Materials

& Design
Materials and Design 27 (2006) 287–307
www.elsevier.com/locate/matdes

Hip implants: Paper V. Physiological effects

a,1 b,*
A. Sargeant , T. Goswami
a
Department of Biological Sciences, Ohio Northern University, USA
b
Department of Mechanical Engineering, Ohio Northern University, Ada, OH 45810, USA

Received 28 May 2004; accepted 27 October 2004

Available online 15 December 2004

Abstract

Osteolysis and aseptic loosening are the major causes of failure of total hip replacements. Particles released from the implant
during articulation have been found to be a clinically relevant size (0.1–10 lm). Wear debris particles and toxic metal ions activate
macrophages and stimulate the release of pro-inflammatory mediators IL-1, IL-6, TNF-a and PGE2, which activate osteoclasts and
lead to bone resorption. Data is compiled to reflect trends in the amounts of wear debris produced by different articulating surfaces.
This article overviews the results of clinical studies, in vivo, and laboratory studies, in vitro, and provides a critical discussion of
results.

2004 Elsevier Ltd. All rights reserved.

Keywords: Total hip replacement; Articulating surfaces; Wear mechanisms; Debris; Fixation; Particles; Ultra-high molecular weight polyethelene

1. Introduction costly and increasingly harmful. Currently, 11% of peo-

ple in the younger than 40 age group receive hip im-
This report presents a review of pathophysiological plants. By 2010, the population in this age group is
aspects of a hip implant. A large number of articles were expected to be 40 million. Similarly, by 2030 the popula-
selected for this purpose, of which only [1–133] were tion of this age group is expected to be 80 million, which
used in preparing this review. Therefore, references to is likely to create a need for implants much higher than
articles appear in random sequence. Over 50,000 total current needs. Because of the increasing number of hips
hip replacements (THRs) are performed in the United needed, THRs must be studied further to improve their
Kingdom [7] and 500,000 implants in the United States durability, and thereby decrease the need for revision
each year. While 10 percent of the implanted prosthesis surgeries.
requires revision within 10 years, the average life span of Osteolysis and aseptic loosening, or loosening in the
an implant is only about 15 years. This short life span absence of infection, are the major causes of failure of
creates a problem for the growing number of people THRs. Particles released from the implant as the femo-
receiving implants that are younger than 40 years old. ral head articulates against the acetabular cup during
The problem is compounded since younger people are movement have been found to be of a clinically relevant
generally more active, and more active patients face size (0.1–10 lm) that activates macrophages. Macro-
multiple revision surgeries in their lifetimes that are phages are monocytes, which are a type of leukocyte,
that mature and settle in tissues instead of remaining
*
in the bloodstream. They are components of the immune
Corresponding author. Tel.: +1 419 772 2385.
E-mail address: t-goswami@onu.edu (T. Goswami).
system and synthesize cytokines and growth factors
1
Alene is a third year Biological Sciences (major) student and holds that initiate inflammation and bone resorption. The
a Presidential Scholarship. enzymatic components stimulated by macrophages,

0261-3069/$ - see front matter
2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2004.10.028
288 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

therefore, cause osteolysis and aseptic loosening, which such as neutrophils, macrophages, and natural killer
lead to failure of the implant [21]. (NK) cells. Cytokines are cellular proteins that mediate
Much research has been done to determine the spe- inflammation and communication between cells of the
cific characteristics of wear debris released and the cor- immune system. Adaptive, or specific, immunity is con-
responding physiological response in the area of the hip trolled by lymphocytes and occurs after exposure to a
implant. The objective of current research, therefore, is foreign body. Memory cells play a major role in adap-
to collect data pertaining to the wear debris and metal tive immunity, which includes humoral, or B cell, adap-
ions released from a variety of implant models and artic- tive immunity and T cell-mediated adaptive immunity
ulating surfaces and to summarize data about their [105].
pathophysiological effects. Salient features identified in
this research may lead to the better design of a hip im-
plant that produces the least amount of debris, metal 2. Wear debris
ions, or foreign body response, a model which would
lead to improved long term success of hip implants. The reaction to wear debris particles is a Type IV T
The pathophysiological response caused by foreign Cell-mediated hypersensitivity reaction as a result of
bodies is an involved process containing many immuno- excessive responses against foreign antigens. Because
genic materials. The basic components of the immune wear particles cannot be destroyed after phagocytosis,
response are leukocytes. Leukocytes, or white blood chronic inflammation results. The proinflammatory
cells, consist of 5 different types – 3 of which are granu- mediators produced by macrophages add to weakening
locytic leukocytes and 2 of which are agranulocytic. The of the connective tissue and bone surrounding the im-
three granulocytic leukocytes are neutrophils, eosinoph- plant and results in osteolysis and aseptic loosening
ils, and basophils since they have granules in their nu- [10]. In this type of reaction, helper (TH) and cytotoxic
clei. The two agranulocytic leukocytes are monocytes (TC) T cells are responsible for the reactions. TH cells
and lymphocytes. Monocytes mature, leave the blood- cause delayed-type hypersensitivity and damage tissues
stream, and settle in body tissues, where they become tis- by macrophage activation and cytokine-mediated
sue macrophages and make up the bodyÕs first line of inflammation. TC cells provide T cell-mediated cytolysis
defense. Lymphocytes are subdivided into B cells, which by direct target cell lysis and cytokine-mediated inflam-
mature in bone marrow, and T cells, which mature in mation [105].
the thymus. B cells mature into plasma cells and release When a wear particle is released from the implant
antibodies, while T cells form cytoxic or helper T cells. surfaces into the synovial joint cavity, a macrophage
Only lymphocytes can recognize and differentiate be- engulfs the particle, loses its normal surface topogra-
tween specific antigens [105]. phy, and displays a part of it on its surface, becoming
The immune response is divided into innate immunity an antigen presenting cell, or APC [61]. Fig. 1 shows
and adaptive immunity. Innate, or natural, immunity is the change in surface topography of a macrophage as
responsible for the initial defense against foreign parti- it engulfs a wear debris particle [61]. The macrophage
cles. Its components consist of physical and chemical also has DNA with genes that encode for self-proteins
barriers, blood proteins, cytokines, and phagocytic cells that are known as the MHC, or major histocompatibil-

Fig. 1. SEM photography shows the engulfment of a particle by a macrophage (left) and the change in the surface topography of the macrophage
that results (right) [61].
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 289

ity complex. The MHC is also displayed on the surface grade the organic and inorganic parts of bone, but do
of the APC. The APC produces IL-1, which is a cyto- not cause bone resorption [39].
kine that attracts helper T cells. Interleukins (ILs) are Osteoclasts degrade bone by attaching via a mem-
structurally defined cytokines. IL-1 also mediates host brane extension to create a sealing zone between the
inflammatory responses in innate immunity and stimu- bone and the osteoclast that acts as a lysosome. Proton
lates chemokine production. TH cells have a receptor pumps produce an acidic pH that causes the hydroxyap-
for the wear particle and a receptor for the MHC atite and collagen making up the bone to be dissolved
and binds to the APC. CD4 is a glycoprotein displayed and broken down [106]. Hydrogen ions also directly
on the TH cell that facilitates binding of the antigen stimulate primary afferent nociceptive fibers, especially
and MHC to their respective receptors. Once bound, after previous sensitization by prostaglandins, which is
the TH cell releases IL-2 and interferon-c, which are an action that may be responsible for the pain associated
chemicals that activate cytotoxic, suppressor, and with the inflammatory response. After the bone has been
memory T cells. IFN-c is a cytokine that also activates resorbed, the osteoclasts cease their activity until they
macrophages. IL-2 promotes T-cell and NK cell prolif- attach to a different bone resorptive site [39]. Since the
eration and induces T cell-mediated immune responses. hypersensitivity reaction from the wear debris causes
TC cells inject toxins, release perforins that punch holes macrophages to release cytokines that induce inflamma-
in the cell membrane, and initiate apoptosis of the tion that in turn induces bone resorption, osteolysis re-
APC in an attempt to destroy the wear particle. Sup- sults. Biophosphanates are a class of drugs that inhibit
pressor T cells are activated to prevent cytotoxic T cells osteolysis by becoming incorporated into the crystal
from destroying normal body cells. The exact mecha- structure of hydroxyapatite on new bone surfaces. When
nism by which the foreign body reaction occurs, how- osteoclasts dissolve the hydroxyapatite crystals, they re-
ever, remains unknown [105–107]. lease the biophosphanates that prevent further degrada-
In addition to their above functions, TNFa and IFN- tion by osteoclasts [60].
c stimulate nitric oxide (NO) [58]. NO is an effector mol- Wear debris, Fig. 2, is produced when surface strains
ecule, produced by macrophages, that is involved in accumulate and form ripples in the lining of the joint.
inflammation, immune function, bone metabolism, and Fibrils are created from the rippled protrusions and a
apoptosis [57]. This small, stable free radical is produced wear particle is formed when the fibril ruptures. Produc-
by the enzyme nitric oxide synthase (NOS) and acts as a tion of a wear particle is a three-step process in which
vasodilator to cause inflammation. Apoptosis, especially surface nodules are formed, then surface fibrils form,
in human osteoblast progenitor cells, occurs when NO and finally a debris particle is released from the partially
reacts with superoxide and hydroxyl radicals to form detached fibril. The size and morphology of the particle
peroxynitrate. Peroxynitrate has strong cytotoxic effects depend on the integrity of the polyethylene and other
by causing single-stranded breaks in DNA [57]. TNFa materials that are used in the implant [38].
and IFN-c stimulate NO by increasing the transcription The size and morphology of debris is significant be-
of inducible NOS. This process is controlled by cyto- cause particles must be a critical size to cause macro-
kines and occurs when they activate transcription factor phages to activate cytokines [11]. The biologically
NFkb in macrophages, which in turn increases the tran- active particle size falls in the range of 0.1–10 lm [33]
scription of inducible NOS [58]. and macrophages only opsonize, or ingest, particles in
One result of the inflammatory reaction is bone this specific size range. In a study by Howling et al.
resorption, which is the process of breaking down bone. [21], particles larger than 10 lm in size showed little ef-
Usually, it is a naturally occurring process that works in fect on the production of inflammatory cytokines, while
conjunction with bone formation to replace bones in the cytokine production was significantly increased by parti-
body. Excessive resorption, however, leads to osteolysis cles less than 10 lm [21]. Since the size of the debris
and failure of the prosthesis. The activity of osteoblasts determines its ability to induce a foreign body response,
and osteoclasts, and the process by which their progen- much research, therefore, has been done on patients
itor cells proliferate and differentiate, are controlled by having revision surgeries due to failed hip replacements
the soluble products released by activated macrophages to analyze the size of particles found in tissues surround-
[39]. Osteoblasts are fibroblasts that are responsible for ing the prosthesis. According to a study done by Kobay-
bone formation, while osteoclasts are large multinuclear ashi et al. [63], most of the particles found in the tissue
cells containing many mitochondria that derive from surrounding the hip implant were less than 1 lm, a find-
monocytes and cause bone resorption. Monocytes be- ing that signifies their ability to induce a physiological
come osteoclasts when cells with a molecule called response. Similarly, most of the particles produced by
RANKL (RANK ligand) bind to RANKL receptors multidirectional movement on surfaces made of ultra-
on monocytes and secrete monocyte-colony stimulating high molecular weight polyethylene (UHMWPE) have
factor (M-CSF) that also binds to a receptor on mono- a large number and greater percentage volume of parti-
cytes. After this fusion, the osteoclast precursors de- cles in the joint that are 0.1–10 lm in size.
290 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

Fig. 2. Epistemological size scale of wear debris from materials compared with cellular components and microbes.

