoprotein
constituent was isolated from the gross protein component and shown to contain
boundary lubricating ability equivalent to that of whole synovial fluid (Swann &
Radin 1972; Swann, Sotman et al. 1977; Swann, Hendren et al. 1981). Because
plasma proteins were not known to lubricate, this one was considered to be
unique to the joint and was termed a lubricating glycoprotein (LGP) or Lubricin
(Swann, Slayter et al. 1981).

Following thorough efforts to characterise this new glycoprotein, it was established
that carbohydrate constituents represent approximately 44% (w/w) and amino acid
95
constituents approximately 43% (w/w) of the molecule. However, 9.2% -13% of
the molecule remained unknown (Swann & Radin 1972; Swann, Sotman et al.
1977; Swann & Mintz 1979; Swann, Hendren et al. 1981; Swann, Slayter et al.
1981; Swann, Silver et al. 1985). Although the lubrication analyses of the LGP
fractions were performed under boundary conditions, the ability of the LGP to bind
to a surface had not been demonstrated until Swann et al carried out binding studies
using iodinated LGP to investigate the ability of the LGP to adsorb to cartilage.
This study indicated that 14% of the radioactive LGP was able to bind to the
cartilage.

The implication of Lubricin as a potential boundary lubricant for AC was discussed
by several authors (Swanson 1979; Swann, Bloch et al. 1984; Furey & Burkhardt
1997; Jay 2004), but no attempts were made to verify the presence of Lubricin on
the AC surface (Sarma, Powell et al. 2001).

At the time when the LGP molecule was being characterised, Davis and co-workers
were studying the detailed mechanisms involved in boundary lubrication for the
excised disks of bovine nasal cartilage (Davis, Lee et al. 1979). They used an
apparatus that was designed to discourage hydrodynamic effects and proposed the
existence of a thin, viscous structured hydration shell at the articular surface with
multiple layers of LGP held together by an alternating sequence of hydrophobic-
hydrophobic and hydrophilic-hydrophilic bonds. Davis et al proposed that one
portion of the LGP was adsorbed to the surface of the cartilage (the behaviour one
would expect of an amphipathic molecule at an interface). These mutual
electrostatic repulsive forces between the charged polysaccharide components
could enhance the boundary lubrication process. This model of the LGP molecules
being bound in layers to the articular surface has now reached a reasonable level of
acceptance in both the physical sciences and biology (DeHaven 1990; Jay 1990;
Mow, Ateshian et al. 1993).

4.2.5.3 Lipids
The presence of traces of fat in bovine synovial fluid was reported over a century
ago (Frerichs, 1846, cited by (Bole & Peltier 1962)). Stockwell (Stockwell 1965)
96
estimated the total fat content of articular cartilage to be about 1-2% of the wet
mass and McCutchen (McCutchen 1978) has commented on the oily nature of
the cartilage surface. Despite the accepted existence of lipid within the joint, there
has been little interest in it in terms of lubrication, despite the fact that fats are well
known to be slippery. The first study to investigate the role of lipid in joint
lubrication was that of Little et al (Little, Freeman et al. 1969). By using the hip
joint as a pivot of a Stanton pendulum, they showed that synovial fluid was a
lubricant but it was not the only lubricant in the joint. The team felt that there may
also be an intrinsic factor within the cartilage which would aid lubrication. Simple
tests were designed utilising synovial fluid and Ringers solution followed by a
study of the oscillation pattern of the joint. These showed linear decay of free
oscillation independent of the load or the rate of movement. This was a strong
indicator for the lubrication mechanism being boundary in type.

Little et al noted two factors during the experiments that indicated the presence of a
lubricant in, or adsorbed to, articular cartilage. (1) A very low could be obtained
using Ringers solution (rather than synovial fluid) on the cartilage and, (2) was
greatly increased following soaking of the cartilage surfaces with a fat solvent (2:1
chloroform/methanol). The extraction of fat left the cartilage unchanged both
histologically (as assessed using the following stains; haematoxylin, eosin,
toluidine blue and alcian blue) and in gross appearance. Further, the permeability
and compressive stiffness of the cartilage was not altered.

Little et al showed that fat could be readily demonstrated histologically in cartilage.
Staining cartilage with Sudan Black and Sudan 111 revealed fat at the superficial
layer of the articular cartilage. Preliminary work treating cartilage surfaces with
wheat germ lipase and also carbon tetrachloride produced increases in the
coefficient of friction similar to rinsing the cartilage surface with fat solvent.
Littles team concluded that the lubrication was boundary in nature, even in the
presence of synovial fluid, and that the lipid present in articular cartilage enhanced
the lubrication mechanism by lowering the frictional characteristics of the two
surfaces. Unfortunately, despite the clear implications provided for a role of lipid as
the active boundary lubricating ingredient in the synovial fluid, this study into the
97
role of lipids stood virtually on its own until the 1980s. Hills reignited the interest
in the role of lipids, in particular for lubrication within the body in the early 1980s
as best described in his book The Biology of Surfactant (Hills 1988). His work on
lipids carries on to this day.

More recently Benz et al confirmed some of Littles findings in that indicated a
strong dependence on the lipids present on the cartilage surface and the removal of
which (with a 2:1 mixture of chloroform: methanol) caused an increase in (Benz,
Chen et al. 2005). Amongst other things, their results confirmed the predominant
role played by the surfaces rather than the fluids between them in joint lubrication.

Another recent study concluded that: A layer of phospholipids is present on the
surface of articular cartilage. This layer can clearly be viewed in the SEM and
OM.The lipid layer acts as a boundary lubricant and is critically important to the
proper functioning of synovial joints (Ballantine & Stachowiak 2002). They
suggested that proper functioning of synovial joints requires the presence of the
articular cartilage surface lipid layer and that removal or damage to this layer is a
key factor in the onset of OA.

4.2.5.4 Surface-active Phospholipid
One of the types of lipid present in the synovial fluid that could be readily acting as
a lubricant is phospholipid (3.4.5.) which constitutes at least one third of the lipids
present in the synovial fluid of the normal healthy joint (Rabinowitz, Gregg et al.
1984). Phospholipids are used in industry to enhance the boundary lubricating
abilities of oils primarily to reduce both the break-out force and, in consequence,
wear (Munro 1964; Larson & Larson 1969). Although few studies have been
performed on the lubricating abilities of surface-active agents in biology, those that
have been completed have shown promising results. Gvozdanovic et al (1975)
investigated the formation of lubricating monolayers at the cartilage surface during
their search for suitable synthetic synovial fluids (Gvozdanovic, Wright et al.
1975). They compared the effect of several surface-active substances with synovial
mucin using the cartilage on glass system to assess friction. An anionic surfactant
(sodium lauryl sulphate) behaved in a similar manner to the mucin at pH 7-10, but
98
a cationic surfactant (cetyl-3-methyl ammonium bromide) was not as effective and
required a pH of 6. In each case the initial coefficient of friction was low, typically
around 0.003-0.004, but it increased with the time of loading, the increase being
faster with the anionic surface-active chemical. It was concluded that electrostatic
forces alone could not account for the observations and that monolayers of
boundary lubricant formed from the surface-active materials were bound to the
cartilage more strongly than the synovial mucin. Gvozdanovic et al (1975) had
noted that cartilage surfaces needed prolonged washing before the lubricating
ability was eliminated. The same was found following the trials of surface-active
substances. When using the surface-active solutions, the value of mu decreased as
loads increased, far exceeding normal physiological loads.

In their studies, Gvozdanovic et al (1975) made no mention of the highly surface
active phospholipid which is already present in the synovial fluid; yet
phospholipids offer much potential for providing boundary lubrication in vivo. In
fact the unidentified 9.213% portion of LGP discussed in 4.2.5.2 has been
identified to be lipidic in nature (Schwarz & Hills 1998) and it seems more than
coincidental that this is approximately the same amount of Lubricin that was shown
to adsorb to the cartilage surface. Hence these lipids have been implicated as the
active boundary lubricant in synovial fluid and labelled Surface-Active
Phospholipid (SAPL). A review of boundary lubrication by SAPL will be
discussed in the following chapter that will show that the surface-active
phospholipids found in vivo have the capability of providing good boundary
lubrication at high load, as is necessary at times in the articular joint.

4.3 Artificial Joint Lubrication: A Review
Much of the discussion on the tribology of natural joints in the previous sections
applies equally here with the exception being that the replacement bearing does not
enjoy the unique properties bestowed by the original bearing material cartilage.
Therefore, the theories that relied upon the porous, complying nature of cartilage to
operate are not feasible in the artificial joint where the articular surfaces are
replaced with hard man-made materials.
99

In comparison to the lubrication of the natural joint the artificial joint has received
little attention. Wang concluded in his review that one of the areas in artificial joint
tribology that is still the least understood is the mechanism of lubrication (Wang,
Essner et al. 1998). The majority of tribology studies on artificial joints are wear
related and will be discussed in (4.3.4.).

There has been relatively little work on the determination of the lubrication regime
for joint prostheses (Dumbleton 1981; Jalali-Vahid, Jagatia et al. 2001). This is
because it has been felt that lubrication is, so to speak, the dependent variable,
being specified by the design of the prosthesis, the nature of the fluid present, the
patient activity and the materials and surface finish of the sliding surfaces. It has
been felt that the lubrication regime is of the boundary type and that the best that
can be hoped for is the adsorption of surface active molecules on the sliding
surfaces, thus allowing the continued operation of the boundary lubrication
(Dumbleton 1981). In fact, it was that type of reasoning which made metal/plastic
prostheses so attractive, since it was thought that such devices were "self-
lubricating" and so the presence or absence of lubrication from the joint fluid was
of no importance. Replacement joints usually operate in the presence of a fluid
similar to healthy synovial fluid, but it is important to note that the original design
intent for those joints made of a metal-on-plastic combination of materials was so
they could perform adequately as dry bearings (Dowson, Wright et al. 1969). As
Unsworth mentioned in his review on the tribology of artificial joints, Design for
lubrication using the bodys own lubricant has not, to date, been a feature of
artificial joints (Unsworth 1991). Recently there have been studies which indicate
that lubrication of the "mixed" type may be expected. Although the idea of fluid-
film lubrication in artificial joints has been explored, it faces more challenges than
fluid-film lubrication that has been shown in the natural joint due to the hard, non-
compliant materials used in total joint replacements.

In consideration of the artificial joint as a replacement bearing of the original, the
major and only difference is the bearing materials. The loading and operating
conditions remain the same and the same indigenous lubricant remains. The
100
bearing materials used in artificial joints have historically been selected for their
intrinsically low friction which typically means hard, metallic or ceramic materials
which are radically different to the original articular cartilage. The way that the
indigenous lubricant interacts with these surfaces is of interest as it is this
interaction that will determine the tribological performance and longevity of the
TJR.

Essentially the best that prosthesis designers with current technology can reasonably hope
for is boundary lubrication (Dumbleton 1981). Boundary lubrication may not be the
dominant mechanism in the natural joint but it is instrumental to the lubrication of
the joint under certain conditions as shown in the previous sections. Given this
knowledge TJRs will require the services of the identified boundary lubricant to
operate effectively in the boundary lubrication regime. Surfaces that can interact
favourably with the boundary lubricant are of utmost importance.

Many of the major problems encountered in prosthetic design are tribological in
nature. In the first place the rate of wear has to be sufficiently low to enable the
joint to operate satisfactorily as a bearing for the remaining life of the human
machine. Secondly, the friction force developed between the sliding surfaces
should be minimised, not only to provide freedom of movement for the patient, but
also to limit the severity of the problem of implant fixation.

Most replacement joints are either of the metal-on-metal or metal-on-plastic
varieties. If two metals are used it is customary to employ like materials to avoid
corrosion problems in the hostile environment of the body. Such a combination of
metals normally leads to high friction and is generally avoided in engineering
bearings. In the human joint many successful metal-on-metal bearings based upon
chrome-cobalt alloys have been designed and used. They have a relatively low
wear-rate but the friction forces and torques are high compared with metal-on-
plastic bearings. The friction coefficient for metal-on-metal prostheses has been
reported as high as 0.3 and as low as 0.04 for a Charnley type metal-on-plastic
prosthesis (Unsworth 1991). This value reported for a Charnley TJR is quite
remarkable and has not been seen to be as low by other researchers. Figure 4.14
101
indicates a range of values collated and their comparison to the natural joint.
Regardless of the actual values for the coefficient of friction, metal-on-plastic
prostheses offer the better frictional performance but unfortunately wear is still
known to occur with consequential effects.

Figure 4.14. Dependence of the efficiency on the friction coefficient in natural
and artificial joints. Source: (Gavrjushenko 1993)

More recent attempts to improve the lubrication of artificial joints have included
attempting to mimic the articular cartilage with elastomeric bearing surfaces and
even attempts to regrow the original bearing surfaces with a biological equivalent
via tissue engineering (Shi 2004).

It is well established that boundary or at best mixed lubrication regimes exist in
artificial joints. However all models will be presented here.

4.3.1 Fluid Film Models
Several groups have investigated the possibility of hydrodynamic lubrication in
artificial joints (Scholes & Unsworth 2000; Jalali-Vahid, Jagatia et al. 2001; Jin,
Medley et al. 2002). The development of a fluid film is possible in artificial joints
102
under certain conditions; however, it has been concluded that the calculated film
thickness is not greater that the asperities and hence asperity contact will still exist
even though reduced by a fluid film in UHMWPE hip joint replacements (Jalali-
Vahid, Jagatia et al. 2000).

Other groups have promoted EHL through the use of elastomeric bearing surfaces
that mimic the mechanical properties of articular cartilage but have faced
difficulties with attaining the desired lubrication under all operating conditions (Jin,
Dowson et al. 1997; Oka, Kumar et al. 2000; Virdee, Wang et al. 2003; Scholes,
Unsworth et al. 2005).

The search is continuing for biomaterials that can effectively mimic the properties
of cartilage and provide efficient lubrication under all operating conditions of the
joint. One such group (Williams, Powell et al. 1995) has attempted to improve the
poor tribological performance of these elastomeric bearing surfaces when a fluid-
film is not present by modifying the surface to adsorb DPPC. This is the
development of a model that covers more of the operating conditions of the joint.
Fluid-film lubrication via EHL when the velocities in the joint a sufficiently high
enough and boundary lubrication via surfactant at low joint velocities. This can be
regarded as a hybrid lubrication model combining the principles of both fluid film
lubrication and boundary lubrication which by nature will cover the middle ground
of mixed lubrication also.

4.3.2 Boundary Lubrication Models
It is well established that boundary lubrication or mixed lubrication is the best that
can be hoped for in the artificial joint (Unsworth 1975; 1978; Dumbleton 1981).

Boundary lubrication is more likely to occur in TJR because the bearing surface is
no longer cartilaginous but hard and therefore non-compliant and non-porous.
Because of this the lubrication regimes that relied upon the properties of cartilage
struggle to exist in the artificial joint; for example, interstitial pressurisation,
weeping, osmotic and boosted, essentially all the regimes that were fluid-film
103
related. As discussed in the previous section the calculated film thicknesses for
artificial joints is not great enough to cover the asperities on the bearing surfaces
and hence the last line of defence in lubrication falls to boundary lubrication.
Thankfully effective boundary lubrication is known to exist in the natural joint
(4.2.3.) and the artificial joint is also able to benefit from the bodys provision of a
boundary lubricant. In a study the first of kind, Purbach et al were able to
demonstrate the presence of surface active phospholipids on the surface of
retrieved hip implants (Purbach, Hills et al. 2002), thus proving that an indigenous
boundary lubricant is present even in arthritic joints and that the synovial cells
continue to produce lubricant.

An interesting study by Kobayashi et al directly observed the lubricating behaviour
of various joint surfaces indicating that, in fact, the cartilaginous surfaces did
benefit from a fluid-film between the surfaces whereas the artificial materials did
not (Kobayashi & Oka 2003). They predicted that boundary lubrication was all that
could be expected as direct contact of the bearing surfaces did occur. They
concluded that to improve the quality of artificial joints the characteristics of the
implant material surface and the synovial macromolecules must be considered i.e.
boundary lubrication.

Several groups have attempted to modify the surface of biomaterials to enhance
boundary lubrication (Williams, Powell et al. 1995; Williams III, Gilbert et al.
1997; Widmer 2002; Benz, Chen et al. 2005; Heuberger, Widmer et al. 2005;
Serro, Gispert et al. 2006). This was achieved by surface modification techniques
discussed in further detail in the next chapter with the goal of attracting the
lubricating molecules present in synovial fluid to the surface of the bearing
materials that would act as a protective film and hence reduce friction.

4.3.3 Mixed Lubrication Models
As discussed in the natural joint lubrication sections mixed lubrication occurs in
the joint when the operating conditions cause the mode of lubrication to change
from boundary to fluid-film. The same applies in artificial joints although as
104
explained above, the likelihood of a fluid-film being generated at all is low and
hence the best that the artificial joint can hope for is a mixed regime, whereby some
of the load is carried by a thin fluid-film and the remainder is handled by the
boundary lubrication mechanism (Unsworth 1995; Jalali-Vahid, Jagatia et al.
2001).

4.3.4 Tribological Studies for total joint replacements
Research is ongoing in the field of tribology of artificial joints with the goal of
improving the quality and durability of prostheses. The main focus is the
development of new biomaterials that have excellent tribological performance. The
research of the tribology for artificial joints is generally conducted in two areas: the
testing of suitable biomaterials for their frictional performance and the testing of
prostheses for their wear determination.

To study the tribology of biomaterials simple laboratory testing machines are
employed called tribometers (Figure 4.16.).

Figure 4.15. Geometric configurations of various tribometers. Source:
(Dumbleton 1981)
(a) pin-on-flat (c) annulus-on-flat
(b) pin-on-disc (d) disc-on-plate

105
To study the tribology of prosthetic joints in conditions similar to those prevailing
in the human body, joint simulators are necessary. The simulator study should
produce wear mechanisms, wear rates and wear debris similar to those seen
clinically. It is difficult to evaluate various simulators because of the lack of
consistent test parameters and the fact that even different tribocouples react
differently to the same lubricant (Brown & Clarke 2006).

