- Email: [email protected]
Failure of an uncemented acetabular prosthesis – a case study
Engineering Failure Analysis 13 (2006) 163–169 www.elsevier.com/locate/engfailanal
Failure of an uncemented acetabular prosthesis – a case study P. Heaton-Adegbile a
a,b
, B. Russery a, L. Taylor b, J. Tong
a,*
Department of Mechanical and Design Engineering, University of Portsmouth, Anglesea Building, Anglesea Road, Portsmouth PO1 3DJ, UK b King Edwards VII Hospital, Midhurst GU29 OBL, UK Received 1 October 2004; accepted 11 October 2004 Available online 7 April 2005
Abstract Severe wear and aseptic loosening in an uncemented acetabular prosthesis have been observed in a revision surgery carried out at King Edwards VII hospital by L. Taylor and P. Heaton-Adegbile, twelve years following the primary total hip replacement operation. The superior-lateral wall of the polyethylene liner and part of the titanium cup were found to be completely worn out, such that the ceramic head was in direct articulation with the titanium cup. A three-dimensional finite element model was developed. The polyethylene liner was modelled with the outer surface of the liner fully constrained to represent the much stiffer metal cup. Contact analyses were performed between the articulating surfaces under physiological loading conditions, including normal walking, climbing upstairs and downstairs, using the finite element software ANSYS. The results show high initial contact pressure along the periphery of the liner due to the oversize of the femoral head. The maximum contact pressure was found in the superior–posterior quadrant, which correlates well with the location and the direction of the wear. Both wear particles and stress shielding may have contributed to the periprosthetic bone loss and ultimately the late loosening. Reduction of the interference between the liner and the femoral head seems to be effective in the reduction of the initial contact pressure. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Acetabular loosening; Contact pressure; Finite element analysis; Wear
1. Introduction Wear has been identified as one of the main limiting factors affecting the long-term stability of total hip replacement (THR) [1–3]. Although the actual causal link between the wear of the polyethylene and the loosening process has not been fully established, it is believed that loosening may be initiated by the adverse *
Corresponding author. Tel.: +44 23 92842326; fax: +44 23 92842351. E-mail address: [email protected] (J. Tong).
1350-6307/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.engfailanal.2004.10.010
164
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
tissue response to wear debris which may induce periprosthetic bone loss and ultimately loosening of the prosthesis. We report a case study of an uncemented acetabular prosthesis that appears to support the above hypothesis. The initial contact pressure was obtained from a 3D contact finite element analysis under physiological loading conditions and the results were discussed with respect to the effects of the interference between the articulating surfaces. 2. Clinical background A 67-years-old male patient was referred to King Edwards VII Hospital with a history of continuous pain of approximately 6–8 weeks in the right hip, following a right total hip replacement some 12 years ago. The patient received an uncemented Biomet 54 mm acetabular titanium shell with a polyethylene liner, the ceramic femoral head was larger than the polyethylene liner with a negative clearance of 0.458 mm between the liner and the head. Radiographs showed abnormal alignment of the acetabular cup to the femoral head, indicating superior-lateral migration of the prosthesis. A large bony defect was also observed in the lower illium, which was confirmed to be a degenerative cyst by a bone scan. Subsequent revision revealed that the superior-lateral wall of the plastic liner as well as part of the metal cup was completely worn out, as shown in Fig. 1. The ceramic head was in direct articulation with the metal cup, emitting audible squeaks on mobilising. Massive periprosthetic bone loss was observed in the medial, posterior and inferior walls with extensive tissue stains. The acetabular component was clinically loose, although the femoral prosthesis was intact. 3. Finite element analysis A three-dimensional finite element model was developed based on the measurement of the retrieved polyethylene liner, the titanium shell and the ceramic head in the metrology laboratory. The inner diameter
Fig. 1. The retrieved uncemented prosthesis. Note the extensive wear in the polyethylene liner and titanium cup, as well as the wear stains on the ceramic head.
