- Email: [email protected]
Failure analysis of retrieved UHMWPE tibial insert in total knee replacement
Engineering Failure Analysis 18 (2011) 1415–1423
Contents lists available at ScienceDirect
Engineering Failure Analysis journal homepage: www.elsevier.com/locate/engfailanal
Failure analysis of retrieved UHMWPE tibial insert in total knee replacement S. Liza a, A.S.M.A. Haseeb a,⇑, A.A. Abbas b, H.H. Masjuki a a b
Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia Department of Orthopaedic Surgery, Faculty of Medicine Building, University of Malaya, 50603 Kuala Lumpur, Malaysia
a r t i c l e
i n f o
Article history: Received 22 November 2010 Received in revised form 7 April 2011 Accepted 9 April 2011 Available online 22 April 2011 Keywords: Polyethylene Failure analysis Total knee replacement Wear Delamination
a b s t r a c t This study involves the failure analysis of an ultra high molecular polyethylene (UHMWPE) tibial insert from ApolloÒ Total Knee System, which was removed after 10 years of service from 70 years old female patient. The tibial insert was investigated by using a stereoscope, scanning electron microscope (SEM), infinite focus microscope (IFM) and energy disperse spectroscopy (EDS) to characterize the morphology and composition of the bearing surface. Differential scanning calorimetry (DSC) and Fourier transform spectroscopy (FTIR) were employed to characterize the degradation and crystallinity of the component. Gel-permeation chromatography (GPC) was used to measure the polyethylene tibial insert molecular weight. Results showed that the failure of total knee replacement (TKR) was associated with high grade wear and oxidation degradation. Surface delamination, scratch marks, pitting, folding, and embedded third body particles were observed on the retrieved UHMWPE tibial surface. Pit depth as large as 60 lm was measured on the surface. The damage features observed on the UHMWPE tibial insert suggested degradation is due to fatigue related wear and is oxidation-induced. Overall results show that the UHMWPE tibial insert which was retrieved from a patient who is active and but not overweight underwent degradation of material properties and high grade wear during 10 years of service. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction Ultra high molecular weight polyethylene (UHMWPE) has been used as a load bearing and articulating counterface in total knee replacement (TKR) for the past 40 years [1]. Polyethylene wear is considered a major limitation of long term success of TKR [2,3]. Analysis of retrieved polyethylene components has been carried out by a number of researchers to reveal the evidence of the TKR failure [4–9]. Their work has helped to characterize and analyse the failure due to wear of polyethylene tibial insert against metal femoral component. Recent studies have dealt with a number of polyethylene tibial inserts from the TKR. The groups of retrieved tibial insert were analyzed by using Hood’s grading scale system. The surface of tibial insert was divided in 10 regions and the wear damage on the surface was measured based on the extent and severity of the seven damage modes (pitting, scratching, burnishing, embedded particulate debris, abrasion, permanent deformation and surface delamination). Then, the total damage was achieved by sum the score for all seven modes across all the ten regions. The results then were compared by considering different factors which could cause their failure. Previous study by Engh et al. [4] compared the wear mode of two different designs of tibial insert, namely mobile and fixed bearing knee inserts. They found that TKR with increased contact areas between metal and polyethylene components can result in reduced contact stresses and decreased wear. Ashraf et al. [10], in their retrieval analysis, determined the wear ⇑ Corresponding author. Tel.: +60 3 7967 5204; fax: +60 3 7967 5317. E-mail address: [email protected] (A.S.M.A. Haseeb). 1350-6307/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.engfailanal.2011.04.001
