Wear 266 (2009) 349355

Contents lists available at ScienceDirect

Wear

journal homepage: www.elsevier.com/locate/wear

Short communication
Nanometer size wear debris generated from ultra high molecular weight
polyethylene in vivo
Monika Lapcikova
a
, Miroslav Slouf
a,
, Jiri Dybal
a
, Eva Zolotarevova
b
, Gustav Entlicher
b
,
David Pokorny
c
, Jiri Gallo
d
, Antonin Sosna
c

a
Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, v.v.i., Heyrovskeho namesti 2, 162 06 Prague 6, Czech Republic
b
Faculty of Sciences, Charles University, Hlavova 8, 12840 Prague 2, Czech Republic
c
Orthopaedics Clinic, Faculty Hospital Motol, V Uvalu 84, 156 06 Prague 5, Czech Republic
d
Department of Orthopaedics, Faculty of Medicine and Dentistry, Palacky University, I.P. Pavlova 6, 775 20 Olomouc, Czech Republic

a r t i c l e i n f o
Article history:
Received 20 December 2007
Received in revised form 21 May 2008
Accepted 9 June 2008
Available online 17 July 2008

Keywords:
Ultra high molecular weight polyethylene
Nanometer size wear debris
Morphology of wear particles
Total hip replacement





1. Introduction
a b s t r a c t
Ultra high molecular weight polyethylene ( UHMWPE) wear debris is a major cause of long-term failure
of total hip replacements. UHMWPE wear particle sizes range from submicron to several millimeters, but
the particles below 10 m exhibit the highest biological activity. Some in vitro wear particles, produced
in joint simulators, were shown to be smaller than 0.2 m and recently even in vitro particles as small as
several tens of nanometers have been detected. This study brings the rst evidence that nano-sized wear
particles wi t h sizes below 0. 05 m are produced in vivo. UHMWPE wear nanoparticles were revealed by
high-resolution, el d emission gun scanning electron microscopy (FEGSEM) in the periprosthetic tissues
of two different patients. Purity of the isolated wear nanoparticles was conrmed by energy-dispersive
analysis of X-rays (EDS) and infrared spectroscopy (IR). Morphology of wear nanoparticles was determined
by image analysis of FEGSEM micrographs. The average equivalent diameters of wear particles in the rst
and the second patient were 18.5 and 21.2 nm, respectively. Nanoscale wear debris could only be reliably
detected if the isolation protocol included intensive sonication and if higher-than-usual magni cati ons
were employed during FEGSEM analysis.
2008 Elsevier B.V. All rights reserved.


thus the most dangerous wear particles have sizes below 10 m
[8]. A few years ago in vitro experiments on wear testing machines
Ultra high molecular weight polyethylene (UHMWPE) has been
used as a bearing material in total joint replacements (TJR) for
more than four decades. The success of UHMWPE in TJRs is due
to its excellent bulk biocompatibility, very good friction proper-
ties, sufcient mechanical performance and high wear resistance
[1]. Therefore, UHMWPE is regarded as the gold standard for given
application [2,3]. Nevertheless, articulation between polymer and
metallic/ceramic components of TJR leads to generation of tiny
UHMWPE wear particles, which are released from the joint space
and cause complex inammatory reactions leading to osteolysis
and aseptic loosening [4].
Since UHMWPE wear particles were recognized as one of the
major causes of TJR failures, a number of studies describing their
isolation [5], characterization [6] and biological activity [7] have
been published. The particle sizes range from submicron to sev-
eral millimeters. It was shown that the most biologically active and

Corresponding author. Tel.: +420 296809310; fax: +420 296809410.
E-mail address: slouf@imc.cas.cz (M. Slouf).
0043-1648/$ see front matter 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.wear.2008.06.005
proved that a signicant fraction of wear particles can be smaller
than 0.2 m [9]. Recently in vitro wear particles of nanometer sizes
below 0.1 m were observed [10]. The present study proves, for the
rst time, that nanometer-sized wear particles are also generated
in vivo.
In our previous work, we focused on development of a reli-
able technique for sampling, isolation and quantication of in vivo
UHMWPE wear particles in periprosthetic tissues [1115]. More
than 100 samples of periprosthetic tissues have been processed
and both total amount and morphology of wear particles were
determined. In two cases we found that the periprosthetic tissues
contained mostly nanometer-sized wear debris.