The lubricant is a crucial parameter in the tribological studies for both the material
testing using tribometers and the simulator studies with prosthetic joints (Ahlroos
2001) and is one consideration in this thesis. Given that synovial fluid is the only
source of lubricant in the body and this is what natural and artificial joints must
operate with, it would follow that this is what should be used in tribo tests for
artificial joints. Unfortunately, as shown by Brown & Clarke in their excellent
review of the lubrication conditions for tribo testing, synovial fluid, even though
being the biological lubricant, it is present in far too small of quantities to be used
exclusively in tribo tests (Brown & Clarke 2006). Animal synovial fluid such as
bovine synovial fluid and equine synovial fluid have been used but are also
impractical due to not only the lack of quantity but the collection process which is
vulnerable to contamination. In addition, body fluids from other species may not
correlate to the human body and disease may also play a role (Coller 2002). Thus,
researchers have had to adopt alternative fluids for use as a lubricant in tribo tests.
So-called pseudo-synovial fluids (PSF) choices have included de-ionised water,
mineral oil, gelatin solutions, physiological saline, bovine serum, plasma, and
artificial lubricants (Brown & Clarke 2006). Bovine serum had emerged as the
lubricant of choice for wear simulator studies and the others have been deemed
inappropriate for use as tribo testing lubricants (Brown & Clarke 2006). Bovine
serum had been selected purely for the reason that it produces results most closely
matching clinical results with polyethylene bearings. Conversely the other
lubricants were deemed not appropriate because the wear results they produced did
not match clinical results. For example Ahlroos found that DPPC produced
negligible wear but yet concluded that DPPC was not important to the tribology of
prosthetic joints because the zero wear results did not match the much higher
clinical wear results! (Ahlroos 2001).
106

The key to selecting a suitable lubricant for tribo testing is to provide a lubricant as
close as possible to the biological lubricant. Several boundary lubricants have been
proposed and used in tribo testing as indicated in Table 4.1.

Table 4.1. Boundary lubricants within SF suitable for tribo tests. Source:
(Brown & Clarke 2006)

Brown & Clarke concluded that proteins may be essential in lubricating metals and
ceramic bearings but that they actively promote and increase the wear of polymeric
surfaces.

A standardised test lubricant has been proposed by several researchers including
Liao et al (Liao, Benya et al. 1998) which would certainly help to qualify test
results across the world.

107
4.4 Summary
In applying the principles of lubrication engineering to the natural joint it becomes
apparent that not one but all lubrication mechanisms known to the engineer would
be needed to explain the excellent tribological performance of this biological
bearing. The extremely low friction achieved by the joint would suggest fluid-film
lubrication but the relatively slow joint velocities, high velocities and stop/start
motion do not support this but rather a boundary lubrication regime. Currently
there is no single comprehensive theory that covers all the friction and lubrication
characteristics existing within a joint cavity. The tribological efficiency of synovial
joints is most probably the result of a series of complicated dynamic interactions
occurring between the synovial fluid, macromolecular components, the articular
cartilage surface and matrix and the interstitial water. The unique lubricating
properties are not solely due to either the synovial fluid, or to the articular cartilage
but to some complex synergistic interaction of both.

It is clear that a combination of lubrication regimes exist when considering the
operating nature and loading of the joint within an ambulation cycle. The
lubrication for the periods of extended standing, heel strike and toe off can only be
explained in lubrication engineering by a boundary lubrication mechanism. The
swing through and weight transfer phase may be answered by a fluid-film
mechanism or a mixed lubrication regime. Plainly boundary lubrication has an
instrumental role to play in the lubrication of the natural bearing if only for those
times of high loads and very low velocities because the near frictionless
performance of the bearing continues through these times. This indicates the
presence of some substance that keeps the bearing surfaces sliding over each
other so effortlessly. This substance or protective film/coating is referred to as the
boundary lubricant. For effective boundary lubrication to occur an effective
boundary lubricant must exist. One such boundary lubricant identified in the joint
is surface active phospholipids (SAPL).

Artificial joints do not enjoy the same excellent tribological performance of the
natural joint due to the replacement of at least one of the cartilage bearing surfaces
108
with a hard, metallic or ceramic surface. The lubrication of the replacement bearing
falls into the mixed regime at best and boundary lubrication becomes more crucial
in controlling wear and hence the long term performance of the artificial bearing.
Thankfully the lubricating ability of pathologic synovial fluid is not lost due to OA
and surface active phospholipids are available to potentially benefit the lubrication
of the total joint replacement.

Even though much research has focused on designing implants to encourage fluid
film lubrication, there is little doubt that the surfaces will make contact throughout
the walking cycle. It is at these times that boundary lubrication will occur and a
boundary lubricant will be necessary to facilitate this. Some groups have studied
compliant bearing surfaces in aid of elastohydrodynamic lubrication but still resort
to providing a surface that the boundary lubricant (SAPL) can adsorb to.
Chapter 5: Literature Review Boundary Lubrication for Artificial Joints
109
Chapter 5
Literature Review Boundary
Lubrication for Artificial Joints
This chapter gives further consideration to boundary lubrication, boundary
lubricants and their relationships to surfaces like those in artificial joints. It will
discuss surface chemistry and tribochemistry with a focus on the role of
phospholipids for lubrication. The design of the chapter is to give an understanding
to the surface interaction of the lubricating fraction of synovial fluid.

Boundary lubrication is the antithesis of full-film lubrication in that it is the quality
of the substances comprising the sliding surfaces and the quality of the fluid used
as a lubricant which are the essential features; the shape of the surfaces and
viscosity of the fluid are unimportant. An example of the lack of correlation to
viscosity is the fact that a watery solution of oleic acid is a better lubricant for glass
sliding on glass than pure glycerine, though the latter is many times more viscous.
In this type of lubrication, the property of 'oiliness' is fundamental; however what is
most important is not the 'oiliness' of the lubricant itself but the property of oiliness
when applied to the sliding surface. Thus methylated spirit has no more suggestion
of oiliness than water when tested between the fingers, but for rubber it is a better
lubricant than water. This illustrates the important fact that in boundary lubrication
a lubricant has an affinity for the surface it lubricates so that when motion takes
place between two such lubricated surfaces it takes place between monomolecular
films of lubricant chemically adhered to the underlying surface. It is obvious that
monomolecular films which are bound to the sliding surfaces are less likely to
rupture under heavy loads than films of a lubricant which is inert towards the
substance of the surface, because the integrity of such a film depends only on the
intermolecular attraction in the fluid itself and the intermolecular attraction of a
fluid is obviously less than that of a solid. The molecules of a fluid which are
bound to the surface of a solid lose the physical properties of a liquid. In the case of
mineral oils, which have no chemical affinity for metals, the addition of small
110
quantities of fatty acids greatly enhances their lubricating properties; this is
explained by soaps being formed between the fatty acids and the metal with the
result that motion takes place between two layers of molecules all with their -
COOH groups attached to the metal surface and the fatty chain projecting like the
bristles of a brush (Charnley 1959).

Surface chemistry, boundary lubrication and tribochemistry are highly inter-related
and it is difficult to discuss any one without the others. Friction is the largest real
world example/use of surface chemistry. Boundary lubrication in artificial joints is
thought to be achieved by surface active lipids; lipids fall under the categories of
surfactants, colloids and monolayers which are themselves a subcategory of surface
chemistry.

Boundary lubrication by detailed definition is tribochemistry and tribochemistry is
the application of surface chemistry for tribological purposes. Boundary lubrication
may be the most complex of all the lubrication regimes as it involves these other
sciences that deal with the nanoscale interactions of surfaces and the lubricants,
with the purpose of determining the interactions that predominate and which forces
are at play. Boundary lubrication at this level may well be beyond the knowledge
of the engineering tribologist and maybe more fitting for a surface, tribo or even a
colloidal chemist. This chapter will cover the science but in consideration of the
engineer and his role to apply this science.

Considering that the mode of lubrication is most likely boundary in nature for
artificial joints and that an active lubricant of the joint has been identified, it is now
of utmost interest to investigate the interaction of this lubricant and the biomaterial
surfaces. As explained previously boundary lubrication is not about the bulk
properties of the lubricant but rather the ingredients of the lubricant and the
interaction of each constituent with the surfaces with which they come into contact.

In the previous chapters, lipids and, more specifically surface active phospholipids
(SAPL) have been identified as an active boundary lubricant in both the natural and
111
artificial joint; this chapter will discuss the interaction of SAPL and the bearing
surface.

5.1 Surface Chemistry
Surface chemistry can be broadly defined as the study of chemical reactions at
interfaces. It is the study of how certain substances interact with a surface. The
adhesion of gas or liquid molecules to the surface is known as adsorption. This can
be due to either chemisorption or by physisorption.

5.1.1 Surfaces and Surface Energy
Living material contains many boundaries. The boundaries are usually maintained
by membranes interfacing with each of the fluid compartments that they separate;
others are the natural interfaces that would exist between different phases even if
no membrane were to exist. Such phases include fat, aqueous fluids, various solids
and semi-solids, and the air or water of the external environment.

Each molecule at one of these interfaces possesses additional energy by virtue of its
location because it is surrounded by similar molecules on one side only. Thus, it
experiences an imbalance of intermolecular forces in contrast to the uniform field
of forces that would apply were it surrounded by a homogenous medium. For
liquids in air, this interfacial energy is manifest as surface tension, the property
that, for example, enables a steel needle to be floated on still water or enables a
pond skater insect to walk on water. Although some surfaces may demonstrate
surface energy in more obvious ways than others, all surfaces possess energy, even
solids (Hills 1988).

Most substances can modify the surface energy of an interface to some degree but
some are much more effective than others and these are termed surfactants (surface
active agents). Surfactants tend to locate at air-liquid interfaces because this
reduces the surface energy, or boundary tension, of the interface (Adamson & Gast
1997). To be effective at an interface involving a solid, the surfactant needs to be
112
directly adsorbed (bound) to the solid surface. When surfactants are adsorbed to
solid surfaces they impart many highly desirable properties (Barnes & Gentle
2005) some of which have immediate physiological applications (Hills 1988).
Some of these attributes will be covered in greater detail later. When adsorbed to a
solid surface, surface-active agents completely change the nature of a surface, i.e.
hydrophilic surfaces will be rendered more hydrophobic.

5.1.2 Hydrophobic vs Hydrophilic
Since water is the predominant substance in all tissues and fluids in the body, the
affinities of various substances tend to be expressed relative to water. If they are
highly compatible with water they are termed hydrophilic from the Greek meaning
water loving. When substances repel water, they are termed hydrophobic from the
Greek meaning water hating.

An important note is that the liquid can behave very differently when close to a
surface in comparison to the bulk and that the water or liquid structure near a single
surface is changed, often quite significantly, when another surface is near-by
(Benz, Chen et al. 2005).

Water comprises 50-80% of most living creatures. Hence, water is almost
invariably one of the phases present wherever interfaces occur in vivo and
consequently wherever surfactants are acting. The fact that water is incorporated
into a structure or that the material has an affinity for water is an important
indication that its surfaces are likely to be hydrophilic and will therefore tend to be
compatible with aqueous solutions or, if a solid, its surfaces should be readily
wetted by aqueous solutions. As a simple indication of compatibility, a droplet of
water will spontaneously spread over a very hydrophilic surface. By contrast, the
same droplet would not wet a hydrophobic surface but would bead up, as observed
when a raindrop falls on a cabbage leaf. In reality there is a spectrum of degrees of
wettability ranging from perfectly hydrophilic to very hydrophobic which can be
very conveniently quantified by this simple example as the contact angle (5.1.5.)
113
imposed upon the droplet by the surface. Thus, the interfacial energy is very low
for compatible phases, but becomes higher as the surfaces become less compatible.

5.1.3 Surface Tension and its Measurement
The molecular forces which impart energy to an interface simply because it is an
interface have already been mentioned, along with an outline of the role of
surfactants in modifying the resultant interfacial energy. To quantify the effects of
surfactants, it is necessary to be able to measure the energy of an interface in the
absence of surfactant.

Definitions
The energy per unit area associated with an interface is usually denoted by E with
the two phases in contact denoted by suffixes for air or any gaseous phase (A),
liquid (L) or water or any aqueous phase (W), oil or lipid (0) and solid or
membrane (S).

Surface energy may be relatively easy to conceptualise with diagrams of molecular
forces but, in the real world, it is much easier to perceive surface tension as the
force retaining a pond skater insect on the surface of a pool of water. It has long
been recognised that surface tension () is a manifestation of surface energy (E)
and can be equated under isothermal conditions (Adamson & Gast 1997). Both E
and have the same dimensions. E is usually expressed in erg cm
-2
and in dyne
cm
-1
which is actually the same because, by definition, 1 erg is the work needed to
move a force of 1 dyne by 1 cm (Barnes & Gentle 2005). Before any further
discussion of the methods of measuring surface tension, an explanation is
necessary of the relationship between the interfacial energies at points where three
phases meet.

5.1.4 The Young Equation
Many methods of measuring surface tension create a periphery at which the liquid
grips the solid. In cross section this is seen as the triple point where solid, liquid
114
and air meet (Figure 5.1.) where is known as the contact angle, i.e. the angle
between solid:air and solid:liquid phases at that point. can have any value
between 0 and 180 depending on the nature of the phases (Phillips & Riddiford
1967). Resolving forces in the plane of the solid surface, which is not deformable,
there is a force of
SA
pulling to the left. Pulling to the right is the solid:liquid
tension
SL
plus the component of the liquid:air surface tension (
LA
) resolved in
the plane in which the balancing forces have a net contribution of
LA
cos. For
equilibrium, the forces pulling to the left will equal those pulling to the right when:

Equation 5.1.
SA
=
SL
+
LA
cos

This is generally known as the Young equation and states that the competition
between the cohesive forces of a liquid to itself and the adhesive forces between the
liquid and solid surface result in a contact angle which, at equilibrium, is constant
and specific to the particular system (Gellman & Spencer 2005). The importance of
the Young equation is that it enables solid surface energies to be determined using
the contact angle technique (Israelachvili 1992; Shpenkov 1995; Adamson & Gast
1997; Barnes & Gentle 2005; Gellman & Spencer 2005; Stachowiak & Batchelor
2005).

Figure 5.1. The triple point in cross section. Depicting the balance of forces at
the edge of a droplet where the liquid, solid and air all meet to subtend a
contact angle ().
115
5.1.5 The Contact Angle
Equation 5.1 can be rearranged to express the difference between the dry and the
wet surface tension of the solid (
SA
-
SL
) as
LA
cos. Since
LA
is the liquid-air
interface tension, i.e. the readily measured surface tension of the liquid, the term
LA
cos is easily determined if is known. The contact angle can be measured
directly via a goniometer; this enables (
SA
-
SL
) to be determined and, hence, the
energy difference (E
SA
- E
SL
). This difference is also known as the work of
adhesion (Adamson & Gast 1997; Barnes & Gentle 2005) since it represents the
work done in breaking an adhesive bond affected by the liquid. The difference is
also an important parameter for the following reason.

A solid surface which is incompatible with water is termed hydrophobic. Thus it
has a high interfacial energy when in contact with water (E
SL
) while it is
compatible with air, so that the interface now has a lower surface energy (E
SA
).
This means that the difference (E
SA
- E
SL
) decreases substantially as the surface
becomes more hydrophobic. If the liquid is water and its surface tension remains
constant, the Young equation predicts that cos will decrease and even go negative
if E
SL
>E
SA
. Since the angle increases as the cosine decreases, this explains the
larger contact angles observed on more hydrophobic surfaces. The contact angle is
a good measure of the everyday experience whereby water tends to bead up on
waxed surfaces and, in the light of the Young equation, represents a good index of
surface hydrophobicity (Hills 1988).
5.1.6 Surfactants
The interfacial energy can be greatly modified by substances with amphipathic
properties. Molecules with combinations of moieties (ends) which have an affinity
for the phase in which they are dissolved and moieties which tend to be repelled by
the medium are termed amphipathic. Amphiphilic by definition is a molecule
having a polar, water-soluble group attached to a nonpolar, water-insoluble
hydrocarbon chain. These ends are covalently bound to each other, one being
hydrophobic and the other hydrophilic. When located at an interface the molecule
will locate with its ends orientated to minimise interfacial energy. Certain
116
chemical groups can be selected as likely for the moiety with an affinity for the
medium and another set for the antipathic moiety (Barnes & Gentle 2005). These
are bound by a strong covalent bond, i.e. it can only be broken by strong chemical
agents.
5.1.7 Electrical Charge
The hydrophilic moieties fall into four categories, with respect to electrical charge.
They may carry no net charge (nonionic), a negative charge (anionic) or a positive
charge (cationic). In the fourth case, the moieties carry both charges; the
phosphoric acid end of the phospholipid molecule can still ionise giving a negative
charge which is comparable in magnitude to the positive charge of the amine or
quaternary ammonium ion, to produce a charge dipole and a molecule classified as
a zwitterion. Zwitterionic is best explained by a zwitterion which is a dipolar ion
that is capable of carrying both a positive and negative charge simultaneously
(Barnes & Gentle 2005).
5.1.8 Adsorption
To change the surface tension of a solid the surfactant molecule attaches to the
surface by any of a variety of chemical and physical bonds known as adsorption.
Adsorption is a common process, especially the weak physical (Van der Waals)
type (physisorption). The much stronger type, chemical adsorption (alias
chemisorption), can be effected when one of the chemical groups (moieties) in the
molecule forms a reversible bond with the surface which is chemical in nature. The
widespread occurrence of adsorption is generally not obvious to an observer as like
tends to attract like; so, in the case of adsorption of the usual homopathic molecule,
it presents a surface not unlike the one it is covering. If, however, the molecule is
highly amphipathic, especially in its affinities for water, adsorption can be readily
recognised (Hills 1988).

The attachment of a surfactant molecule by electrostatic attraction between the
polar moiety and a fixed charge on the surface orientates the molecule with its non
polar group facing outward. This presents a new surface with a total change in
personality from the one covered. In the case of hydrophobic solids the
117
hydrocarbon moieties will be attracted and adsorbed to the surface to leave the
polar ends outwards rendering the surface hydrophilic. Such agents are then acting
as wetting agents. On the other hand, hydrophilic surfaces will attract the polar
ends which then adsorb to the surface. This orientates the hydrocarbon end
outwards rendering the surface hydrophobic. Thus, droplets of water which would
have spread spontaneously on the hydrophilic surface now bead up to display a
contact angle which, as previously discussed, provides a convenient index of the
change in surface energy upon the surface.

Often adsorption of synthetic surfactants in vitro does not stop at the monolayer
state but proceeds to multilayers (Dowson, Priest et al. 1998; Barnes & Gentle
2005). When these layers are well defined and widely spaced, they can be effective
boundary lubricants. This process is known as lamellated solid lubrication
(Stachowiak & Batchelor 2005).