1416
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
rate of 18 retrieved polyethylene tibial inserts by measuring the linear penetration and volumetric wear using a coordinate measuring machine. They found a linear relationship between both the volumetric and linear penetration with the time of implantation. Apart from retrieval study in a group, there was also a case study on one retrieved polyethylene tibial insert. Diabb et al. [11], investigated the cause of failure of a 6 month implanted tibial inserts. In their study, they focused more on the changes of the properties of polyethylene that lead to the failure of polyethylene tibial insert. The present study attempted to establish the causes that led to the surface damage of polyethylene tibial insert from a knee prosthesis which was retrieved after 10 years in vivo. The patient was an elderly housewife who was a community ambulator, and was active and not overweight. Analysis on the worn surface polyethylene will provide information about the mechanism and process of the polyethylene degradation after 10 years of service. 2. Material and methods 2.1. Retrieved specimens In 1999, a 70 years old female patient received a fixed bearing, posterior stabilized ApolloÒ Total Knee System (Sulzer Orthopaedics, Austin, Texas) for the treatment of severe osteoarthritis of the right knee. Ten years postoperatively, the patient returned to the clinic complaining of pain which was increasing in severity. She also complained of increasing instability and deformity of the knee during walking. Radiographs of the knee performed at this time showed that the tibial tray was in a severe varus position, with reduction of the medial joint space (Fig. 1a). The right total knee replacement was revised shortly after the consultation. Intra-operatively, the tibial prosthesis was found to be loose. There was no evidence of infection. The failed knee replacement was replaced with the new implant as shown in Fig. 1b. Fig. 2 shows the retrieved knee prosthesis consisting of three components which are femoral component, UHMWPE tibial insert and tibial tray. The femoral component and tibial tray were made of cobalt–chromium (CoCr) alloys and tibial insert was made from highly cross-linked UHMWPE. 2.2. Surface degradation evaluation Surface degradation features of UHMWPE component were evaluated using stereomicroscope, scanning electron microscope (Quanta 200 FESEM, FEI) and Alicona infinite focus microscope (IFM). IFM not only provided high resolution topographical and true colour information but could also measure the roughness average (Ra) of the polyethylene. A comparison of the Ra values from both lateral and medial regions was done. Scanning electron microscope (Quanta FEGSEM with Oxford EDX), used to investigate the worn surface characterization of the retrieved sample. Energy dispersive X-ray (EDX) microanalysis was performed to find out the component of the surface features. 2.3. Oxidation characterization Fourier transform infrared (FTIR) spectroscopy was performed on a Perkin-Elmer Spectrum 100 FTIR Spectrometer. FTIR was used to characterize the extent of oxidative degradation in the polyethylene tibial insert. The sample for FTIR
Fig. 1. Radiographs of the right knee of the patient (a) at presentation 10 years after primary total knee replacement (arrow pointing at decreased joint space at medial joint line with tibial tray in varus position) and (b) anteroposterior and lateral radiographs of the right knee post revision knee replacement.
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
1417
Femoral Component
UHMWPE Tibial Insert
Tibial Tray
Fig. 2. Knee prosthesis components after retrieval from the patient (cleaned).