2. Materials and methods

2.1. Isolation of wear particles

2.1.1. Sampling
UHMWPE wear particles were detected in periprosthetic (gran-
uloma) tissues obtained during revisions of TJRs in Orthopaedics




350 M. Lapcikova et al. / Wear 266 (2009) 349355

Table 1
Particle separation on low-concentration membranes was further
Summary of periprosthetic tissues studied in this work
improved by sonication as documented below.
Patient Sex Birth Weight (kg) Implant duration (years) Implant type

H1 M 1937 82 8.5 Balgrist/CF 30 2.2. Characterization of wear particles
H2 M 1957 70 8.8 ABG
2.2.1. Field-emission gun scanning electron microscopy (FEGSEM)
Clinic, Faculty Hospital Motol, Prague, Czech Republic and in the
Orthopaedic Clinic of Faculty Hospital Olomouc, Czech Republic. A
summary of periprosthetic tissues analyzed in this study is given
in Table 1.
High-resolution scanning electron microscope Quanta 200 FEG
(FEI), equipped with a eld-emission gun, was used to visual-
ize wear debris. Polycarbonate (PC) membranes with isolated
polyethylene (PE) wear particles were xed onto metallic sup-
port with a conductive silver paste, and the specimen was sputter
coated with a 10-nm thick platinum layer using a vacuum sput-
2.1.2. Purication of chemicals ter coater SCD 050 (Balzers). This xation was found necessary
Distilled water and isopropyl alcohol (iPrOH) were puried to eliminate sample drift and sample damage under the electron
before use by successive ltration through 10 and 0.1 m poly- beam at very high magnications, which were necessary to detect
carbonate (PC) Isopore membranes (Millipore, Ireland). More polyethylene nanoparticles. The micrographs were obtained with
aggressive chemicals, such as 65% HNO3 and 12 mol/l KOH were l - a secondary electron detector at an accelerating voltage of 10 kV
tered through 10 and 0.1 m Teon (PTFE) Omnipore membranes and the smallest possible spot size. In the rst step, homogeneity
(Millipore, Ireland). of PE wear particles on each PC membrane was investigated at low
magnications (500 to 5000). In the second step, selected rep-
2.1.3. Delipidation of samples
Typically, 0.3 g of a freeze-dried tissue sample was cut into small
pieces and extracted twice with 10 ml of a chloroform/methanol
resentative locations on the lters were analyzed at three different
magnications: medium magnication (7500), high magnica-
tion (12,000) and very high magnication (50, 000). In this paper
mixture (2:1, v/v) for 12 h. The solvents were decanted. After two
we use exclusively the above three magnications (medium, high
decantations the tissue samples were dried in a stream of ltered
and very high) for sake of simplicity and easy comparison of all
air and by heating for 2 h at 60