5.2 Tribochemistry
Tribochemistry is basically a sister science to both boundary lubrication and
surface chemistry. It takes the principles of surface chemistry and applies them to
the field of friction, lubrication and wear. Thus, in combining tribology and
chemistry, it is the area where these two sciences overlap that becomes
tribochemistry.

Tribochemistry generally refers to the chemistry that occurs between the lubricant
(and/or the environment) and the rubbing surfaces under boundary lubrication
conditions. This includes specific reactions that occur only under rubbing
conditions and reactions that would occur independently under the temperatures
and pressures in the contact. The reactions that take place only during rubbing
usually involve direct chemical interactions with the surface. The reactions that
would occur independently can be defined as contact chemistry (oxidation, thermal
degradation, catalysis, polymerisation). The two sets of reactions are intimately
intertwined and one affects the other.

118
There are two possible sources of tribochemistry: the mechanically induced
chemistry (fresh nascent surface, electron emission) and the thermally induced
chemistry at the asperity tips due to flash temperatures. Specific attribution of a
reaction to these two possible sources has not been delineated. Speculations abound
in the literature but experimental difficulties prevent a clear definition of this
important issue.

The definition of tribochemistry is not very precise, even though most
researchers declare that their motivation for studying tribochemistry is to
understand the boundary lubrication mechanism. Under the banner of
tribochemistry, researchers have studied the nature of friction polymers, surface
oxidation reactions and environmental reactions with water.(Hsu, Zhang et al.
2002)

Surface tribochemistry refers to the physical adsorption or chemisorption of the
boundary lubricant and its relationship to wear (Pawlak 2003). Much of this has
already been covered in the surface chemistry sections and further discussion will
follow in subsequent sections.

Tribochemistry is a new area of research. It is useful to the engineer considering
the finer details of boundary lubrication such as the mechanism of lubrication for
the synovial joint, in particular the artificial replacement.

5.3 Surfactants for Boundary Lubrication
Surfactants, both anionic and cationic, are frequently used in many aspects of
industry. Probably the largest source of surfactant use in industry is as detergents
(Shaw 1992; Barnes & Gentle 2005). Common soap is the anionic surfactant best
known as a detergent for washing ourselves and clothing. Soap is essentially the
sodium salt of the fatty acid, most commonly stearic (C
18
) but also palmitic (C
16
)
and some unsaturated fatty acids (Hills 1988). Unfortunately the detergency of the
salts of these fatty acids, which are derived from fat, suffers due to the insolubility
of their calcium salts. The great economic incentive to find acids whose calcium
119
salts were soluble created a surge in surfactant technology which led to the
development of synthetic surfactants, many of which were found to have other
highly desirable properties, creating an enormous industry which touches almost
every other industry.

Another large user of surfactants is the food industry which uses Lecithin
(phospholipids) in substances requiring a natural emulsifier, release agent and/or
lubricant. For example, lecithin is the emulsifier that keeps cocoa and cocoa butter
from separating in chocolate.

Although it was the need for good detergents and emulsifiers which prompted the
development and production of synthetic surfactants early in the last century, it was
soon found that these new products needed little modifying for applications in
other areas. These included lubrication, protection from wear, inhibition of
corrosion, water repellency, modification of permeability, defence against
biological invasion, release (anti-stick) action and viscosity modification (Larson &
Larson 1969; Biresaw 1989).

Several of these applications, especially lubrication, anti-wear and release are
beneficial within the biological system. Indeed, there is now considerable interest
in the roles of surfactant in biology.
5.3.1 Lubrication via Surfactants
Surfactants are used in industry under conditions where boundary lubrication is
required (Chung, Homolam et al. 1991; Pawlak 2003; Liang 2004). The primary
requirements for an effective boundary lubricant are the following:

(1) Strong binding to the surface as found with chemisorption.
(2) Adsorption in an adequate quantity to form a continuous monolayer.
(3) Strong cohesion of the adsorbed film to prevent penetration by asperities
on the counterface -a primary consideration in reducing wear (Briscoe,
1980- cited by (Hills 1988)).
120
(4) Mechanisms should be provided for replenishing the monolayer or wear
and friction will increase with time.

Fatty acids have been used as lubricants in the form of soaps since ancient times.
Lead soap is used in gear oils (Munro 1964), the intermeshing of gears representing
a classical application of boundary lubrication in mechanical engineering. Fatty
acids are of interest as they represent the hydrocarbon moieties as found in
phospholipids (Figure 5.2.). Studies in vitro of their lubricating properties have
demonstrated that lubrication improves as chain length increases. The best fatty
acids for maintaining good lubrication were found to be palmitic (C
16
) and stearic
(C
18
) (Dowson, Priest et al. 1998; Barnes & Gentle 2005). Studies of fatty-acid
monolayers adsorbed to metal surfaces may seem far removed from any situation in
vivo, but they do provide information about packed hydrocarbon moieties which
constitute the exposed surface in many situations.

Having outlined the basic principles of boundary lubrication (see also 4.1.2), it is
now appropriate to discuss surfactants in vivo which have also displayed boundary
lubrication ex vivo.
5.3.2 Biological Surfactants
It is worthwhile briefly tracing through the history of how research in surface-
active agents and biology came to meet, prior to pursuing the current scientific
interest surrounding surfactants, especially phospholipids. Apart from free-fatty
acids and phospholipids acting as emulsifying agents at oil-water interfaces, the
major interest in surfactants in biology has centred around the lung and surface
activity is only mentioned in most physiological texts with respect to the lung.

A role for surface-active substances in the lung was derived largely by inference.
The earliest implication of surface activity in the lung can be traced to von
Neergaard (1929, cited by (Hills 1988)) who found that the pressure required to
inflate an excised lung with an aqueous fluid was about one-quarter to one-third of
the pressure required to achieve the same volume change using air. He compared
the lungs to soap bubbles, and interpreted his results as though the liquid used for
121
inflation had filled the bubbles, thus eliminating the air-liquid interface and the
associated collapsing pressure. Von Neergaards classical study was seldom
referenced for a number of years following its publication in 1929 and it was some
time before surface activity of the alveolar wall again attracted attention.

Pattle (Pattle 1950) had been interested in the control of foaming in the lungs
following chemical injury using certain agents categorised as anti-foams. Pattle
(Pattle 1955; 1956) found that these agents did not break the foam expressed from
rabbit lungs and was most impressed by the stability of this foam which persisted
for days. It was argued that this extreme stability was only possible if the surface
activity was near zero, since small bubbles with a normal surface tension shrink
rapidly. Pattles observations led him to propose the existence of a powerful
surfactant. This growing interest in the possible presence of a surfactant in the
lungs attracted the interest of (Clements 1957) and (Brown, Johnson et al. 1959)
who, using the standard tools employed by surface chemists, decided to
investigate the surface-activity of lung extracts. They found that the surface tension
of water could be greatly reduced by lung secretions and washings, thereby
confirming the presence of a highly potent surfactant by standard physiochemical
methods. Following extraction of the surfactant from the lungs, (Pattle & Thomas
1961) and (Klaus, Clements et al. 1961) found it to be a lipoprotein rich in
phospholipid. Brown (Brown 1964) later determined the surface-active ingredient
to be dipalmitoyl lecithin, today more commonly termed dipalmitoyl
phosphatidylcholine (DPPC).

Following on from the detection of surfactant in the lungs the role of phospholipids
was studied for their role in other locations in the body. Hills dominated much of
the research in the 80s and 90s with other groups becoming involved over that
time. Surfactants were considered for their role ranging from the boundary
lubrication of the pleurae (Hills, Butler et al. 1982) to gaseous and solute exchange
in the intestinal membrane (DeSchryver-Kecskemeti, Eliakim et al. 1989). Apart
from the work done on surfactant in the lungs the remainder of the work in the
literature is dominated by the role that surfactant plays in the lubrication of surfaces
in the body (eg. (Butler, Lichtenberger et al. 1983; Hills 1984; Hills & Butler 1984;
122
Hills & Butler 1986; Hills & Cotton 1986; Hills 1989; Girod, Zahm et al. 1992;
Williams III, Powell et al. 1993; Bernhard, Postle et al. 1995; Higaki, Murakami et
al. 1998; Ethell, Hodgson et al. 1999; Chen & Hills 2000)).

5.3.3 Types of Lipids
In order to be an effective surfactant, a molecule needs a substantial hydrocarbon
moiety which is readily provided in vivo by fatty acid chains. These are found in
four basic forms of complex lipids: acyiglycerols, phosphoglycerides,
sphingolipids and the waxes (Tro 2003; Muller 2004). These differ in the backbone
structure to which the fatty acids are covalently bound. Neither the acyiglycerols
nor the waxes are amphipathic, leaving just two lipid families of high surface
activity: the sphingolipids containing sphingosine as their backbone and the
phosphoglycerides, loosely referred to as phospholipids or phosphatides (Figure
5.2.).

123

Figure 5.2. General structure of phosphoglycerides, emphasizing their
amphipathic nature (Schwarz 1994). Various groups for X are given in Figure
5.3.

Phosphoglycerides
When one of the hydroxyl groups of glycerol is esterified by phosphoric acid,
phosphatidic acid is produced which provides the backbone for the
phosphoglycerides (PG). Two fatty-acid chains are normally attached to the
glycerol by esterifying the two remaining hydroxyl groups. One of the two
hydroxyl groups on the phosphatidic acid backbone can be replaced by
esterification of an alcohol (X-OH) to the phosphoric acid. X-OH can be any of six
124
or so alcohols: ethanolamine, choline (containing a quaternary ammonium ion),
inositol, serine, glycerol and diglycerol; see Figure 5.3.

Figure 5.3. Various polar head groups for the general phosphoglyceride
depicted in Figure 5.2. Source: (http://en.wikipedia.org/wiki/Membrane_lipids
2005)
Phosphatidylcholine
Phosphatidylcholines (PC) are the dominant species of PG in biology and is present
in most mammalian cell membranes. PCs are also the major constituents of
125
pulmonary surfactant, in particular DPPC (Brown 1964). PCs can either be
saturated or unsaturated. The term saturated and unsaturated phosphatidylcholine
refers to the presence or lack of a double bond in the fatty acid chains. When both
the fatty acid chains are saturated, it is called saturated phosphatidylcholine (SPC).
When one of both of the fatty acid chains are unsaturated, it is called unsaturated
phosphatidylcholine (USPC) (Chen, Hills et al. 2005). DPPC contains two
saturated fatty acid chains and is a SPC. USPCs are distinguished by a kink in at
least one of the fatty acid chains at the point of the additional bond (Figure 5.3.
Oleate). SPCs do not exhibit this kink and appear like Figure 5.2.

The highly surface-active surfactant dipalmitoylphosphatidylcholine (DPPC) found
in synovial fluid contains a highly positively charged quaternary ammonium (QA)
ion at its terminal group that is ideal for binding these small molecules to the
negatively charged surfaces.

Much research has been aimed at determining the types of lipids present in the
human body and more recently in particular the types of lipids present in the
synovial joint (Prete, Gurakar-Osborne et al. 1995; Ballantine & Stachowiak 2002).
Rabinowiz et al were the first to report on the lipid profiles of the tissues of the
knee joint. However they were not able to completely identify all lipid types or
their subgroups (Rabinowitz, Gregg et al. 1979). Similarly Sarma et al identified
some classes of lipids present in articular cartilage but not all (Sarma, Powell et al.
2001).

The term SAPL is used loosely. Although the title suggests surface active
phospholipids, its actual components are phosphatidylcholines which are two
subcategories below phospholipids; see Figure 5.3. It is suggested that SAPL be
retitled with SAPC (surface active phosphatidyl choline) for a more accurate
representation. However, SAPL will be used throughout this thesis even though it
refers to a group of PCs that make up part of phospholipids. SAPL is made up of a
number of PCs and it is has been assumed that DPPC is the dominant component as
it is in the lungs (Brown 1964). The actual detail of all the components of SAPL is
largely unknown.
126

5.3.4 Phospholipid Analysis
Very little phospholipid occurs as individual molecules in solution because their
solubility is so low. They are more likely to form micelles, vesicles, liposomes,
adsorbed layers, lamellar bodies or tubular myelin (Hills 1988). Phospholipid is a
major component of most biological membranes. Various phospholipids can form
associations with numerous substances including protein, bile salts and
polysaccharide, all of which can change the quantity determined by assay
depending upon the ability of the method used to break the chemical or physical
associations and to extract the phospholipid from the membranous material present.
The quantification (certainly estimations) of phospholipid can be achieved through
several methods, two of which are viz solvent extraction and chromatographic
separation of phospholipids. High performance liquid chromatography (HPLC) has
been found to be effective for the identification of lipids (Chen, Hills et al. 2005).

5.3.5 Adsorption in Biology
In the section on adsorption (5.1.8.) there seemed little doubt that adsorption could
occur very readily, at least in industrial situations. Adsorption is a very common
process and there is also substantial evidence for its occurrence at interfaces in the
body. Most surfaces are negatively charged, be they metal or a mucosal surface
with carboxyl or sulphonyl groups incorporated into its structure (Dorinson &
Ludema 1985; Hills 1988). There is a vast range of cationic surfactants which are
effectively adsorbed to negatively charged surfaces (Sharma; Adamson & Gast
1997; Barnes & Gentle 2005) and many of these possess the highly positively
charged quaternary ammonium ion which is very common in the form of the
choline group.

Surfactant monolayers adsorbed to solids can be encouraged to increase the
tightness of packing following the addition of cations. The interspersion of these
positive ions between the negative phosphate ions at the distal end of the polar
moieties would have the effect of pulling those groups together, Figure 5.4., which
127
imparts greater cohesion - a highly desirable feature for effective boundary
lubrication (Hills 1988). This role of cations (Shah & Schulman 1965; Watkins
1968) was demonstrated at the liquid-air interface by the reduction in the surface
area per molecule at the same pressure. Later the same forces have been used to
explain the packing of monolayers adsorbed to solids (Hills 1984; 1984; Hills &
Butler 1984).

Figure 5.4. A molecular model for the adsorption of phospholipid zwitterions
to a negatively charged surface in which cations in the plane of the phosphate
ions pull those ions together, thus enhancing close packing of both polar and
non polar moieties and imparting coherency. (Hills 1988)

One of the desirable properties imparted by adsorption of surfactants in industry,
which also plays a large role in biology, is lubrication.

5.3.6 Biological Surfactants and Lubrication
There are many pairs of surfaces in the body which need to move either by sliding
over each other or by separating to allow air, fluid or water to pass between them.
This movement needs to be achieved with minimal friction and wear and, when
appropriate, be facilitated by good lubrication. Hence, it is very interesting to find
many of the same surface-active phospholipids in many organs and in the fluids
adjacent to surfaces which, if they were to stick or rub, could severely compromise
128
physiological function (Rapport, Lim et al. 1975; Hills 1984; Hills & Butler 1985;
Hills & Butler 1986; Hills & Cotton 1986; Hills 1988). This opens up an interesting
new field for surfactant whose molecules are very similar to many of their synthetic
counterparts already used as lubricants and release agents with great success in
industry (Dorinson & Ludema 1985).

Of particular interest is the reduction of friction and wear in the synovial joint
where wear is known to have serious implications to the longevity of both the
natural and artificial joint (3.5 & 3.6.1.).

5.4 Boundary Lubrication via SAPL
Having outlined the theoretical aspects of surfactant adsorption to membrane
surfaces, it is now appropriate to consider the highly desirable properties which an
adsorbed layer might impart.

The theory of how phospholipids can facilitate sliding basically follows that of
boundary lubrication (a phenomenon experienced after touching a wet bar of soap
and squeezing the water from between ones fingers: the slipperiness remains).The
surface active phospholipids offer much potential for providing boundary
lubrication in vivo for several reasons:

(1) SAPLs are found in vivo adjacent to the moving surfaces of the body in
forms in which they are available in a surface-active form (5.3.6.) and they
have also been shown to display multilayer adsorption to biological
surfaces (Ueda, Kawamura et al. 1985; Hills 1990; Hills 1990; 1990).
(2) Most SAPLS possess polar moieties which readily bind to the negative
sites on most biological surfaces (especially mucosal and epithelial
surfaces) and artificial surfaces to affect strong adsorption (5.3.5.).
Cations interspersed in the plane of the phosphate ions will neutralize
these ions, effectively rendering the phospholipids cationic with
characteristically strong adsorption (Figure 5.4.).
129
(3) The fatty acids in the hydrophobic moieties are mostly of ideal chain
length for lubrication, namely C
16
and C
18
(5.3.1.).
(4) When adsorbed these long hydrocarbon chains are orientated and packed
in much the same way as free fatty acids (5.3.1.) and the lubrication is as
good, but the monolayer is more tightly bound and, therefore, more
durable.
(5) Durability is improved by the insolubility of these layers (Hills 1988).
(6) Phospholipids contain a phosphate group at the middle of the molecule
which is ionized at physiological pH for most locations. These ions
provide an ideal point for pulling neighbouring molecules together if small
cations are placed between them, thus imparting strong cohesion (Figure
5.4) in addition to adhesion (Hills & Butler 1984).

Experimental evidence is also in strong support of phospholipids being able to
lubricate in vivo by Hills and others (Moro-oka, Miura et al. 2000). When tested
using a simple apparatus to measure the effectiveness of surfactant, the reductions
in the coefficient of friction for a glass-on-glass system were DPPC (70%),
dipalmitoyl phosphatidylethanolamine (72%) and sphingomyelin (70%) according
to Hills et al (Hills, Butler et al. 1982). The coefficient of kinetic friction for DPPC
was found to be 0.02 under dry conditions, representing a 99% reduction in friction
(Hills & Butler 1986). Similar values were achieved using DPPC at 37C in the
method of (Radin & Paul 1971) in which lubrication under high stress (up to 13
kg/cm
2
compared with a typical loading in the human knee joint of 3 kg/ cm
2
) was
tested.

Surface-active phospholipid (SAPL) has been found to act as an effective boundary
lubricant in the lungs (pleural surfaces) (Hills, Butler et al. 1982), pericardium
(Hills & Butler 1986), gastrointestinal tract (Hills 1996), and tendons (Uchiyama,
Amadio et al. 1997).