The production of wear particles results in a chronic teristics, while biocompatibility is difficult to determine
physiologic response that leads to premature replace- because of the bodyÕs many substances that interact with
ments. Many types of implant models have been pro- the material. Among other functions, the material deter-
duced using different combinations of articulating mines the rigidity, corrosion characteristics, biocompat-
surfaces. They consist of a plastic socket made up of ibility, and tissue receptivity of the implant [89].
either UHMWPE or cross-linked polyethylene, and a Corrosion and the resulting release of metal ions is
metal or ceramic component on the femoral head. The the source of many adverse pathophysiological effects.
UHMWPE that makes up the plastic socket is the com- The susceptibility of a material to corrosion and the ef-
ponent that leads to the failure of THRs since it releases fect of the corrosion debris on the tissues are used to
wear debris. Cross-linked polyethylenes wear at a slower determine a materialÕs biocompatibility [86]. To improve
rate and are, therefore, thought to improve THRs. the biocompatibility of hip implants, and thus their suc-
However, after time they harden and then release as cess rate, many studies have been conducted to analyze
much wear debris as UHMWPE. corrosion of materials used in hip implants and the ef-
The metals that make up the femoral part of the im- fects of their debris. Corrosion can take place in various
plant are either cobalt–chrome (Co–Cr) or titanium al- ways, including pitting, crevice corrosion, fatigue-corro-
loy. Usually, the anatomic medullary locking (AML) sion, stress corrosion, fretting corrosion, galvanic corro-
stem is made of Co–Cr and the socket is made of tita- sion, and intergranular corrosion [94]. Intergranular
nium alloy. Metal-on-metal hip implants, introduced corrosion is the result of mechanical and heat treat-
in 2002, are popular because they have a very slow wear ments, while crevice corrosion is due to sulphur in ami-
rate. Ceramic-on-ceramic implants are newer and a me- no acids [99]. The majority of implants contain
tal oxide is used as the ceramic for both the ball and aluminum, cobalt, copper, chromium, molybdenum,
socket. Although the release of wear debris is decreased, nickel, titanium and various alloys. Since any metal sur-
the concern with metal ions accumulating in the body rounded by biological systems will suffer corrosion
from the metal-on-metal implants and the risk of the [102,92], toxicity, carcinogenicity, and metal sensitivity
ceramic implants fracturing are design considerations from the release of metal ions must be quantified to de-
to these types of implants [108,111,112]. crease patient risk and failure of implants.
The material used in the prosthesis, therefore, plays
an important role in the success of the implant. As sta-
ted by Gotman [86], the success of any material in the 3. Physiology of wear debris
body is determined by the materialÕs biofunctionality
and biocompatibility. Biofunctionality is easily estab- Debris in the synovial cavity of the hip leads to im-
lished by analyzing its physical and mechanical charac- mune response and osteolysis. The physiological re-
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 291

sponse is mediated by proinflammatory cytokines and amount of metal ions that were thought to cause metal-
activated macrophages. Byproducts of the physiological losis and aseptic loosening.
response cause inflammation, apoptosis, necrosis, and Because of these problems, metal-on-polyethylene
bone resorption that lead to aseptic loosening and oste- combinations, especially the Charnley model, were
olysis [66]. Larger amounts of wear debris lead to greater introduced. Engineering deficiencies and problems with
aseptic loosening and osteolysis. UHMWPE acetabular materials used in these high friction implants were later
cup liners result in the largest production of wear debris decided to be primary causes of loosening with the first
[97]. metal-on-metal implants [13,27,31]. A study done by
Metal-on-metal implant models release little debris Higuchi et al. [14] involving a 20-year follow-up of the
but yield increased amounts of ions in the body. All met- McKee Farrar model, whose socket, ball, and stem were
als in a biological environment like that of the body un- made of cast CoCrMo Vitallium by Howmedica,
dergo corrosion, which results in the release of ions showed excellent clinical results. The head diameter of
[84,92]. The behavior of each metal ion is independent the stem was 35 mm and the ball had an equatorial bear-
of other metals present [85]. Duration of the implant ing in the acetabular cup [14]. They found the previous
and health of the patient depend on the fixation materi- failure of metal on metal implants was caused by the
als and articulating surfaces used in the prosthesis [89]. migration of the acetabular component due to the equa-
The behavior and performance of materials are affected torial bearing instead of osteolysis. Similarly, Schmalz-
by the chemical, mechanical, biological, and bioelectri- ried et al. [30] concluded that no problems were
cal events that occur in the environment of the body directly attributable to metal-on-metal bearings. The
[95]. Biofunctionality and biocompatibility of a material metallurgy, diametric bearing clearance, sphericity, and
must be determined to predict its effects in the body. surface finish of metal-on-metal implants may be the
most important engineering factors that account for
3.1. Clinical survey the success of these implants [13].
The metal-on-metal THR was reintroduced in l988 as
Many articulating surfaces; including metal-on-me- the Metasul (Sulzer Medica, Winterthur, Switzerland)
tal, metal-on-polyethylene, ceramic-on-polyethylene, [27]. A cobalt–chrome metal inlay sits in the conven-
and ceramic-on-ceramic were studied in this paper. Ta- tional plastic polyethylene insert. This second genera-
ble 1 compiles such data from a number of clinical tion of metal-on-metal protheses is also marketed
studies and compares wear debris data. It shows that under the name of Sikomet [31]. Many advancements
increased femoral head diameter in uncemented metal have been made in the manufacturing and materials
backed cups leads to increased wear debris and osteol- used in these implants to improve the problems of geom-
ysis, while increased femoral head diameter in ceramic- etry, tolerance, or metallurgy associated with the origi-
on-ceramic implants causes less wear debris released nal metal-on-metal models and to reduce the amount
and less osteolysis [42]. A similar find has been re- of wear, but no long-term evaluations are currently
ported by the senior author earlier [112]. A survey of available [113,27]. A study by Reinisch et al. [36] inves-
8 reports in the literature, conducted by Oparaugo tigated wear in 22 uncemented Metasul implants and
et al. [42] stated the best wear rates (<80 mm3/yr) were determined that abrasive wear was the main mode of
with the 22 and 28 mm femoral head and cemented wear, third-body wear accounted for very little wear,
polyethylene cups [42]. As the volumetric wear rate in- and no manufacturing defects were present in any of
creased, so did the incidence of osteolysis and revision the implants examined [36]. Current concerns regarding
rates. Also, the highest wear rate occurred in studies metal-on-metal implants are the higher systemic release
involving Co–Cr heads [42]. In vivo and in vitro studies of cobalt and chromium as compared to their release in
on the characteristics of metal-on-metal implants are metal-on-polyethylene articulation [17].
first described. In an in vitro study done by Anissian et al. [5] the
Metal-on-metal bearings were first used in total hip wear rates of Metasul (metal-on-metal) and Protasul
arthroplasty over forty years ago [13]. The McKee Far- (metal-on-polyethylene) models were determined using
rar prosthesis was the most common metal-on-metal a 12 channel joint simulator. One million cycles in a sim-
prosthesis. Paterson and Brown were the first scientists ulator was considered equivalent to 1 year of prosthetic
to perform metal-on-metal hip replacements in North use by the patient. The metal used in the femoral heads
America [27]. Initially, studies suggested metal-on-metal of both models was cobalt–chromium. The cup liner in
articulation resulted in carcinogenicity, metal sensitivity the Metasul model was also Co–Cr, while the cup liner
concerns, high infection rates, frictional torque and in- in the Protasul model was polyethylene. The volumetric
creased strain in bone surrounding the implant, and fa- wear rates from metal-on-metal (Metasul) prostheses
tigue fractures of the acetabular floor [17]. Additionally, were almost 100 times less than that of metal-on-poly-
early loosening of the socket presented a problem. The ethylene (Protasul). According to Savarino et al. [17],
high friction of this type of socket released a large in vitro testing showed minimal early wear and a lower
292 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

Table 1
Comparison of wear debris data released by different types of implants from many studies and the correlation of wear debris with osteolysis [42]
A B C D E F G H I
Bankston-i [117] T R-28 CoCr (CE PE) 6-8 28 162 0.05 30.8 A
Bankston-ii [117] T-28 SS (CE PE) 6–8 28 307 0.06 36.9 A
Sochart-i [118] Charnley SS (CE PE) 6–30 22 117 <0.10 <38.9 A 6
Bankston-iii [117] MOSC Ti6-4 (CE PE) 6–8 28 99 0.08 49.2 A
Cates et al. [119] MOSC Ti6-4 (CE PE) 3–10 28 99 0.08 49.2 A
Sochart-ii [118] Charnley SS (CE PE) 6–30 22 36 <0.14 <58.3 A 13
Nashed-i [120] BIAS Ti6-4 (CE PE) 5–10 28 24 0.1 61.5 A 0
Bankston-iv [117] MOSC Ti6-4 (CE-MB) 6–8 28 134 0.11 67.7 A
Cates et al. [119] MOSC Ti6-4 (NC-MB) 3–10 28 134 0.11 67.7 A
Sochart-iii [118] Charnley SS (CE PE) 6–30 22 40 <0.09 <77.7 A 20
Nashed-ii [120] BIAS Ti6-4 (CE-MB) 5–10 28 62 0.13 80 B 31
Thanner et al. [121] PCA Co–Cr (NC-MB) 7–9 28 84 0.15 92.3 B 6
Sochart-iv [118] Charnley SS (CE PE) 6–30 22 22 <0.24 <97.2 B 18
Thanner et al. [121] HGP Ti6-4 (NC-MB) 7–9 28 87 0.16 98.5 B 20
Nashed-iii [120] BIAS Ti6-4 (NC-MB) 5–10 28 74 0.17 104.6 B 24
Devane-i [122] MH Ti6-4 (CE-MB) 4–6 28 12 0.2 123.7 B 17
Bono-i [123] AML + Co–Cr (NC-MB) 2–6 32 72 0.18 144.7 C 21
Nashed-iv [120] BIAS Ti6-4 (NC-MB) 5–10 28 15 0.25 153.9 C 87
Devane-ii [122] MH Ti6-4 (NC-MB) 4–6 28 34 0.3 187.1 C 49
Kim-i [124] AML Co–Cr (NC-MB) 7–11 32 29 0.3 241.2 C 100
Kim-ii [124] AML Co–Cr (NC-MB) 7–11 32 16 0.53 426 C 75
Bono-ii [123] AML + Co–Cr (NC-MB) 2–6 32 15 0.77 619 C 100
Elfick et al. [28] SS (PE) 11.4 22 17 490.3 C
Elfick et al. [28] Co–Cr (PE) 13.9 28 6 819.5 C
Elfick et al. [28] SS,Co–Cr,Ti (PE) 12.8 32 10 1317.9 C
Urban et al. [9] Alumina (CE PE) 17–21 32 64 0.034 28 A 0
Woolson and Murphy [125] Co–Cr (CE PE) 28 0.14
Okumura et al. [126] SS (NC-MB) 22 0.14
Livermore et al. [127] SS (NC-MB) 22 0.13
Livermore et al. [127] Co–Cr (NC-MB) 32 0.1
Livermore et al. [127] SS (NC-MB) 28 0.08
Madey et al. [128] SS (CE PE) 22 0.09
Ohashi et al. [129] Co–Cr (NC-MB) 32 0.04
Ohashi et al. [129] SS (NC-MB) 28 0.04
Saito et al. [45] Alumina (PE) 28 0.1
Sugano et al. [130] Alumina (PE) 28 0.1
Okumura et al. [126] Alumina (PE) 28 0.08
Ohashi et al. [129] Alumina (PE) 28 0.03
Wroblewski et al. [131] Alumina (X-PE) 22 0.03
Wroblewski et al. [35] Charnley SS (CE PE) 15–21 22 93 0.096
Hamadouche [70] Alumina-on-Alumina (CE) 18.5–20.5 32 85 0.025 5.9
Hamadouche [70] Alumina-on-Alumina (NC) 18.5–20.5 32 29 0.025 5.9
Schüller et al. [41] Al2O3-Cer (PE) 9–11 32 33 0.03
Schüller et al. [41] Protasul (PE) 9–11 32 33 0.1
Huo et al. [66] Co–Cr (CE PE) 9.6 28 4 0.104 100
Huo et al. [66] SS (CE PE) 9.6 22 4 0.104 100
Huo et al. [66] Ti (CE PE) 9.6 32 4 0.104 100
CE: cemented, NC: uncemented, PE: polyethylene, MB: metal-backed, X: cross-linked, CoCr: cobalt–chromium, SS: stainless steel, Ti6-4: titanium-
alloy, CW: clinical wear grade, ACS: anatomic cup system, AML: anatomic medullary locking, BIAS: biologic ungrowth anatomic system, HGP:
Harris Galante porous, MH: mallory head, MOSC: Miami orthopedic surgical consultant, PCA: porous-coated anatomic, T 28: trapezoidal, TR 28:
trapezoidal-rectangular, Kim i and ii: 7–10 year duration of follow-up (1999), Sochart i–iv: 4 groups of increasing wear rates in SochartÕs study [118],
Notes on Grade C: the study by Bono et al. [123] was an inferior ACS cup design with 32 mm head size and rim-cracking, Kim et al. studied a high
activity patient group (average 48 years), Devane et al. Õs study was complicated by variations in polyethylene resin content and by design modularity
and fixation [42].
A, reference; B, design, material (fixation); C, follow-up (years); D, femoral head sizes (mm); E, number of hips in study; F, wear-rate mm/yr; G,
volumetric wear rate (mm3/yr); H, CW grade; A, <80 mm3/yr; B, 80–140 mm3/yr; C, >40 mm3/yr; I, osteolysis total (%).
Summary: increased femoral head diameter in uncemented metal-backed cups leads to increased wear debris and osteolysis. Increased femoral head
diameter in ceramic-on-ceramic implants causes less wear debris and therefore lower osteolysis cases. Best wear rates were with 22 and 28 mm femoral
heads and polyethylene (<80 mm3/yr). An increase in volumetric wear rate was related to an increase in osteolysis and revision surgeries. Highest
wear rate was seen in studies involving Co–Cr heads [42].
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 293