C.
samples.

2.1.4. Acid hydrolysis and washing 2.2.2. Energy-dispersive analysis of X-rays (EDS)
The delipidated samples of tissue (obtained from 0.3 g of freeze- The same samples as those used for SEM were observed in the
dried sample) were hydrolyzed with 5 ml of 65% HNO3 at room
microscope Quanta 200 FEG as described in the previous section,
temperature for 24 h. Only the upper 2 ml layer of the suspen- using an EDS detector (EDAX). EDS spectra were taken at 30 kV. The
sion was used for further isolation. The remaining lower part was signal was collected both from the whole areas of SEM micrographs
removed by aspiration and discarded. The upper layer was washed and from selected places containing large agglomerates of wear
twice with 5 ml of 65% HNO3 and twice with 5 ml of distilled
particles.
water: 5 ml of solvent (HNO3 or H2O) were added, the sample was
centrifuged (500 g, 1 min) and the lower 5 ml were removed by 2.2.3. Fourier-transform infrared spectroscopy (FTIR)
aspiration. The resulting washed 2 ml layer was neutralized with
Infrared-spectra were recorded with an IFS-55 spec-
12 mol/l KOH and washed twice with water in the same way as
trometer (Bruker, Germany) equipped with MCT detector
described above. The nal 2 ml of the sample was used for further
(256 scan/spectrum, resolution 4 cm
1
). IR spectra of selected
purication.
PC membranes with PE particles were measured before sputtering
with platinum and investigation by SEM and EDS. The measure-
2.1.5. Separation of wear debris by ltration ment was performed in two ways: (i) several locations on each
The wear debris suspension obtained by acid hydrolysis was membrane were measured by the attenuated total reection (ATR)
mixed with 2 ml of iPrOH and ltered through a 10- m Teon mem- technique using a Golden Gate TM Heated Diamond ATR Top-Plate
brane Omnipore (Millipore, USA). The ltrate was ll ed up to 6 ml (Specac Ltd.) and (ii) the whole membrane was also measured in
with iPrOH. Then these 6 ml of the suspension were divided into the transmission mode.
four parts: (i) 4.5 ml of the suspension was ltered through a 0.1-
m polycarbonate membrane Isopore (Millipore, USA), (ii) 0.2 ml of
2.2.4. Image analysis of FEGSEM micrographs
the suspension was ltered through another 0.1- m polycarbonate
Image analysis of FEGSEM micrographs was performed using
membrane Isopore (Millipore, USA), (iii) another 0.2 ml of the sus-
the Lucia program (Laboratory Imaging, Czech Republic). FEGSEM
pension was sonicated (5 min, 30 kHz, amplitude 112 m, energy
density 125 W/cm
2
) just before the ltrati on through the 0.1 m
micrographs with magnications of 50,000 were used to
distinguish and measure individual wear nanoparticles. Lower
membrane Isopore and (iv) the remaining 1.5 ml of the suspension
magnication micrographs were not used because the two ana-
was left as a reserve.
lyzed samples contained only nanoparticles, as demonstrated
below. Both image analyses were based on more than 500 particles.
2.1.6. Usage of lter membranes containing wear particles The particles were described by means of equivalent circle diam-
All membranes were investigated by all the methods described eter (D), circularity (C) and elongation (E). Equivalent diameter of
below, however, the image analysis results were based exclusively an object is equal to the diameter of a circle with the same area as
on membranes with lower concentrations of particles (0.2 ml of the corresponding object: D=(4 Area/ )
1/2
. Circularity is calcu-
the suspension), whereas infrared spectroscopy and microanalysis lated from the object area and perimeter: C =4 Area/Perimeter
2
;
were performed on all the membranes (both 0.2 and 4.5 ml). The C =1 for circles and C <1 for all other shapes. Elongation is dened
low-concentration membranes showed better particle separation as the ratio of maximum and minimum diameter of the parti-
on electron micrographs, while the high-concentration membranes cle: E =MaxDiameter/MinDiameter; E =1 for circles and E >1 for all
gave higher signals in infrared spectroscopy and microanalysis. other shapes.




M. Lapcikova et al. / Wear 266 (2009) 349355 351

particles (Al, Zr), were found; a typical EDS spectrum is shown in
Fig. 2.

3.2. Separation of nanoparticles by sonication

Fig. 3 shows PC membranes with PE particles isolated from
periprosthetic tissue of the rst patient H1 (Table 1). On the mem-
branes, nanometer-sized wear particles tended to form aggregates
(Fig. 3a, c and e), which could be disintegrated by sonication
of the ltered suspension (Fig. 3b, d and f). The aggregates of
non-sonicated nanoparticles at medium magnications (Fig. 3a;
magnication 7500) and high magnications (Fig. 3c; magni-
cation 12,000) looked as if they were microparticles. Single
nanoparticles within the aggregates were only revealed at very
high magnications (Fig. 3e; magnication 50, 000). Sonication
resulted in nanoparticle separation, which could be observed even
Fig. 1. Comparison of IR spectra of pure PE (from a catalog), blank PC mem-
brane and PC membrane with isolated PE particles. The spectra were measured in
ATR mode (PE = polyethylene, PC = polycarbonate, ATR = attenuated total reection
infrared spectroscopy).