Morphological studies have provided visible evidence of oligolamellar
phospholipid at the articular cartilage surface (Hills 1989; Hills 1990; Guerra,
Frizziero et al. 1996; Hills 2000) the alveolar surface (Ueda, Kawamura et al. 1985)
130
and the mesothelial surface of the visceral pleura (Ueda, Kawamura et al. 1986).
These lamellar structures bear a striking similarity to the structure of graphite and
MoS
2
, recognised solid lubricants used widely as boundary lubricants in industry
(Larson & Larson 1969). Hills suggested that this layer upon layer of SAPL may be
protecting the underlying surface by acting as sacrificial layers (Hills 1989)

Many studies by Hills and others (eg. (Little, Freeman et al. 1969; Gavrjushenko
1993; Higaki, Murakami et al. 1997; Hills & Thomas 1998; Hills 2000; 2002; Hills
& Jay 2002; Hills & Crawford 2003; Ozturk, Stoffel et al. 2004)) has led to the
identification of SAPL as a remarkable load-bearing lubricant in the normal joint
i.e. a vital active ingredient in the lubrication of joints. SAPL is also a very
efficient release (anti-stick) agent, which, as a thin lining, enables the articular joint
surfaces to violate the first principle of lubrication engineering by allowing
surfaces of the same material to slide over each other without binding.

Of particular interest to the lubrication of artificial joints was a study by Purbach et
al who were the first to report on the presence of SAPL on artificial joints
(Purbach, Hills et al. 2002). They rinsed the bearing surface of a number of
removed total hip replacement prostheses obtained at revision. Analysis of these
rinsings showed that sufficient SAPL was present to form oligolamellar layers on
the bearing surfaces. Purbach et al. suggested that, if indeed formed, this layered
structure of phospholipid would act similar to a lamellated-solid lubricant,
protecting the surface from wear and ensuring low coefficients of friction.

The body of studies exploring the lubricating potential of surface-active
phospholipids is steadily increasing, continually adding support to the theory that
surface-active phospholipids may be providing boundary lubrication in vivo.

5.4.1 SAPL and Wear in artificial joints
SAPL lubrication has attracted much attention from the groups studying wear in
prosthetic joints (Ahlroos 2001; Bell, Tipper et al. 2001; Calonius 2002). All
groups have shown remarkably low wear rates of UHMWPE when DPPC was used
131
as a lubricant. Bell et al showed that at higher DPPC concentrations (5% w/v) the
wear was reduced by some 25-50 times as compared to the control bovine serum
lubricant and by at least threefold at relatively low concentrations (0.05% w/v).
Ahlroos reported that in several tests using DPPC as a lubricant, the lubricant
completely prevented polyethylene transfer, practically no wear debris was
generated and that the surfaces looked unchanged under microscopy.

Even Hills proved the anti-wear efficacy of SAPL by subjecting it to a traditional
engineering wear test called a standard four-ball test, a test normally reserved for
qualifying extreme pressure (EP) lubricants. Despite using speeds and loads that
were "extreme" by engineering criteria, i.e., several orders of magnitude in excess
of physiological logical loadings, the indices of wear (mean scar diameters) were
comparable to those of the best industrial lubricants. Hills concluded with This
lubricant imparts not only phenomenal anti-wear properties, but also the
remarkable reduction in friction reported previously(Hills 1995).

5.5 Summary
Surface and tribo chemistry are necessary to understand the boundary lubrication
that is sure to occur in the synovial joint. Joint surfactant, or surface active
phospholipids, has been implicated as the indigenous boundary lubricant for the
body. It is speculated that SAPL will function effectively in the artificial bearing.
However the constitution of SAPL is still largely unknown.

132

Chapter 6: Scientific Paper I

133
Chapter 6
Scientific Paper I - The Role of SAPL as a
Boundary Lubricant in Prosthetic Joints

Lorne R. Gale, BEng(Hons), Rebecca Coller, BEng(Hons), Doug J. Hargreaves,
DPhil, Brian A. Hills
a
, DPhil, D.Sc., Sc.D and Ross W. Crawford, DPhil, MBBS.

School of Mechanical, Manufacturing and Medical Engineering, Queensland
University of Technology, Brisbane, Australia and the
a
Mater Medical Research
Institute, South Brisbane, Australia

Published in: Tribology International, (2007). 40(4): p. 601-606.

Authors Contributions:
Lorne R Gale: All experimental work, data analysis and manuscript composition.
Rebecca Coller: Advisory role (ex-student)
Brian A Hills: Guidance and assistance with contact angle measurements and
adsorption tests. Supervisory role.
Doug J Hargreaves & Ross W Crawford: Supervisory role

Corresponding author:
Lorne R Gale
School of Mechanical, Manufacturing and Medical Engineering,
Queensland University of Technology,
2 George St,
Brisbane, Australia
Tel.: +617 3864 2423
Fax: +617 3864 1469
Email: l.gale@qut.edu.au

Keywords: Boundary lubrication; SAPL; DPPC; Prosthetic joint; Synovial joint;
Biotribology

Chapter 6: Scientific Paper I

134
This article is not available here.
Please consult the hardcopy thesis
available from QUT Library
Chapter 7: Scientific Paper II

141
Chapter 7
Scientific Paper II Boundary lubrication
of Pyrolytic Carbon with Surface Active
Phospholipids: Tribological Assessment
for Artificial Joints

Lorne R. Gale
1
, Brian A. Hills
2
, Doug Hargreaves
1
, Ross Crawford
1
,
Jerry Klawitter
3

1
Medical Engineering, Queensland University of Technology, Brisbane, Australia
2
Mater Medical Research Institute, Brisbane, Australia
3
Ascension Orthopedics, Austin, Texas USA

Revised and resubmitted: Acta Orthopaedica (2007)

Lorne R Gale: All experimental work, data analysis and manuscript composition.
Brian A Hills: Guidance and assistance with adsorption tests. Supervisory role.
Doug J Hargreaves & Ross W Crawford: Supervisory role.
Jerry Klawitter: Provided PyC specimens

Lorne R Gale
Medical Engineering,
2 George St,
Brisbane, Australia
Tel.: +617 3864 2423
Fax: +617 3864 1469

Keywords: Boundary lubrication; SAPL; DPPC; Prosthetic joint; PyC; Pyrolytic
Carbon; Biotribology
Chapter 7: Scientific Paper II

142

Chapter 8: Scientific Paper III

149
Chapter 8
Scientific Paper III Boundary
lubrication of joints: Characterisation of
Surface-Active Phospholipids found on
retrieved implants

Lorne R. Gale
a
, BEng(Hons), Yi Chen
a
, DPhil, Brian A. Hills
b
, DPhil, D.Sc., Sc.D
and Ross W. Crawford
a,c
, DPhil, MBBS.

a
School of Mechanical, Manufacturing and Medical Engineering, Queensland
University of Technology, Brisbane, Australia

b
Mater Medical Research Institute, South Brisbane, Australia
c
The Prince Charles Hospital, Rode Rd, Chermside, Australia

Published in: Acta Orthopaedica, (2007). 78(3): p. 309-314.

Lorne R Gale: Experimental work, data analysis and manuscript composition.
Yi Chen: HPLC analysis
Brian A Hills: Guidance and assistance with surface rinsings. Supervisory role.
Ross W Crawford: Supervisory role

Lorne R Gale
2 George St,
Brisbane, Australia
Tel.: +617 3864 2423
Fax: +617 3864 1469

Keywords: Boundary lubrication; SAPL; DPPC; Prosthetic joint; Biotribology;
PC; USPC; SPC; phosphatidylcholine
Chapter 8: Scientific Paper III

150

Chapter 9: Scientific Paper IV
157
Chapter 9
Scientific Paper IV - Tribological Testing
of Saturated and Unsaturated Surface
Active Phospholipids: Implications for
artificial joints

Lorne R. Gale
a
, BEng(Hons), Yi Chen
a
, DPhil, Prasad Gudimetla
a
, DPhil, Doug
Hargreaves
a
, DPhil and Ross W. Crawford
a,b
, DPhil, MBBS.

a
School of Engineering Systems, Queensland University of Technology, Brisbane.

b
The Prince Charles Hospital, Rode Rd, Chermside, Australia

Submitted to: Proceedings of the Institution of Mechanical Engineers: Part H;
Journal of Engineering in Medicine

Lorne R Gale: Experimental work, data analysis and manuscript composition.
Prasad Gudimetla: Supervisory role and manuscript composition
Yi Chen: Chemistry role, mixed solutions.
Doug Hargreaves & Ross W Crawford: Supervisory role

Lorne R Gale
2 George St,
Brisbane, Australia
Tel.: +617 3864 2423
Fax: +617 3864 1469

Keywords: Boundary lubrication; SAPL; DPPC; Prosthetic joint; Biotribology;
PC; USPC; SPC; phosphatidylcholine
Chapter 9: Scientific Paper IV

158

Chapter 10: General Discussion
165
Chapter 10
General Discussion
The work presented for examination in this thesis started with a pilot study
(Chapter 6) to investigate the tribological interactions of Pyrolytic carbon, a novel
load-bearing biomaterial, and SAPL the implicated indigenous boundary lubricant
by performing adsorption tests, contact angle measurements and tribo tests.
Common prosthetic materials, stainless steel and UHMWPE were also included for
part of the testing. It was found that a synthetic SAPL in the form of DPPC
promoted much surface activity in all materials and dramatically reduced the
friction between each tribo pair when tested under boundary lubrication conditions.

The prerequisite for boundary lubrication is that the surfactant be strongly adsorbed
to the surface, and ideally, this adsorbed lining be highly cohesive such that an
asperity from the counterface is less likely to penetrate the lubricating lining. The
results from the contact angle measurements indicated high surface activity
bestowed by the application of SAPL to the PyC surfaces. The PyC surface was
transformed from a hydrophobic surface (large contact angle) to a hydrophilic
surface (zero contact angle). In fact, the PyC surface became spontaneously
wettable indicating that DPPC was acting as an excellent wetting agent and that a
DPPC coating/film was obviously present. This adsorbed coating would be of
limited benefit as a boundary lubricant unless it was shown to be tenaciously
adsorbed. The results obtained from the adsorption tests indicated that DPPC is
indeed strongly adsorbed to the PyC surface. More than 80% of the originally
applied DPPC remained after 18 hrs of vigorous shaking in a saline bath meaning
that DPPC is strongly retained by the PyC surface. From a surface chemistry
standpoint these tests intimate that DPPC was attaching strongly to the PyC surface
and providing high surface activity, what remained to be seen for the purpose of
this surfactant being an effective boundary lubricant when in the presence of PyC
was whether this coating/film could in fact reduce the friction between the PyC
surface and another.

166
Pyrolytic carbon has been used successfully for prostheses in the heart and in hand
surgery and is known to be long lasting/wear resistant material in these
applications. It also has potential as a novel load-bearing surface for prosthetic
hips and knees, especially if it can be demonstrated to produce minimal wear under
load. One way to achieve this is to effect excellent boundary lubrication which has
become the accepted mode of lubrication in the artificial joint. Effective boundary
lubrication reduces friction and, hence, abrasive wear. In addition, SAPL, as an
excellent release agent when adsorbed to a surface, will inhibit adhesive wear.

The later stage to the first study (Chapter 6) began by tribo testing the
UHMWPE/SS tribo-couple. The tribo tests were performed at two temperatures,
body temperature and room temperature, to determine if temperature had an effect
upon the frictional performance of the tribo-couple when lubricated with DPPC and
saline and when in a non-lubricated state using physiological saline only. The
results did not show a marked difference in frictional force between the two
conditions and it was deemed that for the purpose of the tribo testing, the data
collected was relevant to the physiological state even though the tests were
conducted at a temperature lower than the bodys. The remainder of the tribo tests
were all performed at room temperature for ease of experimentation.

The pin-on-flat tribometer chosen for the tribo tests was suited for measurements
within the boundary lubrication regime, that is, relatively high loads and slow
speeds. Testing of the SS/PE tribo-couple with DPPC validated the results obtained
in a previous study (Coller, Hargreaves et al. 2004) whereby the friction was
reduced by more than 50% when lubricated with DPPC and saline as compared to a
non-lubricated (physiological saline) state. This alone was a remarkable result
supporting the role of SAPL reducing the friction between a prosthetic material
couple. More outstanding was the results obtained from the tribometer for the
UHMWPE/PyC combination. The friction between the tribo-pair was reduced by
more than 75% with DPPC as the lubricant as compared to the control lubricant
saline.

167
The excellent results obtained from the tribo tests in the first study (Chapter 6)
provided sound evidence that DPPC was certainly aiding the lubrication of the
tribo-couples in what can only be described as boundary lubrication conditions. At
this early stage of the thesis DPPC was assumed to differ little from the native
SAPL whereby DPPC was considered to be the dominant portion of SAPL as it is
in the lung. In addition Purbach et al has identified the presence of SAPL on
retrieved hip prostheses (Purbach, Hills et al. 2002). Therefore if a SAPL
equivalent can act as an effective boundary lubricant in vitro with prosthetic
bearing materials there is little reason to think that the same would not occur in
vivo. The pilot study for Pyrolytic carbon returned encouraging results and it was
suggested that further testing be done to assess the suitability of PyC as a candidate
material for load bearing prosthetic joints.

The second study (Chapter 7) was a follow on study which investigated the
tribological interactions of various forms of PyC (coupled with PyC and
UHMWPE) and SAPL by performing adsorption and tribo tests. The various forms
were largely denoted by their surface roughness. The results from the adsorption
tests showed similar results to the previous study, that is, PyC showed an affinity
for DPPC, however in contrast to previous results the smoothest PyC surface
showed less retention than the similar surface in the previous study. Although the
surface roughness was the same in both cases (0.03m) the specimens were quite
different, the original was a circular disc whereas the specimen in this study was an
actual half crescent shape leaflet prosthetic heart valve. The difference in results
may be explained by two reasons; 1) the PyC heart prosthesis was physically much
smaller than the original circular disc and therefore the tenacity tests became more
difficult to perform and 2) the number of tests was greatly reduced in the 2
nd
study
from n=18 to n=4 hence reducing the accuracy of the results. The greater retention
by the rougher surfaces may also be explained by the larger surface area available
for the molecules to attach to as opposed to the lipids layering upon each other and
being more likely to detach from one another as on a smoother surface.

The results from the tribo tests showed an excellent tribological performance for all
cases where UHMWPE slid against PyC under the boundary lubricating influence
168
of DPPC. The trend in the results indicated that the smoother the PyC surface the
greater the friction reduction achieved by the lubricant. This follows for two
reasons; 1) in engineering generally smoother, harder surfaces are intrinsically
lower friction couples; and, 2) smoother surfaces have less surface area for the
lipids to attach to, therefore encouraging multiple layers which is conducive to
better lubrication (lamellated solid lubrication). Some results from the tribo tests
were inconclusive. In regards to all cases where PyC rubbed against PyC, SAPL
had little to no effect in reducing the friction between the tribo-pairs. This is best
explained in lubrication engineering by analysing the surface roughness of the
specimens and considering the hardness of the material as PyC is a hard material.
In all instances where PyC was coupled with PyC at least one of the surfaces were
relatively rough and hence only the highest asperities make contact under load.
Since little or no deformation can occur due to the high hardness the lubricant can
only protect the small areas of contact which are naturally under higher load. This
demands more than the lubricant can provide. More than likely, the lubricant, in
this case the surface active lipids, are quickly rubbed off from the peaks and
deposited into the valleys where they are of no effect and wear between the
contacting asperities will be likely to occur.

These overall results from the PyC studies indicate that regardless of the tenacity of
the lubricant (DPPC) to the surface which in most cases is quite high and shows
high surface activity, it is the smoothest PyC surfaces that show the greatest
reduction in friction. Although the additional manufacturing process of polishing
PyC is labour intensive and hence expensive, it is strongly suggested that as a
candidate material for future prosthetic joints that polishing be done so that
indigenous SAPL can ideally reduce friction and wear the greatest amount in a PyC
bearing.

Pyrolytic carbon has received large merit as a biomaterial in the body for many
years where it has been used primarily for heart valves. Wear resistance of the
pyrolytic carbon is excellent. The strength, stability and durability of Pyrolytic
carbon are responsible for the extension of mechanical valve lifetimes from less
than 20 years to more than the recipients expected lifetime (More, Haubold et al.
169
2000). This property alone makes PyC attractive for use in the body where wear is
known to be an issue. It is interesting to note that some PyC heart valves have
PyC/PyC hinge joints that allow the valve leaflets to open and close about 40
million times a year and have shown very little wear after 10 20 years of service.
Contact pressures at the PyC/PyC hinge joint are high due to the water hammer
effect at valve closure. Other studies by Ascension Orthopedics (unpublished) have
performed wear tests with PyC and noted remarkable anti-wear properties, superior
to other common prosthetic materials, even when bovine serum was used as the
lubricant. The tribological performance of PyC is excellent and its mechanical
properties are within acceptable ranges suitable for a load bearing joint making
PyC a highly recommended candidate material for future hip, knee and shoulder
prostheses.

It is worthwhile at this point to discuss a possible explanation of the lubricating
mechanism of SAPL and artificial surfaces. Firstly, there is an interesting
difference between what was found with PyC, SS and UHWMPE and how it
differs from traditional studies of adsorbed surfactants. Normally, surface chemists
select the polar 'head' of a surfactant molecule or chemically modify the surface
itself to obtain binding by adsorption. This orientates the non-polar group
outwards such that the surface is typically transformed from hydrophilic to
hydrophobic (Figure 5.4.). However, in the case of these artificial materials we are
finding the exact reverse. We start with a hydrophobic surface and then render it
very hydrophilic/wettable. This implies that the SAPL molecules are attaching
themselves to the surface by means of their hydrophobic/hydrocarbon moieties,
thus orientating the polar ends outwards to provide a highly wettable surface.
Surprisingly, this bonding is quite tenacious. This would also explain subsequent
blood protein adsorption that occurs in PyC heart valves in vivo according to the
literature. Regardless of how these lipids attach themselves to the artificial surface,
it is undeniable that they reduce friction extremely well. A lubrication model that
may explain this behaviour has been developed by considering the work done by
two groups in the literature (Davis, Lee et al. 1979; Benz, Chen et al. 2005). Since
SAPL renders the artificial surface spontaneously wettable there is considerable
attraction for water (the major constituent of synovial fluid) to be present at the
170
surface. Basically these groups show that water in confined spaces takes on
properties markedly different to that of the bulk and may act as a repulsive
lubricant; that is, water and the presence of surface active molecules keep the two
surfaces apart by a repulsive force. This suggests that SAPL may work in
conjunction with water to lubricate the bearing surfaces of artificial joints.

