Académique Documents
Professionnel Documents
Culture Documents
ABSTRACT
The study was undertaken to investigate the association, if any, between the surface roughness of 35 explanted
femoral heads and the clinical wear factors of the corresponding polyethylene sockets. The wear of the socket was
measured using a shadowgraph technique. The surface topography was investigated using a Rodenstock RM 600
non-contacting profilometer, and six parameters were used to characterise the roughness. Further, qualitative inspections of the femoral surface were undertaken using both a Joel JSM-IC-848 scanning electron microscope and a Zeiss
Axiotech microscope with a differential interference contrast facility. Two parameters were found to correlate with the
clinical wear factor, namely the skewness of the amplitude distribution function, Sk, and the arithmetic mean
roughness, Ra. Simple parameters describing the peak heights of the asperities were found not to have a significant
association with the clinical wear factor. The exponent in the power relationship between the arithmetic mean roughness and kclinical was found to be equal to 0.5 (SE: 0.2). This value is significantly smaller than that found in
laboratory experiments and may be due to the non-uniform nature of the roughening of the femoral head, three-body
wear and the effect of other clinical factors on the wear process. Further, extremely strong correlations were detected
between the differing roughness parameters, which would suggest that any attempt to deduce which one is the most
important in affecting the wear of the polymeric counterface is fraught with difficulty. However, further investigation
of those parameters, such as the reduced peak height or the material filled profile peak area, which may better describe
the effect of the counterface on the wear of the UHMWPE surface would appear to be prudent. 1997 IPEM
Published by Elsevier Science Ltd
Keywords: Wear, surface topography, UHMWPE, total hip replacement
Med. Eng. Phys., 1997, Vol. 19, 711719, December
1. INTRODUCTION
The wear of ultra-high molecular weight polyethylene (UHMWPE) components remains a key
issue in joint replacement surgery, especially of
the lower limb1,2. The wear process may lead to
component loosening and the eventual requirement for revision surgery3. Of those parameters
which may lead to excessive wear, changes in the
surface roughness of the femoral head are currently thought to be important and have been
investigated extensively in the laboratory49. As
well as affecting the volume of debris produced,
laboratory experimentation has demonstrated
that particle morphology is influenced by the
roughness of the counterface10,11. In particular,
rougher surfaces tend to produce more particles
with sub-micron dimensions. It is these subCorrespondence to: R. M. Hall, Academic Department of Orthopaedic Surgery, Clinical Sciences Building, St Jamess University Hospital, Beckett Street, Leeds LS9 7TF, UK. Tel: 0113 206 4186,
0113 233 2132; fax: 0113 206 6791.
of the femoral heads in areas which were considered to be in contact with the socket and, in
general, this coincided with the most heavily
scratched region. Each profile consisted of a
1.4 mm evaluation length, le, with a cutoff of
0.2 mm and contained 770 sampling points.
Twenty-two surface parameters were calculated
once the profiles had been acquired. Six parameters were investigated further. Definitions of
these parameters and the reasons for their
inclusion in the study are presented in Table 1. In
particular, those parameters that have been used
in the assessment of wear in the orthopaedic
environment have been included. The excluded
parameters, including the mean peak-to-valley
height, Rz, and the mean spacing of the profile
irregularities, Sm, were thought not to have a
strong influence on the wear of the polymeric surface. The mean of each of the chosen parameters
across the twenty profiles was calculated.
Qualitative assessment of the femoral heads was
undertaken using both scanning electron (Joel
JSM-IC848) and optical (Zeiss Axiotech)
microscopy. A differential interference contrast
(DIC) facility22 was available on the optical microscope and was found to be particularly suitable for
observing small changes in the slope of features
on the surface of the head (e.g. marks left after
polishing). The objective lenses used were 5,
10, 20 and 50, which allowed overall magnifications of 50, 100, 200 and 500 times respectively. Features of interest were recorded on photographic film. Images on the SEM were recorded
using an image capture system.
The changes in the volume of the internal
socket bore, V, were recorded using the shadowgraphic technique which has been explained
extensively elsewhere23. In brief, a mould is taken
of the internal bore of the socket and projected
onto a screen at a given magnification. A profile
is recorded of the wear plane, and the centres of
the femoral head with respect to the worn and
unworn surfaces of the socket are recorded using
a circular template. The distance between the
centres of the circular template in these two positions is the penetration depth, P. The angle of
penetration, , is the angle between the direction
of penetration and the plane defining the open
face of the socket. The total change in the bore
volume, V, can then be calculated from the
modified form of the equation presented by Kabo
et al.24. Simulator investigations of the wear of
Charnley sockets have revealed a creep component to the volume change in the socket bore,
which is of the order of 30 mm3 25. This is small
compared to the mean volume change of approximately 550 mm3 for explanted Charnley prostheses and was therefore neglected in this analysis26. Thus the total volume change in the internal
bore of the socket was assumed to be attributable
to wear alone.