characterization was prepared by cutting the UHMWPE sample into approximately 1 mm thin section. The FTIR measurements were performed on two distinct regions of the thin section. These are the oxidized surface layer and the oxidized bulk region. Scans of thin sections were acquired at 2 cm 1 intervals from 600 cm 1 to 4000 cm 1. 2.4. Crystallinity measurement The crystallinity of the UHMWPE tibial insert was measured using differential scanning calorimeter (DSC) (DSC 820, Mettler-Toledo Inc., Columbus, OH). The sample for DSC characterization was prepared by using a 5 mm inner diameter serrated coring bit at the weight bearing region. Heating runs were conducted from 30 °C to 200 °C at rate of 10 °C/min on samples weighing about 5.6 mg. The sample was sealed in an aluminium sample pan and used for the measurement in the standard DSC mode. Sample crystallinity was determined by comparing the total heat of melting DH (the area under the endotherm) to the total heat of fusion of fully crystalline UHMWPE (DHf = 293 J/g) [12]. The percentage crystallinity was calculated as 100 DH/DHf. 2.5. Molecular weight measurement Gel-permeation chromatography (GPC) analysis (ALLIANCE GPCV-200) equipment was made in order to measure the polyethylene molecular weight of the tibial insert. Molecular weight analysis was conducted with a 50 mg of polyethylene dissolved in trichlorobenzene at 140 °C. 3. Results The wear damage analysis of the articulating surface of the UHMWPE tibial insert showed different types of features. Fig. 3b and c displays the wear surface of retrieved UHMWPE at 40 magnification under the stereoscope. Light scratching marks, oriented in the vertical direction in the micrographs are noted on both the lateral and medial compartments. Further observation was done in the SEM and the micrographs are shown in Fig. 4. Close observation of the surface at different areas revealed a number of features. Scratches (Fig. 4a), pitting (Fig. 4b), folding (Fig. 4c), and surface delamination (Fig. 4d) can be observed as a result of the wear of the component during service. Such features were seen all over the surface. As can be seen from Fig. 4c, there were some ‘white particles’ scattered on the worn surface. The EDX analysis (Fig. 5) confirmed the presence of Ca and P in these particles. Since hydroxyapatite Ca10(PO4)6(OH)2 is an important component of the human bone [13], the presence of Ca and P in these particles suggests that these particles are bone fragments.
1418
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
(a)
(b)
(c)
1 cm
1 cm
Fig. 3. (a) General view of UHMWPE and optical microscopic images of the surface of tibial insert, (b) lateral compartment and (c) medial compartment.
(a)
(b)
Pit
(c)
(d)
Delamination
Fig. 4. SEM images of UHMWPE tibial plate surface indicating different features of wear damage: (a) scratches, (b) pits, (c) folding and (d) delamination.
Images of the sample surface were taken by using infinite focus microscope (IFM) (Fig. 6). These images (Fig. 6a and b) revealed the occurrence of the features described above in a more clear way. In addition, IFM also allows the quantification of surface data and provide the dimensional surface profiles (Fig. 6c and d). For example, deep pits were revealed at the bottom right corner of Fig. 6c and upper left corner of Fig. 6d. Pit depths as large as 60 lm and 35 lm were obtained. The surface roughness of UHMWPE tibial component was also measured by IFM. The surface roughness data, which are listed in Table 1, suggest that there is no significant difference between lateral and medial roughness.
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
1419
Fig. 5. (a) SEM image showing white particles scattered on the worn surface of UHMWPE tibial insert, (b) a typical EDX spectrum on the white particles indicating the presence of calcium (Ca) and phosphorus (P).
Fig. 7 shows the FTIR spectra taken on material at the surface and bulk region of the UHMWPE samples. There was no significant difference between the two regions. Prominent peaks due to absorption at 2915 and 2848 cm 1 (CH2 stretching), 1462 and 1471 cm 1 (CH2 and CH3 bending) and 719 and 730 cm 1 (linear chain of at least 4 CH2 groups) are clearly visible in Fig. 7. Additionally, there are a number of small absorption peaks at surface region which can be assigned to oxidation products. These are carboxylic acid (C@O) at 1697.03 and 1300.69 cm 1, ketone (C@O) at 1716 cm 1, ester (C@O) at 1748.26 cm 1 and alcohol (C@O) at 3361.09 cm 1 [14,15]. The band at 1166.54 cm 1 can be mainly assigned to PO 4 [13]. This suggests that the sample contained a small amount of bone fragment which was indicated previously by EDX analysis. In the present study, the crystallinity and molecular weight measurement of the UHMWPE tibial insert were carried out in order to determine whether there was a change after the insert was implanted for 10 years. Fig. 8 shows the melting endotherm for the retrieved UHMWPE tibial insert. The heat of fusion for this sample was 164.85 J/g. It was obtained by integrating the area under the endothermic peak. The value of the average crystallinity obtained from DSC of the UHMWPE tibial insert was 56.6%. Finally, the molecular weight of the retrieved UHMWPE tibial insert was measured by gel-permeation chromatography (GPC). Fig. 9 shows the distribution of molecular weight of UHMWPE tibial insert after the tibial insert was implanted for 10 years. The molecular weight (Mw) of UHMWPE tibial insert was 130,104 g/mol.
