Quantitative comparison of wear debris from
UHMWPE that has and has not been
sterilised by gamma irradiation
A. A. Besong, J. L. Tipper, E. Ingham, M. H. Stone,
B. M. Wroblewski, J. Fisher
From the University of Leeds, England
ltra-high-molecular-weight polyethylene
U (UHMWPE) components for total joint
replacement generate wear particles which cause
adverse biological tissue reactions leading to osteolysis
and loosening. Sterilisation of UHMWPE components
by gamma irradiation in air causes chain scissions
which initiate a long-term oxidative process that
degrades the chemical and mechanical properties of
the polyethylene. Using a tri-pin-on-disc tribometer we
studied the effect of ageing for ten years after gamma
irradiation in air on the volumetric wear, particle size
distribution and the number of particles produced by
UHMWPE when sliding against a stainless-steel
counterface.
The aged and irradiated material produced six
times more volumetric wear and 34 times more wear
particles per unit load per unit sliding distance than
non-sterilised UHMWPE. Our findings indicate that
oxidative degradation of polyethylene after gamma
irradiation in air with ageing produces more wear.
J Bone Joint Surg [Br] 1998;80-B:340-4.
Received 8 January 1997; Accepted after revision 22 December 1997
Osteolysis and subsequent loosening as a result of the
generation of wear particles are now recognised as a major
cause of long-term failure in total joint arthroplasty. Liver-
1 6-8
more, Duane and Murray have shown a positive correla-
A. A. Besong, B Eng, Research Student
J. Fisher, D Eng, Professor
Department of Mechanical Engineering
J. L. Tipper, PhD, Research Fellow
E. Ingham, PhD, Senior Lecturer
Department of Microbiology
University of Leeds, Leeds LS2 9JT, UK.
M. H. Stone, FRCS, FRCS Ed, Consultant Orthopaedic Surgeon
Leeds General Infirmary, Great George Street, Leeds LS1 3EX, UK.
B. M. Wroblewski, FRCS, Consultant Orthopaedic Surgeon, Professor of
Orthopaedic Biomechanics
The John Charnley Research Institute, Wrightington Hospital for Joint
Disease, Hall Lane, Appley Bridge, Wigan, Lancashire WN6 9EP, UK.
Correspondence should be sent to Professor J. Fisher.
©1998 British Editorial Society of Bone and Joint Surgery
0301-620X/98/28234 $2.00
340
tion between the amount of resorption of the proximal part
2-4
of the femur and wear, and tests in vitro have demon-
strated that the macrophage response to particulate wear
debris is dependent on the number and size of particles.
These studies have shown that cells are more reactive to
small phagocytosible particles and irregularly shaped par-
3,4
ticles than to larger and more regularly shaped particles.
It is therefore important to reduce the wear volume of ultra-
high-molecular-weight polyethylene (UHMWPE) and the
number of wear particles to improve the long-term clinical
performance.
The volume, size, morphology and concentration of the
wear particles produced are likely to depend on tribological
factors such as the properties of the material and the loads
and movements experienced at the contact surfaces. The
smooth hard counterface (typically Ra = 0.01 to 0.05
m)
slides against a concave surface producing UHMWPE wear
particles by adhesive and fatigue wear. Many particles of
various sizes are produced in each cycle. The wear rates
and volumes increase dramatically as the femoral counter-
face becomes roughened due to third-body damage; the
counterface asperities remove particles by abrasive actions
5
or by low cycle fatigue. A second wear process that may
5
produce larger polyethylene particles is called ‘macro-
scopic polymer asperity wear’ and is associated with cyclic
deformation and strain accumulation of much larger asper-
ities (up to 10
m high) on the polymer surfaces. Evidence
from explanted acetabular cups has shown that fatigue
failure of the polymer surface just below the asperity
(=5
m deep) can produce long fibres or cigar-shaped
particles of up to 100
m in length. Most polyethylene
9,10
wear particles are submicron in size.