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
165
of the titanium shell was measured as 27 mm while the diameter of the ceramic head was 27.916 mm. The polyethylene liner was assumed to be fully bonded to the titanium shell, while the titanium shell was neglected in the model due to the two orders difference in the elastic modulus in titanium and polyethylene [4]. The liner was consequently modelled with the outer surface fully constrained. A full 3D model was considered due to the multiaxial hip contact force [5] experienced under physiological loading conditions, including normal walking, climbing upstairs and downstairs. Contact analyses were performed between the articulating surfaces with a coefficient of friction of 0.1 assumed. The numerical calculation was carried out using a commercial software ANSYS. A total of 9481 tetrahedral elements, of which 1106 contact elements, were used in the analyses. The elastic modulus and PoissonÕs ratio were chosen to be 380 GPa and 0.26 for ceramic and IGPa and 0.4 for polyethylene. Three load cases were selected in each loading condition and the loads applied are shown in Table 1. The meshed finite element model is shown in Fig. 2. The contact analysis was validated and convergence evaluated using published data [4,6] (see Fig. 3).
Table 1 Three physiological loading conditions used in the analysis Load case
Data selection
Normal walking
Al A2 A3
Fx (N) 155.5 105.6 152.6
502.4 1386.7 525.1
136.9 854 120.5
Climb upstairs
Bl B2 B3
354.3 272 42.6
489 1463.7 258.4
74.3 924 4.6
Climb downstairs
Cl C2 C3
202.37 31.2 821.1
649.25 335.2 1275.4
172.48 55.8 886.2
Fy (N)
Fz (N)
The components of hip contact force are shown for three cases for each loading condition. The original data are from Bergmann [5].
Fig. 2. The 3D finite element model developed in the analysis. Only the polyethylene liner was modelled with the outer surface fully constrained; preload was applied to the ceramic head to maintain contact between the ball and the cup.
166
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
Fig. 3. Physiological loading conditions used in the analysis: normal walking (a); climbing upstairs (b) and downstairs (c). Three cases were examined for each load case. The hip contact force relative to the cup was obtained from Bergmann [5].
The initial contact pressure distribution during the immediate post-operative phase was evaluated and presented in Fig. 4 for three load cases under normal walking, climbing upstairs and downstairs. High contact pressure was found along the periphery of the liner. Cross-sectional contact pressure distributions are presented where angular coordinate 0° corresponds to the superior rim and 180° corresponds to the inferior rim of the plastic liner. The influence of the load case and the level of the applied hip contact force seem to be swamped by the prestress induced by the interference between the oversized head and the plastic liner. Further examination of the circumference of the liner shows that the highest pressure occurred in the superior–posterior quadrant (Fig. 5), irrespective of the load case applied.
4. Discussion Contact pressure has been identified as one of the key indicative parameters in the contact mechanics studies of articulating surfaces subject to fatigue [7]. A number of studies [6–8] have been carried out, often associated with specific bearing systems. In the current case study, very high initial contact pressure was obtained along the periphery of the polyethylene liner, associated with the high initial wear rate. This high pressure clearly cannot be sustained due to the low creep resistance of the polyethylene and the immediate wear following the operation. The maximum contact pressure correlates well with the location and the direction of the wear from the retrieved prosthesis. This is consistent with the results from an elastic–plastic wear analysis [8] where the highest contact stresses were found in the superior–posterior quadrant during gait. The contact pressure would be immediately modified post-operation, although the level of the contact pressure must have been sufficiently high to sustain the development of wear damage well into the titanium shell. The evolution of the wear damage may be facilitated by the non-uniform distribution of the contact pressure due to the change of the geometry of the liner, as a result of both wear and creep deformation. Consequently high local contact stress near the worn area would encourage further deformation. Under dynamic loading conditions, subsurface cracking may be developed in the highly strained regions, which may accelerate the failure and removal of materials from these regions [9]. By the time of revision surgery, the femoral head had re-established itself at a new recess created superiorly in the titanium shell. The measurement of volumetric wear was not attempted due to the difficulties in establishing the onset of the pen-
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
167
Fig. 4. Cross-sectional contact pressure distribution. Angular coordinate 0° corresponds to the superior rim of the liner while 180° the inferior rim of the liner for (a) normal walking; (b) climbing upstairs and (c) climbing downstairs.