4. Discussion Orthopaedic literature contains many articles describing the clinical performance of knee replacement with various designs and materials [4–8]. According to Hood et al. [6], there are seven surface wear damage modes on the articulating surface of retrieved tibial insert. The wear damage modes are pitting, scratching, burnishing, embedded particulate debris, abrasion, permanent deformation and surface delamination. In the present study the most common features on the articulating surface of the retrieved polyethylene tibal insert was delamination. This was followed by scratches, pitting, folding, and third body particles (bone fragment). Ho et al. [16] suggested that high-grade wear consisted of pitting, scratching, and delamination. Thus, the wear damage on the 10 years implanted UHMWPE tibial insert in this current study can be classified as a high-grade wear. Delamination and pitting which are related to fatigue wear were reported by Bradford et al. [17] as the most common feature of damage for total knee replacement. Medel et al. [18] suggested that the fatigue wear mechanism was predominant in the knee implant system due to lower conformity and higher contact stress between the femoral and tibial component. Studies show that the formation of large flakes of wear debris known as delamination was initiated by cracks from the cyclic stress at the contact surface of polyethylene [19]. Similarly, Burnett et al. [20] suggested that the fatigue wear modes (delamination and cracking) are resulted from the combination of higher stresses and reduced mechanical properties. They also reported that the reduction in mechanical properties such as loss of toughness is due to oxidation. It has been observed that the oxidative degradation of polyethylene can lead to a decrease in wear resistance and mechanical properties [15,21,22]. Collier et al. [8] observed that delamination is correlated with zone of high oxidation. They found that subsurface zone which highly oxidized could be related to the delamination. They proposed that the reduction in mechanical properties such as decrease in tensile strength and low elongation, led to the formation of visible subsurface white zone and resulted delamination and cracking. In the present study, the presence of oxidation product such as acid carboxylic, ketone, ester and alcohol from FTIR analysis confirms that oxidation degradation had occurred in the UHMWPE insert. Repetitive cyclic loading on the insert during daily activities over 10 years have resulted in the delamination and pitting in the present retrieved sample.
1420
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
Fig. 6. IFM images taken on wear surface of UHMWPE tibial inset; (a and b) surface images and (c and d) 3D surface images corresponding dimensional surface profiles.
Table 1 Surface roughness of wear UHMWPE component surface.
Lateral Medial
Top (lm)
Left (lm)
Bottom (lm)
Right (lm)
Average
1.931 1.843
1.753 2.204
1.242 1.343
1.722 1.454
1.662 1.711
1421
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
-1
3361.29 cm -1
3361.09 cm
-1
1697.24 cm -1 1716.2 cm -1 1748.67 cm
-1
1697.03 cm -1 1716 cm -1 1748.26 cm
1300.51 cm -1
1300.69 cm
-1
1166.85 cm -1
1166.54 cm
-1
Bulk region
Surface region
Fig. 7. IR spectrum recorded on the surface and bulk region of UHMWPE samples.
Fig. 8. A plot of heat flow versus temperature for UHMWPE tibial insert.
Fig. 9. The distribution of molecular weight of UHMWPE tibial insert after being implated for 10 years.