3. Results

3.1. Purity of isolated particles

Three facts indicated that observed nanoparticles were not
artifacts, but pure UHMWPE. Firstly, we used a well-established
isolation technique, which had already been employed and per-
fected in our previous studies [1215]. Secondly, the purity of the
isolated PE particles on PC membranes was investigated by means
of FTIR: comparison of IR spectra of pure PE, unused PC mem-
brane and PC membrane with PE nanoparticles showed that all
samples exhibited only peaks of PE and/or PC (Fig. 1). Thirdly, we
performed EDS analysis of all membranes, which were investi-
gated by FEGSEM. In all cases, we found only peaks of C (mostly
from PC membrane and partially from PE particles), O (from PC
membrane), Pt (10 nm thick platinum layer from vacuum sput-
ter coater), Ag and Cl (from conductive silver paste that xed the
membranes to the support). No traces of other elements, which
would indicate bone fragments (Ca, P), dust (Si), organic impu-
rities (N, S), metallic wear particles (Cr, Mo, Ti) or ceramic wear
















Fig. 2. EDS spectrum of UHMWPE wear particles on a PC membrane. The peaks of
C and O come from the PC membrane and/or UHMWPE particles. The peaks of Pt,
Ag and Cl result from the fact that the membrane was sputter coated with platinum
and xed with a conductive silver paste containing chlorine before investigation
in an electron microscope (PC = polycarbonate, EDS = energy-dispersive analysis of
X-rays).
at medium magnications (Fig. 3b). At high magnications (Fig. 3d),
the single nanoparticles could be clearly resolved. The micrographs
of sonicated nanoparticles at very high magnications (Fig. 3f) were
suitable for image analysis.

3.3. Image analysis of separated nanoparticles

Fig. 4 shows nal FEGSEM micrographs of PC membranes with
in vivo PE wear particles, which were isolated from periprosthetic
tissues of both patients (Table 1). Both suspensions of the isolated
particles were sonicated immediately before ltrati on through the
membrane to separate particles. It is worth noting that FEGSEM
micrographs displayed neither bigger particles (>0.1 m) nor impu-
rities (such as bone fragments with their characteristic texture or
dust particles with sharp edges).
Image analysis of the wear particles of each patient was based
on nine FEGSEM micrographs from different parts of the mem-
brane. All nine micrographs in each set have the same magnication
(50, 000; the same as in Fig. 4e and f) and each set of micrographs
contained more than 700 well-separated particles for analysis. The
wear particles from patients H1 and H2 exhibited average sizes
18.5 and 21.2 nm, respectively. Complete histograms (Fig. 5) proved
that in both cases the wear particles were quite small, with rel-
atively narrow size distributions. Complete numerical results of
image analysis (Table 2) re-conrmed the small particle size, con-
rmed narrow size distributions and indicated that all particles
were spherical in shape.

3.4. Nanoparticles in other samples

In the previous sections we have demonstrated that 2 samples
out of 100 contained mostly nanoparticles. Detailed microscopic
analysis of the two samples suggested that they contained
almost no microparticles. At higher magnications, practically all
microparticles were identied as the agglomerates of nanoparticles
as documented by FEGSEM (Figs. 3 and 4) and conrmed by image
analysis (Fig. 5 and Table 2).
The other samples exhibited no exceptional features and con-
tained mostly microparticles, in agreement with numerous previous
studies (e.g. [1120]). Particles coming from one of our typical
samples of periprosthetic tissue are shown in Fig. 6. The differ-
ence between the two samples containing mostly nanoparticles
(Figs. 3 and 4) and the typical samples containing mostly micropar-
ticles (Fig. 6) was visible even at medium magnication FEGSEM
micrographs (cf. Figs. 3a and b, 4a and b and 6a). In the two
samples containing nanoparticles there were no particles >0.1 m,
whereas a typical sample contained a number of particles with
dimensions >1 m. This was the main reason why we focused
our attention on the two exceptional samples and revealed wear




352 M. Lapcikova et al. / Wear 266 (2009) 349355










































Fig. 3. FEGSEM micrographs of UHMWPE wear particles on 0.1 m PC membrane. All particles come from patient H1 in Table 1. Left column (a, c and e): particles without
sonication before l trati on; right column (b, d and f): particles with sonication before ltration. The selected micrographs were taken at medium (a and b), high (c and d)
and very high (e and f) magnication.