Until recently, sterilisation of UHMWPE was by gamma
irradiation in the presence of air causing the generation of
free radicals which then initiated a series of long-term
oxidation reactions. The degree of oxidation in polyethyl-
11
ene has been shown to depend on the dose of radiation,
12
the post-irradiation age and the environmental condi-
11,13,14
tions. Studies have indicated that it causes deteriora-
tion of important mechanical properties such as tensile
15 16
strength, impact strength, toughness, fatigue
16,17
strength and Young’s modulus. These time-dependent
changes have been shown to affect the volumetric
18,19
wear.
THE JOURNAL OF BONE AND JOINT SURGERY
 COMPARISON OF WEAR DEBRIS FROM UHMWPE IN RELATION TO STERILISATION BY GAMMA IRRADIATION
Our aim was to compare quantitatively the wear debris
generated from UHMWPE that had been sterilised by
gamma irradiation and aged in air, with that generated from
UHMWPE which had not been irradiated.
Materials and Methods
UHMWPE (grade RCH 1000 GUR1120) wear pins were
machined from acetabular cups (DePuy International Ltd,
Leeds, UK) that had been gamma irradiated at the standard
2.5 Mrad and left to age in a sterile environment of air for
ten years. Wear pins were also machined from acetabular
cups made from the same material matched for type and
specification which had not been irradiated. Eight pins were
machined from each type of material: six were used as wear
pins, and two as soak control pins to monitor changes due
to the hydrophilic nature of polyethylene. After initial
cleaning, the pins were left to soak in deionised water for
14 days to stabilise their moisture content. They were
prepared so that the position of the wear surface at the start
of the test was 1 mm below the initial wear surface of the
acetabular cup in clinical use. Previous oxidation profiles
for irradiated and aged UHMWPE, determined by FTIR
spectroscopy, have shown that the region of maximum
oxidative degradation is at a depth of between 0.5 mm and
17,19
2 mm into the acetabular cup.
18
We used a tri-pin-on-disc unidirectional tribometer.
The cylindrical polyethylene wear pins had a conical tip
with a 3 mm diameter flat wear surface. In each test, three
pins of the same material were run on a slightly roughened
(Ra = 0.07
m) stainless-steel counterface. Six tests were
performed for each material and the lubricating fluid was
collected and stored for the isolation and quantification of
debris. The duration of each test was between 24 and 30
hours for the aged irradiated material and 48 hours for the
non-irradiated material. The longer duration test was for
material with a lower wear rate but had no effect on the size
of particles produced. Bovine serum diluted to 25% (v/v)
by adding 0.1% (w/v) sodium azide solution was used as
the lubricant for all tests. The load on each pin was 80 N,
giving a stress of 12 MPa at the contact surfaces. The
surfaces were sliding at a speed of 80 mm/s. Wear factors
3
(wear volume (mm) ) per unit load (N) per unit sliding
distance (m)) were determined by weight loss as previously
20
described.
The wear particles were isolated from the lubricant by a
multistage digestion method using potassium hydroxide,
21,22
chloroform/methanol extraction and centrifugation.
This allowed the supernatant containing the wear debris to
be filtered sequentially through preweighed polycarbonate
filters of 10
m and 0.1
m. The filters were dried and
weighed to obtain the mass of debris in each of the two size
ranges. This method was designed to collect all the wear
particles greater than 0.1
m onto the filters, the mem-
branes of which were then coated with gold and scanned by
SEM. Magnifications of from 3000 up to 20 000 were used
VOL. 80-B, NO. 2, MARCH 1998
341
to detect the boundaries of the very small particles. The
distribution of particle size was obtained using image anal-
ysis (Image pro-Plus; Media Cybernetics, Datacell Ltd,
Maidenhead, UK). SEM micrographs were taken from at
least three distinct regions on each filter membrane. Partic-
les in each region were analysed and a mean particle size
distribution was calculated. An average of 2000 particles
was counted from each sample. The mass and frequency
distributions as a function of size were determined from the
SEMs by two-dimensional particle analysis. Initial analysis
used an estimate of the total surface area of particles on the
0.1
m filter membrane and the mass of debris on the filter
membrane to estimate the mean thickness of 0.4
m of the
wear particles. The total number of particles was deter-
mined using this mean thickness, the mean area of partic-
les, the density of polyethylene and the mass of debris on
the filter membrane as follows:
M
N=
Āt̄
where N is total number of particles on the filter membrane,
M is the mass of debris on the filter membrane,  is the
density of polyethylene, Ā is the mean area of the wear
particles and t̄ is the mean thickness of the particles.