1422
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
The gamma sterilization method has significant effect on delamination of the polyethylene surface. Severe delamination was found on the retrieved polyethylene surface with gamma radiation in air and in nitrogen [8,23]. No delamination was found on the gas plasma sterilized polyethylene insert [4]. The degradation does not occur immediately after sterilization but over time. It has been reported that degradation occurs over a period prior to implantation (shelf ageing) and in vivo services [21,24]. Sterilization by gamma irradiation in air has been shown to have the potential to accelerate the oxidation of polyethylene components [25]. It is suggested that gamma sterilization have played a role in the degradation of the retrieved insert. A correlation between implantation time and presence of delamination was found by Collier et al. [8]. Retrieved samples with short implantation duration, less than 4 years, displayed signs of delamination or cracking on 17–27% of cases. In contrast, 65% of the retrieved tibial components displayed a signs of delamination or cracking for durations greater than 4 years. Apart from delamination, pitting, scratching and folding were also observed in the present sample. Garcia et al. [5] have found that pitting and scratching were the most predominant damage modes on the surface damage of 40 rotating-platform TKAs. Third body particles have been reported by Crowninshield et al. [7] and Muratoglu et al. [26] as a main cause of formation multiple surfaces folding, pitting and scratching. Crowninshield et al. [7] found that the third body particles were embedded in and dislodged from the polyethylene surface, causing the formation of pitting during the motion and loading at knee prosthesis. In addition, scratches were formed when third body particles were pushed along the articular surface and leaving tracks of deformation within the polyethylene surface. Hirakawa et al. [27] suggested that small debris particles attributed to the formation of scratches and larger debris particles which produce by polyethylene fatigue attributed to the delamination formation. These findings clearly showed the formation of surface damage modes were related to each other. The absence of surface burnishing on the UHMWPE surface distinguishes this work from the previous reports. The absence of burnishing wear can be caused by the design of knee implant. Engh et al. [4] have found that the knee implant with fixed bearing system have small percentage of the surface being burnished when compared with the mobile bearing system due to small contact area. In this present study, the ApolloÒ Total Knee system which is fixed bearing system did not exhibit burnishing on the polyethylene surface. In the present study, the crystallinity and molecular weight measurement of the UHMWPE tibial insert were carried out in order to determine whether there is a change after the insert was implanted for 10 years. Kurtz [28] suggested that the degree of crystallinity for medical grade polyethylene should be within 39–75%. Thus in this study, the crystallinity falls within the suggested range. Since the crystallinity data of UHMWPE tibial insert prior to implantation was also not available, no comment can be made on the role of crystallinity on the performance of the present implant. However, it is expected that there is a change in the molecular weight of UHMWPE after implanted for 10 years. In the present case, the molecular weight of UHMWPE was measured as 130, 104 g/mol. This result is approximately ten times lower than the actual average molecular weight of a medical grade polyethylene which is 1.5 106 g/mol [28]. Diabb [11] also reported on the reduction of molecular weight after retrieval. They found that the molecular weight of the polyethylene tibial insert just after 7 months was 166,855 g/mol. The degradation in molecular weight of retrieved UHMWPE tibial insert in the present case suggest that polyethylene properties have changed. This is believed to have contributed to the wear damage. 5. Conclusions The following conclusions can be drawn from the present study. High-grade wear and oxidation degradation are associated with the failure of the 10 years implanted UHMWPE tibial insert. The most prominent wear damage mode identified in the retrieved tibial insert is delamination. This is followed by pitting, folding and scratching. The pit depth as large as 60 lm was measured on the surface. Third body particles also occur on the insert surface. The retrieved UHMWPE tibial insert showed low molecular weight, indicating that polyethylene properties have changed in vivo which contributed to wear damage.
Acknowledgements The authors would like to thank University Malaya Medical Centre (UMMC) for supplying the sample. This research was supported by Postgraduate Research Grant (PPP), Grant No: PS079-2009B from University of Malaya. References [1] Santavirta S, Konttinen YT, Lappalainen R, Anttila A, Goodman SB, Lind M, et al. Materials in total joint replacement. Curr Orthop 1998;12:51–7. [2] O’Rourke MR, Callaghan JJ, Goetz DD, Sullivan PM, Johnston RC. Osteolysis associated with a cemented modular posteriorcruciate-substituting total knee design: five to eight years follow up. J Bone Joint Surg Am 2002;84:1362–71.