The third paper presented in this thesis (Chapter 8) was the first study to date that
has completely characterised the constituents of the SAPL adsorbed to the surface
of load-bearing artificial joints. These retrieval studies were also the first to identify
the presence of SAPL on the surface of knee implants. Previously SAPL had only
been shown on the surface of retrieved hip implants (Purbach, Hills et al. 2002) so
this finding adds further support to the role of SAPL as a boundary lubricant in vivo
for artificial joints. The results from the HPLC analysis of the adsorbed SAPL
identified eight different species of phosphatidylcholines with the majority being
made up by four of the PCs. Basically 90% of SAPL was equally made up of three
unsaturated PCs (USPC) and the remaining 10% was DPPC a saturated PC (SPC).
This finding closely matches the profile that was detected on other biological
surfaces outside of the lung including articular cartilage (Chen & Hills 2004) and
reinforces the fact that all non-lung locations are dominated by USPC. This is an
interesting finding as all previous studies for friction and wear have used DPPC
which was thought to be the dominant species within SAPL given that that was the
case in the lung. The reason for the difference in SAPL profiles between lung and
non-lung tissues has been proposed to be due to differences in the ability of
saturated and unsaturated PCs to reduce surface tension (Bernhard, Postle et al.
2001). Furthermore, DPPC has a gel-liquid crystal transition temperature of 41.5C
which makes it effectively a rigid molecule at body temperature and thus better
able to reduce surface tension. However, unsaturated PCs have phase transition
temperatures far below body temperature, which enables them to adsorb more
easily.

This third study (Chapter 8) also showed that SAPL, in particular the specific PCs
mentioned above, were detected on all prosthetic surfaces analysed. That is the
same profile that was noted on the various materials of CoCr, Ti, SS and
171
UHMWPE. The materials were divided into two groups for analysis: metallic and
polymeric. The statistical analysis did not show a significant difference between the
two groups indicating that the profile of SAPL was the same for both types of
materials. This interesting finding must be considered under the context of the
study which was to identify the profile of the constituents (their relative
concentrations) rather than the amounts of SAPL present. It is known that SAPL
has an affinity for particular surfaces (as shown by the PyC surface) and therefore
this finding does not mean that SAPL attaches equally to all prosthetic materials
but that a similar profile of SAPL is present on all surfaces. SAPL may well prefer
a metallic surface as could be expected but only a study on the amounts of SAPL
present on the respective surfaces would reveal that. It is difficult to analyse the
amounts of SAPL present on a prosthetic surface given the large range of prosthetic
types and sizes and make a meaningful comparison.

There has been a need for a lubricant suitable for the laboratory testing of
prostheses and their materials for quite some time. As stated by a prominent group
that studies the wear of prosthetic joints Unfortunately, wear studies are hampered
by the lack of a reliable, stable lubricant that accurately replicates joint fluid.
(Ahlroos & Saikko 1997). Often the results obtained from in vitro tests do not
match the results obtained in clinical studies for prosthetic joints. One major factor
identified in the literature that may play a role in the disparaging results is the
lubricant used for the tribo tests. Several lubricants are used but none that mimic
the lubricating portion of synovial fluid. It is agreed that in vitro testing should be
performed with a lubricant as close as possible to what the body provides and given
the impracticality of using synovial fluid an equivalent pseudo-synovial fluid (PSF)
is required. The literature and these studies give strong support for SAPL being the
lubricating factor in the joint. The data obtained in this study emphasizes that
POPC, SLPC, PLPC and DPPC are the major constituents of the SAPLs adsorbed
to joint surfaces and not just DPPC as previously assumed. It is clear that a
combination rather than a single SAPL constitutes the boundary lubricant of
diarthrodial joints. Thus, in formulating a lubricant whose composition closely
resembles the proportion of unsaturated and saturated PCs found in the joint, a
suitable artificial joint fluid was tendered for future friction and wear studies.
172
Cost and availability of large quantities of the individual PC species for wear
studies should not pose a problem as they are readily available for harvesting from
various sources such as egg yolk, bananas and their peels. This PSF may have
implications beyond the laboratory in providing a joint supplement similar to HA
viscosupplementation but for total joint replacement patients and even possibly be
useful in the development of pharmaceutical products for OA sufferers. In addition
a suitable PSF may reduce the variability in results noted between in vivo and in
vitro studies.

HA injections (viscosupplementation) have been shown to have some positive
effects even though HA is no longer considered to be the lubricant in the joint. This
causes one to wonder if HA affects more than the viscosity in the joint and the
literature seems to agree. HA may be acting as a carrier for the lubricant or in
synergy with SAPL. The same has been suggested for Lubricin and hence both
should be considered in the development of a PSF that may have implications
towards a pharmaceutical product.

Previous work (Chen, Hills et al. 2005) seemed to suggest that USPC species may
lubricate better than SPC species. DPPC is a saturated PC that has been shown to
reduce friction between two artificial surfaces quite remarkably and the suggestion
that a combination of USPCs and DPPC (the true SAPL profile) may reduce the
friction further was of keen interest. The fourth study (Chapter 9) in this thesis
investigated this suggestion and unfortunately did not find the same results as
previously reported. The PSF did not lubricate as well as DPPC by itself when used
as lubricants for a UHMWPE/SS tribo-pair. DPPCs tribological performance was
superior and reduced friction by more than 50% over the control lubricant whereas
the PSF reduced friction by up to 40% over the control. This may be explained by
the packing order of the USPC species versus the SPC species when adsorbed to
the surface. USPC only differ in structure by an extra bond in the fatty acid chains
which causes the fatty acid chain to kink part way along its length. This causes the
USPC molecules to pack less densely when compared to the straight chains of the
SPC species which can pack tighter and provide a more cohesive surface layer and,
173
hence, a better boundary against friction. Regardless, the PSF still lubricated well
and can still be regarded as an active boundary lubricant for the joint.

An additional study (unpublished) in this thesis, a first of its kind, verified the
action of SAPL on a prosthesis directly by measuring the friction of an explanted
ball from an artificial hip. Specimens were taken directly from the operating room
and tested on a tribometer to determine the frictional performance of the surface
adsorbed substances. The metallic balls were slid against a stainless steel surface
and showed that the friction was 50% less prior to cleaning the prosthetic surface
with a strong organic solvent. This is the same result that the first study showed
when UHMWPE was slid against SS when lubricated with DPPC. This finding is
excellent confirmation of the role of SAPL in the boundary lubrication of artificial
joints.

10.1 Conclusions
It is still largely unknown how and with what the human joint is lubricated. Surface
Active Phospholipid (SAPL) has been implicated as the boundary lubricant in the
human synovial joint, where such conditions exist that boundary lubrication is
crucial to the integrity of the natural bearing. More so, a previous study has shown
the presence of SAPL on retrieved hip implants suggesting that SAPL may play an
integral role in the boundary lubrication of the artificial joint.

This thesis has verified the action of a synthetic SAPL (DPPC) on common
prosthetic materials in vitro. SAPL does effectively lubricate prosthetic materials in
vitro under boundary lubrication conditions.

In addition, the action of SAPL with a novel load-bearing biomaterial, Pyrolytic
carbon, under boundary lubricated conditions returned an excellent tribological
performance. The interaction between SAPL and PyC is very favourable
suggesting that PyC receive further consideration for future prosthetic designs.

174
This thesis contained a study which was the first to identify the presence of SAPL
on retrieved knee implants and the first to characterise the constituents of the
adsorbed SAPL. This evidence that indigenous SAPL adsorbs to artificial surfaces
in vivo adds to the strong support that artificial joints are lubricated by SAPL. The
composition of SAPL found on retrieved implants is made up of four main species
of PC: namely PLPC, POPC, SLPC and DPPC. USPC species are dominant. Future
tribo tests (friction and wear trials) should adopt a lubricant similar to this detected
composition of the indigenous lubricant. The discovery of the constituents of SAPL
is invaluable for not only future tribo testing but also for artificial joint developers
and for the development of effective cures for several disease processes in the
natural joint where lubrication may play a role. Treatments for OA that inject a so-
called lubricant (viscosupplementation) into the joint should adopt a lubricant that
mimics the lubricating portion of SF that has been detected on TJRs.

The effectiveness of adsorbed SAPL to an artificial joint as a boundary lubricant
was proven when an explanted prosthesis was shown to have considerably less
friction when tested on a tribometer in its ex vivo state prior to being cleaned with a
strong organic solvent. Therefore artificial joints are boundary lubricated in vivo
and the indigenous boundary lubricant has been identified as SAPL. This adds to
the strong support that the same occurs in the natural joint and has implications to a
model for the pathogenesis for osteoarthritis.

Regardless of the material, future prosthetic designs need to take into consideration
the surface active nature of the indigenous boundary lubricant and provide surfaces
that are conducive to SAPL attachment in order to reduce friction and, ultimately,
wear. One such surface ideal for this application is the PyC surface.

10.2 Future Work
Determine actual amounts of SAPL attached to retrieved implants and
hence determine what surfaces are preferred in vivo. This would also
indicate the concentration or amount of SAPL to be applied to surfaces for
in vitro testing.
175
Visualisation is key to understanding the mechanism whereby SAPL
lubricates a surface for both the natural joint and the artificial joint. AFM
and SFA may be useful for this.
Modify the surface of PyC (and the other prosthetic materials for that
matter) to enhance SAPL attachment.
Continue Pyrolytic Carbon along the development path as a very real
candidate for future load bearing prostheses by performing wear
simulations.
Search for the presence of HA and Lubricin on retrieved implants to either
prove or disprove their role in the lubrication of artificial joints.
Further studies into the tribology of the natural and artificial joint will be
invaluable to the quality of life for millions of individuals if OA can be
cured.
176

177
References
1. Adamson, A.W. and Gast, A.P. (1997) Physical Chemistry of Surfaces. 6th
ed, New York / Chichester / Weinheim / Brisbane / Singapore / Toronto:
John Wiley & Sons Inc. 784.
2. Ahlroos, T. (2001) Effect of Lubricant on the Wear of Prosthetic Joint
Materials, in Laboratory of Machine Design, Helsinki University of
Technology: Helsinki. 26.
3. Ahlroos, T. and Saikko, V. (1997). "Wear of prosthetic joint materials in
various lubricants." Wear 211(1): 113-119.
4. American Academy of Orthopaedic Surgeons (2002) FAQ's about
Osteoarthritis. Your Orthopaedic Connection,
5. Archibald, F.R. (1969) Squeeze films, in Handbook of Lubrication, J.J.
O'Connor, Boyd, J., Avallone, E.A., Editor, McGraw-Hill: New York. p. 7-
1 - 7-8.
6. Australian Orthopaedic Association (2002) National Joint Replacement
Registry. 82.
7. Axn, N., Hogmark, S., and Jacobson, S. (2001) Friction and Wear
Measurement Techniques, in Modern Tribology Handbook, B. Bhushan,
Editor, CRC Press: New York. p. 493-510.
8. Bader, D. and Lee, D. (2000) Structure - properties of soft tissues: articular
cartilage, in Structural Biological Materials: Design and Structure-
Property Relationships, M. Elices, Editor, Elsevier Science Ltd.: Oxford. p.
75-103.
9. Baier, R.E., Gott, V.L., and Feruse, A. (1970). "Surface chemical
evaluation of thromboresistant materials before and after venous
implanttion." Trans Am Soc Artif Intern Organs 16: 50-7.
10. Balazs, E.A. (1968). "Viscoelastic properties of hyaluronic acid and
biological lubrication." Univ Mich Med Cent J: 255-9.
11. Ballantine, G.C. and Stachowiak, G.W. (2002). "The effects of lipid
depletion on osteoarthritic wear." Wear 253(3-4): 385-393.
12. Barnes, G.T. and Gentle, I.R. (2005) Interfacial Science - An Introduction.
1st ed, New Tork: Oxford. 247.
13. Barnett, C.H. and Cobbold, A.F. (1962). "Lubrication within living joints."
Journal of Bone and Joint Surgery 44B: 662.

178
14. Batchelor, A.W. and Stachowiak, G.W. (1996). "Arthritis and the
interacting mechanisms of synovial joint lubrication. Part II - Joint
lubrication and its relation to arthritis." J. Orthop. Rheumatol. 9: 11-21.
15. Bell, J., Tipper, J.L., et al. (2001). "The influence of phospholipid
concentration in protein-containing lubricants on the wear of ultra-high
molecular weight polyethylene in artificial hip joints." Proceedings of the
Institution of Mechanical Engineers, Part H: Journal of Engineering in
Medicine 215(2): 259-263.
16. Bennett, A. and Higginson, G.R. (1970). "Hydrodynamic lubrication of soft
solids." Journal of Mechanical Engineering Science 12(3): 218-222.
17. Benz, M., Chen, N., et al. (2005). "Static forces, structure and flow
properties of complex fluids in highly confined geometries." Ann Biomed
Eng 33(1): 39-51.
18. Bernhard, W., Postle, A., et al. (1995). "Composition of phospholipid
classes and phosphatidylcholine molecular species of gastric mucosa and
mucus." Biochim Biophys Acta 1255: 99-104.
19. Bernhard, W., Postle, A., et al. (2001). "Pulmonary and gastric surfactants.
A comparison of the effect of surface requirements on function and
phospholipid composition." Comp Biochem Physiol A Mol Integr Physiol
(129): 173-182.
20. Bhushan, B. (2001) Modern Tribology Handbook. Vol. 1, New York: CRC
Press.
21. Biresaw, G., ed. (1989) Tribology and the Liquid-Crystalline State.
American Chemical Society. Vol. 441: Miami Beach, Florida, Maple Press,
York. 133.
22. Black, J. and Hastings, G., eds. (1998) Handbook of Biomaterial
Properties. Melbourne, Chapman and Hall.
23. Blanchard, C.R. (1995) Biomaterials: Body Parts of the Future. [cited;
Available from: http://www.swri.edu/3pubs/ttoday/fall95/implant.htm.
24. Bokros, J.C. (1969) Deposition, structure, and properties of pyrolytic
carbon, in Chemistry and Physics of Carbon. p. 1-118.
25. Bole, G.G., Jr. and Peltier, D.F. (1962). "Synovial fluid lipids in normal
individuals and patients with rheumatoid arthritis." Arthritis Rheumat. 5:
589-601.
26. Bowsher, J.G. and Shelton, J.C. (2001). "A hip simulator study of the
influence of patient activity level on the wear of crosslinked polyethylene
under smooth and roughened femoral conditions." Wear 250(1-12): 167-
179.

179
27. Brandt, K.D., Smith, G.N., Jr., and Simon, L.S. (2000). "Intraarticular
injection of hyaluronan as treatment for knee osteoarthritis: what is the
evidence?" Arthritis Rheum 43(6): 1192-203.
28. Broom, N.D. (1988) The collagen framework of articular cartilage: its
profound influence on normal and abnormal load-bearing function, in
Collagen, M.E. Nimni, Editor, CRC Press: Boca Raton. p. 244-264.
29. Brown, E.S. (1964). "Isolation and Assay of Dipalmityl Lecithin in Lung
Extracts." Am J Physiol 207: 402-6.
30. Brown, E.S., Johnson, R.P., and Clements, J.A. (1959). "Pulmonary surface
tension." J Appl Physiol 14: 717-20.
31. Brown, S.S. and Clarke, I.C. (2006). "A Review of Lubrication Conditions
for Wear Simulation in Artificial Hip Replacements." Tribology
Transactions 49(1): 72.
32. Buckwalter, J.A. (1983). "Articular cartilage." Instr Course Lect 32: 349-
70.
33. Buckwalter, J.A. and Lane, N.E. (1997). "Athletics and osteoarthritis." Am
J Sports Med 25(6): 873-81.
34. Butler, B.D., Lichtenberger, L.M., and Hills, B.A. (1983). "Distribution of
surfactants in the canine gastrointestinal tract and their ability to
lubricate." Am J Physiol 244(6): G645-51.
35. Calonius, O. (2002) Tribology of Prosthetic Joints - Validation of Wear
Simulation Methods, in Department of Mechanical Engineering, Helsinki
University of Technology: Espoo.
36. Caplan, A.I., Elyaderani, M., et al. (1997). "Principles of cartilage repair
and regeneration." Clinical Orthopaedics and Related Research 342: 254-
269.
37. Caygill, J.C. and West, G.H. (1969). "The rheological behaviour of
synovial fluid and its possible relation to joint lubrication." Medical and
Biological Engineering 7: 507-516.
38. Charnley, J. (1959). "The lubrication of animal joints." Institution of
Mechanical Engineers: Symposium on Biomechanics 17: 12-22.
39. Charnley, J. (1966). "An artificial bearing in the hip joint: implications in
biological lubrication." Fed Proc 25(3): 1079-81.
40. Charnley, J. (1982) Long Term Results of Low-Friction Arthroplasty. in The
Hip: Proceedings of the Tenth Open Scientific Meeting of the Hip Society.
1982. St Louis: C. V. Mosby.
41. Chen, Y. and Hills, B. (2004) Unsaturated phosphatidylcholines and uses
thereof: International Patent Application - Australia. PCT/AU2004/001290.

180
42. Chen, Y. and Hills, B.A. (2000). "Surgical Adhesions: Evidence for
Adsorption of Surfactant to Peritoneal Mesothelium." ANZ Journal of
Surgery 70(6): 443-447.
43. Chen, Y., Hills, B.A., and Hills, Y.C. (2005). "Unsaturated
phosphatidylcholine and its application in surgical adhesion." ANZ Journal
of Surgery 75(12): 1111-1114.
44. Chikama, H. (1985). "The role of protein and hyaluronic acid in the
synovial fluid in animal joint lubrication." Nippon Seikeigeka Gakkai
Zasshi 59(5): 559-72.
45. Chung, A.C., Shanahan, J.R., and Brown, E.M., Jr. (1962). "Synovial fluid
lipids in rheumatoid and osteoarthritis." Arthritis Rheum 5: 176-83.
46. Chung, Y., Homolam, A.M., and Bryan Street, G., eds. (1991) Surface
Science Investigations in Tribology. York, PA, Maple Press. 253.
47. Clements, J.A. (1957). "Surface tension of lung extracts." Proc Soc Exp
Biol Med 95(1): 170-2.
48. Coller, R. (2002) Friction Testing Using Surfactant as the Boundary
Lubricant, Queensland University of Technology. 85.
49. Coller, R., Hargreaves, D.J., et al. (2004). "Is SAPL the Boundary Lubricant
in Prosthetic Joints: Friction Testing and Surface Rinsing." Australian
Journal of Mechanical Engineering 1: 63-71.
50. Comper, W.D. and Laurent, T.C. (1978). "Physiological function of
connective tissue polysaccharides." Physiol Rev 58(1): 255-315.
51. Cook, S.D., Thomas, K.A., and Kester, M.A. (1989). "Wear characteristics
of the canine acetabulum agianst different femoral prostheses." The Journal
of Bone and Joint Surgery 71-B(2): 188-197.
52. Cooke, A.F., Dowson, D., and Wright, V. (1978). "The rheology of synovial
fluid and some potential synthetic lubricants for degenerate synovial
joints." Engineering in Medicine 7: 66-72.
53. Craig, L.E. and LaBerge, M. (1994) Effect of DPPC concentration on the
boundary lubrication of rigid bearings: An arthroplasty model. in
Proceedings of the 15th Annual Conference of the Canadian Biomaterials
Society. 1994. Quebec City, Quebec.
54. Currey, J., Unsworth, A., and Hall, D.A. (1981) Properties of bone,
cartilage & synovial fluid., in Introduction to Joints and Joint Replacement,
D. Dowson and V. Wright, Editors, Mechanical Engineering Publications
Ltd: London. p. 103-119.
55. Davis, W.H., Jr., Lee, S.L., and Sokoloff, L. (1978). "Boundary lubricating
ability of synovial fluid in degenerative joint disease." Arthritis Rheum
21(7): 754-6.