occurrence of osteolysis compared to metal-on- balt and chromium have the same affinity for proteins,
polyethylene. nickel significantly competes for cobalt and chromium
binding areas [88].
Once a metal is bound to a protein, it can be trans-
3.2. Electrochemical issues ported within the body or excreted. Cobalt is trans-
ported from tissues to the blood and eliminated in
Metal-on-metal implants thus, may present a prob- the urine within 48 h, while chromium builds up in
lem because they release metal ions into the body. It the tissues and red blood cells [16]. The only ion taken
has been well documented that patients receiving me- up intracellularly by red blood cells following corrosion
tal-on-metal hip replacements have increased metal ions of stainless steel is Cr6+, which is then rapidly con-
in blood and urine analysis [27]. Many researchers have verted to Cr3+. Intracellular Cr3+ exerts mutagenic
examined metal ion release to determine their effect on and carcinogenic properties by interacting with DNA
the immune system, carcinogenesis, toxicity, and metal [16,98]. Cobalt is also mutagenic and carcinogenic.
sensitivity of the patient. Although metal ions are essen- Morais et al. [91] also found that chromium and nickel
tial elements in the body, they can become toxic in high- are retained in bone marrow. Nickel is very small and
er concentrations. The division between normal and has a low affinity for blood cells. Cobalt binds to both
toxic amounts of metals is unclear [27]. In addition, all red blood cells and white blood cells. Although only
metal implants undergo corrosion in the body at a rate very small quantities of Cr3+ bind to cells, Cr6+ binds
determined by their surface area. Large diameter me- very strongly to red blood cells and white blood cells
tal-on-metal bearings, therefore, result in a higher re- [84,85].
lease of ions than bearings of smaller diameter [13]. Because of the toxicity of metal ions, specific values,
Corrosion at the surface of the implant and the wear known as EKA values (Expositionäquivalente für
debris produced cause the release of metal ions [13,97]. Krebserzeugende Arbeitsstoffe), have been developed
When metal exists in a solid form, it exhibits no ad- to relate the allowable air concentrations of substances
verse effects. All metal implants, however, corrode in in the workplace to concentrations in biologic fluids.
the body at a rate determined, in part, by their surface The EKA values of cobalt and chromium in air, blood,
area [13,92,102]. When particles are released from the and urine for cobalt are: air = 0.1 mg/m3, blood = 5 lg/l,
implant, acute inflammation, necrosis, and a chronic urine = 60 lg/l, while for chromium, they are air = 0.1
inflammatory response follows [97]. Biological risks of mg/m3, blood = 17 lg/l, urine = 20 lg/l. Schaffer et al.
metal ions include wear debris, colloidal organometallic [16] found that 24.14% of the cases studied exceeded
complexes, free metal ions, and inorganic metal salts or the blood cobalt EKA threshold limit. The maximum
oxides, which promote adverse effects in the tissue [17]. concentration was 11.5 lg/l. 10.34% of cases exceeded
Metal ions can cause shifts in the pH of the body by the EKA threshold limit for cobalt in urine, in which
participating in oxidation-reduction reactions that ex- the maximum value was 130 lg/l. In addition, some sub-
change electrons [77,90]. They also cause hydrolysis (a jects showed increased chromium urine concentrations
proton exchange) of oxide-hydrates to form metal-or- higher than EKA values with a maximum concentration
ganic complexes. Contact with the foreign body results being 51.8 lg/l [16].
in denatured tissue, which is seen in the redox reactions Comparison between patients with metal prostheses
involving stainless steel, gold, and most other metals ex- to those who have never had a metallic implant surgi-
cept for titanium and tantalum [102]. In CarterÕs study, cal procedure showed increased levels in nickel, cobalt,
iron, copper, chromium, and cobalt ions combined with and chromium in many studies. In a study by Pazza-
oxygen species, including molecular dioxygen, superox- glia et al. [104] the level of nickel in blood, plasma,
ide, and hydrogen peroxide to produce many different and urine and the level of chromium in the plasma
redox reactions. Chromium VI is reduced by glutathione significantly increased. Another study by MacDonald
to form lower oxidation states of chromium that interact et al. [27], showed increased levels of cobalt, chro-
with DNA, possibly creating a serious adverse effect. mium, and titanium in the blood and urine after an
Metal ions bind to proteins to form organometallic average of 3.2 years of implantation. Table 2 shows
complexes. Proteins are Zwitter ions, or ions that can the concentrations of cobalt and chromium in the
become either positively or negatively charged. Based blood and urine released from either a metal-on-metal
on their isoelectric points, most proteins are negatively or metal-on-polyethylene hip implant. In the metal-on-
charged in the bodyÕs pH of 7.4. Positively charged me- metal implants, a 7.9-fold increase was seen in eryth-
tal ions, including cobalt, chromium, and nickel, there- rocyte (red blood cell) cobalt, a 2.3-fold increase in
fore, bind to proteins, changing the pH of albumin erythrocyte chromium, and a 1.7-fold increase in
solutions. Metal ions, especially cobalt and chromium, erythrocyte titanium from preoperative to postopera-
dissolve more in the presence of protein, which increases tive values, indicating elevated plasma levels of these
the corrosion rate of the implant [83,87]. Although co- metal ions [27].
294 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

Table 2
Concentrationsa of cobalt, chromium, and titanium in the blood and urine [27]
Duration of # of Hips Metal-on-metal Metal-on-polyethylene Head Cobalt Chromium Titanium
implant (years) diameter (mm)
Urine Blood Urine Blood Urine Blood
0 22 Co28Cr6M0.08C 28 0.42 0.14 0.26 1.09 0.15 1.04
0 18 UHMWPE 28 0.39 0.11 0.23 0.58 0.09 1.23
2 22 Co28Cr6M0.08C 28 14.7 1.1 4.53 2.5 0.39 1.8
2 18 UHMWPE 28 0.29 0.17 0.3 1.3 0.38 1.5
a
The blood (erythrocyte) concentrations are given in micrograms/liter (equivalent to nanograms per milliliter or parts per billion), while the urine
concentrations are given in micrograms/day (equivalent to nanograms per milliliter or parts per billion) [27].

For the hips with metal-on-polyethylene articulating metal ion concentrations in their blood and urine at
surfaces, a 1.5-fold increase in blood cobalt and a 2.2-fold their two-year follow-up evaluation [27]. In a study by
increase in blood chromium from pre-postoperative val- Huo et al. [66], the concentrations of cobalt, chromium,
ues were observed. The amount of titanium in the blood and barium released from either cobalt–chrome, stain-
from metal-on-polyethylene, however, did not show a less steel, or titanium bearings articulating against ce-
significant difference. The amount of cobalt between the mented polyethylene were determined in tissues [66].
group of metal-on-metal bearings and the group of me- Table 3 shows that the concentrations (lg/gm) of cobalt
tal-on-polyethylene bearings showed a 6.5-fold difference and chromium in the tissues were the highest in tissues
in erythrocyte Co in which the amount released was surrounding titanium bearings compared to cobalt–
higher in the metal-on-metal implants. A significant dif- chrome and stainless steel materials after the same dura-
ference was also seen between the two groups for chro- tion of the implant [66].
mium levels but not for titanium levels [27]. Ions also impair normal osteogenesis and are able to
In the metal ion analysis of urine concentrations, a disturb the differentiation of osteoblasts and the produc-
35.1-fold increase of cobalt, a 17.4-fold increase in chro- tion of bone resorbing agents, except for Cr3+ [17,91].
mium, and a 2.6-fold increase in titanium were observed Hexavalent chromium was labeled as a class-I human
over the two-year span between the preoperative and carcinogen by the International Agency for Research
postoperative values of metal-on-metal prostheses. In on Cancer [17], signifying carcinogenesis as a potential
the metal-on-polyethylene group, no significant increase long-term biological effect in patients with metal-on-me-
in the urine cobalt or chromium concentrations was seen tal hip replacements. Cases of malignant tumors at sites
between pre- and postoperative values. A 4.2-fold in- of metal orthopedic implants have been reported in hu-
crease, however, was seen between the pre- and postop- mans and in animals, making the occurrence of sarcoma
erative values of urine titanium levels for metal-on- at the implantation site a rare complication of implanted
polyethylene bearings. No significant difference between orthopedic prostheses [81]. The presence of kidney and
the two groups of articulating surfaces was observed for prostate cancers, lymphoma, and leukemia is also in-
any of the urine metal concentrations [27]. creased in patients receiving implants [93].
The results of this study showed that more ions were Metal particles have also been correlated with the re-
released from metal-on-metal than metal-on-polyethyl- lease of inflammatory mediators that cause the inflam-
ene articulating surfaces. Another conclusion of this matory response. In a study conducted by Haynes
study was that 59% of patients with metal-on-metal et al. [6], rat peritoneal macrophages were introduced to
prostheses showed decreasing serum and urine metal either titanium–aluminum–vanadium (TiAlV) particles
ion concentrations by their two-year postoperative eval- or cobalt–chromium (Co–Cr) alloy particles obtained
uations. This decrease in concentrations over time is de- from the metal components of hip prostheses. Because
fined as the wearing in phenomenon. Although this both types of these metal particles were approximately
phenomenon suggests that the implant releases less me- the same size and shape, identified differences were di-
tal ions after it has become accustomed to the body, it is rectly related to the chemical properties of the metal al-
inconclusive since the other 49% of subjects had higher loys. Fig. 3 shows that at many different concentrations

Table 3
Comparison of cobalt, chromium, and barium concentrations in tissues with either cobalt–chrome, stainless steel, or titanium bearings articulating
against cemented polyethylene [66]
Author Material Fixation Duration (years) Head diameter (mm) Number of implants Co (lg/gm) Cr (lg/gm) Ba (lg/gm)
Huo et al. [66] Co–Cr (CE PE) 9.6 28 4 77.4 45.2 1260
Huo et al. [66] SS (CE PE) 9.6 22 4 5.97 5.6 861.1
Huo et al. [66] Ti (CE PE) 9.6 32 4 1262.7 733.3
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 295

Fig. 5. Graph showing the effect of titanium–aluminum–vanadium

Fig. 3. Graph showing the effect of titanium–aluminum–vanadium particles (open circles) and cobalt–chromium particles (solid circles) on
particles (open circles) and cobalt–chromium particles (solid circles) on the release of interleukin-1. Exposure to Ti-alloy particles >8 · 106
the viability of rat peritoneal macrophages. Exposure to Co–Cr particles/ml increased IL-1 levels. Each point represents the mean of a
particles >106 particles/ml caused cell death, while Ti-alloy particles least nine experiments. Analysis was done using a paired Student t-test.
had little effect on cell viability. Each point represents the mean of a One asterisk, p < 0.05 and two asterisks, p < 0.005 [6].
least nine experiments. Analysis was done using a paired Student t-test.
Two asterisks, p < 0.005 [6].

greater than 106 particles per milliliter, Co–Cr particles

significantly decreased the percentage of viable cells,
while exposure to TiAlV particles showed no toxicity
even at high concentrations of particles per milliliter
[6]. Cobalt chrome is toxic because macrophages con-
taining metal particles accumulate in the tissues sur-
rounding the prosthesis in response to Co–Cr.
Ulceration and necrosis of the synovial tissue is the
result [12].
Although contact with TiAlV particles did not cause Fig. 6. Graph showing the effect of titanium–aluminum–vanadium
particles (open circles) and cobalt–chromium particles (solid circles) on
cell necrosis in the study by Haynes et al. [6], TiAlV in-
the release of interleukin-6. Exposure to Ti-alloy particles >8 · 106
creased the release of prostaglandin E2 (PGE2), Interleu- particles/ml but not Co–Cr particles increased IL-6 levels. Each point
kin-1 (IL-1), Interleukin-6 (IL-6), and tumor necrosis represents the mean of a least nine experiments. Analysis was done
factor a (TNF-a) [6,15]. Compared with UHMWPE, using a paired Student t-test. One asterisk, p < 0.05 and two asterisks,
TiAlV particles caused a 15-fold increase in the levels p < 0.005 [6].
of IL-1, IL-6, and PGE2 [15]. Fig. 4 shows that at con-
centrations of 1.8 · 106 particles per milliliter or more,
TiAlV particles caused increased release of PGE2, mea- PGE2 release at levels greater than 2 · 106 particles/ml
sured in nanomolars per milliliter (nM/ml), compared [6]. In Fig. 5, TiAlV particles at concentrations larger
with control values, while Co–Cr particles reduced than 8 · 106 particles/ml increased IL-1 activity mea-
sured in units/ml, while Co–Cr particles had no effect
on IL-1 activity [6]. Similarly, in Fig. 6, exposure to
TiAlV particles, but not Co–Cr particles, increased IL-
6 activity [6]. In addition, Fig. 7 shows that the release
of TNF-a was increased with concentrations greater
than 107 particles per milliliter of TiAlV similar to the
pattern of IL-1 [6]. The release of TiAlV into tissues sur-
rounding the implant may be potentially worse than the
same amount of Co–Cr particles because the cells re-
main viable and are able to produce more inflammatory
mediators [6,15].
A study conducted by Rae [82] found that some met-
Fig. 4. Graph showing the effect of titanium–aluminum–vanadium als were toxic because of their high solubility in body
particles (open circles) and cobalt–chromium particles (solid circles) on fluids. Cultures of fibroblasts were poisoned by cobalt
the release of prostaglandin E2. Exposure to Ti-alloy particles
and vanadium, but were not affected under the same
>1.8 · 106 particles/ml increased PGE2 release, while Co–Cr particles
reduced PGE2 release. Each point represents the mean of a least nine conditions by nickel, chromium, titanium or aluminum.
experiments. Analysis was done using a paired Student t-test. One The most harmful components in alloys were deter-
asterisk, p < 0.05 and two asterisks, p < 0.005 [6]. mined to be cobalt from the Co–Cr alloy, nickel from
296 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