Particle size and shape were defined by length, area and
aspect ratio (length/width). We also used qualitative
descriptions.
Results
The aged irradiated polyethylene material wore signifi-
cantly (six times, p < 0.05) more than the non-irradiated
polyethylene acetabular cup as shown by the mean wear
factors in Figure 1.
The percentage mass distribution of the wear particles
between the two ranges of size (0.1 to 10
m, > 10
m)
was similar for both materials. Particles less than 10
m in
size accounted for 78% of the mass of the aged irradiated
Fig. 1
The mean wear factors (+ 95% confidence limit) for the aged and
irradiated (Ai) and non-irradiated (Ni) material.
342 A. A. BESONG, J. L. TIPPER,
Fig. 2
Mean percentage mass (+ 95% confidence limit) as a function of particle
length.
Fig. 4
Mean (+ 95% confidence limit) percentage mass as a function of particle
area.
debris and for 71% of the mass of the debris from the non-
irradiated material. We compared the mass of the wear
debris recovered after filtration with that expected to be
present in the lubricant from the calculations of the weight
loss of the pins. The total mass loss due to polyethylene
wear in a test station containing three wear pins was 2 mg
for the aged irradiated material and 0.5 mg for the unsteri-
lised material. For the non-irradiated material the mean
proportion of debris recovered was 75% while for the aged
irradiated material it was 70% (60% to 100%).
Initial qualitative analysis of micrographs of the wear
debris on the 0.1
m and 10
m filters suggested that partic-
les produced from both materials were similar in morphol-
ogy. The smaller particles were regularly shaped granules,
while the larger particles (>10
m) appeared to be a cohesion
of many small particles and very irregular in shape.
E. INGHAM, ET AL
Fig. 3
Mean (+ 95% confidence limit) percentage distribution of the total number
of particles as a function of particle length. Less than 0.1% of the particles
were greater than 10
m in length and have not been included.
Fig. 5
Mean (+ 95% confidence limit) percentage distribution of the total number
of particles as a function of particle area.
The calculated mean aspect ratios (length/width) of the
wear particles were 2.0 and 2.6 for the unsterilised and
aged irradiated UHMWPE, respectively. The mass and
number distributions with respect to size of the wear
particles for both materials are shown in Figures 2 to 5. The
size distributions ranged from 0.1 to 0.5 and the percentage
number in this size range was higher (85%) for the aged
irradiated material than for the non-irradiated material
(65%). In terms of particle size and particle area both
analyses showed that in the lowest size range the per-
centage mass and percentage number of particles were
greater for the aged irradiated material.
Figures 6 and 7 show the number of particles in the size
3
ranges per unit volume (mm ) of wear debris as a function
of the length and area distributions, respectively. For the
aged irradiated material the number of particles per unit
THE JOURNAL OF BONE AND JOINT SURGERY
 COMPARISON OF WEAR DEBRIS FROM UHMWPE IN RELATION TO STERILISATION BY GAMMA IRRADIATION
Fig. 6
3 3
Mean distribution of the total number of particles per unit volume (mm )
of wear debris as a function of particle length.
Fig. 8
Mean3 (+ 95% confidence limit) number of particles per unit volume
(mm ) of wear debris for aged and irradiated (Ai) and non-irradiated (Ni)
polyethylene.
9
volume of wear was estimated to be (24 ± 6)
10 per
3
mm which was significantly higher (5.5 times, p < 0.05)
9 3
than that ((4.4 ± 0.7)
10 per mm ) for the non-irradiated
material (Fig. 8).