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
1423
[3] Whaley AL, Trousdale RT, Rand JA, Hanssen AD. Cemented long-stem revision total knee arthroplasty. J Arthroplast 2003;18:592–9. [4] Engh GA, Zimmerman RL, Parks NL, Engh CA. Analysis of wear in retrieved mobile and fixed bearing knee inserts. J Arthroplast 2009;24(6):28–32. [5] Garcia RM, Kraay MJ, Messerschmitt PJ, Golberg VM, Rimnac CM. Analysis of retrieved ultra high molecular polyethylene tibial components from rotating-platform total knee arthroplasty. J Arthroplast 2009;24:131–8. [6] Hood RW, Wright TM, Burstein AH. Retrieval analysis of total knee protheses: a method and its application to 48 total condylar protheses. J Biomed Mater Res 1983;17:829–42. [7] Crowninshield RD, Wimmer MA, Jacobs JJ, Rosenberg AG. Clinical performance of contemporary tibial polyethylene components. J Arthroplast 2008;21(5):754–61. [8] Collier JP, Sperling DK, Currier JH, Sutula LC, Saum KA, Mayor MB. Impact of gamma sterilization on clinical performance of polyethylene in the knee. J Arthroplast 1996;11(4):377–89. [9] William IR, Mayor MB, Collier JP. The impact of sterilization method on wear in knee arthroplasty. Clin Orthop 1998;356:170–80. [10] Ashraf T, Newman JH, Desai VV, Beard D, Nevelos JE. Polyethylene wear in a non congruous unicompartmental knee replacement: a retrieval analysis. Knee 2004;11:177–81. [11] Diabb J, Juarez-Hernandez A, Reyes A, Gonzalez-Rivera C, Hernandez-Rodriguez MAL. Failure analysis for degradation of a polyethylene knee prosthesis component. Eng Fail Anal 2009;16:1770–3. [12] Lopéz JM, Gómez-Barrenata E. Estructura y morfología de las regiones degradadas del polietileno de ultra alto peso molecular en prótesis articulares. Biomecánica 1999;7(13):2–11. [13] Slouf M, Sloufora I, Horak Z, Stepanek P, Entlicher G, Krejcik M, et al. New fast method for determination of number of UHMWPE wear particles. J Mater Sci Mater Med 2004;15:1267–78. [14] Costa L, Luda MP, Trossarelli L, Brach del Prever EM, Crova M, Gallinaro P. Oxidation in orthopaedic UHMWPE sterilized by gamma-radiation and ethylene oxide. Biomaterials 1998;19:659–68. [15] Goldman M, Lee M, Gronsky R, Pruitt L. Oxidation of ultrahigh molecular weight polyethylene characterized by Fourier transformation infrared spectrometry. J Biomed Mater Res 1997;37(1):43–50. [16] Ho FY, Ma HMJJ, Liau, Yeh CR, Huang CH. Mobile-bearing knees reduce rotational asymmetric wear. Clin Orthop Relat Res 2007;462:143–9. [17] Bradford L, Baker DA, Graham J, Chawan A, Ries MD, Pruitt LA. Wear and surface cracking in early retrieved highly cross-linked polyethylene acetabular liners. J Bone Joint Surg Am 2004;86:1271–82. [18] Medel FJ, Kurtz SM, Parvizi J, Klein GR, Kraay MJ, Rimnac CM. In vivo oxidation contributes to delamination but not pitting in polyethylene components for total knee arthroplasty. J Arthroplast in press. [19] Blunn GW, Joshi AB, Lilley PA, Engelbrecht E, Ryd L, Lidgren L, et al. Polyethylene wear in unicondylar knee prostheses. 106 Retrieved Marmor, PCA, and St. Georg tibial components compared. Acta Orthop Scand 1992;63(3):247–55. [20] Burnett RSJ, Biggerstaff S, Currier BH, Collier JP, Barrack RL. Unilateral tibial polyethylene liner failure in bilateral total knee arthroplasty–bilateral retrieval analysis at 8 years. J Arthroplast 2007;22(5):753–8. [21] Kurtz SM, Hozack W, Marcolongo M, Turner J, Rimnac C, Edidin A. Degradation of mechanical properties of UHMWPE acetabular liners following longterm implantation. J Arthroplast 2003;18(1):68–78. [22] Wannomae KK, Christensen SD, Freiberg AA, Bhattacharyya S, Harris WH, Muratoglu OK. The effect of real-time aging on the oxidation and wear of highly cross-linked UHMWPE acetabular liner. Biomaterials 2006;27:1980–7. [23] Brandt JM, Medley JB, MacDonald SJ, Bourne RB. Delamination wear on two retrieved polyethylene inserts