conditions also shown to be associated with the degra-

dation products of implants [92]. In the absence of infec-
tion, macrophage phagocytosis and activation occurred
without significant involvement of the immune system
[12,40,65].
Although studies have determined the response to
wear debris produced by implants to be a Type IV T-cell
mediated hypersensitivity reaction [79], other studies
have shown the involvement of humoral immunity as
well. The presence of metal-protein complex-specific
Fig. 7. Graph showing the effect of titanium–aluminum–vanadium antibodies was detected in serum, indicating the forma-
particles (open circles) and cobalt–chromium particles (solid circles) on tion of antibodies against metals [96]. Thus both innate
the release of TNF-a. Exposure to Ti-alloy particles >107 particles/ml and acquired immunity have been shown to be involved
significantly released TNF-a, while Co–Cr particles >1.8 · 106 parti- in the physiological reactions against foreign wear
cles/ml only slightly increased TNF-a release. Each point represents
the mean of a least nine experiments. Analysis was done using a paired
debris.
Student t-test. One asterisk, p < 0.05 and two asterisks, p < 0.005 [6]. The most popular material used in orthopedic im-
plants today is UHMWPE. Over the past 30 years,
many studies have been conducted concerning a stain-
stainless steel and vanadium from the titanium alloy less steel/UHMWPE bearing developed by Sir John
[82]. Doran et al. [100] ascertained that soluble forms Charnley [13,27,31,35,54,118]. Charnley implants are
of chromium, cobalt, and nickel caused significant in- based on the concept of a low friction arthroplasty made
creases in cell transformation in fibroblasts of mice while of either a hard metal or ceramic femoral head that
aluminum, vanadium, or titanium did not induce this ef- articulates against an UHMWPE acetabular cup liner
fect. The transforming ability was directly related to tox- with or without polymethylmethacrylate (PMMA) [59].
icity of metal [100]. Permanent deformation due to creep occurred because
Another study, however, suggested that fibroblasts of sustained body temperature conditions in the poly-
may be resistant to the toxic effects of cobalt and cited ethylene cups. Creep was found to be most important
this as a possible explanation for the persistence of metal in the first post-operative years, but wear became a more
particles in fibrous tissue [12]. Fibroblasts extracted significant factor with time [41]. The long-term use of
from interfacial membranes of patients with failed UHMWPE has shown the formation of large amounts
THRs and from normal tissue released bone-resorbing of wear debris that elicit biological reactions and cause
metalloproteinases and mediators that decreased the osteolysis and aseptic loosening of prostheses. A study
production of collagen by osteoblasts when they were by Dowd et al. [50] revealed that the production of true
exposed to titanium-alloy particles less than 3 lm in size wear was constant and showed a linear pattern as shown
[48]. Fibroblasts also showed a proliferative response to in Fig. 8. Evaluation of early true wear rates may there-
titanium-alloy particles in low concentrations [49,55]. fore be a predictor of late osteolysis [50].
Since each metal behaves independently with respect Several studies have analyzed the tissues surrounding
to the behavior of other metals [84], the side effects di- failed hip implants. Many found a synovial membrane-
rectly relate to the particles that are generated [12,13,78]. like interface tissue (SMLIT) in aseptic loosening of
In the cases of late infections, the accumulation of THRs. The histological features of the SMLIT were
metallic wear particle as well as products of corrosion similar to tissues of rheumatoid arthritis and foreign
may cause a localized reaction and a reduction in the im- body reactions, and cells released cytokines and protein-
mune response. Nickel and cobalt reduced the effective- ases that led to loosening [52,67]. The presence of bone-
ness of the immune response by fifty per cent in killing resorbing cytokines secreted by synovial capsules into
bacteria. Chromium had no effect [103]. A depression the synovial fluid of patients with failed THAs and oste-
of the immune system was correlated with an increase olysis may activate osteoclasts to degrade bone [56].
in metal ions released in the blood in a study done by Maloney et al. [49] demonstrated in an in vitro study
Donati et al. [79]. The toxic action of metal ions, in con- that a fibrous membrane formed around implants in-
junction with the compartmentalization of lymphocytes serted without cement as a result of the metallic debris
in periprosthetic tissues, is responsible for the observed released. This membrane was thought to act as a conduit
decrease in the components of the immune system [79]. allowing polyethylene debris into the space between the
Patients with infection in the periprosthetic tissue implant and bone.
around their implants showed an increased rate of aller- Other researchers, however, have analyzed the tissue
gic reactions to cobalt and nickel compared to patients layers surrounding the implant and have had different
with implants, but no infection, and to persons without observations. Tissues from patients who developed asep-
implants [80]. Dermatitis, urticaria, and vasculitis are tic loosening showed morphological changes including
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 297

peroxy radicals [72]. Collagenase, a destructive proteo-

lytic enzyme present in the inflamed synovial cavity, also
creates a strong oxidative environment [72,10].
Gamma radiation, used to sterilize prostheses, pro-
duces free radicals by breaking the covalent bonds be-
tween carbon and hydrogen in the polymer chains
making up UHMWPE. Oxidation increases the surface
area of UHMWPE by altering its crystalline structure,
and thereby, making the material more susceptible to
biodegradation and the formation of debris [72]. Fiorito
et al. [4] showed that the concentration of free radicals
on polyethylene significantly increased by 30% following
exposure to cultures with inflamed synovial tissues [4].
In very inflamed synovial cell cultures, however, the
number of free radicals on polyethylene decreased be-
cause the highly oxidative environment caused the rapid
decay of free radicals. On the other hand, no free radi-
cals were present in samples of unsterilized UHMWPE
that had not had gamma radiation [72].
In an in vitro study by Yamamoto et al. [38], the ef-
Fig. 8. Graph of predicted head penetration compared to measured
head penetration shows a linear penetration rate [50].
fects of different doses of radiation on UHMWPE ace-
tabular cups liners and the morphology of debris were
quantified following testing in a hip simulator. The higher
markedly cellular tissue that was one to four centimeters the radiation dose, the more crosslinking of polyethyl-
thick, numerous giant cells interspersed in a fibrous stro- ene occurs. Nodules, fibrils, and ripples, which form
ma, and sheets of macrophages. Morphologically, this on the cup surfaces and lead to release of debris, were
tissue had three layers. Large polygonal cells one to increased in non-irradiated cup surfaces. Therefore,
three layers thick with their nuclei polarized away from the surfaces of noncrosslinked polyethylenes were very
the interface made up the layer adjacent to the implant. disrupted after in vitro testing in a hip simulator, while
The intermediate layer contained numerous macro- greater crosslinking from increasing radiation doses
phages and giant cells, and it was the most cellular. In made it harder to extract fibrils from the surface nodules
the layer adjacent to the bone, sheets of macrophages [38]. Table 4 shows that as the radiation dose increased,
infiltrated fibrous marrow and areas of active bone ECD and aspect ratio surface fibrils, nodule density, fi-
resorption were seen [47,63–66]. Fibroblast proliferation bril density, and fibril size decreased, while nodule den-
was also increased in activated mesenchymal tissue pres- sity/mm2 increased or stayed the same [38].
ent around loosened total hip prostheses [53,74]. Crosslinked polyethylenes released lower amounts of
Large numbers of osteoclasts were present in the wear debris, but they produced smaller and more biolog-
resorption cavities. In areas beside these resorption cav- ically active particles that counteracted the benefits of
ities, active bone formation was attempting to create a the low wear [37]. Although low wear rates are seen ini-
barrier to invading tissue. When prostheses were rigidly tially, aging with time decreases the density and hardens
fixed and there was no osteolysis present, no soft tissue the material, resulting in rates similar to the high wear
was present at the cement–bone interfaces. Particulate rates seen with UHMWPE. Wear of moderately cross-
foreign matter was absent and the tissue was predomi- linked polyethylene, however was 500-fold higher than
nately fibrous with few histiocytes. The surface layer alumina-on-alumina wear in microseparation [37,112].
of this tissue did not show organization into the syno- Wear and aging of UHMWPE generates particles,
vial-like membrane that has been described by many ranging in size from 0.1 to 10 lm, that activate macro-
researchers to be present in loose prostheses [63–66]. phages and lead to an immune response [28,54,71]. A
Although the resistance of UHMWPE to friction en- study done in China analyzed the mechanism by which
hances the biofunctionality of this material, its biocom- polyethylene wear debris was produced and led to peri-
patibility is questionable. Like metals, UHMWPE is prosthetic fibrosis and loosening. The results indicated
well tolerated in bulk form but harmful when it is de- that a three-body abrasion was the main model of poly-
graded. Since it is a polymer, it is subject to oxidation ethylene wear. In this model, the three bodies included
in the body, a process that produces many free radicals, the abrasions of the two bearing surfaces plus polynu-
including nitric oxide, hydrogen peroxide, and hydrox- clear giant cells interposed in the surrounding tissues.
ide. In oxygen-rich environments, oxygen diffuses into These giant nuclear cells were considered the most
polymers and reacts with existing free radicals to form important element in the prosthetic loosening [68].
298 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

Table 4
Descriptive statistics for surface fibrils and wear debris [38]
Radiation dose (Mrad) ECD (lm) Aspect ratio (lm) CSF (lm) Nodule Fibril Fibril
density (mm2) density (mm2) size (lm)
Surface Wear Surface Wear Surface Wear
fibrils debris fibrils debris fibrils debris
0 1.89 0.75 39.5 1.5 0.071 0.793 250,000 45,000 >20
2.5 0.76 0.34 5.17 1.66 0.426 0.848 250,000 15,000 1–5
50 0.21 0.3 1.5 1.45 0.711 0.854 >2,000,000 <5000 Rare
100 0.17 0.28 1.48 1.32 0.804 0.942 2,000,000 <5000 Rare
150 0.16 0.44 1.33 1.41 0.889 0.878 <2,000,000 <5000 Rare

4. Biological response [22]. Another in vivo study conducted by Elfick et al.

[28] showed similar results. The ability of particles to mi-
In order to predict the biological responses to wear grate was inversely proportional to their size [28].
debris, all sizes of polyethylene particles in the tissues The means by which wear debris is transported are
must be analyzed [54]. According to Green et al. [11], believed to be cellular transport and diffusion. In cellular
the composition, number, size, surface area, shape, transport, macrophages remove debris by carrying it to
and volume influence the foreign body response medi- the lymphatic system for disposal. These cells, however,
ated by macrophages. After quantification of many are unable to transport any debris that is larger than 10
characteristics of polyethylene particles, it was postu- lm. Multi-nuclear giant cells surround particles larger
lated that submicron particles, or those within the size than 10 lm, making the particles stay closer to the joint
range of 0.1–10 lm, were the most clinically relevant [28]. Another study by Noble et al. [29] determined that
since this is the clinically active size range [76]. Although a relationship exists between particle size and the rate of
the most biologically active UHMWPE particles were diffusion. Smaller particles diffused more rapidly causing
0.24 lm, larger particles had no ability to stimulate mac- them to migrate, while the larger particles remained clo-
rophages to induce bone resorption since the particles ser to the joint [29].
were too big to be phagocytized. They also determined Numerous studies have described the ability of sub-
that the volume and size of particles are critical in the micron-sized debris particles from UHMWPE to mi-
activation of macrophages. A greater number of foreign grate through tissue. In the study by Elfick et al. [28],
bodies in an area cause a greater recruitment of macro- the smallest particle isolated was above 0.1 lm and less
phages to the region, and physical restrictions of macro- than 10 lm. The particle numbers were found to be high
phages make the size of the particle critical in their at all locations in the periprosthetic tissue, where
activation. chronic foreign-body reactions were also occurring.
The particle size of wear debris has been compared to The size of particles disseminated was not related to
the sizes of other antigens and cellular components in the diameter of the femoral head [28].
Fig. 2, establishing an epistemological size scale [110]. Particles from knee and shoulder implants have been
The size scale, ranging from a few Angstroms to 250 determined to be larger than those from hip peripros-
lm, begins with DNA and ends with UHMWPE parti- thetic tissue. This factor may account for the lower inci-
cles. The size range of importance to studies on hip im- dence of osteolysis associated with knee implants than
plants is the clinically active size range, which is with hip implants [23]. Hips show higher degrees of mul-
expanded in the figure and ranges from 0.1 to 10 lm. tidirectional movements than knees, which create a dif-
Many particles from different types of hip prostheses, ference in kinematics between the two types of
including Charnley, UHMWPE, and titanium-alloy par- prostheses. With increased multidirectional movement,
ticles fall within this size range, which is also the size a greater number of debris particles in the size range
range of prokaryotes, the main stimuli for macrophage of 0.1–10 lm are released through the body. This factor
activation in organisms. may also account for the difference in the frequency of
An in vivo study done by Mabrey et al. [22] evaluated osteolysis between knee and hip implants. Recent stud-
the size and morphology of UHMWPE particles ies, however, have shown greater variations in knee
from the hip capsule, synovial fluid, and femoral canal kinetics [21].
tissue. The largest particles were found in the hip capsule The exact method by which UHMWPE particles in
with the size decreasing in the synovial fluid, while the the submicron size lead to osteolysis and bone resorp-
smallest particles were found in the femoral canal tissue tion, has been the subject of much investigation. Studies
because the capsule tissue acted like a trap holding show that these particles activate primary macrophages
the larger particles close to the implant, but allowing causing the macrophages to increase their production of
the smaller particles to migrate to more distant areas the osteolytic cytokines IL-6, IL-1, and TNF-a Since
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 299