Combining the increased wear rate and the number of
particles per unit wear volume allowed the total number of
particles produced per unit load per unit sliding distance to
be calculated. This was defined as the wear particle factor;
3
for the aged irradiated material it was 30
10 particles/
Nm which was 34 times higher (p = 0.07) than the value for
non-irradiated material (884 particles/Nm; Fig. 9).
Discussion
Our study has compared the wear debris released in vitro
from polyethylene materials which have had different ster-
ilisation treatments. We have shown that gamma-irradiated
and aged UHMWPE generated many times more wear
particles than UHMWPE which had not been sterilised.
VOL. 80-B, NO. 2, MARCH 1998
343
Fig. 7
Mean distribution of the total number of particles per unit volume (mm )
of wear debris as a function of particle area.
Fig. 9
Mean (+ 95% confidence limit) number of particles generated per load per
unit sliding distance for aged and irradiated (Ai) and non-irradiated (Ni)
polyethylene.
The increase in the number of particles produced was
attributed to both the increase in wear rate and the different
particle size distribution which led to a significantly greater
number of wear particles per unit wear volume. Analysis of
the percentage frequency distribution of the particle size, as
has been common practice in the literature, did not show a
substantial difference in the particles. It was necessary to
determine the mass distribution and total number of partic-
les produced per unit wear volume in order to establish this
difference. We also defined a wear particle number factor
(number/Nm) to allow comparison of the number of wear
particles produced per unit load per unit sliding distance
from different wear tests.
Larger particles accounted for a reasonable proportion
(26%) of the mass of the debris. Although they contributed
little to the percentage frequency distribution, they did
influence the total number of particles produced per unit
volume of wear debris. This method of analysis shows that
the total number of particles produced is sensitive to both
344 A. A. BESONG, J. L. TIPPER,
the mass and size distribution of the debris.
A simple configuration wear tester cannot fully replicate
the time-dependent loading and multidirectional friction
force found in hip prostheses in vivo. In our study we
performed the simple configuration wear tests in a format
to reproduce realistic wear rates. We used slightly rough-
ened counterfaces which produced wear factors of the order
-6 3 aged UHMWPE tibial inserts. 21st Annual meeting, Society for Bio-
of 10 mm /Nm, which are representative of clinical wear
rates. The slightly roughened counterface (Ra = 0.07
m)
represented femoral heads that had been damaged in vivo.
The particle size distributions produced in our tests had
modes in the range 0.1 to 0.5
m which were similar to
distributions found for particles from retrieved tissues of
21
failed prostheses.
Conclusion. We have shown that for the same sliding
distance and load, UHMWPE that had been sterilised by
gamma irradiation and aged for ten years produced greater
wear volume (six times) and more wear particles (34 times)
than non-irradiated UHMWPE.
A. A. Besong was supported by an Overseas Research Studentship (ORS)
and Leeds University Tetley/Lupton Scholarships. This work was sup-
ported in part by Brite Euram Project BE7928, the Arthritis and Rheuma-
tism Council UK, and the UK Department of Trade and Industry.
No benefits in any form have been received or will be received from a
commercial party related directly or indirectly to the subject of this
article.
References
1. Livermore J, Duane L, Murray B. Effect of femoral head size on the
wear of the polyethylene acetabular component. J Bone Joint Surg
[Am] 1990;72-A:518-28.
2. Glant TT, Jacobs JJ, Molnar G, et al. Bone resorption activity of
particulate-stimulated macrophages. J Bone Miner Res 1993;8:
1071-9.
3. Shanbhag AS, Jacobs JJ, Black J, Galante JO, Glant TT. Macro-
phage/particle interactions: effect of size, composition and surface
area. J Biomed Mater Res 1994;28:81-9.
4. Green T, Fisher J, Ingham E. Polyethylene particles of a ‘critical
size’ are necessary for the induction of IL-6 by macrophages in vitro.
Trans 43rd Annual Meeting, ORS Vol. 22. 1997:733.
5. Cooper JR, Dowson D, Fisher J. Macroscopic and microscopic wear
mechanisms in ultra-high molecular weight polyethylene. Wear
1993;162:378-84.