nanoparticles, which become clearly visible at higher magnica- 4. Discussion
tions (Figs. 3cf and 4cf).
Nevertheless, further analysis of the other samples, which con- In the previous section it has been demonstrated for the rst
tained mostly microparticles, showed certain small amounts of time that UHMWPE wear nanoparticles are generated in vivo. We
wear nanoparticles as well (Fig. 6b and c). Although FEGSEM should re-emphasize that in this study the term wear nanopar-
micrographs of typical samples re-conrmed clear predominance ticles denotes UHMWPE particles with sizes below 0.05 m, i.e.
of wear microparticles, small amounts of wear nanoparticles the particles whose size is even smaller than the lowest pore
could be detected (Fig. 6c), nanoparticles marked with white size in commonly used lters [9]. Three independent methods,
arrows. This indicated that a certain number of wear nanopar- FEGSEM, EDS and IR suggested that the observed nanoparticles
ticles might be present in all periprosthetic tissues. Accurate are a polyethylene and not an artifact. Moreover, the nanoparticles
determination of volume fraction of wear nanoparticles in typi- with average sizes 18.5 and 21.2 nm were observed in two indepen-
cal samples of periprosthetic tissues is a subject of our subsequent dent samples. On the other hand, many previous studies reported
study. much larger average sizes of in vivo UHMWPE wear nanoparticles.




M. Lapcikova et al. / Wear 266 (2009) 349355 353










































Fig. 4. Final FEGSEM micrographs of PC membranes with UHMWPE wear particles of both studied patients, H1 and H2 (Table 1). Left column (a, c and e): particles from H1;
right column (b, d and f): particles from H2. The selected micrographs were taken at medium (a and b), high (c and d) and very high (e and f) magnications.

Why were similar wear nanoparticles not found in previous
works?
The rst reason why wear nanoparticles were not reported
before might consist in the fact that we investigated more than
100 samples and detected signicant volumes of wear nanopar-
ticles just in 2 of them. In the other samples the microparticles
predominated and the amounts of nanoparticles were either
small or negligible. Elck et al. [16] summarized the average
sizes of in vivo wear particles from various studies; their sizes
varied around 0.5 m. Also in vitro wear particles from joint
simulators showed average sizes of several tenths of microm-
eters [17]. Thus it seems that standard wear particles are
microparticles, whereas wear nanoparticles might be an excep-
tion.
The second reason why wear nanoparticles escaped the atten-
tion of previous investigators could be sonication. Wear particles
are usually isolated from periprosthetic tissues by means of
acid/base digestion method. Then the suspension of the particles
is ltered through a membrane and the particles are analyzed by
means of scanning electron microscopy. Our results showed that
the nanoparticles on the membrane tend to form agglomerates
looking like single microparticles (Fig. 3). Intensive sonication of
the suspension just before its l trati on through a membrane is nec-
essary to achieve nanoparticle separation. As this sonication was
not always used in other studies, the agglomerates of nanoparticles
could have been misinterpreted as single microparticles in partic-
ular at low and medium magnication micrographs. In this study
the nanoparticles on the membranes occurred as loose agglomer-
mated

standard

deviation

(

=

(1/n

i ( xi) ), where xi is the ith element

354 M. Lapcikova et al. / Wear 266 (2009) 349355

The fourth reason why wear nanoparticles were not found on
the membranes after isolation might consist in the isolation and
characterization methods themselves. Over the course of time, the
methods of isolation of wear particles from periprosthetic tissues
have been gradually improved. Some of the isolation protocols
were far from perfect, as discussed in our previous work [13].









Fig. 5. Histogram showing size distributions of UHMWPE wear particles isolated
from patient H1 (full columns) and patient H2 (striped columns). Size of the particles
was described by means of equivalent diameter. Number of particles used for the
analysis was >700 in both cases.