TNF-a stimulates the proliferation of osteoclast progen- heads with smaller diameters, such as 22 or 26 mm, be-
itors and activates osteoclasts to resorb bone via osteo- cause two of the three fractures that occurred in the
blasts, TNF-a is thought to be one of the principle study of 68 implants were in models with smaller head
cytokines involved in osteolysis from particles [11]; in diameters [20]. A recent paper by the senior author dis-
this study [11], it was present in higher levels than any cusses this and other clinical studies elsewhere [116].
other cytokine. The pro-inflammatory cytokines IL-1, Use of alumina ceramic implants in patients with sec-
IL-6, TNF-a, and PGE2, the mediator of inflammation, ondary osteoarthrosis, a condition caused by congenital
were determined to recruit and activate macrophages in hip dysplasia or dislocation in patients who were rela-
vivo [11]. tively young at the time of need for a hip implant, was
A similar study compared the effects particles of dif- examined by Saito et al. [45]. The Bioceram prosthesis,
ferent sizes have on the release of the proinflammatory which has a bearing mechanism consisting of an acetab-
cytokines IL-1, IL-6, and PGE2. Table 5 shows that ular component of UHMWPE and an alumina ceramic
the amount of proinflammatory cytokines released de- head, was used in this study. Factors such as decreased
pended on particle size and the ratio of particles to cells wear and decreased component loosening were dis-
[109]. The volume of particles was related to the cell cussed. The surface roughness of the Bioceram prosthe-
number ratio at two different concentrations, either 10 sis has been determined to be less than 0.2 lm for the
lm3:1 or 100 lm3:1. Particles in the 0.24 ± 0.094 lm size Bioceram. This measurement is considered minimal
range caused the greatest amount of IL-1, IL-6, and and allows a lower rate of wear than has been docu-
PGE2 to be released when the ratio of particles to cells mented for the Charnley prosthesis. Also, two types of
was 10 lm3:1. At the higher ratio of particles to cells, narrow stems can be used with the Bioceram prosthesis
the larger sized particles caused greater release of proin- making it highly suitable for patients with anatomic dis-
flammatory cytokines. The size and dose of clinically rel- tortions because the femoral component can be easily
evant UHMWPE particles, therefore, influenced the implanted with minimal risk of femoral fracture [45].
production of pro-inflammatory cytokines [109]. Table 6 shows the results of several studies on the
The osteolysis caused by polyethylene wear debris incidence of acetabular and femoral component loosen-
thus has been determined to be a primary factor in lim- ing [45]. Each successive study shows a reduction in the
iting the longevity of THRs. This problem has led to fur- percentage of acetabular loosening [45]. The Harris
ther investigation of alternative bearing surfaces, prosthesis examined by Gerber and Harris [132] had
including ceramic-on-polyethylene. Alumina ceramic 21.3% acetabular component loosening after a 7-year
was introduced as a bearing surface in the 1970s. This follow-up. Linde et al. [133] and McQueary and John-
material was considered as an alternative because of ston [44] performed approximately 9 year follow up
its inertness, coefficient of friction, wettability, and hard- studies on acetabular loosening in Charnley implants
ness [19,45,70,112]. and found that 12.4% and 8.2%, respectively, of the im-
Since alumina is a brittle material, one of the main plants suffered acetabular component loosening. The
concerns has been its ability to fracture. Since 1977, sur- study by Saito et al. [45] showed that only 7% of the Bio-
gical grade dense alumina ceramic with a low risk of ceram implants examined after 6.2 years failed due to
fracture has been available due to improvement in the acetabular loosening [45]. In the Bioceram prosthesis,
manufacturing process. Out of 707 alumina-on-alumina the ceramic head rather than improved cementing tech-
implants created between 1977 and 1989, prostheses niques was considered the primary cause for the reduc-
made during 1977–1979 were the only ones that suffered tion in acetabular loosening since cementing
femoral head fractures [20]. In a study by Nizard et al. techniques are usually applied to the femoral side and
[20], femoral head diameters of 32 mm were better than not the acetabular side. Two types of narrow stems

Table 5
Volumes of proinflammatory cytokines released in response to different sizes and doses of UHMWPE particles [109]
Particle size (lm ± SD) Particle volume to cell number ratio
10 lm3:1 100 lm3:1
IL-1 IL-6 PGE2 IL-1 IL-6 PGE2
0.24 ± 0.094 0.85 ± 0.2* 5.96 ± 0.47* 24.04 ± 3.53* 0.32 ± 0.08 0.218 ± 0.31 16.29 ± 2.57*
0.45 ± 0.22 0.51 ± 0.23 1.72 ± 1.31 11.97 ± 0.88* 0.61 ± 0.17 6.68 ± 1.9* 21.26 ± 2.54*
1.71 ± 0.99 0.63 ± 0.23 2.58 ± 0.24* 12.06 ± 0.47* 0.85 ± 0.2* 2.8 ± 0.8* 25.34 ± 3.08*
7.62 ± 7.04 0.33 ± 0.08 0.74 ± 0.81 8.71 ± 1.62* 0.32 ± 0.07 1.3 ± 0.65 14 ± 1.3*
88 ± 29 0.32 ± 0.09 0.73 ± 0.89 3.5 ± 1.2 0.32 ± 0.06 0.28 ± 0.06 3 ± 0.5
Culture medium 0.31 ± 0.05 0.33 ± 0.3 2.73 ± 0.63
LPS 1 gxml 1.27 ± 0.58* 8.78 ± 2.5* 25.04 ± 4.03*
300 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

Table 6
Comparison of studies on acetabular loosening of cemented alumina ceramic heads articulating against polyethylene cup liners [45]
Authors # of Hip implants Average follow-up Prosthesis Acetabular component Femoral component
period (year) looseninga (%) looseninga (%)
Gerber and Harris [132] 47 7.1 Harris 21.30 4.30
Linde et al. [133] 129 9 Charnley 12.40 2.30
McQueary and Johnston [134] 61 8.5 Charnley 8.20 4.90
Saito et al. [45] 57 6.2 Bioceram 7.00 3.50
a
Component loosening was defined as component migration, cement fixation, component fracture, or a radiolucent zone over 2 mm wide
involving the entire circumference of the component.

can be used with the Bioceram prosthesis making it than metal-on-polyethylene bearings in a hip simula-
highly suitable for patients with anatomic distortions tor. In addition, ceramic-on-polyethylene articulation
because the femoral component can be easily implanted produced 20 times less wear than the metal-on-poly-
with minimal risk of femoral fracture [45]. ethylene bearings [9]. In this study, the mean linear
Although alumina-on-alumina implants may release and volumetric wear rates of ceramic-on-polyethylene
a very small amount of wear and have little effect on bearings were 0.034 mm/yr and 28 mm3/yr, respec-
the body, they may also result in catastrophic effects tively [9]. These in vitro results correlated with the
and severe wear [43]. According to Plitz [31] wear on lowest reported in vivo ceramic-on-polyethylene wear
alumina prostheses is unavoidable because of the intrin- rates. In a similar study by Tipper [1] involving micro-
sic nature of the material. Few studies have been done to separation of alumina bearings, the rate was 2 mm3/
determine the biological effects of alumina particles in million cycles (or per year) for alumina ceramic-on-
vitro. The studies that have been done used alumina ceramic compared to 30–100 mm3/million cycles for
powder of questionable clinical relevance [31]. To ad- UHMWPE [1].
dress these concerns, Hatton et al. [8] conducted another Zirconia ceramic on polyethylene articulations have
in vitro study. These researchers compared the ability to also been studied [62]. Zirconia ceramics have higher
incite osteolytic cytokine production by human mono- resistance to corrosion and scratching and have an insig-
nuclear phagocytes from donors between commercially nificant amount of ion release compared to metal
available alumina powder particles and alumina wear surfaces. In addition, after 10 million cycles of the pin-
particles produced by microseparation. A hip joint sim- on-disk wear test with PMMA cement, no detectable
ulator was used to produce the alumina wear particles wear or surface roughness were recorded [62]. A com-
by causing the joint to separate in a medial-lateral meth- parison of cobalt–chromium and zirconia femoral heads
od, mimicking human wear [8]. In ceramic-on-ceramic (28 and 22 mm in diameter) articulating against polyeth-
hip implants, joint laxity and microseparation have led ylene was performed by Cales et al. [62]. The polyethyl-
to increased wear in vivo and in vitro. Joint laxity allows ene wear rate was 2–3 times lower for zirconia than for
for separation of the head and the cup during the swing Co–Cr femoral heads [10].
phase of walking causing contact between the head on For any implant to be successful, histological and
the superior rim of the cup [37]. It is believed that during biomechanical/biocompatibility criteria must be met.
heel strike, micrometer sized particles are generated by Histologically, the implant and the bone must be in
intergranular fracture of the ceramic material [8]. close proximity without any gaps or inclusion of fi-
Both types of alumina particles, those from alumina brous tissue. Biomechanically, there must be a connec-
powder and those generated by microseparation, were tion between implant and bone that is functional
able to induce osteolytic cytokine production. A much enough to transmit the weight on the socket without
greater amount of particles from microseparation than motion between the implant and bone. In addition,
from alumina powder were required to promote osteol- the patient must be satisfied without complaints of
ysis. It was considered highly unlikely that the large vol- pain. When these criteria are met, osseointegration is
ume required would be produced even under severe achieved on a longer-term basis. Branemark [94] coined
microseparation conditions because of the low wear the term osseointegration and defined several criteria
rates of ceramic-on-ceramic prostheses. The wear rates necessary for it to occur. The fixation surface of the
of ceramic-on-ceramic have been found to be less than implant must have the capacity to become osseointe-
4 mm3 per million cycles, which is about ten times less grated, there must be contact between the implant
than UHMWPE acetabular cups that wear at 30–100 and viable host bone, and the surgical technique must
mm3 per million cycles. A million cycles has been deter- provide implant stability to the bone [26]. To increase
mined to be equivalent to one year of wear [8]. skeletal fixation, powder metallurgy is used to attach
In a study by Urban et al. [9], however, ceramic-on- within porous metallic surfaces and increase interstitial
ceramic bearings showed up to 100 times less wear bone growth [94].
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 301

5. Fixation methods – bone cement and debris issues optimum shape, and developing technology to maintain
a uniform cement-mantle thickness [65].
In 1959, surgical bone cement made of PMMA was Improvements leading to a solid fixation, such as bet-
introduced to total joint arthroplasty to enhance fixa- ter cementing techniques and hydroxyapatite coatings,
tion [65]. Despite the success of acrylic cement, fragmen- were thought to prevent the spread of particles into
tation of the acrylic cement and the biological the synovial joint cavity and therefore, prevent distal
consequences caused by the acrylic particles challenged osteolysis [2,73,75]. Surface coatings, such as polytetra-
the longevity of joint replacements. Charnley grouted fluoroethylene (Teflon) were supposed to prevent the
the components of hip implants with acrylic bone ce- spread of particles by forming a secondary biological
ment and believed that this cement was well tolerated seal when they become integrated into the bone in the
with loosening occurring because of mechanical factors implant [3]. However, Teflon has shown detrimental ef-
[35]. Other problems dealing with acrylic fragmentation fects leading to increased wear debris. Porous coatings,
typically relate to the differences in the elasticity and the such as those made of Co–Cr alloy, were also popular
tendency of the cement to fracture under heavy loads because of their stabilizing effect that was thought to
[65]. lead to longer duration of fixation [101].
Jasty et al. [65] demonstrated that a tissue response Surface-engineered coatings may also be made of
could be triggered by PMMA alone and could occur metals, such as titanium nitride (TiN), chromium nitride
independently of polyethylene particulate. Although se- (CrN), and chromium carbon nitride (CrCN). The
curely fixed cemented components showed no foreign McMinn design used a peripherally expanded hydroxy-
body tissue reaction, there was a significant reaction apatite coated cup and a cemented metal head [32]. In an
when the cemented femoral components were loose in vitro study by Williams et al. [7], metal coatings were
and acrylic fragmentation had occurred. Since PMMA found to release fewer particles than Co–Cr bearings,
particulate was present in the tissues, they concluded resulting in a tenfold decrease in wear rate [7]. Since
that the transformation of the acrylic polymer to partic- the particles were less than 40 nm in length compared
ulate form was a critical event in the loosening of hip im- to 20 nm sizes of Co–Cr particles, a decreased cytotox-
plants [65]. icity resulted. The decreased wear debris and cytotoxic-
To analyze the physiological response, particulate ity associated with surface engineered coatings may
PMMA was subcutaneously injected into a group of indicate improved clinical performance in comparison
fully immunocompetent mice and into other strains with metal-on-metal implants [7].
of mice with progressively increasing degrees of im- According to recent studies, however, coatings may
mune deficiencies. The PMMA led to a foreign body also cause the release of third body particles into the
response in all the strains of mice. PMMA incited pro- synovial cavity of cementless implants because of hydro-
duction of PGE2 and collagenase, both of which play a lysis, debonding, and delamination of the porous and
role in bone resorption. These findings suggest that the hydroxyapatite coatings [25]. Bloebaum [25] found that
biological reactions to acrylic cement can be instigated 70% of the total particles observed from cemented and
and sustained without a significant contribution by the porous coated implants contained hydroxyapatite coat-
immune system [65]. Similarly, in a study by Kim et al. ing because cement fixation forms a seal that prevents
[18], both cemented and cementless membranes re- particle movement. If cement fragments, porous coating
leased collagenase, PGE2, and IL-1 that led to bone particles, metals, or hydroxyapatite coatings are shed
resorption and aseptic loosening. An in vitro study from the implant component, they can either travel to
by Horowitz et al. [51] indicated macrophages respond the articulating surface of polyethylene, increasing wear
to mechanical failure of cement by phagocytosis of par- of polyethylene, or spread throughout the body [25].
ticles less than 12 lm. TNF-a was produced, but not A 5-year follow-up study of 426 cementless surface
PGE2 upon stimulation by PMMA, indicating a possi- coated hip implants conducted by McAuley [73], showed
ble specificity in mediator production [51]. The occur- bone ingrowths of the fixation materials in 96% of the
rence of bone resorption around non-cemented femoral parts and that the fixation did not weaken with
prosthesis, however, may indicate causes of osteolysis time. Emerson et al. [115] compared hips with circumfer-
other than cement [46]. ential plasma-spray titanium porous coatings to hips
To minimize the bodyÕs response to acrylic fragments, with noncircumferential plasma-spray titanium coating
many advances have been made in cementing tech- for osteolysis [115]. Although the in vitro linear wear
niques. Cleaning the bone and pressurizing the cement rate of polyethylene was the same in both groups of hips
using canal plugs and cement guns have increased pene- (0.189 and 0.187 mm/yr, respectively), the amount of
tration of the cement into the bone. Newer techniques osteolysis was much higher (40% > 10%) in the group
also include reducing the porosity, improving the with the noncircumferential coating. Therefore, circum-
strength of the interface by precoating and texturing, ferential plasma-spray coatings inserted without cement
redesigning femoral components to accomplish a more create a barrier to prevent the interaction between wear
302 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