6. Shanbhag AS, Jacobs JJ, Glant TT, et al. Composition and morphol-
ogy of wear debris in failed uncemented total hip replacement. J Bone
Joint Surg [Br] 1994;76-B:60-7.
7. Lee J-M, Salvati EA, Betts F, et al. Size of metallic and polyethylene
debris particles in failed cemented total hip replacements. J Bone Joint
Surg [Br] 1992;74-B:380-4.
E. INGHAM, ET AL
8. Rose RM, Radin EL. Wear of polyethylene in the total hip prosthesis.
Clin Orthop 1982;170:107-15.
9. Campbell P, Ma S, Yeom B, et al. Isolation of predominantly
submicron-sized UHMWPE wear particles from periprosthetic tissues.
J Biomed Mater Res 1995;29:127-31.
10. McKellop HA, Campbell P, Park SH, et al. The origin of submicron
polyethylene wear debris in total hip arthroplasty. Clin Orthop 1995;
311:3-20.
11. Trieu HH, Avent RT, Paxson RD. Effects of sterilisation on shelf-
materials 1995:109.
12. Furman B, Li S. The effects of long term shelf life aging of ultra-high
molecular weight polyethylene. 21st Annual Meeting, Society for
Biomaterials 1995: 114.
13. Rimnac CM, Klein RW, Betts F, Wright TM. Post-irradiation aging
of ultra-high molecular weight polyethylene. J Bone Joint Surg [Am]
1994;76-A:1052-6.
14. Sun DC, Schmidig G, Stark C, Dumbleton JH. On the origins of a
subsurface oxidation maximum and its relationship to the performance
of UHMWPE implants. 21st Annual Meeting, Society for Biomaterials
1995;362.
15. Pascaud RS, Evans WT, McCullagh PJJ. An investigation into the
tensile/compressive properties of UHMW-PE as used in total joint
replacements. 21st Annual Meeting, Society for Biomaterials 1995:
108.
16. Ries MD, Rose RM, Greer J, Weaver JD, Saver WI. Sterilization-
induced effects on UHMWPE performance properties. 62nd Annual
Meeting. American Academy of Orthopaedic Surgeons. 1995.
17. Goldman M, Ranganathan R, Pruitt L, Gronsky R. Character-
ization of structure and fatigue resistance of aged and irradiated
UHMWPE. 21st Annual Meeting, Society for Biomaterials 1995:111.
18. Besong AA, Hailey JL, Ingham E, et al. A study of the combined
effects of shelf ageing following irradiation in air and counterface
roughness on the wear of UHMWPE. J Biomed Mat Eng 1997;7:
59-65.
19. Fisher J, Reeves EA, Isaac GH, Saum KA, Sanford WM. Compar-
ison of the wear of aged and non-aged ultra-high molecular weight
polyethylene sterilized by gamma irradiation and by gas plasma. J Mat
Sci: Mat in Med 1997;8:375-8.
20. Fisher J, Dowson D, Hamdzah H, Lee HL. The effect of sliding
velocity on friction and wear of UHMWPE for use in total artificial
joints. Wear 1994;175:219-25.
21. Tipper JL, Ingham E, Hailey JL, et al. Quantitative comparison of
polyethylene wear debris, wear rate and head damage in retrieved hip
prostheses. Trans 43rd Annual Meeting, ORS 1997; 22:355.
22. Fisher J, Barbour PSM, King M, et al. Wear of ultrahigh molecular
weight polyethylene in artificial joints: a new approach to quantifica-
tion of wear debris. Proc Mat 1997;2:543-4.
23. Trieu HH, Buchanan DJ, Needham DA, et al. Accelerated fatigue
wear of gamma-sterilized UHMWPE tibial components. EORS trans
1997;7:106.
24. Brach Del Prever E, Costa L, Gallinaro P. Ethylene oxide sterilisa-
tion of polyethylene cups versus gamma radiation: different behaviour
in vivo. EFORT trans 1997:585.
25. Kurtz SM, Rimnac CM, Bartel DL. Degradation rate of ultra-high
molecular weight polyethylene. J Orth Res 1997;15:57-61.
THE JOURNAL OF BONE AND JOINT SURGERY