ates (Figs. 3 and 4). If the sonication was even more intensive, the
nanoparticles might be completely separated and, consequently,
they might go through the membrane without being detected. It
seems that just sufcient sonication is needed to catch the nanopar-
ticles.
The third reason why wear nanoparticles were not observed
might be associated with a combination of sonication, high magni-
cations and the use of FEGSEM instead of standard SEM microscopy.
The effect of sonication has already been discussed in the previ-
ous paragraph. The effect of high magnication is as follows: an
ISO standard for quantitative image analysis of wear nanoparticles
[18] recommends using SEM micrographs with magnications of
at least 5000. Such medium magnications (5000 to 10,000)
were used in most of the previous studies, whereas here we
have demonstrated that non-sonicated wear nanoparticles were
observed only at very high magnications (50,000). In a standard
SEM microscope, it is possible to achieve such a high magnication
with a good signal and resolution only if an intensive electron beam
and acceleration voltage around 30 kV are used. We have veried
that under such conditions the PC membranes with PE particles
were burnt by the electron beam and, as a result, the nanoparticles
could not be observed. In a FEGSEM microscope, the magnication
50,000 with good resolution could be achieved at 10 kV and a
small probe size, which did not destroy the specimens. Therefore,
somewhat unusual combination of sonication, very high magni-
cation and FEGSEM microscopy seems to be necessary for detection
of wear nanoparticles. At the highest magnications the nanoparti-
cles could be observed even without sonication, but they were not
so well separated from each other.

Table 2
Results of the image analysis of UHMWPE nanoparticles

Patient Eq. diameter (D) Circularity (C) Elongation (E)

(nm) (nm) () () () ()

H1 18.47 5.29 0.97 0.07 1.29 0.13
H2 21.20 8.01 0.93 0.09 1.35 0.29

Each numerical value in this table was calculated from a distribution obtained by
image analysis of nine FEGSEM micrographs with magnication 50,000 and real
width 1.26 m. The mean of the distribution, , was calculated as the arithmetic
mean

(

=

1/n

i
xi
)

and

the

width

of

the

distribution,

,

was

calculated

as

esti-

2
of the distribution, n is the number of all elements in the distribution and summation
Fig. 6. FEGSEM micrographs of PC membranes with UHMWPE wear particles of a
typical patient, whose periprosthetic tissue contained mostly microparticles. The
micrographs were taken at medium (a), high (b) and very high (c) magnications.
A few nanoparticles are visible at very high magnications (marked with white
runs through all elements.
arrows).




M. Lapcikova et al. / Wear 266 (2009) 349355 355

In some other studies, the authors used techniques, which were
not able to detect the smallest particles. For example, some parti-
cle counters detected only particles with sizes above 0.3 m [19],
in transmission electron microscopy the nanoparticles might be
hardly observable due to insufcient contrast as their size may be
lower than the thickness of investigated ultrathin section (typically
around 60 nm) [20] and in the methods based on light scatter-
ing the signal from bigger particles may entirely predominate
[11,12]. Our results suggest that a combination of isolation method
(veried in previous studies) and good detection method (sonica-
tion +FEGSEM) are necessary to catch wear nanoparticles. In future
studies the in vivo UHMWPE wear nanoparticles might be observed
more and more frequently as the sampling, isolation and detection
methods are further improved.

5. Conclusion

Methods for isolation and visualization of UHMWPE wear par-
ticles are still being improved. As a result, it has been possible
to isolate and detect smaller and smaller wear particles. A few
years ago Scott et al. [9] proved that a signicant fraction of in
vitro UHMWPE wear particles can be smaller than 0.2 m. Recently
Galvin et al. [10] found in vitro UHMWPE wear nanoparticles with
sizes below 0.1 m. In the present work we found, for the rst time,
in vivo UHMWPE wear nanoparticles with sizes below 0.05 m.
These nanoparticles were isolated from periprosthetic tissues of
two different patients and had average particle size of 18.5 and
21.2 nm in the rst and the second case, respectively.

Acknowledgement

Financial support through grant 2B06096 (Ministry of Edu-
cation, Youth and Sports of the Czech Republic) is gratefully
acknowledged.