debris and surrounding bone. Cementless straight- Osteolysis has been well documented to be a particle-
stemmed Co–Cr extensively coated femoral prostheses related phenomenon. The particulate wear has been
are also preferred because they provide long-term fixa- implicated in numerous studies as the instigator of the
tion, good clinical results, and high patient satisfaction ensuing osteolysis which leads to the aseptic loosening.
[2]. Although low wear rates of microscopic debris particles
In alumina-on-alumina prostheses, fixation of the are tolerable, high wear rates lead to adverse physiolog-
acetabular component has caused many mechanical fail- ical effects. When a patient has a high early wear rate
ures over the past 20 years. Three different types of sock- (greater than 0.3 mm per year) and a linear head pene-
ets used have been cemented plain alumina cups that tration pattern within the first five years following
may fracture when a mismatch between the bone, ce- implantation, the risk of osteolysis is greater. Patients
ment, and ceramic occurs, press-fit grooved plain alu- with these high risk factors, therefore, should have more
mina cups, and smooth screw-in shells with an frequent follow-up evaluations [50].
alumina core in which the screws migrate [24]. Cement- The joint fluid in patients with osteolysis following
less metal-backed sockets are used with polyethylene to THRs contains high levels of bone-resorbing cytokines
get a stable fixation via integration into bone. Since they produced by the synovium capsule and macrophages
reduce polyethylene creep, press-fit metal-backed sock- stimulated by wear debris. Cytokines play a critical role
ets improve fixation of the acetabular component in alu- in the bone resorption stimulated by osteoclasts. The
mina-on-alumina THRs [24]. A self-locking principle most critical factor associated with bone resorption
combined with alumina ceramic that is produced by was the presence of particles small enough to be phago-
all manufacturing companies in the world, is also favor- cytized by macrophages. The use of biophosphates,
able, especially in younger and more active patients [34]. which have been used successfully to treat osteoporosis
Another source of debris that worsens the foreign and PagetÕs disease, may be effective in preventing wear
body reaction is the release of plastic from intramedul- debris-induced osteolysis without significant effects on
lary brushes used in modern cementing techniques. In the biomechanical and compositional properties of
a study by Germain et al. [33], the debris released by bone.
brushes was found in the cement–bone interface and The process of osteolysis has commonly been consid-
99.5% were less than 1 lm in size. Although the femoral ered a chronic inflammatory response that has an insid-
canal is irrigated after brushing, not all of the particles ious onset. Studies have shown, however, that
are removed and they are able to elicit a foreign body re- particulate debris may instigate an acute inflammatory
sponse. In a comparison between new brushes and auto- response apart from, or in addition to, the chronic
claved brushes, autoclaved and reused brushes produced inflammation that results. Preventing the development
more particles [33]. of an acute inflammatory process may be a possible area
of intervention in future cases.
The low-frictional torque of the Charnley implant
6. Closing statements may be a reliable design if component fixation can be
maintained and particulate wear debris is minimized.
Aseptic loosening continues to be the primary cause No relationship was found between the depth of wear
of the long-term failure of THRs. Hip prostheses are of- and the patientÕs weight in the research reviewed on
ten implanted in inflamed osteoarthritic joint cavities, the Charnley design. The amount of wear of polyethyl-
and the enzymatic substances and reactive oxygen spe- ene was found to be the major factor responsible for
cies produced by the inflamed synoviocytes can institute loosening of the socket, limiting the use of the Charnley
a breaking of the polymer chains and increase the level model in the very young and the very active. In all de-
of free radicals at the onset of implantation. This tissue signs of metal-on-polyethylene, a smaller femoral head
response can promote the failure of the implant by pro- diameter articulating against a cemented all-polyethyl-
moting osteolysis at the time of the initial surgery. A ene cup gave better results and lower incidences of
synoviectomy may be necessary at the time of surgery, osteolysis.
therefore, to eliminate inflammatory products and to Metal-on-metal THRs offer lower wear rates than
prevent the spread of inflamed synovial tissue outside metal-on-polyethylene THRs because of less wear debris
the joint cavity. generated, and thus, less risk for osteolysis. Significant
All prosthetic components including metal, ceramic, differences in designs, materials, methods of manufac-
polyethylene, plastic brushes, surface coatings, and fixa- turing, and patient tolerances between the metal-on-me-
tion materials have been demonstrated to generate par- tal implants exist. Many issues with metal-on-metal
ticulate wear debris. A successful THR must have a low implants remain unresolved including ideal head size
wear rate and low biological activity in response to the and the effect of patient activity level. Increased metal
particles. The polyethylene bearing surface has resulted ions in the blood and urine in patients with metal-on-
in the greatest production of wear debris. metal implants have been well documented.
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 303

A difference in the cellular response to different types using guided bone regeneration. Clin Orthop Related Res
of metal-alloy particles of the same size has been demon- 2000;355:192–204.
[4] Fiorito S, Goze C, Adrey J, Magrini L, Goalard C, Bernier P.
strated in several studies. Even though cobalt–chro- Increase in free radicals on UHMWPE hip prostheses compo-
mium particles have been found to be the most toxic nents due to inflamed synovial cell products. Biomed Mater Res
to tissue, particles from less toxic metal-alloys may be 2001;57:35–40.
worse because of their ability to cause the release of [5] Anissian LH, Stark A, Gustafson A, Good V, Clarke IC. Metal-
inflammatory mediators. Although many studies have on-metal bearing in hip prosthesis generates 100-fold less wear
debris than metal on polyethylene. Acta Orthop Scand
shown potentially adverse pathophysiological effects 1999;70(6):578–82.
related to metal ions including Co, Cr, Ti, Al, and Ni [6] Haynes DR, Rogers SD, Hay S, Pearcy MJ, Howie DW. The
in the human body, the definite effects have yet to be differences in toxicity and release of bone-resorbing mediators
determined. Toxicity, carcinogenicity, and metal allergy induced by titanium and cobalt–chromium-alloy wear particles.
are the most significant concerns. J Bone Joint Surg A 1993;75(6):825–34.
[7] Williams S, Tipper JL, Ingham E, Stone MH, Fisher J. In vitro
Compared to the other types of bearings, ceramics analysis of the wear, wear debris and biological activity of
bearing surfaces have the lowest wear rates. Because of surface-engineered coatings for use in metal-on-metal total hip
these low wear rates and the improvement in the quality replacements. Proc Inst Mech Eng 2003;217(H):155–63.
of the ceramic material to reduce the risk of fracture, the [8] Hatton A, Nevelos JE, Mathews JB, Fisher J, Ingham E. Effects
use of ceramics in hip arthroplasties, especially for youn- of clinically relevant alumina ceramic wear particles on TNF-a
production by human peripheral blood mononuclear phago-
ger and more active patients, may provide the best cytes. Biomaterials 2003;24:1193–204.
opportunity for the long-term success of hip implants. [9] Urban JA, Garvin KL, Boese CK, Bryson L, Pederson DR,
Ceramic-on-ceramic implants can be expected to last Callaghan JJ, et al. Ceramic-on-polyethylene bearing surfaces in
twenty years. Concerns with the use of ceramic-on-cera- total hip arthroplasty. J Bone Joint Surg 2001;83A(11):1688–94.
mic implants, however, are the weight of the patient and [10] Lassus. Aseptic loosening of orthopaedic joint prosthesis. Clin
Orthop Related Res 1998;352:13–4.
acetabular cup loosening. Research has shown that the [11] Green TR, Fisher J, Stone M, Wroblewski BM, Ingham E.
volume of material degraded is proportional to the mass Polyethylene particles of a Ôcritical sizeÕ are necessary for the
of the patient and the patientÕs activity level. induction of cytokines by macrophages in vitro. Biomaterials
The need to render the surfaces of implant compo- 1998;19:2297–302.
nents clean and free of debris has been well documented [12] Howie DW, Roberts V. The synovial response to intraarticular
cobalt–chrome wear particles. Clin Orthop Related Res
through research. The need to design implants that re- 1988;232:244–53.
lease minimal debris is necessary to reduce the number [13] Clarke MT, Lee A. Arora Levels of metal ions after small- and
of revision surgeries. Surface engineering technology large-diameter metal-on-metal hip arthroplasty. J Bone Joint
has been used to address these issues as well as to im- Surg 2003;85(6):913–7.
prove wear performance through the development of [14] Higuchi F, Inoue A, Semlitsch. Metal-on-metal Co–Cr–Mo
McKee–Farrar total hip arthroplasty: characteristics from a
surface coatings. long-term follow up study. Arch Orthop Trauma Surg
Despite continuing research and improvements in 1997;116:121–4.
components of hip implants, an ideal hip prosthesis [15] Yang I, Kim S, Rubash HE, Shanbhag AS. Fabrication of
has yet to be created. Metal releases potentially harmful submicron titanium-alloy particles for biological response stud-
ions, ceramic is susceptible to fracture, and polyethylene ies. Biomed Mater Res 1999;48:220–3.
[16] Schaffer AW, Pilger A, Engelhardt C, Zweymueller K, Ruediger
yields high amounts of wear debris that lead to adverse HW. Increased blood cobalt and chromium after total hip
physiological effects. Since the number of hip prostheses replacement. Clin Toxicol 1999;37(7):839–44.
implanted per year is increasing, especially in the youn- [17] Savarino L, Granchi D, Ciapetti G, Cenni E, Pantoli A
ger age groups, more long term studies need to be con- Nardi, Rotini R, et al. Ion release in patients with metal-on-
ducted and the biofunctionality and biocompatibility of metal hip bearings in total joint replacement: a comparison
with metal-on-polyethylene bearings. Biomed Mater Res
new materials need to be assessed to improve the success 2002;63:467–74.
rate of THRs. [18] Kim KJ, Rubash HE, Wilson SC, DÕAntonio JA, McClain EJ. A
histological and biochemical comparison of the interface tissues
in cementless and cemented hip prostheses. Cementless Cemen-
ted Hip Prostheses 1993;287:142–51.
References [19] Hamadouche M, Boutin P, Dassange J, Bolander ME, Sedel L.
Alumina-on-alumina total hip arthroplasty a minimum 18.5
[1] Tipper JL, Hatton A, Nevelos JE, Ingham E, Doyle C, Streicher years follow-up study. J Bone Joint Surg 2002;84A:663–5.
R, et al. Alumina–alumina artificial hip joint. Part II. Charac- [20] Nizard RS, Sedel L, Christel P, Meunier A, Soudry M, Witvoet
terization of the wear debris from in vitro hip joint simulations. J. Ten-year survivorship of cemented ceramic-ceramic total hip
Biomaterials 2002;23:3441–8. prostheses. Ceramic–Ceramic THA 1992;282:54–60.
[2] McAuley JP, Culpepper WJ, Engh CA. Total hip arthroplasty [21] Howling GI, Barnett PI, Tipper JL, Stone MH, Fisher J, Ingham
concerns with extensively porous coated femoral components. E. Quantitative characterization of polyethylene debris isolated
Clin Orthop Related Res 1998;355:182–8. from periprosthetic tissue in early failure knee implants and early
[3] Bhumbra RS, Walker PS, Berman AB, Emmanual J, Barrett DS, and late failure Charnley hip implants. Biomed Mater Res
Blunn GW. Prevention of loosening in total hip replacements 2001;58:415–20.
304 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