after gamma sterilization in nitrogen. Knee 2011;18(2):125–9. [24] H. Oonishi, S.C. Kim, M. Kyomoto, M. Iwamoto, M. Ueno, Comparison of in-vivo wear between polyethylene inserts articulating against ceramic and cobalt–chrome femoral components in total knee prostheses, Bioceramics and Alternative Bearings in Joint Arthroplasty 12th BIOLOXÒ Symposium Seoul, Republic of Korea, 2007, pp. 149–159. [25] Costa L, Jacobson K, Bracco P, Brach del Prever EM. Oxidation of orthopaedic UHMWPE. Biomaterials 2002;23:1613–24. [26] Muratoglu OK, Ruberti J, Melotti S, Spiegelberg SH, Greenbaum ES, Harris WH. Optical analysis of surface changes on early retrievals of highly crosslinked and conventional polyethylene tibial inserts. J Arthroplast 2003;18(7):42–7. [27] Hirakawa K, Bauer TW, Yamaguchi M, Stulberg BN, Wilde AH. Relationship between wear debris particles and polyethylene surface damage in primary total knee arthroplasty. J Arthroplast 1999;14(2):165–71. [28] Kurtz S. The UHMWPE Handbook: Ultra-High Molecular Weight Polyethylene in Total Joint Replacement. first ed. New York: Elsevier Inc.; 2004.
Contents lists available at ScienceDirect
Engineering Failure Analysis journal homepage: www.elsevier.com/locate/engfailanal
Failure analysis of retrieved UHMWPE tibial insert in total knee replacement S. Liza a, A.S.M.A. Haseeb a,⇑, A.A. Abbas b, H.H. Masjuki a a b
Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia Department of Orthopaedic Surgery, Faculty of Medicine Building, University of Malaya, 50603 Kuala Lumpur, Malaysia
a r t i c l e
i n f o
Article history: Received 22 November 2010 Received in revised form 7 April 2011 Accepted 9 April 2011 Available online 22 April 2011 Keywords: Polyethylene Failure analysis Total knee replacement Wear Delamination
a b s t r a c t This study involves the failure analysis of an ultra high molecular polyethylene (UHMWPE) tibial insert from ApolloÒ Total Knee System, which was removed after 10 years of service from 70 years old female patient. The tibial insert was investigated by using a stereoscope, scanning electron microscope (SEM), infinite focus microscope (IFM) and energy disperse spectroscopy (EDS) to characterize the morphology and composition of the bearing surface. Differential scanning calorimetry (DSC) and Fourier transform spectroscopy (FTIR) were employed to characterize the degradation and crystallinity of the component. Gel-permeation chromatography (GPC) was used to measure the polyethylene tibial insert molecular weight. Results showed that the failure of total knee replacement (TKR) was associated with high grade wear and oxidation degradation. Surface delamination, scratch marks, pitting, folding, and embedded third body particles were observed on the retrieved UHMWPE tibial surface. Pit depth as large as 60 lm was measured on the surface. The damage features observed on the UHMWPE tibial insert suggested degradation is due to fatigue related wear and is oxidation-induced. Overall results show that the UHMWPE tibial insert which was retrieved from a patient who is active and but not overweight underwent degradation of material properties and high grade wear during 10 years of service. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction Ultra high molecular weight polyethylene (UHMWPE) has been used as a load bearing and articulating counterface in total knee replacement (TKR) for the past 40 years [1]. Polyethylene wear is considered a major limitation of long term success of TKR [2,3]. Analysis of retrieved polyethylene components has been carried out by a number of researchers to reveal the evidence of the TKR failure [4–9]. Their work has helped to characterize and analyse the failure due to wear of polyethylene tibial insert against metal femoral component. Recent studies have dealt with a number of polyethylene tibial inserts from the