181
56. Davis, W.H., Lee, S.L., and Sokoloff, L. (1979). "A Proposed Model of
Boundary Lubrication by Synovial Fluid: Structuring of Boundary Water."
Journal of Biomechanical Engineering 101: 185-192.
57. de Medicis, R., Reboux, J.F., and Lussier, A. (1976). "[Synovial fluid.
Review of international literature from 1972 to 1975 (author's transl)]."
Pathol Biol (Paris) 24(9): 641-56.
58. DeHaven, K.E. (1990). "Decision-making factors in the treatment of
meniscus lesions." Clin Orthop Relat Res (252): 49-54.
59. DeSchryver-Kecskemeti, K., Eliakim, R., et al. (1989). "Intestinal
surfactant-like material. A novel secretory product of the rat enterocyte." J
Clin Invest 84(4): 1355-61.
60. Dinnar, U. (1975). "Lubrication theory in synovial joints." CRC Critical
Reviews in Bioengineering 2(2): 159-177.
61. Dintenfass, L. (1963). "Lubrication in synovial joints." Nature 197: 496-
497.
62. Dintenfass, L. (1963). "Lubrication in synovial joints: a theoretical analysis
- a rheological approach to the problems of joint movement and joint
lubrication." Journal of Bone and Joint Surgery 45A: 1241-1256.
63. Donnelly, W.J., et al. (1997). "Radiological and survival comparison of
four methods of fixation of a proximal femoral stem." Journal of Bone and
Joint Surgery 79-Br(3): 351-360.
64. Dorinson, A. and Ludema, K.C. (1985) Lubricant additive action. I - Basic
catergories and mechanisms., in Mechanics and Chemistry in Lubrication,
A. Dorinson and K.C. Ludema, Editors, Elsevier: Amsterdam. p. 198-254.
65. Dorinson, A. and Ludema, K.C. (1985) Lubricant additive action. II -
Chemical reactivity and additive functionality., in Mechanics and
Chemistry in Lubrication, A. Dorinson and K.C. Ludema, Editors, Elsevier:
Amsterdam. p. 255-307.
66. Dowson, D. (1966-67). "Modes of lubrication in human joints - Paper 12."
Proceedings of the Institution of Mechanical Engineers 191(3-J): 45.
67. Dowson, D. (1973). "Lubrication and wear of joints." Physiotherapy 59(4):
104-106.
68. Dowson, D. (1990) Biotribology of natural and replacement synovial joints,
in Biomechanics of Diarthroidal Joints, V.C. Mow, A. Ratcliffe, and S.L.Y.
Wood, Editors, Springer-Verlag: New York. p. 305-345.
69. Dowson, D., Fisher, J., et al. (1991). "Design considerations for cushion
form bearings in artificial hip joints." Proc Inst Mech Eng [H] 205(2): 59-
68.

182
70. Dowson, D. and Jin, Z. (1986). "Micro-elastohydrodynamic lubrication of
synovial joints." Proceedings of the Institution of Mechanical Engineers.
Part H - Journal of Engineering in Medicine 15(2): 63-65.
71. Dowson, D., Longfield, M.D., et al. (1968). "An investigation of the friction
and lubrication in human joints." Proceedings of the Institution of
Mechanical Engineers 182(London): 68-76.
72. Dowson, D., Priest, M., et al., eds. (1998) Lubrication at the frontier : the
role of the interface and surface layers in the thin film and boundary
regime. Tribology and Interface Engineering Series, No. 36, ed. D.
Dowson, Elsevier. 893.
73. Dowson, D., Unsworth, A., et al. (1981) Lubrication of Joints, in An
Introduction to the Biomechanics of Joints and Joint Replacement, D.
Dowson and V. Wright, Editors, Mechanical Engineering Publication Ltd.:
London. p. 120-145.
74. Dowson, D. and Wright, V. (1973) Bio-Tribology, in The Rheology of
Lubricants, T.C. Davenport, Editor, Barking : Applied Science Publishers
on behalf of the Institute of Petroleum: Longman London. p. 81-88.
75. Dowson, D., Wright, V., and Longfield, M.D. (1969). "Human joint
lubrication." Biomedical Engineering 4: 160-165.
76. Dumbleton, J.H. (1981) Tribology of Natural and Artificial Joints,
Amsterdam: Elsevier Scientific Publishing.
77. Erdemir, A. (2001) Solid lubricants and self-lubricating solids, in Modern
Tribology Handbook, B. Bhushan, Editor, CRC Press: New York. p. 787-
825.
78. Ethell, M.T., Hodgson, D.R., and Hills, B.A. (1999). "The synovial
response to exogenous phospholipid (synovial surfactant) injected into the
equine radiocarpal joint compared with that to prilocaine, hyaluronan and
propylene glycol." New Zealand Veterinary Journal 47(4): 128-132.
79. Eyre, D.R., Wu, J.J., and Apone, S. (1987). "A growing family of collagens
in articular cartilage: identification of 5 genetically distinct types." J
Rheumatol 14 Spec No: 25-7.
80. Faber, J.J., Williamson, G.R., and Feldman, N.T. (1967). "Lubrication of
joints." J Appl Physiol 22(4): 793-799.
81. Felson, D.T. (1990). "Osteoarthritis." Rheum Dis Clin North Am 16(3):
499-512.
82. Felson, D.T. (1998) Epidemology of osteoarthritis, in Osteoarthritis, K.D.
Brandt, M. Doherty, and L.S. Lomander, Editors, Oxford University Press:
New York. p. 13-22.

183
83. Felson, D.T. and Radin, E.L. (1994). "What causes knee osteoarthrosis: are
different compartments susceptible to different risk factors?" J Rheumatol
21(2): 181-3.
84. Ferguson, J. and Nuki, G. (1973) The Rheology of a Synthetic Joint
Lubricant, in The Rheology of Lubricants, T.C. Davenport, Editor, Barking
: Applied Science Publishers on behalf of the Institute of Petroleum:
Longman London. p. 89-95.
85. Fisher, J. (2000) Materials for Improved Wear Resistance of Total Artificial
Joints, M.o. Understanding, Editor, University of Leeds: Leeds.
86. Fisher, J. and Dowson, D. (1991). "Tribology of total artificial joints."
Proceedings of the Institution of Mechanical Engineers. Part H - Journal of
Engineering in Medicine 205(2): 73-79.
87. Flores, R.H. and Hochberg, M. (1998) Definition and classification of
osteoarthritis, in Osteoarthritis, K.D. Brandt, M. Doherty, and L.S.
Lomander, Editors, Oxford University Press: New York. p. 1-12.
88. Fox, R.I., Lotz, M., and Carson, D.A. (1989) Structure and function of
synviocytes, in Arthritis and Allied Conditions, a Textbook of
Rheumatology, D.J. McCarty, Editor, Lea & Febiger: Philadelphia. p. 273-
288.
89. Fraser, J.R. and Laurent, T.C. (1989). "Turnover and metabolism of
hyaluronan." Ciba Found Symp 143: 41-53; discussion 53-9, 281-5.
90. Freeman, M.A.R. and Meachim, G. (1974) Ageing, degeneration and
remodelling of articular cartilage, in Adult Articular Cartilage, M.A.R.
Freeman, Editor, Grune and Stratton, Inc.: New York. p. 287-329.
91. Freeman, M.A.R., Swanson, S.A.V., and Manley, P.T. (1975). "Stress-
lowering function of articular cartilage." Medical and Biological
Engineering: 245-251.
92. Fukuda, Y., Takai, S., et al. (2000). "Impact load transmission of the knee
joint - influence of leg alignment and the role of meniscus and articular
cartilage." Clinical Biomechanics 15: 516-521.
93. Furey, M.J. (2000) Joint Lubrication, in The Biomedical Engineering
Handbook, Second Edition, J.D. Bronzino, Editor, CRC Press: Boca Raton.
p. 1-27.
94. Furey, M.J. and Burkhardt, B.M. (1997). "Biotribology: Friction, wear and
lubrication of natural synovial joints." Lubric Sci 9(3): 273-281.
95. Furey, M.J. and Burkhardt, B.M. (1997) Biotribology: Friction, Wear, and
Lubrication of Natural Synovial Joints, in Lubrication Science. p. 255-271.

184
96. Gale, L.R., Chen, Y., et al. (2006). "Boundary lubrication of joints:
Characterisation of Surface-Active Phospholipids found on retrieved
implants." Acta Orthopaedica In Press.
97. Gale, L.R., Coller, R., et al. (2007). "The role of SAPL as a boundary
lubricant in prosthetic joints." Tribology International 40(4): 601-606.
98. Gavrjushenko, N.S. (1993). "Recommendations with respect to the
improvement of lubricating qualities of synovial fluid in artificial joints."
Proc Inst Mech Eng [H] 207(2): 111-4.
99. Gellman, A.J. and Spencer, N.D. (2005) Surface chemistry in tribology, in
Wear : materials, mechanisms and practice, G.W. Stachowiak, Editor,
Wiley: Chichester. p. 458.
100. Ghadially, F.N. (1978) Fine structure of joints, in The Joints and Synovial
Fluid, L. Sokoloff, Editor, Academic Press: New York. p. 105-176.
101. Ghadially, F.N. (1983) Fine structure of synovial joints. A Text and Atlas
of the Ultrastructure of Normal and Abnormal Synovial Joints, England:
Butterworth & Co Ltd.
102. Ghadially, F.N., Lalonde, J.M., and Wedge, J.H. (1983). "Ultrastructure of
normal and torn menisci of the human knee joint." J Anat 136(Pt 4): 773-
91.
103. Girod, S., Zahm, J.M., et al. (1992). "Role of the physiochemical properties
of mucus in the protection of the respiratory epithelium." Eur Respir J 5(4):
477-87.
104. Glenister, T.W. (1976). "An embryological view of cartilage." J Anatomy
122: 323-330.
105. Glynn, L.E. (1977). "Primary lesion in osteoarthrosis." Lancet 1(8011):
574-5.
106. Gray, H. (1918) Anatomy of the Human Body, Philadelphia: Lea &
Febiger.
107. Greenwald, R.A., Moy, W.W., and Seibold, J. (1978). "Functional
properties of cartilage proteoglycans." Semin Arthritis Rheum 8(1): 53-67.
108. Guerra, D., Frizziero, L., et al. (1996). "Ultrastructural identification of a
membrane-like structure on the surface of normal articular cartilage." J
Submicrosc Cytol Pathol 28(3): 385-93.
109. Gvozdanovic, D., Wright, V., and Dowson, D. (1975). "Formation of
lubricating monolayers on the cartilage surface." Ann Rheum Dis
34(Suppl. 2): 100-101.

185
110. Hale, J.E., Rudert, M.J., and Brown, T.D. (1993). "Indentation assessment
of biphasic mechanical property deficits in size-dependant osteochondral
defect repair." J Biomechanics 26: 1319-1325.
111. Hall, R.M., Unsworth, A., et al. (1997). "The friction of explanted hip
prostheses." British J Rheumatology 36(1): 20-26.
112. Hamerman, D., Rosenberg, L.C., and Schubert, M. (1970). "Diarthrodial
joints revisited." J Bone Joint Surg Am 52(4): 725-74.
113. Hanson, S.R. (1998) Blood-Material Interactions, in Handbook of
Biomaterial Properties, J. Black and G.R. Hastings, Editors, Chapman and
Hall: London. p. 545-555.
114. Henderson, B. and Pettipher, E.R. (1985). "The synovial lining cell: biology
and pathobiology." Semin Arthritis Rheum 15(1): 1-32.
115. Henderson, B., Revell, P.A., and Edwards, J.C. (1988). "Synovial lining cell
hyperplasia in rheumatoid arthritis: dogma and fact." Ann Rheum Dis
47(4): 348-9.
116. Heuberger, M.P., Widmer, M.R., et al. (2005). "Protein-mediated boundary
lubrication in arthroplasty." Biomaterials 26(10): 1165-73.
117. Higaki, H., Murakami, T., and Nakanishi, Y. (1997). "Lubricating Ability of
Langmuir-Blodgett films as boundary lubricating films on articular
surfaces." JSME Int J 40(4): 776-781.
118. Higaki, H., Murakami, T., et al. (1998). "The Lubricating Ability of
Biomembrane Models with Dipalmitoyl Phosphatidylcholine and gamma-
Globulin." Proceedings of the Institution of Mechanical Engineers, Part H:
Journal of Engineering in Medicine 212(5): 337-346.
119. Higginson, G.R. (1978). "Elastohydrodynamic lubrication in human joints."
120. Hills, B.A. (1984). "Analysis of eustachian surfactant and its function as a
release agent." Arch Otolaryngol 110(1): 3-9.
121. Hills, B.A. (1984). "Hydrophobic lining of the eustachian tube imparted by
surfactant." Arch Otolaryngol 110(12): 779-82.
122. Hills, B.A. (1984). "Surfactant as a release agent opposing the adhesion of
tumor cells in determining malignancy." Med Hypotheses 14(1): 99-110.
123. Hills, B.A. (1988) The Biology of Surfactant. Vol. 321, Cambridge:
Cambridge University Press.
124. Hills, B.A. (1989). "Oligolamellar lubrication of joints by surface active
phospholipid." J Rheumatol 16(1): 82-91.

186
125. Hills, B.A. (1990). "Multiple roles for surface-active phospholipid in
hypertension." Medical Hypotheses 33(4): 275-281.
126. Hills, B.A. (1990). "Oligolamellar Nature of the Articular Surface." Journal
of Rheumatology 17(3): 349-356.
127. Hills, B.A. (1990). "Surface-active phospholipid in muscle lymph and its
lubricating and adhesive properties." Lymphology 23(1): 39-47.
128. Hills, B.A. (1995). "Remarkable anti-wear properties of joint surfactant."
Ann Biomed Eng 23(2): 112-5.
129. Hills, B.A. (1996). "Lubrication of visceral movement and gastric motility
by peritoneal surfactant." Journal of Gastroenterology and Hepatology 11:
797-803.
130. Hills, B.A. (2000). "Boundary lubrication in vivo." Proceedings of the
Institution of Mechanical Engineers. Part H - Journal of Engineering in
Medicine 214(1): 83-94.
131. Hills, B.A. (2000). "Role of surfactant in peritoneal dialysis." Peritoneal
Dialysis International: Journal Of The International Society For Peritoneal
Dialysis 20(5): 503-515.
132. Hills, B.A. (2002). "Surface-active phospholipid: a Pandora's box of
clinical applications. Part II. Barrier and lubricating properties." Intern
Med J 32(5-6): 242-251.
133. Hills, B.A. and Butler, B.D. (1984). "Identification of surfactants in
synovial fluid and their ability to act as boundary lubricants." Ann. Rheum.
Dis. (48): 51-57.
134. Hills, B.A. and Butler, B.D. (1984). "Surfactants identified in synovial fluid
and their ability to act as boundary lubricants." Ann Rheum Dis 43(4):
641-8.
135. Hills, B.A. and Butler, B.D. (1985). "Phospholipids identified on the
pericardium and their ability to impart boundary lubrication." Ann Biomed
Eng 13(6): 573-86.
136. Hills, B.A. and Butler, B.D. (1986). "Surfactants identified on the
pericardium and their ability to impart boundary lubrication." Annals of
Biomedical Engineering 13: 573-586.
137. Hills, B.A., Butler, B.D., and Barrow, R.E. (1982). "Boundary lubrication
imparted by pleural surfactants and their identification." Journal of Applied
Physiology 53(2): 463-9.
138. Hills, B.A. and Cotton, D.B. (1986). "Release and lubricating properties of
amniotic surfactants and the very hydrophobic surfaces of the amnion,
chorion, and their interface." Obstet Gynecol 68(4): 550-4.

187
139. Hills, B.A. and Crawford, R.W. (2003). "Normal and prosthetic synovial
joints are lubricated by surface-active phospholipid: a hypothesis." The
Journal of Arthroplasty 18(4): 499-505.
140. Hills, B.A. and Jay, G.D. (2002). "Identity of the Joint Lubricant." J
Rheumatol 29(1): 200-1.
141. Hills, B.A. and Monds, M. (1998). "Enzymatic identification of the load-
bearing boundary lubricant in the joint." Rheumatology 37(2): 137-142.
142. Hills, B.A. and Thomas, K. (1998). "Joint Stiffness and 'articular gelling':
inhibition of the fusion of articular surfaces by surfactant." British Journal
of Rheumatology 37: 532-538.
143. Hlavacek, M. (1995). "The role of synovial fluid filtration by cartilage in
lubrication of synovial joints--IV. Squeeze-film lubrication: the central film
thickness for normal and inflammatory synovial fluids for axial symmetry
under high loading conditions." Journal of Biomechanics 28(10): 1199-
1205.
144. Hou, J.S., Holmes, M.H., et al. (1989). "Boundary conditions at the
cartilage-synovial fluid interface for joint lubrication and theoretical
verifications." J Biomech Eng 111(1): 78-87.
145. Hou, J.S., Mow, V.C., et al. (1992). "An analysis of the squeeze-film
lubrication mechanism for articular cartilage." Journal of Biomechanics
25: 247-259.
146. How, M.J., Long, V.J., and Stanworth, D.R. (1969). "The association of
hyaluronic acid with protein in human synovial fluid." Biochim Biophys
Acta 194(1): 81-90.
147. Hsu, S.M. and Gates, R.S. (2001) Boundary lubrication and boundary
lubricating films, in Modern Tribology Handbook, B. Bhushan, Editor,
CRC Press: New York. p. 455-492.
148. Hsu, S.M., Zhang, J., and Yin, Z. (2002). "The Nature and Origin of
Tribochemistry." Tribology Letters 13(2): 131-139.
149. http://en.wikipedia.org/wiki/Membrane_lipids. (2005) [cited.
150. http://www.utahhipandknee.com/history.htm. (2004) [cited.
151. Huber, M., Trattnig, S., and Lintner, F. (2000). "Anatomy, biochemistry,
and physiology of articular cartilage." Invest Radiol 35(10): 573-80.
152. Hutchings, I.M., ed. (2003) Friction, Lubrication, and Wear of Artificial
Joints. London, Professional Engineering Publishing. 133.
153. Ingham, E. and Fisher, J. (2000). "Biological reactions to wear debris in
total joint replacement." Proceedings of the Institution of Mechanical
Engineers. Part H - Journal of Engineering in Medicine 214(1): 21-37.