[22] Mabrey JD, Afsar-Keshmiri A, McClung II II, Pember II GA. [42] Oparaugo PC, Clarke IC, Malchau H, Herberts P. Correlation
Comparison of UHMWPE particles in synovial fluid and tissues of wear debris-induced osteolysis and revision with volumetric
from failed THA. Biomed Mater Res 2001;58:196–202. wear-rates of polyehtylene. Acta Ortho Scand 2001;72(1):22–8.
[23] Wirth MA, Agrawal CM, Mabrey JD, Dean DD, Blanchard [43] Prudhammeaux F, Nevelos J, Meunier A, Hamadouche M,
CR, Miller MA. Isolation and characterization of polyethylene Doyle C, Sedel L. Wear of alumin-on-alumina total hip
wear debris associated with osteolysis following total shoulder arthroplasties at a mean 11-year followup. Clin Orthop Related
arthroplasty. J Bone Joint Surg 1999;81A:29–37. Res 2000;379:113–21.
[24] Bizot P, Larrouy M, Witvoet J, Sedel L, Nizard R. Press-fit [44] McQueary FG, Johnston RC. Coxarthrosis after congenital
metal-backed alumina sockets a minimum 5-year follow up dysplasia. Treatment by total hip arthroplasty without acetab-
study. Clin Orthop Related Res 2000;379:134–42. ular bone-grafting. J Bone Joint Surg Am 1988;70(8):1140–4.
[25] Bloebaum RD, Zou L, Bachus KN, Shea KG, Hofmann AA, [45] Saito M, Saito S, Ohzono K, Takaoka K, Ono K. Efficacy of
Dunn HK. Analysis of particles in acetabular components from alumina ceramic heads for cemented total hip arthroplasty. Clin
patients with osteolysis. Clin Orthop Related Res Orthop Related Res 1990;283:171–6.
1997;338:109–18. [46] Mohanty M. Cellular basis for failure of joint prostheses.
[26] Maloney WJ, Anderson M, Harris WH, Kraay M, Rubash HE, Biomed Sci Instrum 1996;32:119–25.
Galante JO, et al. Acetabular fixation in primary total hip [47] Tucci M, Tsao Hughes Jr J. Analysis of capsular tissue from
arthroplasty fixation, polyethylene wear, and pelvic osteolysis in patients undergoing primary and revision total hip arthroplasty.
primary total hip replacement. Clin Orthop Related Res J Bone Minor Res 1995;10(9):1417–27.
1999;369:157–64. [48] Yao J, Glant TT, Lark MW, Mikeez K, Jacobs JJ, Hutchinson
[27] MacDonald SJ, McCalden RW, Chess DG, Bourne RB, NI, et al. The potential role of fibroblasts in periprosthetic
Rorabeck CH, Cleland D, et al. Metal-on-metal vs. polyethyl- osteolysis. Fibroblast Response Titanium Particles 1992;285:
ene in hip arthroplasty: a randomized clinical trial. Clin Related 116–28.
Res 2003;406:282–96. [49] Maloney MJ, Smith L, Castro F, Schurman DJ. Fibroblast
[28] Elfick AP, Green SM, Krikler S, Unsworth A. The nature and response to metallic debris in vitro. J Bone Joint Surg
dissemination of UHMWPE wear debris retrieved from peri- 1993;75(6):835–43.
prosthetic tissue of THR. Biomed Mater Res 2003;65A:95–108. [50] Dowd J, Sychterz C, Young A, Engh C. Characterization of
[29] Noble J, Jones AG, Davies MA, Sledge CB, Kramer RL, Livni long-term femoral-head-penetration rates. J Bone Joint Surg
E. Leakage of radioactive particle systems from a synovial joint 2000;82(8):1102–7.
studied with a gamma camera. Its approach to a radiation [51] Horowitz S, Doty S, Lane J, Burstein A. Studies of a mechanism
synoviectomy. J Bone Joint Surg 1983;65A:381–9. by which the mechanical failure of polyethylmethacrylate leads
[30] Schmalzried TP, Fowble VA, Ure KJ, Amstutz HC. Metal-on- to bone resorbtion. J Bone Joint Surg 1993;75(6):802–13.
metal surface replacement of the hip, technique, fixation and [52] Li T, Warris V, Ma J, Lassus J, Yoshida T, Santavirta S, et al.
early results. Clin Orthop Related Res 1996;329:5106–14. Distribution of tenscin-X in different synovial samples and
[31] Plitz W, Veihelmann A, Pellengahr C. Metal-on-metal pairs for synovial membrane-like interface tissue from aseptic loosening
hip prostheses. Der Orthopade 2003;23(1):17–22. of total hip replacement. Rheumatol Int 2000;19:177–83.
[32] McMinn D, Treacy R, Lin K, Pynsent P. Metal-on-metal surface [53] Santavirta S, Xu J, Hietanen J, Ceponis A, Sorsa T, Kontio R,
replacement of the hip. Clin Orthop Related Res et al. Activation of periprostetic connective tissue in asceptic
1996;329:589–98. loosening of total hip replacements. Clin Orthop Related Res
[33] Germain MA, Tipper JL, Ingham E, Makeram R, Wroblewski 1998;352:16–24.
BM, Fisher J. Analysis of debris from brushing the femoral canal [54] Tipper J, Ingham JL, Hailey A, Besong A, Fisher J, et al.
with a plastic brush – a potential cause of loosening in total hip Quantitative analysis of polyethylene wear debris, wear rate and
replacement. Inst Mech Eng 1999;213:503–6. head damage in retrieved Charnley hip prostheses. J Mater Sci:
[34] Mittelmier H, Heisel J. Sixteen-yearsÕ experience with ceramic Mater Med II 2000:117–24.
hip prostheses. Clin Orthop Related Res 1992;282:64–71. [55] Shanbhag A, Jacons J, Black J, Galante J, Glant T. Effects of
[35] Wroblewski BM. 15–21 year results of the charnley low-friction particles on fibroblast proliferation and bone resorbtion in vitro.
arthroplasty. Clin Orthop Related Res 1986;211. Clin Orthop Related Res 1997;342:205–17.
[36] Reinisch G, Judmann KP, Lhotka C, Lintner F, Zweymuller [56] Kim K, Hijikata H, Itoh T, Kumegawa M. Joint fluid from
KA. Retrieval study of uncemented metal–metal hip prostheses patients with failed total hip arthroplasty stimulates pit formtion
revised for early loosening. Biomaterials 2003;24:1081–91. by mouse osteoclasts on dentin slices 1997:234–40.
[37] Williams S, Butterfield M, Stewart T, Ingham E, Stone M, [57] Puskas B, Menke N, Huie P, Song Y, Kier E, Trindade M, Smith
Fisher J. Wear and deformation of ceramic-on-polyethylene R, Goodman S. Expression of nitric oxide peroxynitrate, and
total hip replacements with joint laxity and swing phase apoptosis in loose total hip replacements. Biomed Maters Res
microseperation. Inst Mech Eng 2003;217(H):147–53. 2003;66:541–9.
[38] Yamamoto K, Clarke IC, Masoaka T, Oonishi H, Williams PA, [58] Stea S, Visentin M, Donati M, Granchi D, Ciapetti G, Sudanese
Good VD, et al. Microwear phenomena of ultrahigh molecular A, et al. Nitric oxide synthase in tissues around failed hip
weight polyethylene cups and debris morphology related to c prostheses. Biomaterials 2002;23:4833–8.
radiation dose in simulator study. Biomed Mater Res [59] Green T, Fisher J, Mathews JB, Stone M, Ingham E. Effect of
2001;56:65–73. size and dose on bone resorbtion activity of macrophages by in
[39] Lassus J, Jiranek WA, Nevalainen J, Horák P, Salo J, Santavirta vitro clinically relevant ultra high molecular weight polyethylene
S, et al. Macrophage activation results in bone resorption. Clin particles 2000:490–6.
Orthop Related Res 1998;352:7–15. [60] Wang X, Shanbhag A, Rubash H, Agrawal C. Mauli short-term
[40] Lee SH, Brennan FR, Jacobs JJ, Urban RM, Ragasa DR, Glant effects of bisphosphonates on the biomechanical properties of
TT. Human monocyte/macrophage response to cobalt–chro- canine bone. Biomed Mater Res 1999;44:456–60.
mium corrosion products and titanium particles in patients with [61] Boynton EL, Waddell J, Meek E, LaBow RS, Edwards V,
total joint replacements. Biomed Mater Eng 1996;6(3):165–72. Santerre JP. The effect of polyethylene particle chemistry on
[41] Schüller Hans M, Marti René K. Ten year socket wear in 66 hip human monocyte-macrophage function in vitro. Biomed Mater
arthroplasties. Acta Orthop Scand 1990;61(3):240–3. Res 2000;52:239–45.
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 305

[62] Cales wear behavior of zirconia femoral heads. Clin Orthop [83] Merritt K, Brown SA. Effect of proteins and pH on fretting
Related Res 2000;379:101–9. corrosion and metal ion release. J Biomed Mater Res
[63] Kobayashi A, Bonfield W, Kadoya Y, Yamac T, Freeman M, 1988;22(2):111–20.
Scott G, et al. The size and shape of particulate polyethylene [84] Merritt K, Brown SA, Sharkey NA. The binding of metal salts
wear debris in total joint replacement. Inst Mech Eng and corrosi on products to cells and proteins in vitro. J Biomed
1997;211(H):11–5. Mater Res 1984;18(9):1005–15.
[64] Chen F, Scher D, Clancy R, Vera0Yu A, Cesare P. In vitro and [85] Merritt K, Brown SA, Sharkey NA. Blood distribution of nickel,
in vivo activation of polymorphonuclear leukocytes in response cobalt, and chromium following intramuscular injection into
to particulare debris. Biomed Mater Res 1999;48:904–12. hamsters. J Biomed Mater Res 1984;18(9):991–1004.
[65] Jasty M, Jiranek W, Harris W. Acrylic fragmentation in total hip [86] Gotman I. Characteristics of metals used in implants. J Endourol
replacements and its biological consequences. Clin Orthop 1997;11(6):383–9.
Related Res 1992;285:116–27. [87] Clark GC, Williams DF. The effects of proteins on metallic
[66] Huo M, Salvati E, Lieberman J, Betts F, Bansai M. Metallic corrosion. J Biomed Mater Res 1982;16(2):125–34.
debris in femoral endosteolysis in failed cemented total hip [88] Yang J, Black J. Competitive binding of chromium, cobalt, and
arthroplasties. Clin Orthop Related Res 1992;276:157–9. nickel to serum proteins. Biomaterials 1994;15(4):262–8.
[67] Stea S, Visentin M, Granchi D, Melchiorri C, Soldati S, [89] Simon JP, Fabry G. An overview of implant materials. Acta
Sudanese A, et al. Montanaro wear debris and cytokine Orthop Belg 1991;57(1):1–5.
production in the interface membrane of loosened prostheses. [90] Carter DE. Oxidation–reduction reactions of metal ions. Envi-
Biomater Polym Ed 1999;10(2):247–57. ron Health Perspect 1995;103(Suppl 1):17–9.
[68] Wany Y, Dai K, Zhang P. Polyethylene wear in periprosthetic [91] Morais S, Dias N, Soussa JP, Fernandes MH, Carvalho GS. In
tissue of failed total hip arthroplasties. Zhonghua Wai Ke Za Zhi vitro osteoblastic differentiation of human bone marrow cells in
1997;35(8):459–61. the presence of metal ions. J Biomed Mater Res 1999;44(2):
[69] Willert H, Bertram H, Buchhorn G. Osteolysis in alloarthropl- 176–90.
asty of the hip. The role of ultra-high molecular weight [92] Hallab N, Merritt K, Jacobs JJ. Metal sensitivity in patients with
polyethylene wear particles. Clin Orthop Related Res orthopaedic implants. J Bone Joint Surg Am 2001;83A(3):
1990:95–103. 428–36.
[70] Hamadouche M, Boutin P, Daussange J, Bolander M, Sedel L. [93] Berkenstock OL. Issues concerning possible cobalt–chromium
Alumina-on alumina total hip arthroplasty. A minimum 18.5- carcinogenicity: a literature review and discussion. Contemp
year follow-up study. J Bone Joint Surg 2002;84A(1):69–77. Orthop 1992;24(3):265–78.
[71] Bell J, Tipper JL, Ingham E, Stone MH, Wroblewski BM, Fisher [94] Lopez GD. Biodeterioration and corrosion of metallic implants
J. Quantitative analysis of UHMWPE wear debris isolated from and prostheses. Medicina (B Aires) 1993;53(3):260–74.
the periprosthetic femoral tissues from a series of Charnley total [95] Bundy KJ. Corrosion and other electrochemical aspects of
hip arthroplasties. J Bone Joint Surg Br 2001;83(7):1075–81. biomaterials. Crit Rev Biomed Eng 1994;22(3–4):139–251.
[72] Fiorito S, Goze C, Adrey J, Magrini L, Goalard C, Bernier P. [96] Yang J, Merritt K. Detection of antibodies against corrosion
Increase in free radicals on UHMWPE hip prostheses compo- products in patients after Co–Cr total joint replacements. J
nents due to inflamed synovial cell products. Biomed Mater Res Biomed Mater Res 1994;28(11):1249–58.
2001;58(4):415–20. [97] Howie DW, Rogers SD, McGee MA, Haynes DR. Biologic
[73] McAuley JP, Culpepper WJ, Engh CA. Total hip arthroplasty. effects of cobalt–chrome in cell and animal models. Clin Orthop
Concerns with extensively porous coated femoral components. J 1996;329(Suppl):S217–32.
Biomed Mater Res 1998;43(3):234–40. [98] Merritt K, Brown SA. Release of hexavalent chromium from
[74] Santavirta S, Xu JW, Hietanen J, Ceponis A, Sorsa T, Kontio R, corrosion of stainless steel and cobalt–chromium alloys. J
et al. Activation of periprosthetic connective tissue in aseptic Biomed Mater Res 1995;29(5):627–33.
loosening of total hip replacements. Acta Orthop Scand Suppl [99] Traisnel M, le Maguer D, Hildebrand HF, Iost A. Corrosion of
1998;278:1–16. surgical implants. Environ Health Perspect 1981;40:207–26.
[75] Kadoya Y, Kobayashi A, Ohashi H. Wear and osteolysis in total [100] Doran A, Law FC, Allen MJ, Rushton N. Neoplastic
joint replacements. J Arthroplasty 1997;12(8):946–9. transformation of cells by soluble but not particulate forms
[76] Campbell P, Ma S, Yeom B, McKellop H, Schmalzried TP, of metals used in orthopaedic implants. Biomaterials
Amstutz HC. Isolation of predominantly submicron-sized 1998;19(7–9):751–9.
UHMWPE wear particles from periprosthetic tissues. J Biomed [101] Placko HE, Brown SA, Payer JH. Effects of microstructure on
Mater Res 1995;29(1):127–31. the corrosion behavior of Co–Cr porous coatings on orthopedic
[77] Bundy KJ, Kelly RG, Brown EL, Delahunty CM. pH shifts and implants. J Biomed Mater Res 1998;39(2):292–9.
precipitation associated with metal ions in tissue culture. [102] Steinemann SG. Metal implants and surface reactions. Injury
Biomaterials 1985;6(2):89–96. 1996;27(Suppl 3):SC16–22.
[78] McNamara A, Williams DF. Scanning electron microscopy of [103] Rae T. The action of cobalt, nickel, and chromium on
the metal-tissue interface. II. Observations with lead, copper, phagocytosis and bacterial killing by human polymorphonuclear
nickel, aluminum, and cobalt. Biomaterials 1982;3(3):165–76. leucocytes; its relevance to infection after total joint arthroplasty.
[79] Donati ME, Savarino L, Granchi D, Ciapetti G, Cervellati M, Biomaterials 1983;4(3):175–80.
Rotini R, Pizzoferrato A. The effects of metal corrosion debris [104] Pazzaglia UE, Minoia C, Ceciliani L, Riccardi C. Metal
on immune system cells. Chir Organi Mov 1998;83(4):387–93. determination in organic fluids of patients with stainless steel
[80] Hierholzer S, Hierholzer G. Allergy to metal following osteo- hip arthroplasty. Acta Orthop Scand 1983;54(4):574–9.
synthesis. Unfallchirurgie 1982;8(6):347–52. [105] Abbas, Lichtman, Pober. Cellular and molecular immunology.
[81] Sunderman Jr FW. Carcinogenicity of metal alloys in orthopedic 4th ed. New York: WB Saunders Company; 2000.
prostheses: clinical and experimental studies. Fundam Appl [106] Ganong. Review of medical physiology review of medical
Toxicol 1989;13(2):205–16. physiology. 21st ed. New York: Lange Medical Books/
[82] Rae T. The Toxicity of metals used in orthopaedic prostheses. McGraw-Hill; 2003.
An experimental study using cultured human synovial fibro- [107] Vander, Sherman, Luciano. Human physiology, the mechanisms
blasts. J Bone Joint Surg Br 1981;63B(3):435–40. of body function. Boston: McGraw Hill; 2004.
306 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307