168
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
Fig. 5. Circumferential contact pressure distribution of the polyethylene liner for three loading conditions (case 2 only, as indicated in Fig. 3).
etration into the metal shell. The patient weighed 96 kg, although this seems to have little influence on the initial contact pressure from the FE results, consequently might not be critical towards wear, a fact noted by Charnley [10], supported by Jasty et al. [11]. The interference between the articulating surfaces plays a vital role in the wear process. Much reduced contact pressure may be obtained by reducing the interference between the liner and the head. Fig. 6 shows selected cross-sectional results for contact pressure as a function of the interference. Optimal diametrical clearance has been reported for metal-on-metal prosthesis [8], although a general rule is to avoid being too tight or too loose between the articulating surfaces. Too tight would imply large prestress superimposed on to the hip contact force, hence giving rise to high wear damage. Too loose
Fig. 6. Contact pressure as a function of the interference between the liner and the ball. Negative values indicate that the size of the ball is larger than the liner.
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
169
means large clearance where the contact area would be smaller hence a higher local contact stress which also leads to wear. Polyethylene wear debris is known to cause osteolysis due to the peripheral ingress of particulate debris that leads to inflammatory periprosthetic bone loss in total hip replacement [1,2,12,13]. The mechanism of osteolysis [13] is believed to consist of creation of particles, migration of particles and cellular response to the particles. Periprosthetic osteolysis may be induced by either metal or polyethylene particles. In the present case, revision surgery revealed stained tissues, possibly due to the tissue response to both titanium and polyethylene particles. Massive periprosthetic bone loss was also observed, an observation that appears to support the hypothesis of wear-particle-induced bone loss [2]. On the other hand, stress shielding due to the stiff metal cup would be substantial, and this may also contribute to the cyst in the superior of the acetabulum.
5. Conclusions A case of wear and late acetabular loosening has been reported in this study. High initial contact pressure was found along the periphery of the polyethylene liner with the highest pressure found in the superior–posterior quadrant, consistent with the location and the direction of wear from the retrieved prosthesis. Subsequent development of the wear damage may be encouraged by the non-uniform distribution of the contact pressure due to wear and creep of the polyethylene liner. Extensive bone loss may be the result of wear particle-induced osteolysis, although stress shielding may also play a part. Reducing the interference between the articulating surfaces would be effective in reducing the initial contact pressure hence preventing the onset of wear.
References [1] Schmalzried TT, Kwong LM, Jasry M, Sedlacek RC, Haire TC, OÕConnor DO et al. The mechanism of loosening of cemented acetabular components in total hip arthroplasty. Clin Orthop Rel Res 1992;274:60–78. [2] McGee MA, Howie DW, Costi K, Haynes DR, Wildenauer CI, Pearcy MJ et al. Implant retrieval studies of the wear and loosening of prosthetic joints: a review. Wear 2000;241:158–65. [3] Edidin AA, Rimnac CM, Goldberg VM, Kurtz SM. Mechanical behaviour, wear surface morphology, and clinical performance of UHMWPE acetabular components after 10 years of implantation. Wear 2001;250:152–8. [4] Mak MM, Jin Z-M. Analysis of contact mechanics in ceramic-on-ceramic hip joint replacements. Proc Inst Mech Eng, Part H, Eng In Med 2002;216:231–6. [5] Bergmann G. HIP98. Berlin: Freie Universitaet; 2001. [6] Liu F, Jin ZM, Grigoris P, Hirt F, Rieker C. Contact mechanics of metal-on-metal hip implants employing a metallic cup with a UHMWOE backing. Proc Inst Mech Eng, Part H, Eng In Med 2003;217:207–13. [7] Jin ZM, Dowson D, Fisher J. A parametric analysis of the contact stress in ultra-high molecular weight polyethylene acetabular cups. Med Eng Phys 1994;16:398–405. [8] Teoh SH, Chan WH, Thampuran R. An elasto-plastic finite element model for polyethylene wear in total hip arthroplasty. J Biomech 2002;35:323–30. [9] Cooper JR, Dowson D, Fisher J. Macroscopic and microscopic wear mechanisms in ultra-high molecular weight polyethylene. Wear 1993;162–164:378–84. [10] Charnley J. Low friction arthroplasty of the hip. Berlin: Springer-Verlag; 1979. [11] Jasty M, Goetz DD, Bragdon CR, Lee KR, Hanson AE, Elder JR et al. Wear of polyethylene acetabular components in total hip arthroplasty. J Bone Joint Surg 1997;79-A:349–58. [12] Livermore J, Ilstrop D, Morrey B. Effect of femoral head size on wear of the polyethylene acetabular component. J Bone Joint Surg 1990;72A:518–28. [13] Harris WH. Osteolysis and particle disease in hip replacement. Acta Orthop Scand 1994;65:113–23.