188
154. Israelachvili, J.N. (1992) Intermolecular and surface forces. 2nd ed, London
; New York: Academic Press. 450p.
155. Jalali-Vahid, D., Jagatia, M., et al. (2000) Elastohydrodynamic lubrication
analysis of UHMWPE hip joint replacements. in Thinning films and
tribological interfaces. 2000. Leeds: Elsevier.
156. Jalali-Vahid, D., Jagatia, M., et al. (2001). "Prediction of lubricating film
thickness in UHMWPE hip joint replacements." J Biomech 34(2): 261-6.
157. Jay, G.D. (1990) Joint lubrication: A physicochemical study of a purified
lubricating factor from bovine synovial fluid, State University of New York
at Stony Brook: United States -- New York.
158. Jay, G.D. (2004). "Lubricin and surfacing of articular joints." Current
Opinion in Orthopedics 15(5): 355-359.
159. Jay, G.D. and Cha, C.J. (1999). "The effect of phospholipase digestion upon
the boundary lubricating ability of synovial fluid." J Rheumatol 26(11):
2454-7.
160. Jay, G.D., Elsaid, K.A., et al. (2004). "Lubricating ability of aspirated
synovial fluid from emergency department patients with knee joint
synovitis." J Rheumatology 31(3): 557-64.
161. Jayson, M.I. and Dixon, A.S. (1970). "Intra-articular pressure in
rheumatoid arthritis of the knee. 3. Pressure changes during joint use." Ann
Rheum Dis 29(4): 401-8.
162. Jazrawi, L.M., Kummer, F.J., and DiCesare, P.E. (1998). "Alternative
bearing surfaces for total joint arthroplasty." Journal of the American
Academy of Orthopaedic Surgeons 6(4): 198-203.
163. Jin, Z., Dowson, D., and Fisher, J. (1993) Fluid film lubrication in natural
hip joints. in Thin Films in Tribology: Tribology Series 25. 1993.
University of Leeds, U.K.: Elsevier.
164. Jin, Z.M., Dowson, D., and Fisher, J. (1997). "Analysis of Fluid Film
Lubrication in Artificial Hip Joint Replacements with Surfaces of High
Elastic Modulus." Proceedings of the Institution of Mechanical Engineers,
Part H: Journal of Engineering in Medicine 211(3): 247-256.
165. Jin, Z.M., Medley, J.B., and Dowson, D. (2002) Fluid film lubrication in
artificial hip joints. in Tribological Reserach and Design for Engineering
Systems: Tribology Series, 41. 2002. Leeds, UK: Elsevier.
166. Jones, E.S. (1934). "Joint lubrication." Lancet: 1426-1427.
167. Jones, E.S. (1936). "Joint lubrication." Lancet: 1043-1044.
168. Jones, H.P. and Walker, P.S. (1968). "Casting techiques applied to study of
human joints." Scientific Technology 14: 57-68.

189
169. Kawano, T., Miura, H., et al. (2003). "Mechanical effects of the
intraarticular administration of high molecular weight hyaluronic acid plus
phospholipid on synovial joint lubrication and prevention of articular
cartilage degeneration in experimental osteoarthritis." Arthritis Rheum
48(7): 1923-9.
170. Klaus, M.H., Clements, J.A., and Havel, R.J. (1961). "Composition of
surface-active material isolated from beef lung." Proc Natl Acad Sci U S A
47: 1858-9.
171. Knight, A.D. and Levick, J.R. (1983). "The density and distribution of
capillaries around a synovial cavity." Q J Exp Physiol 68(4): 629-44.
172. Kobayashi, A., et al (1997). "Early radiological observations may predict
the long term survival of femoral hip prostheses." Journal of Bone and Joint
Surgery 79-B(4): 583-589.
173. Kobayashi, M. and Oka, M. (2003). "The lubricative function of artificial
joint material surfaces by confocal laser scanning microscopy. Comparison
with natural synovial joint surface." Biomed Mater Eng 13(4): 429-37.
174. Kuettner, K.E., Aydelotte, M.B., and Thonar, E.J. (1991). "Articular
cartilage matrix and structure: a minireview." J Rheumatol Suppl 27: 46-8.
175. LaGrange, L.D., Gott, V.L., et al. (1969) Compatibility of carbon and
blood. in Artificial Heart Program Conference Proceedings. 1969.
Washington, DC.: US Government Printing Office.
176. Lai, W.M., Kuei, S.C., and Mow, V.C. (1978). "Rheological Equations for
Synovial Fluids." Journal of Biomechanical Engineering 100: 169-186.
177. Lane, N.E. and Buckwalter, J.A. (1993). "Exercise: a cause of
osteoarthritis?" Rheum Dis Clin North Am 19(3): 617-33.
178. Larsen, R.G. and Perry, G.L. (1950) Chemical Aspects of Wear and
Friction, in Mechanical Wear, American Society of Metals, J.T. Burwell,
Editor. p. 73-94.
179. Larson, C.M. and Larson, R. (1969) Lubricant additives, in Standard
Handbook of Lubrication, J. O'Connor.J, Boyd, J., Avallone, E.A., Editor,
McGraw-Hill: New York. p. 14-1 - 14-24.
180. Laurent, T.C., Laurent, U.B., and Fraser, J.R. (1996). "The structure and
function of hyaluronan: An overview." Immunol Cell Biol 74(2): A1-7.
181. Levick, J. (1987) Synovial fluid and trans-synovial flow in stationary and
moving normal joints, in Joint Loading, H.J. Helminen, et al., Editors,
Wright: Bristol.
182. Levick, J. (1989). "Synovial Fluid Exchange - A Case of Flow Through
Fibrous Mats." News Physiol Sci 4(5): 198-202.

190
183. Levick, J.R. (1995). "Microvascular architecture and exchange in synovial
joints." Microcirculation 2(3): 217-33.
184. Lewis, P.R. and McCutchen, C.W. (1959). "Experimental evidence for
weeping lubrication in mammalian joints." Nature 184: 1285.
185. Liang, H. (2004) Mechanical Tribology: Materials, Characterization and
Applications, ed. G.E. Totten, Seattle, WA USA: Marcel Dekker Inc. 496.
186. Liao, Y.-S., Benya, P.D., and McKellop, H.A. (1998). "Effect of protein
lubrication on the wear properties of materials for prosthetic joints."
Journal of Biomedical Materials Research 48(4): 465 - 473.
187. Ling, F.F., Klaus, E.E., and Fein, R.S., eds. (1969) Boundary Lubrication:
An Appraisal of World Literature. ASME Research Committe on
Lubrication: New York, United Engineering Center. 576.
188. Linn, F.C. (1967). "Lubrication of Animal Joints. I. The arthrotripsometer."
The Journal of Bone and Joint Surgery 49(A): 1079-1098.
189. Linn, F.C. (1968). "Lubrication of animal joints II: The mechanism."
Journal of Biomechanics 1: 93-205.
190. Linn, F.C. and Radin, E.L. (1968). "Lubrication of animal joints: III. The
effect of certain chemical alterations of the cartilage and the lubricant."
Arthritis & Rheumatism 11: 674.
191. Linn, F.C. and Sokoloff, L. (1965). "Movement and Composition of
Interstitial Fluid of Cartilage." Arthritis Rheum 8: 481-94.
192. Little, T.D., Freeman, M.A.R., and Swanson, S.A. (1969) Experiments on
friction in the human hip joint, in Lubrication and Wear in Joints,, V.
Wright, Editor, Lippincott Publisher,: Philadelphia. p. 110114.
193. Lohmander, S. (1988). "Proteoglycans of joint cartilage. Structure,
function, turnover and role as markers of joint disease." Baillieres Clin
Rheumatol 2(1): 37-62.
194. Lu, Y., Levick, J.R., and Wang, W. (2005). "The mechanism of synovial
fluid retention in pressurized joint cavities." Microcirculation 12(7): 581-
95.
195. Ludema, K.C. (2001) Friction, in Modern Tribology Handbook, B.
Bhushan, Editor, CRC Press: New York. p. 205-233.
196. MacConaill, M.A. (1932). "The function of intra-articular fibrocartilages,
with special reference to the knee and inferior radio-ulnar joints." Journal
of Anatomy 66: 210-227.
197. Malcolm, L.L. (1976) An experimental investigation of the frictional and
deformational responses of articular cartilage interfaces to static and
dynamic loading., University of California: San Diego.

191
198. Mankin, H.J. and Brandt, K.D. (1984) Biochemistry and metabolism of
cartilage in osteoarthritis., in Osteoarthritis: Diagnosis and Management,
R.W. Moskowitz, et al., Editors, WB Sauders Co.: USA. p. 43-79.
199. Mankin, H.J. and Lippiello, L. (1969). "The turnover of adult rabbit
articular cartilage." J Bone Joint Surg Am 51(8): 1591-600.
200. Mansour, J.M. and Mow, V.C. (1977). "On the Natural Lubrication of
Synovial Joints: Normal and degenerate." Journal of Lubrication
Technology: 163-171.
201. Marcinko, D.E. and Dollard, M.D. (1986). "Physical and mechanical
properties of joints (the pathomechanics of articular cartilage
degeneration)." J Foot Surg 25(1): 3-13.
202. Maroudas, A. (1967). "Hyaluronic Acid Films." Proceedings of the
Institution of Mechanical Engineers 3J(181): 122-124.
203. Maroudas, A. (1976). "Transport of solutes through cartilage: permeability
to large molecules." J Anat 122(Pt 2): 335-47.
204. Maroudas, A. and Bullough, P. (1968). "Permeability of articular
cartilage." Nature 219(5160): 1260-1.
205. Maroudas, A. and Venn, M. (1977). "Chemical composition and swelling of
normal and osteoarthrotic femoral head cartilage. II. Swelling." Ann
Rheum Dis 36(5): 399-406.
206. Maroudas, A.I. (1976). "Balance between swelling pressure and collagen
tension in normal and degenerate cartilage." Nature 260(5554): 808-9.
207. Marshall, K.W. (2000). "Intra-articular hyaluronan therapy." Curr Opin
Rheumatol 12(5): 468-74.
208. McCutchen, C.W. (1959). "Mechanism of animal joints. Sponge-hydrostatic
and weeping bearings." Nature 184: 1284-1285.
209. McCutchen, C.W. (1962). "The frictional properties of animal joints."
Wear 5: 1-17.
210. McCutchen, C.W. (1966). "Boundary lubrication by synovial fluid:
demonstration and possible osmotic explanation." Fed. Am. Soc. Exp. Biol.
25: 1061-1068.
211. McCutchen, C.W. (1967). "Physiological lubrication." Proceedings of the
Institution of Mechanical Engineers 181(Part 3J): 55-62.
212. McCutchen, C.W. (1978) Lubrication of joints, in The Joints and Synovial
Fluid, L. Sokoloff, Editor, Academic Press, Inc: New York. p. 437-483.
213. McCutchen, C.W. (1982). "Cartilage is poroelastic, not viscoelastic
(including an exact theorem about strain energy and viscous loss, and an

192
order of magnitude relation for equilibration time)." J Biomechanics 15(4):
325-327.
214. McCutchen, C.W. (1983) Lubrication of and by articular cartilage, in
Cartilage, B.K. Hall, Editor, Academic Press Inc.: New York. p. 340.
215. McIlwraith, C.W. and Vachon, A. (1988). "Review of pathogenesis and
treatment of degenerative joint disease." Equine Vet J Suppl (6): 3-11.
216. McKibbin, B. and Maroudas, A. (1979) Nutrition and Metabolism, in Adult
Articular Cartilage, M.A.R. Freeman, Editor, Pitman Medical Publishing
Co.: London. p. 461-486.
217. Meachim, G. and Brooke, G. (1984). "The synovial response to intra-
articular acrylic cement particles in guinea pigs." Biomaterials 5(2): 69-
74.
218. Meachim, G. and Stockwell, R.A. (1974) The matrix, in Adult Articular
Cartilage, M.A.R. Freeman, Editor, Grune and Stratton, Inc.: New York. p.
1-50.
219. Medley, J.B., Dowson, D., and Wright, V. (1984). "Transient
elastohydrodynamic lubrication models for the human ankle joint."
Engineering in Medicine 13: 137-151.
220. Momberger, T.S., Levick, J.R., and Mason, R.M. (2005). "Hyaluronan
secretion by synoviocytes is mechanosensitive." Matrix Biol 24(8): 510-9.
221. More, R.B., Haubold, A.D., and Bokros, J.C. (2000) Pyrolytic Carbon for
Long-Term Medical Implants, Medical Carbon Research Institute: Austin,
Texas.
222. Mori, S., Naito, M., and Moriyama, S. (2002). "Highly viscous sodium
hyaluronate and joint lubrication." Int Orthop 26(2): 116-21.
223. Moro-oka, T., Miura, H., et al. (2000). "Mixture of hyaluronic acid and
phospholipid prevents adhesion formation on the injured flexor tendon in
rabbits." J Orthop Res 18(5): 835-40.
224. Mow, V.C., Ateshian, G.A., and Spilker, R.L. (1993). "Biomechanics of
diarthrodial joints: a review of twenty years of progress." J Biomech Eng
115(4B): 460-7.
225. Mow, V.C., Fithian, D.C., and Kelly, M.A. (1988) Fundamentals of
articular cartilage and meniscus biomechanics, in Articular Cartilage and
Knee Joint Function: Basic Science and Arthroscopy, J.W. Ewing, Editor,
Raven Press: New York. p. 1-18.
226. Mow, V.C. and Hayes, W.C. (1997) Basic Orthopaedic Biomechanics. 2nd
ed, Philadelphia: Lippincott-Raven.

193
227. Mow, V.C., Holmes, M.H., and Lai, W.M. (1984). "Fluid Transport and
Mechanical Properties of Articular Cartilage: A Review." Journal of
Biomechanics 17(5): 377-394.
228. Mow, V.C., Kuei, S.C., et al. (1980). "Biphasic Creep and Stress
Relaxation of Articular Cartilage in Compression: Theory and
Experiments." Journal of Biomechanical Engineering 102: 73-84.
229. Mow, V.C. and Mak, A.F. (1987) Lubrication of Diarthrodial Joints, in
Handbook of Bioengineering, McGraw-Hill. p. 5.1-5.34.
230. Muir, I.H.M. (1980) The chemistry of the ground substance of joint
cartilage, in The Joints and Synovial Fluid, L. Sokoloff, Editor, Academic
Press, Inc.: New York. p. 27-94.
231. Muller, F.J., Pita, J.C., et al. (1989). "Centrifugal characterization of
proteoglycans from various depth layers and weight-bearing areas of
normal and abnormal human articular cartilage." J Orthop Res 7(3): 326-
34.
232. Muller, M. (2004) Chemistry and the Building Blocks of Life. [cited.
233. Munro, L.A. (1964) Chemistry in Engineering, Englewood Cliffs, NJ.:
Prentice-Hall Inc. 159-190.
234. Murakami, T., Higaki, H., et al. (1998). "Adaptive multimode lubrication in
natural synovial joints and artificial joints." Proceedings of the Institution
of Mechanical Engineers. Part H - Journal of Engineering in Medicine 212:
23-35.
235. Myers, S.L. and Christine, T.A. (1983). "Hyaluronate synthesis by synovial
villi in organ culture." Arthritis Rheum 26(6): 764-70.
236. Nakano, T., Momozono, S., and Aizawa, S. (2000). "Tribological analysis
of bacterial flagellar motor." Japanese journal of tribology 45(1): 97-103.
237. Neame, P.J. and Barry, F.P. (1994). "The link proteins." Exs 70: 53-72.
238. Neame, R.L., Muir, K., et al. (2004). "Genetic risk of knee osteoarthritis: a
sibling study." Ann Rheum Dis 63(9): 1022-7.
239. Nitzan, D.W. (2001). "The process of lubrication impairment and its
involvement in temporomandibular joint disc displacement: a theoretical
concept." J Oral Maxillofac Surg 59(1): 36-45.
240. Nitzan, D.W., Mahler, Y., and Simkin, A. (1992). "Intra-articular pressure
measurements in patients with suddenly developing, severely limited mouth
opening." J Oral Maxillofac Surg 50(10): 1038-42; discussion 1043.
241. Nixon, A.J. (1991). "Articular cartilage surface structure and function."
Equine Vet Education 3: 72-75.