[108] Huddleston HD. ‘‘Hip implant designs and materials – arthritis techniques of cementing. The results after a minimum of fifteen
hip surgery info – total hip replacement.’’ 2003. 28 October 2003. years of follow-up. J Bone Joint Surg Am 1997;79:53–64.
Available from http://www.hipsandknees.com/hip/hipimplants. [129] Ohashi T, Inoue S, Kajikawa K. The clinical wear rate of
htm. acetabular component accompanied with alumina ceramic head.
[109] Green TR, Fisher J, Matthews JB, Ingham E. Effect of size and In: Oonishi H, Aoli H, Sawai K, editors. Bioceramics: proceed-
dose on bone resorption activity of macrophages by in vitro ings of the first international symposium on bioceramics. St.
clinically relevant ultra high molecular weight polyethylene Louis: Ishiyaku EuroAmerica; 1989. p. 278.
particles. J Biomed Mater Res (Appl Biomater) 2000;53:490–7. [130] Sugano N, Nishii T, Nakata K, Masuhara K, Takaoka K.
[110] Cell biology: Size of Cells.’’ 9 Sep 2003. Wikibooks. 16 Oct 2003. Polyethylene sockets and alumina ceramic heads in cemented
Available from http://wikibooks.org/w/wiki/phtml?title;= total hip arthroplasty. A ten-year study. J Bone Joint Surg Br
Cell_biology:Size_of_cells. 1995;77:548–56.
[111] Buford A, Goswami T. Materials and Design 2004;25:385–93. [131] Wroblewski BM, Siney PD, Dowson D, Collins SN. Prospective
[112] Slonaker M, Goswami T. Materials and Design 2004;25: clinical and joint simulator studies of a new total hip arthro-
395–405. plasty using alumina ceramic heads and cross-linked polyethyl-
[113] Weber BG, Rieker BC. The history of metasul. Acta Chir ene cups. J Bone Joint Surg Br 1996;78:280–5.
Orthop Traumatol Cech 2002;69(5):277–84. [132] Gerber SD, Harris WH. Femoral head autografting to augment
[114] Dorr LD, Wan Z, Longjohn DB, Dubois B, Murken R. Total acetabular deficiency in patients requiring total hip replacement.
hip arthroplasty with use of the Metasul metal-on-metal artic- A minimum five-year and an average seven-year follow-up study.
ulation. Four to seven-year results. J Bone Joint Surg Am J Bone Joint Surg Am 1986;68(8):1241–8.
2001;83-A(5):783–5. [133] Linde F, Jensen J, Pilgaard S. Charnley arthroplasty in osteo-
[115] Emerson Jr RH, Sanders SB, Head WC, Higgins L. Effect of arthritis secondary to congenital dislocation or subluxation of
circumferential plasma-spray porous coating on the rate of the hip. Clin Orthop 1988;227:164–71.
femoral osteolysis after total hip arthroplasty. J Bone Joint Surg
Am 1999;81(9):1291–8.
[116] Slonaker M, Goswami T. Wear Mechanisms in ceramic hip
implants. J Surg Orthop Adv 2004;13(2):1–12. N terminologies
[117] Bankston AB, Faris PM, Keating EM, Ritter MA. Polyethylene
wear in total hip arthroplasty in patient-matched groups. A Antibody: a type of glycoprotein molecule produced by B-lymphocytes
comparison of stainless steel, cobalt–chrome, and titanium- that binds antigens, often with a high degree specificity and affinity
bearing surfaces. J Arthroplasty 1993;8(3):315–22. Antigen: a molecule that binds to an antibody or a TCR
[118] Sochart DH. Relationship of acetabular wear to osteolysis and Antigen presenting cell: a cell that displays peptide fragments of pro-
loosening in total hip arthroplasty. Clin Orthop 1999;363: tein antigens in association with MHC molecules on its surface
135–50. and activates antigen-specific T cells
[119] Cates HE, Faris PM, Keating EM, Ritter MA. Polyethylene Apoptosis: a process of cell death characterized by DNA cleavage, nu-
wear in cemented metal-backed acetabular cups. J Bone Joint clear condensation and fragmentation, and plasma membrane
Surg (Br) 1993;75(2):249–53. blebbing that leads to phagocytosis of the cell without inducing
[120] Nashed RS, Becker DA, Gustilo RB. Are cementless acetabular an inflammatory response
components the cause of excess wear and osteolysis in total hip B-lymphocyte: the only cell type capable of producing antibody mole-
arthroplasty? Clin Orthop 1995;317:19–28. cules and therefore the central cellular component of humoral im-
[121] Thanner J, Kärrholm J, Malchau H, Herberts P. Poor outcome mune responses
of the PCA and Harris–Galante hip prostheses. Randomized Cytokines: proteins produced by many different cell types that mediate
study of 171 arthroplasties with 9-year follow-up. Acta Orthop inflammatory and immune reactions; principle mediators of com-
Scand 1999;70(2):155–62. munication between cells of immune system
[122] Devane PA, Robinson EJ, Bourne RB, Rorabeck CH, Nayak Cytolysis: the dissolution or destruction of a cell
NN, Horne JG. Measurement of polyethylene wear in acetabular Cytotoxic T lymphocyte (CTL): a type of T lymphocyte whose major
components inserted with and without cement. A randomized effector function is to recognize and kill host cells infected with
trial. J Bone Joint Surg (Am) 1997;79(5):682–9. viruses or other intracellular microbes; CTL killing of infected cells
[123] Bono JV, Sanford L, Toussaint JT. Severe polyethylene wear in involves the release of cytoplasmic granules whose contents include
total hip arthroplasty. Observations from retrieved AML PLUS membrane pore-forming proteins and enzymes
hip implants with an ACS polyethylene liner. J Arthroplasty Erythrocyte: a cell in the blood of vertebrates that transports oxygen
1994;9(2):119–25. and carbon dioxide to and from tissues
[124] Kim YH, Kim JS, Cho SH. Primary total hip arthroplasty Helper T cell: the functional subset of T lymphocytes whose main
with the AML total hip prosthesis. Clin Orthop 1999;360: effector functions are to activate macrophages in cell-mediated im-
147–58. mune responses and promote B cell antibody production in humor-
[125] Woolson ST, Murphy MG. Wear of the polyethylene of Harris– al immune responses
Galante acetabular components inserted without cement. J Bone Histiocyte: a relatively inactive, immobile macrophage found in nor-
Joint Surg Am 1995;77:1311–4. mal connective tissue; fixed macrophage
[126] Okumura H, Yamamuro T, Kumar T. Socket wear in total hip Hydroxyapatite: the principal bone salt, Ca5(PO4)3OH, which pro-
prosthesis with alumina ceramic head. In: Oonishi H, Aoli H, vides compressional strength of vertebrate bone
Sawai K, editors. Biomedics: proceedings of the first interna- Hypersensitivity diseases: disorders that result from uncontrolled or
tional symposium on bioceramics. St Louis: Ishiyaku Euro- excessive responses against foreign antigens, such as microbes
America; 1989. p. 284. and allergens; tissue damage that occurs in hypersensitivity diseases
[127] Livermore J, Ilstrup D, Morrey B. Effect of femoral head size on is due to same effector mechanisms used by the immune system to
wear of the polyethylene acetabular component. J Bone Joint protect against antigens
Surg Am 1990;72:518–28. Inflammation: a complex reaction of the innate immune system in vas-
[128] Madey SM, Callaghan JJ, Olejniczak JP, Goetz DD, Johnston cularized tissues that involves the accumulation and activation of
RC. Charnley total hip arthroplasty with use of improved leukocytes and plasma proteins at a site of infection, toxin expo-
 A. Sargeant, T. Goswami / Materials and Design 27 (2006) 287–307 307

sure, or cell injury; although inflammation serves a protective func- sponses to kill microbe-infected cells by direct lytic mechanisms
tion in controlling infections and promoting tissue repair, it can and by secreting IFN-c
also cause tissue damage and disease Nitric oxide: a biologic effector molecule with a broad range of activ-
Interferon-c (IFN-c): a cytokine produced by T lymphocytes and NK ities that in macrophages functions as a potent microbicidal agent
cells whose principal function is to activate macrophages in both to kill ingested organisms
innate immune responses and adaptive cell-mediated immune Nitric oxide synthase (NOS): a member of the family of enzymes
responses that synthesize the vasoactive and microbicidal compound nitric
Interleukin-1 (IL-1): a cytokine produced mainly by activated mono- oxide from L-arginine; macrophages express an inducible form
nuclear phagocytes whose principle function is to mediate host of this enzyme upon activation by various microbial or cytokine
inflammatory responses in innate immunity stimuli
Interleukin-6 (IL-6): a cytokine produced by many cell types, includ- Nuclear factor jB (NF-jB): a family of transcription factors that are
ing activated mononuclear phagocytes, endothelial cells, and fibro- important in the transcription of many genes in both innate and
blasts, that functions in both innate and adaptive immunity; adaptive immune responses
stimulates the growth of antibody-producing B lymphocytes Osteoblast: a cell from which bone develops; a bone-forming cell
Leukocyte: any of various blood cells that have a nucleus and cyto- Osteoclast: a large multinucleate cell found in growing bone that re-
plasm, separate into thin white layer when whole blood is centri- sorbs bony tissue, as in the formation of canals and cavities
fuged, and help protect body from infection and disease; white Osteogenesis: the formation and development of bony tissue
blood cells include neutrophils, eosinophils, basophils, lympho- Phagocytosis: the process by which certain cells of the innate im-
cytes, and monocytes mune system, including macrophages and neutrophils, engulf par-
Macrophages: a tissue based phagocytic cell derived from blood ticles by surrounding the particle with extensions of its plasma
monocytes that plays important roles in innate and adaptive im- membrane
mune responses; activated macrophages phagocytose and kill Prostaglandins: a class of lipid inflammatory mediators derived from
microorganisms secrete proinflammatory cytokines and present arachadonic acid in many cell types via a cyclooxygenase pathway
antigens to helper T cells T lymphocyte: the cell type that mediates cell-mediated immune re-
Major histocompatibility complex (MHC) molecule: a heterodimeric sponses in the adaptive immune system; they express antigen recep-
membrane protein encoded in the MHC locus that serves as a pep- tors that recognize antigens
tide display molecule for recognition by T lymphocytes. Tumor necrosis factor (TNF): a cytokine produced mainly by acti-
Monocytes: a type of bone marrow-derived circulating blood cell that vated mononuclear phagocytes that functions to stimulate the
is the precursor of tissue macrophages; monocytes are actively re- recruitment of neutrophils and monocytes to sites of infection
cruited into inflammatory sites, where they differentiate into and to activate these cells to eradicate microbes; TNF stimulates
macrophages vascular endothelial cells to express new adhesion molecules, in-
Natural killer (NK) cells: a subset of bone marrow-derived lympho- duces macrophages and endothelial cells to secrete chemokines,
cytes, distinct from B or T cells, that function in innate immune re- and promotes apoptosis of target cells