Failure of an uncemented acetabular prosthesis – a case study P. Heaton-Adegbile a
a,b
, B. Russery a, L. Taylor b, J. Tong
a,*
Department of Mechanical and Design Engineering, University of Portsmouth, Anglesea Building, Anglesea Road, Portsmouth PO1 3DJ, UK b King Edwards VII Hospital, Midhurst GU29 OBL, UK Received 1 October 2004; accepted 11 October 2004 Available online 7 April 2005
Abstract Severe wear and aseptic loosening in an uncemented acetabular prosthesis have been observed in a revision surgery carried out at King Edwards VII hospital by L. Taylor and P. Heaton-Adegbile, twelve years following the primary total hip replacement operation. The superior-lateral wall of the polyethylene liner and part of the titanium cup were found to be completely worn out, such that the ceramic head was in direct articulation with the titanium cup. A three-dimensional finite element model was developed. The polyethylene liner was modelled with the outer surface of the liner fully constrained to represent the much stiffer metal cup. Contact analyses were performed between the articulating surfaces under physiological loading conditions, including normal walking, climbing upstairs and downstairs, using the finite element software ANSYS. The results show high initial contact pressure along the periphery of the liner due to the oversize of the femoral head. The maximum contact pressure was found in the superior–posterior quadrant, which correlates well with the location and the direction of the wear. Both wear particles and stress shielding may have contributed to the periprosthetic bone loss and ultimately the late loosening. Reduction of the interference between the liner and the femoral head seems to be effective in the reduction of the initial contact pressure. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Acetabular loosening; Contact pressure; Finite element analysis; Wear
1. Introduction Wear has been identified as one of the main limiting factors affecting the long-term stability of total hip replacement (THR) [1–3]. Although the actual causal link between the wear of the polyethylene and the loosening process has not been fully established, it is believed that loosening may be initiated by the adverse *
Corresponding author. Tel.: +44 23 92842326; fax: +44 23 92842351. E-mail address: [email protected] (J. Tong).
1350-6307/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.engfailanal.2004.10.010
164
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
tissue response to wear debris which may induce periprosthetic bone loss and ultimately loosening of the prosthesis. We report a case study of an uncemented acetabular prosthesis that appears to support the above hypothesis. The initial contact pressure was obtained from a 3D contact finite element analysis under physiological loading conditions and the results were discussed with respect to the effects of the interference between the articulating surfaces. 2. Clinical background A 67-years-old male patient was referred to King Edwards VII Hospital with a history of continuous pain of approximately 6–8 weeks in the right hip, following a right total hip replacement some 12 years ago. The patient received an uncemented Biomet 54 mm acetabular titanium shell with a polyethylene liner, the ceramic femoral head was larger than the polyethylene liner with a negative clearance of 0.458 mm between the liner and the head. Radiographs showed abnormal alignment of the acetabular cup to the femoral head, indicating superior-lateral migration of the prosthesis. A large bony defect was also observed in the lower illium, which was confirmed to be a degenerative cyst by a bone scan. Subsequent revision revealed that the superior-lateral wall of the plastic liner as well as part of the metal cup was completely worn out, as shown in Fig. 1. The ceramic head was in direct articulation with the metal cup, emitting audible squeaks on mobilising. Massive periprosthetic bone loss was observed in the medial, posterior and inferior walls with extensive tissue stains. The acetabular component was clinically loose, although the femoral prosthesis was intact. 3. Finite element analysis A three-dimensional finite element model was developed based on the measurement of the retrieved polyethylene liner, the titanium shell and the ceramic head in the metrology laboratory. The inner diameter
Fig. 1. The retrieved uncemented prosthesis. Note the extensive wear in the polyethylene liner and titanium cup, as well as the wear stains on the ceramic head.
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
165
of the titanium shell was measured as 27 mm while the diameter of the ceramic head was 27.916 mm. The polyethylene liner was assumed to be fully bonded to the titanium shell, while the titanium shell was neglected in the model due to the two orders difference in the elastic modulus in titanium and polyethylene [4]. The liner was consequently modelled with the outer surface fully constrained. A full 3D model was considered due to the multiaxial hip contact force [5] experienced under physiological loading conditions, including normal walking, climbing upstairs and downstairs. Contact analyses were performed between the articulating surfaces with a coefficient of friction of 0.1 assumed. The numerical calculation was carried out using a commercial software ANSYS. A total of 9481 tetrahedral elements, of which 1106 contact elements, were used in the analyses. The elastic modulus and PoissonÕs ratio were chosen to be 380 GPa and 0.26 for ceramic and IGPa and 0.4 for polyethylene. Three load cases were selected in each loading condition and the loads applied are shown in Table 1. The meshed finite element model is shown in Fig. 2. The contact analysis was validated and convergence evaluated using published data [4,6] (see Fig. 3).