References
[1] G. Lewis, Properties of crosslinked ultra-high-molecular-weight polyethylene,
Biomaterials 22 (2001) 371401.
[2] Alternatives to conventional UHMWPE for hip arthroplasty, in: S.M. Kurtz (Ed.),
The UHMWPE Handbook, Elsevier, Amsterdam, 2004, p. 93 (Chapter 6).
[3] S.M. Kurtz (Ed.), The Clinical Performance of UHMWPE in Knee Replacements.
The UHMWPE Handbook, Elsevier, Amsterdam, 2004, p. 182 (Chapter 8).
[4] G. Lewis, Polyethylene wear in total hip and knee arthroplasties, J. Biomed.
Mater. Res. 38 (1997) 5575.
[5] J.L. Tipper, E. Ingham, J.L. Hailey, A.A. Besong, J. Fisher, B.M. Wroblewski, M.H.
Stone, Quantitative analysis of polyethylene wear debris, wear rate and head
damage in retrieved Charnley hip prostheses, J. Mater. Sci.: Mater. Med. 11
(1999) 117124.
[6] J.D. Mabrey, A. Afsar-Keshmiri, G.A. Engh, C.J. Sychterz, M.A. Wirth, C.A.
Rockwood, C.M. Agrawal, Standardized analysis of UHMWPE wear parti-
cles from failed total joint arthroplasties, J. Biomed. Mater. Res. 63 (2002)
475548.
[7] E. Ingham, J. Fisher, Proceedings of the Institution of Mechanical Engineers. Part
HJ. Eng. Med. 214 (2000) 2137.
[8] J.B. Matthews, A.A. Besong, T.R. Green, M.H. Stone, B.M. Wroblewski, J. Fisher,
E. Ingham, Evaluation of the response of primary human peripheral blood
mononuclear phagocytes to challenge with in vitro generated clinically rel-
evant UHMWPE particles of known size and dose, J. Biomed. Mater. Res. 52
(2000) 296307.
[9] M. Scott, K. Widding, S. Jani, Do current wear particle isolation procedures
underestimate the number of particles generated by prosthetic bearing com-
ponents? Wear 251 (2001) 12131217.
[10] A.L. Galvin, J.L. Tipper, E. Ingham, J. Fischer, Nanometre size wear debris gen-
erated from crosslinked and non-crosslinked ultra high molecular weight
polyethylene in articial joints, Wear 259 (2005) 977983.
[11] M. Slouf, I. Sloufova, G. Entlicher, Z. Horak, M. Krejc, P. Stepanek, T. Radonsky, D.
Pokorny, A. Sosna, New fast method for determination of numbers of UHMWPE
wear particles, J. Mater. Sci.: Mater. Med. 15 (2004) 12671278.
[12] D. Pokorny, M. Slouf, Z. Horak, D. Jahoda, G. Entlicher, S. Eklova, A. Sosna, Method
for assesment of distribution of UHMWPE wear particles in periprosthetic tis-
sues in total hip arthroplasty, Acta Chirurgiae Othopaedicae et Traumatologiae
Cechosl 73 (2006) 243250.
[13] M. Slouf, S. Eklova, J. Kumstatova, S. Berger, H. Synkova, A. Sosna, D. Poko-
rny, M. Spundova, G. Entlicher, Isolation, characterization and quantication of
polyethylene wear debris from periprosthetic tissues around total joint replace-
ments, Wear 262 (2007) 11711181.
[14] M. Slouf, D. Pokorny, G. Entlicher, J. Dybal, H. Synkova, M. Lapcikova, Z. Fej-
farkova, M. Spundova, F. Vesely, A. Sosna, Quantication of UHMWPE wear
in periprosthetic tissues of hip arthroplasty: description of a new method
based on IR and comparison with radiographic appearance, Wear 265 (2008)
674684.
[15] E. Zolotarevova, Z. Fejfarkova, G. Entlicher, M. Lapcikova, M. Slouf, D. Pokorny, A.
Sosna, Can centrifugation affect the morphology of polyethylene wear debris?
Wear (in press) doi:10.1016/j.wear.2008.04.005.
[16] A.P.D. Elck, S.M. Green, S. Krikler, A. Unsworth, The nature and dissemination
of UHMWPE wear debris retrieved from periprosthetic tissue of THR, J. Biomed.
Mater. Res. 65A (2003) 95108.
[17] O. Calonius, V. Saikko, Analysis of polyethylene particles produced in different
wear conditions in vitro, Clin. Orthop. Rel. Res. 399 (2002) 219230.
[18] ISO 17853:2003 (E).
[19] K.J. Margevicius, T.W. Bauer, J.T. McMahon, S.A. Brown, K. Merritt, Isolation and
characterization of debris in membranes around total joint prostheses, J. Bone
Joint Surg. 76-A (1994) 16641675.
[20] M. Visentin, S. Stea, S. Squarzoni, B. Antonietti, M. Reggiani, A. Toni, A new
method for isolation of polyethylene wear debris from tissue and synovial uid,
Biomaterials 25 (2004) 55315537.