In keeping with the in vitro analysis, a clinical
wear factor, kclinical, was derived using the empirical wear equation:
V = kclinical Ldx
0
712
(1)
Definition
Ra =
1
zdx
l
0
Rq =
1 2
z dx
l
0
Distance between the highest peak and lowest May be used as a measure of counterface
valley within one evaluation length
imperfections, but does not distinguish
between peaks and valleys
Peak height, Rp
Mean distance between the highest peak and Less prone to exceptional peaks than Rp. The
the mean line in five consecutive sampling
mean peak height can be used in providing
lengths
information on profile shape in conjunction
with the mean peak to valley height
Skewness, Sk
1
Sk =
(Rq)3
z3p(z)dz
I = Ldx = 2.376NMr
(2)
713
Median value
(interquartile range)
2.3 (0.73.4)
608 (274920)
1.69 107 (1.032.09 107)
3.05 108 (1.634.13 108)
1.51 106 (1.092.51 106)
Figure 2 DIC micrograph image of a heavily scratched area on the surface of an explanted head.
714
Figure 3 DIC micrograph of polishing pits observed on a portion of the heads. Note the tails eminating from the pits and that their direction
appears to be oriented with the polishing marks.
Figure 4 DIC micrograph of raised features observed on two explanted femoral heads which may be due to the grain structure of the stainless steel.
715
Figure 5 SEM micrograph of the material build-up either side of a scratch. It was noted that the build-up was not uniform along the length
of the scratch.
Table 3 Summary of those parameters gained from the measurement of surface texture
Parameter
Median value
(interquartile range)
62 (4180)
96 (75145)
1080 (8201550)
500 (400690)
240(200390)
0.28(1.340.38)
Figure 6 Relationship between the arithmetic mean surface roughness, Ra, and the root mean square roughness, Rq.
Table 4 Matrix displaying the correlations (Spearmans rank) between the different surface roughness parameters
Ra
1.0000
Rq
0.9351
(0.000)
0.7218
(0.000)
0.7344
(0.000)
0.8311
(0.000)
0.0644
(0.705)
Ra
Rt
Rp
Rpm
Sk
716
1.0000
0.8127
(0.000)
0.8234
(0.000)
0.8727
(0.000)
0.1248
(0.461)
Rq
1.0000
0.8184
(0.000)
0.8184
(0.000)
0.1293
(0.446)
Rt
1.0000
0.8838
(0.000)
0.1347
(0.429)
Rp
1.0000
0.0253
(0.882)
Rpm
1.0000
Sk
the softer polymeric material are a weighted integration of the effects of all the individual scratches
with some sort of orientational parameter
attached. The polishing pits observed on the less
damaged areas of the head appeared not to have
any lip around their edges, and therefore it could
be assumed they may not have any significant
affect on the wear of the UHMWPE socket.
Isaac et al.12 investigated retrieved acetabular
components and noted that over fifty per cent had
cement embedded within the articulating surface.
Further, it was demonstrated that bone cement
had a detrimental effect on the roughness of the
metallic counterface. There is a high probability
that the material embedded in the surfaces of
sockets investigated in this study was bone cement.
Bone may also ingress into the surface of the polymeric bearing and cortical bone, in particular, has
been shown to be of sufficient hardness to scratch
the surface of a stainless steel counterface30.