Table 1 Three physiological loading conditions used in the analysis Load case
Data selection
Normal walking
Al A2 A3
Fx (N) 155.5 105.6 152.6
502.4 1386.7 525.1
136.9 854 120.5
Climb upstairs
Bl B2 B3
354.3 272 42.6
489 1463.7 258.4
74.3 924 4.6
Climb downstairs
Cl C2 C3
202.37 31.2 821.1
649.25 335.2 1275.4
172.48 55.8 886.2
Fy (N)
Fz (N)
The components of hip contact force are shown for three cases for each loading condition. The original data are from Bergmann [5].
Fig. 2. The 3D finite element model developed in the analysis. Only the polyethylene liner was modelled with the outer surface fully constrained; preload was applied to the ceramic head to maintain contact between the ball and the cup.
166
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
Fig. 3. Physiological loading conditions used in the analysis: normal walking (a); climbing upstairs (b) and downstairs (c). Three cases were examined for each load case. The hip contact force relative to the cup was obtained from Bergmann [5].
The initial contact pressure distribution during the immediate post-operative phase was evaluated and presented in Fig. 4 for three load cases under normal walking, climbing upstairs and downstairs. High contact pressure was found along the periphery of the liner. Cross-sectional contact pressure distributions are presented where angular coordinate 0° corresponds to the superior rim and 180° corresponds to the inferior rim of the plastic liner. The influence of the load case and the level of the applied hip contact force seem to be swamped by the prestress induced by the interference between the oversized head and the plastic liner. Further examination of the circumference of the liner shows that the highest pressure occurred in the superior–posterior quadrant (Fig. 5), irrespective of the load case applied.
4. Discussion Contact pressure has been identified as one of the key indicative parameters in the contact mechanics studies of articulating surfaces subject to fatigue [7]. A number of studies [6–8] have been carried out, often associated with specific bearing systems. In the current case study, very high initial contact pressure was obtained along the periphery of the polyethylene liner, associated with the high initial wear rate. This high pressure clearly cannot be sustained due to the low creep resistance of the polyethylene and the immediate wear following the operation. The maximum contact pressure correlates well with the location and the direction of the wear from the retrieved prosthesis. This is consistent with the results from an elastic–plastic wear analysis [8] where the highest contact stresses were found in the superior–posterior quadrant during gait. The contact pressure would be immediately modified post-operation, although the level of the contact pressure must have been sufficiently high to sustain the development of wear damage well into the titanium shell. The evolution of the wear damage may be facilitated by the non-uniform distribution of the contact pressure due to the change of the geometry of the liner, as a result of both wear and creep deformation. Consequently high local contact stress near the worn area would encourage further deformation. Under dynamic loading conditions, subsurface cracking may be developed in the highly strained regions, which may accelerate the failure and removal of materials from these regions [9]. By the time of revision surgery, the femoral head had re-established itself at a new recess created superiorly in the titanium shell. The measurement of volumetric wear was not attempted due to the difficulties in establishing the onset of the pen-
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
167
Fig. 4. Cross-sectional contact pressure distribution. Angular coordinate 0° corresponds to the superior rim of the liner while 180° the inferior rim of the liner for (a) normal walking; (b) climbing upstairs and (c) climbing downstairs.
168
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
Fig. 5. Circumferential contact pressure distribution of the polyethylene liner for three loading conditions (case 2 only, as indicated in Fig. 3).
etration into the metal shell. The patient weighed 96 kg, although this seems to have little influence on the initial contact pressure from the FE results, consequently might not be critical towards wear, a fact noted by Charnley [10], supported by Jasty et al. [11]. The interference between the articulating surfaces plays a vital role in the wear process. Much reduced contact pressure may be obtained by reducing the interference between the liner and the head. Fig. 6 shows selected cross-sectional results for contact pressure as a function of the interference. Optimal diametrical clearance has been reported for metal-on-metal prosthesis [8], although a general rule is to avoid being too tight or too loose between the articulating surfaces. Too tight would imply large prestress superimposed on to the hip contact force, hence giving rise to high wear damage. Too loose
Fig. 6. Contact pressure as a function of the interference between the liner and the ball. Negative values indicate that the size of the ball is larger than the liner.