194
242. Nixon, J.S., Bottomley, K.M., et al. (1991). "Potent collagenase inhibitors
prevent interleukin-1-induced cartilage degradation in vitro." Int J Tissue
React 13(5): 237-41.
243. O'Hara, B.P., Urban, J.P., and Maroudas, A. (1990). "Influence of cyclic
loading on the nutrition of articular cartilage." Ann Rheum Dis 49(7):
536-9.
244. Ogston, A.G. and Stanier, J.E. (1953). "Some effects of hyaluronidase on
the hyaluronic acid of ox synovial fluid, and their bearing on the
investigation of pathological fluids." J Physiol 119(2-3): 253-8.
245. Oka, M., Kumar, P., et al. (2000). "Development of Artificial Articular
Cartilage." Proceedings of the Institution of Mechanical Engineers, Part H:
Journal of Engineering in Medicine 214(1): 59-68.
246. Ozturk, H.E., Stoffel, K.K., et al. (2004). "The Effect of Surface-Active
Phospholipids on the Lubrication of Osteoarthritic Sheep Knee Joints:
Friction." Tribology Letters 16(4): 283-289.
247. Panjabi, P.M. and White, A.A. (2001) Joint Friction, Wear and
Lubrication, in Biomechanics in the Musculoskeletal System, R. Zorab,
Editor, Churchill Livingstone: Philadelphia. p. 151-166.
248. Panush, R.S. (1990). "Does exercise cause arthritis? Long-term
consequences of exercise on the musculoskeletal system." Rheum Dis Clin
North Am 16(4): 827-36.
249. Pattle, R.E. (1950). "The control of foaming. I. The mode of action of
chemical anti-foams." J. Soc. Chem. Ind. 69: 363-368.
250. Pattle, R.E. (1955). "Properties, function and origin of the alveolar lining
layer." Nature 175(4469): 1125-6.
251. Pattle, R.E. (1956). "A test of silicone antifoam treatment of lung oedema in
rabbits." J Pathol Bacteriol 72(1): 203-9.
252. Pattle, R.E. and Thomas, L.C. (1961). "Lipoprotein composition of the film
lining the lung." Nature 189: 844.
253. Pawlak, Z. (2003) Tribochemistry of Lubricationg Oils. Tribology and
Interface Engineering Series, No. 45, ed. B.J. Briscoe: Elsevier. 368.
254. Phillips, M.C. and Riddiford, A.C. (1967) Contact angles and the free
surface energies of solids, in Wetting, Staples Printers Ltd. p. 31-56.
255. Pickard, J.E., Fisher, J., et al. (1998). "Investigation into the effects of
proteins and lipids on the frictional properties of articular cartilage."
Biomaterials 19(19): 1807-1812.

195
256. Prete, P.E., Gurakar-Osborne, A., and Kashyap, M.L. (1993). "Synovial
fluid lipoproteins: review of current concepts and new directions." Semin
Arthritis Rheum 23(2): 79-89.
257. Prete, P.E., Gurakar-Osborne, A., and Kashyap, M.L. (1995). "Synovial
fluid lipids and apolipoproteins: a contemporary perspective." Biorheology
32(1): 1-16.
258. Purbach, B., Hills, B.A., and Wroblewski, B.M. (2002). "Surface-active
phospholipid in total hip arthroplasty." Clinical Orthopaedics and Related
Research 396: 115-118.
259. Rabinowitz, J.L., Gregg, J.R., and Nixon, J.E. (1984). "Lipid composition of
the tissues of human knee joints. II. Synovial fluid in trauma." Clinical
Orthopaedics and Related Research 190: 292-298.
260. Rabinowitz, J.L., Gregg, J.R., et al. (1979). "Lipid Composition of the
Tissues of Human Knee Joints. I. Observations in normal joints (articular
cartilage, meniscus, ligaments, synovial fluid, synovium, intra-articular fat
pad and bone marrow)." Lipids of Normal Knee Joints 143: 260-265.
261. Radin, E., Swann, D., and Weisser, P. (1970). "Separation of a
hyaluronate-free lubricating fraction from synovial fluid." Nature
228(269): 377-8.
262. Radin, E.L. and Paul, I.L. (1971). "Response of joints to impact loading."
Arthritis & Rheumatism 14: 356-362.
263. Radin, E.L., Paul, I.L., and Pollock, D. (1970). "Animal joint behaviour
under excessive loading." Nature 226: 554-555.
264. Radin, E.L., Paul, I.L., et al. (1971). "Lubrication of synovial membrane."
Ann Rheum Dis 30(3): 322-5.
265. Rapport, P.N., Lim, D.J., and Weiss, H.S. (1975). "Surface-active agent in
Eustachian Tube Function." Arch Otolaryngol 101(5): 305-11.
266. Reilly, D.T. and Burstein, A.H. (1974). "Review article. The mechanical
properties of cortical bone." J Bone Joint Surg Am 56(5): 1001-22.
267. Reilly, D.T., Burstein, A.H., and Frankel, V.H. (1974). "The elastic
modulus for bone." J Biomech 7(3): 271-5.
268. Ropes, M.W., Muller, A.F., and Bauer, W. (1960). "The entrance of glucose
and other sugars into joints." Arthritis Rheum 3: 496-514.
269. Ropes, M.W., Robertson, W.B., et al. (1947). "Synovial fluid mucin,." Acta
Med. Scand. 196(Suppl.): 700714.
270. Ropes, M.W., Rossmeisl, E., and Bauer, W. (1939). "The Relationship
between the Erythrocyte Sedimentation Rate and the Plasma Proteins." J
Clin Invest 18(6): 791-8.

196
271. Rorvik, A.M. and Grondahl, A.M. (1995). "Markers of osteoarthritis: a
review of the literature." Vet Surg 24(3): 255-62.
272. Saikko, V. and Ahlroos, T. (1997). "Phospholipids as boundary lubricants
in wear tests of prosthetic joint materials." Wear 207(1-2): 86-91.
273. Sandy, J.D., Brown, H.L., and Lowther, D.A. (1978). "Degradation of
proteoglycan in articular cartilage." Biochim Biophys Acta 543(4): 536-
44.
274. Sarma, A.V., Powell, G.L., and LaBerge, M. (2001). "Phospholipid
composition of articular cartilage boundary lubricant." Journal of
Orthopaedic Research 19(4): 671-676.
275. Sayles, R.S., Thomas, T.R., et al. (1979). "Measurement of the surface
microgeometry of articular cartilage." J Biomech 12(4): 257-67.
276. Scholes, S.C. and Unsworth, A. (2000). "Comparison of Friction and
Lubrication of Different Hip Prostheses." Proceedings of the Institution of
Mechanical Engineers, Part H: Journal of Engineering in Medicine 214(1):
49-57.
277. Scholes, S.C., Unsworth, A., et al. (2005). "Design aspects of compliant,
soft layer bearings for an experimental hip prosthesis." Proc Inst Mech Eng
[H] 219(2): 79-87.
278. Schreppers, G.J., Sauren, A.A., and Huson, A. (1991). "A model of force
transmission in the tibio-femoral contact incorporating fluid and mixtures."
Proc Inst Mech Eng [H] 205(4): 233-41.
279. Schurz, J. and Ribitsch, V. (1987). "Rheology of Synovial Fluid."
Biorheology 24(4): 385-399.
280. Schwarz, I.M. (1994) The isolation of the lubricating factor in the synovial
joint : implications for the development of an artificial synovial fluid,
University of New England: Armidale.
281. Schwarz, I.M. and Hills, B.A. (1998). "Surface-active phospholipid as the
lubricating component of lubricin." Br J Rheumatol 37(1): 21-6.
282. Scmalzried, T.P. and Callaghan, J.J. (1999). "Wear in total hip and knee
replacements." Journal of Bone and Joint Surgery 81-A(1): 136-155.
283. Scott, D.L. and Walton, K.W. (1984). "The significance of fibronectin in
rheumatoid arthritis." Semin Arthritis Rheum 13(3): 244-54.
284. Serro, A.P., Gispert, M.P., et al. (2006). "Adsorption of albumin on
prosthetic materials: Implication for tribological behavior." J Biomed
Mater Res A 78(3): 581-9.

197
285. Shah, C.O. and Schulman, J.H. (1965). "Binding of metal ions to
monolayers of lecithins, plasmalogen, cardiolipin and dicetyl phosphate." J
Lipid Res 6: 341-349.
286. Sharma, R., ed. Surfactant Adsorption and Solubilization. York, PA, Maple
Press. 401.
287. Shaw, D.J. (1992) Chapter 4, Liquid-gas and liquid-liquid interfaces, in
Introducton to Surface and Colloid Chemistry, Butterworth Hienemann:
Oxford. p. 64-114.
288. Shi, B. (2004) Tribological comparison of materials, University of Alaska
Fairbanks: United States -- Alaska.
289. Shpenkov, G.P. (1995) Friction surface phenomena. Tribology and
Interface Engineering Series, No. 29: Elsevier. 358.
290. Simkin, P.A. (1991). "Physiology of normal and abnormal synovium."
Semin Arthritis Rheum 21: 179-183.
291. Simkin, P.A. (1993) Synovial physiology, in Arthritis and Allied
Conditions, a Textbook of Rheumatology, D.J. McCarty and W.J. Koopman,
Editors, Lea & Febiger: Philadelphia. p. 199-212.
292. Simmons, C.A., Shaker, M.A., and Pilliar, R.A. (2001). "Differences in
osseointegration rate due to implant surface geometry can be explained by
local tissue strains." Journal of Orthopaedic Research 19: 187-194.
293. Skotheim, J.M. and Mahadevan, L. (2004). "Soft lubrication." Phys Rev
Lett 92(24): 245509.
294. Small, D.M., Cohen, A.S., and Schmid, K. (1964). "Lipoproteins of
Synovial Fluid as Studied by Analytical Ultracentrifugation." J Clin Invest
43: 2070-9.
295. Smith, C., Habermann, E., and Hamerman, D. (1979). "A technique for
investigating the antigenicity of cultured rheumatoid synovial cells." J
Rheumatol 6(2): 147-55.
296. Stachowiak, G. and Batchelor, A.W. (2005) Engineering Tribology. 3rd ed.
Tribology and Interface Engineering Series, No. 24: Elsevier.
297. Stachowiak, G.W., ed. (2005) Wear - Materials, Mechanisms and Practice.
Chichester, John Wiley & Sons Ltd. 458.
298. Stachowiak, G.W. and Podsiadlo, P. (1997). "Analysis of wear particle
boundaries found in sheep knee joints during in vitro wear tests without
muscle compensation." Journal of Biomechanics 30(4): 415-419.
299. Stockwell, R.A. (1965). "Lipid in the matrix of ageing articular cartilage."
Nature 207(995): 427-8.

198
300. Sumner, D.R., et al. (1998). "Functional adaption and ingrowth of bone
vary as a function of hip implant stiffness." Journal of Biomechanics 31:
909-917.
301. Swann, A.D. (1978) Macromolecules of synovial fluid, in The Joints and
Synovial Fluid, L. Sokoloff, Editor, Academic Press Inc.: New York. p.
407-435.
302. Swann, A.D., Silver, H.F., et al. (1985). "The molecular structure and the
lubricating activity of lubricin isolated from bovine and human synovial
fluids." Journal of Biochemistry 225: 195-201.
303. Swann, A.D., Sotman, S.L., et al. (1977). "The isolation and partial
characterization of the major glycoprotein (LGP-I) from the articular
lubricating fraction from bovine synovial fluid." Biochem J 161(3): 473
485.
304. Swann, D. and Mintz, G. (1979). "The isolation and properties of a second
glycoprotein (LGP-II) from the articular lubricating fraction from bovine
synovial fluid." Biochem J 179(3): 465-71.
305. Swann, D.A., Bloch, K.J., et al. (1984). "The lubricating activity of human
synovial fluids." Arthritis Rheum 27(5): 552-6.
306. Swann, D.A., Hendren, R.B., et al. (1981). "The lubricating activity of
synovial fluid glycoproteins." Arthritis Rheum 24(1): 22-30.
307. Swann, D.A. and Radin, E.L. (1972). "The Molecular Basis of Articular
Lubrication. I. Purification and properties of a lubricating fraction from
bovine synovial fluid." J. Biol. Chem. 247(24): 8069-8073.
308. Swann, D.A., Radin, E.L., et al. (1974). "Role of hyaluronic acid in joint
lubrication." Ann Rheum Dis 33(4): 318-26.
309. Swann, D.A., Slayter, H.S., and Silver, F.H. (1981). "The molecular
structure of lubricating glycoprotein-I, the boundary lubricant for articular
cartilage." J. Biol. Chem. 256(11): 5921-5925.
310. Swanson, S.A.V. (1979) Friction, Wear and Lubrication, in Adult Articulat
Cartilage, M.A.R. Freeman, Editor, Pitman Medical: UK. p. 415-460.
311. Szeri, A.Z. (2001) Hydrodynamic and elastohydrodynamic lubrication, in
Modern Tribology Handbook, B. Bhushan, Editor, CRC Press: New York.
p. 383-453.
312. Tandon, P.N., Bong, N.H., and Kushwaha, K. (1994). "A new model for
synovial joint lubrication." International Journal of Bio-Medical Computing
35(2): 125-140.
313. Tanner, R.I. (1966). "An alternative mechanism for the lubrication of
synovial joints." Physics in Medicine and Biology 11: 119.

199
314. Tew, W.P. (1980). "Synovial fluid particle analysis in equine joint disease."
Mod Vet Pract 61(12): 993-7.
315. Tro, N.J. (2003) Chapter 19 Biochemistry, in Introductory Chemistry,
Prentice-Hall Inc.
316. Tsukamoto, Y., Yamamoto, M., et al. (1983). "Boundary lubricating
property of synovial fluid on artificial material and lubrication of artificial
joints." Nippon Seikeigeka Gakkai Zasshi 57(1): 91-9.
317. Uchiyama, S., Amadio, P.C., et al. (1997). "Boundary lubrication between
the tendon and the pulley in the finger." Journal of Bone and Joint Surgery
79A: 213-220.
318. Ueda, S., Kawamura, K., et al. (1985). "Ultrastructural studies on surfaces
lining layer of the lungs. Part IV. Resected human lung." J Jpn Med Soc
Biol Interface 16: 36-60.
319. Ueda, S., Kawamura, K., et al. (1986). "Morphological studies on surface
lining layer of the lungs. Part VI. Surfactant-like substance in other organs
(plueral cavity, vascular lumen and gastric lumen) than lungs." J Jpn Med
Soc Biol Interface 17: 132-156.
320. Unsworth, A. (1975). "Endoprosthetic design." Physiotherapy 61(10): 296-
301.
321. Unsworth, A. (1978). "The effects of lubrication in hip joint prostheses."
Physics in Medicine and Biology 23(2): 253-268.
322. Unsworth, A. (1991). "Tribology of human and artificial joints."
Proceedings of the Institution of Mechanical Engineers. Part H - Journal of
323. Unsworth, A. (1995). "Recent developments in the tribology of artificial
joints." Tribology International 28(7): 485-495.
324. Unsworth, A., Dowson, D., and Wright, V. (1975). "Some new evidence on
human joint lubrication." Ann Rheum Dis 34: 277-285.
325. Virdee, S.S., Wang, F.C., et al. (2003). "Elastohydrodynamic lubrication
analysis of a functionally graded layered bearing surface, with particular
reference to 'cushion form bearings' for artificial knee joints." Proceedings
of the Institution of Mechanical Engineers, Part H: Journal of Engineering
in Medicine 217(3): 191-198.
326. Walker, E.R., Boyd, R.D., et al. (1991). "Morphologic and morphometric
changes in synovial membrane associated with mechanically induced
osteoarthrosis." Arthritis Rheum 34(5): 515-24.
327. Walker, P.S., Dowson, A.D., et al. (1968). ""Boosted Lubrication" in
Synovial Joints by Fluid Entrapment and Enrichment." Ann Rheum Dis 27:
512-520.

200
328. Walker, P.S., Dowson, D., et al. (1969). "Lubrication of human joints." Ann
Rheum Dis 28(2): 194.
329. Walker, P.S., Sikorski, J., et al. (1970). "Features of the synovial fluid film
in human joint lubrication." Nature 225: 956-957.
330. Wang, A., Essner, A., et al. (1998). "Lubrication and wear of ultra-high
molecular weight polyethylene in total joint replacements." Tribology
International 31(1-3): 17-33.
331. Watkins, J.C. (1968). "The surface properties of pure phospholipids in
relation to those of lung extracts." Biochim Biophys Acta 152(2): 293-306.
332. Weinberger, A. and Simkin, P.A. (1989). "Plasma proteins in synovial
fluids of normal human joints." Semin Arthritis Rheum 19(1): 66-76.
333. Widmer, M.R. (2002) Modified molecular friction in artificial hip joints,
Eidgenoessische Technische Hochschule Zuerich (Switzerland):
Switzerland.
334. Wildner, M. and Sangha, O. (2000) Epidemiological and economic aspects
of osteoarthritis, in Osteoarthritis: Fundamentals and Strategies for Joint-
Preserving Treatment, J. Grifka and D.J. Ogilvie-Harris, Editors, Springer-
Verlag: Berlin. p. 1-8.
335. Wilkins, J.F. (1968). "Proteolytic destruction of synovial boundary
lubrication." Nature 219: 1050-1051.
336. Williams III, P.F., Gilbert, J.A., et al. (1997). "Evaluation of the Frictional
Properties of an Elastomer with Enhanced Lipid-Adsorbing Ability."
Proceedings of the Institution of Mechanical Engineers, Part H: Journal of
337. Williams III, P.F., Powell, G.L., and LaBerge, M. (1993). "Sliding friction
analysis of phosphatidylcholine as a boundary lubricant for articular
cartilage." Proc Inst Mech Eng [H] 207(1): 59-66.
338. Williams, J. (2005) Engineering Tribology: Cambridge University Press.
488.
339. Williams, P.F. (1998) General Concepts of Biocompatibility, in Handbook
of Biomaterial Properties, J. Black and G.R. Hastings, Editors, Chapman
and Hall: London. p. 481-489.
340. Williams, P.F., 3rd, Powell, G.L., et al. (1995). "Fabrication and
characterization of dipalmitoylphosphatidylcholine-attracting elastomeric
material for joint replacements." Biomaterials 16(15): 1169-74.
341. Wise, C.M., White, R.E., and Agudelo, C.A. (1987). "Synovial fluid lipid
abnormalities in various disease states: review and classification." Semin
Arthritis Rheum 16(3): 222-30.

201
342. Wobig, M., Dickhut, A., et al. (1998). "Viscosupplementation with hylan G-
F 20: a 26-week controlled trial of efficacy and safety in the osteoarthritic
knee." Clin Ther 20(3): 410-23.
343. Wright, V. and Dowson, D. (1976). "Lubrication and cartilage." Journal of
Anatomy 121: 107-118.
344. Wyn-Jones, G. (1986). "In search of the causes and pathogenesis of
lameness." Equine Vet J 18(3): 163-4.
345. Yehia, S.R. and Duncan, H. (1975). "Synovial fluid analysis." Clin Orthop
Relat Res (107): 11-24.
346. Yielding, K.L., Tomkins, G.M., and Bunim, J.J. (1957). "Synthesis of
hyaluronic acid by human synovial tissue slices." Science 125(3261): 1300.

Lorne Gale Thesis Lubrication

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Lorne Gale Thesis Lubrication

Загружено:

Авторское право:

Доступные форматы

Biotribological Assessment for

Artificial Synovial Joints: The Role

Вам также может понравиться