In terms of correlations between the clinical
wear factor and the roughness parameter, only
two were observed to have any significant association. The power function fit of k against Ra produced an exponent which was significantly different from zero but considerably lower than the
values gained from the studies by Dowson et al.8
or Weightman and Light4. This difference in
exponent may be due to a number of factors,
including the previously mentioned observation
that the surface roughness of the explanted heads
is not uniform. Further, the lubricating regimes
may be very different between simple wear
machines and the in vivo environment. The other
amplitude roughness parameters produced no
exponent values that were significantly different
from zero. The marginally significant association
Table 5 Degree of association between the surface roughness parameters and the clinical wear factor measured by the Spearmans rank
correlation coefficient and the value of the exponent in the empirical model used by Dowson et al.8 (the P-values are given in parentheses)
Roughness parameter
Ra
Rq
Rt
Rp
Rpm
Sk
0.284
0.248
0.004
0.130
0.211
0.374
(P
(P
(P
(P
(P
(P
=
=
=
=
=
=
0.099)
0.151)
0.981)
0.457)
0.225)
0.027)
Exponent
0.54,
0.40,
0.07,
0.26,
0.47,
SE
SE
SE
SE
SE
0.23
0.26
0.28
0.29
0.30
(P
(P
(P
(P
(P
=
=
=
=
=
0.02)
0.13)
0.79)
0.37)
0.13)
717
718
REFERENCES
1. Howie, D. W., Haynes, D. R., Rogers, S. D., McGee, M.
A. and Pearcy, M. J., The response to particulate debris.
Orthopedic Clinics of North America, 1993, 24, 571581.
2. Harris, W. H., Osteolysis and particle disease in hipreplacementa review. Acta Orthopaedica Scandinavica,
1994, 65, 113123.
3. Wroblewski, B. M., Cementless versus cemented total hiparthroplastya scientific controversy. Orthopedic Clinics of
North America, 1993, 24, 591597.
4. Weightman, B. and Light, D., The effect of the surface
finish of alumina and stainless-steel on the wear rate of
UHMW polyethylene. Biomaterials, 1986, 7, 2024.
5. McKellop, H. and Clarke, I., Wear of artificial joint
materials in laboratory tests. Acta Orthopaedica Scandinavica, 1988, 59, 349352.
6. Dowson, D., Taheri, S. and Wallbridge, N. C., The role
of counterface imperfections in the wear of polyethylene.
Wear, 1987, 119, 277293.
7. Cooper, J. R., Dowson, D. and Fisher, J., The effect of
transfer film and surface roughness on the wear of lubricated ultra-high molecular weight polyethylene. Clin.
Mater., 1993, 14, 295302.
8. Dowson, D., El-Haby Diab, M.M., Gillis, B.J. and Atkinson, J.R., The influence of counterface topography on
the wear of ultra-high molecular weight polyethylene
under wet and dry conditions. In Proc. Int. Symp. on Polymer Wear and its Control, St Louis, 1984, publ. American
Chemical Society, New York, pp. 226232.
9. Fisher, J., Firkins, P., Reeves, E. A., Hailey, J. L. and Isaac,
G. H., The influence of scratches to metallic counterfaces
on the wear of ultra-high molecular weight polyethylene.
Proc. Inst. Mech. Engrs, 1995, 209H, 263264.
10. Cooper, J. R., Dowson, D. and Fisher, J., Macroscopic and
microscopic wear mechanisms in ultra-high-molecularweight polyethylene. Wear, 1993, 162, 378384.
11. Hailey, J. L., Ingham, E., Stone, M., Wroblewski, B. M.
and Fisher, J., Ultra-high molecular weight polyethylene
wear debris generated in vivo and in laboratory tests; the
influence of counterface roughness. Proc. Inst. Mech.
Engrs, 1996, 210H, 310.
12. Isaac, G. H., Atkinson, J. R., Dowson, D. and Wroblewski,
B. M., The role of cement and roughening upon the penetration rate of femoral heads into acetabular cups of
Charnley prostheses. J. Bone Joint Surg. [Br.], 1986, 68,
496.
13. Isaac, G. H., Wroblewski, B. M., Atkinson, J. R. and Dowson, D., A tribological study of retrieved hip prostheses.
Clin. Orthop., 1992, 115125.
14. Atkinson, J. R., Dowson, D., Isaac, J. H. and Wroblewski,
B. M., Laboratory wear tests and clinical observations of
the penetration of femoral heads into acetabular cups
in total replacement hip joints. 3. The measurement of
internal volume changes in explanted Charnley sockets
after 216 years in vivo and the determination of wear
factors. Wear, 1985, 104, 225244.
15. Wroblewski, B. M., Revision Surgery in Total Hip Arthroplasty. Springer-Verlag, London, 1990.
16. Jasty, M., Bragdon, C. R., Lee, K., Hanson, A. and Harris,
W. H., Surface damage to cobaltchrome femoral-head
prostheses. J. Bone Joint Surg. [Br.], 1994, 76B, 7377.
17. Hall, R. M., Unsworth, A., Siney, P. and Wroblewski, B.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
719