P. Heaton-Adegbile et al. / Engineering Failure Analysis 13 (2006) 163–169
169
means large clearance where the contact area would be smaller hence a higher local contact stress which also leads to wear. Polyethylene wear debris is known to cause osteolysis due to the peripheral ingress of particulate debris that leads to inflammatory periprosthetic bone loss in total hip replacement [1,2,12,13]. The mechanism of osteolysis [13] is believed to consist of creation of particles, migration of particles and cellular response to the particles. Periprosthetic osteolysis may be induced by either metal or polyethylene particles. In the present case, revision surgery revealed stained tissues, possibly due to the tissue response to both titanium and polyethylene particles. Massive periprosthetic bone loss was also observed, an observation that appears to support the hypothesis of wear-particle-induced bone loss [2]. On the other hand, stress shielding due to the stiff metal cup would be substantial, and this may also contribute to the cyst in the superior of the acetabulum.
5. Conclusions A case of wear and late acetabular loosening has been reported in this study. High initial contact pressure was found along the periphery of the polyethylene liner with the highest pressure found in the superior–posterior quadrant, consistent with the location and the direction of wear from the retrieved prosthesis. Subsequent development of the wear damage may be encouraged by the non-uniform distribution of the contact pressure due to wear and creep of the polyethylene liner. Extensive bone loss may be the result of wear particle-induced osteolysis, although stress shielding may also play a part. Reducing the interference between the articulating surfaces would be effective in reducing the initial contact pressure hence preventing the onset of wear.
References [1] Schmalzried TT, Kwong LM, Jasry M, Sedlacek RC, Haire TC, OÕConnor DO et al. The mechanism of loosening of cemented acetabular components in total hip arthroplasty. Clin Orthop Rel Res 1992;274:60–78. [2] McGee MA, Howie DW, Costi K, Haynes DR, Wildenauer CI, Pearcy MJ et al. Implant retrieval studies of the wear and loosening of prosthetic joints: a review. Wear 2000;241:158–65. [3] Edidin AA, Rimnac CM, Goldberg VM, Kurtz SM. Mechanical behaviour, wear surface morphology, and clinical performance of UHMWPE acetabular components after 10 years of implantation. Wear 2001;250:152–8. [4] Mak MM, Jin Z-M. Analysis of contact mechanics in ceramic-on-ceramic hip joint replacements. Proc Inst Mech Eng, Part H, Eng In Med 2002;216:231–6. [5] Bergmann G. HIP98. Berlin: Freie Universitaet; 2001. [6] Liu F, Jin ZM, Grigoris P, Hirt F, Rieker C. Contact mechanics of metal-on-metal hip implants employing a metallic cup with a UHMWOE backing. Proc Inst Mech Eng, Part H, Eng In Med 2003;217:207–13. [7] Jin ZM, Dowson D, Fisher J. A parametric analysis of the contact stress in ultra-high molecular weight polyethylene acetabular cups. Med Eng Phys 1994;16:398–405. [8] Teoh SH, Chan WH, Thampuran R. An elasto-plastic finite element model for polyethylene wear in total hip arthroplasty. J Biomech 2002;35:323–30. [9] Cooper JR, Dowson D, Fisher J. Macroscopic and microscopic wear mechanisms in ultra-high molecular weight polyethylene. Wear 1993;162–164:378–84. [10] Charnley J. Low friction arthroplasty of the hip. Berlin: Springer-Verlag; 1979. [11] Jasty M, Goetz DD, Bragdon CR, Lee KR, Hanson AE, Elder JR et al. Wear of polyethylene acetabular components in total hip arthroplasty. J Bone Joint Surg 1997;79-A:349–58. [12] Livermore J, Ilstrop D, Morrey B. Effect of femoral head size on wear of the polyethylene acetabular component. J Bone Joint Surg 1990;72A:518–28. [13] Harris WH. Osteolysis and particle disease in hip replacement. Acta Orthop Scand 1994;65:113–23.