TKR. The groups of retrieved tibial insert were analyzed by using Hood’s grading scale system. The surface of tibial insert was divided in 10 regions and the wear damage on the surface was measured based on the extent and severity of the seven damage modes (pitting, scratching, burnishing, embedded particulate debris, abrasion, permanent deformation and surface delamination). Then, the total damage was achieved by sum the score for all seven modes across all the ten regions. The results then were compared by considering different factors which could cause their failure. Previous study by Engh et al. [4] compared the wear mode of two different designs of tibial insert, namely mobile and fixed bearing knee inserts. They found that TKR with increased contact areas between metal and polyethylene components can result in reduced contact stresses and decreased wear. Ashraf et al. [10], in their retrieval analysis, determined the wear ⇑ Corresponding author. Tel.: +60 3 7967 5204; fax: +60 3 7967 5317. E-mail address: [email protected] (A.S.M.A. Haseeb). 1350-6307/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.engfailanal.2011.04.001
1416
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
rate of 18 retrieved polyethylene tibial inserts by measuring the linear penetration and volumetric wear using a coordinate measuring machine. They found a linear relationship between both the volumetric and linear penetration with the time of implantation. Apart from retrieval study in a group, there was also a case study on one retrieved polyethylene tibial insert. Diabb et al. [11], investigated the cause of failure of a 6 month implanted tibial inserts. In their study, they focused more on the changes of the properties of polyethylene that lead to the failure of polyethylene tibial insert. The present study attempted to establish the causes that led to the surface damage of polyethylene tibial insert from a knee prosthesis which was retrieved after 10 years in vivo. The patient was an elderly housewife who was a community ambulator, and was active and not overweight. Analysis on the worn surface polyethylene will provide information about the mechanism and process of the polyethylene degradation after 10 years of service. 2. Material and methods 2.1. Retrieved specimens In 1999, a 70 years old female patient received a fixed bearing, posterior stabilized ApolloÒ Total Knee System (Sulzer Orthopaedics, Austin, Texas) for the treatment of severe osteoarthritis of the right knee. Ten years postoperatively, the patient returned to the clinic complaining of pain which was increasing in severity. She also complained of increasing instability and deformity of the knee during walking. Radiographs of the knee performed at this time showed that the tibial tray was in a severe varus position, with reduction of the medial joint space (Fig. 1a). The right total knee replacement was revised shortly after the consultation. Intra-operatively, the tibial prosthesis was found to be loose. There was no evidence of infection. The failed knee replacement was replaced with the new implant as shown in Fig. 1b. Fig. 2 shows the retrieved knee prosthesis consisting of three components which are femoral component, UHMWPE tibial insert and tibial tray. The femoral component and tibial tray were made of cobalt–chromium (CoCr) alloys and tibial insert was made from highly cross-linked UHMWPE. 2.2. Surface degradation evaluation Surface degradation features of UHMWPE component were evaluated using stereomicroscope, scanning electron microscope (Quanta 200 FESEM, FEI) and Alicona infinite focus microscope (IFM). IFM not only provided high resolution topographical and true colour information but could also measure the roughness average (Ra) of the polyethylene. A comparison of the Ra values from both lateral and medial regions was done. Scanning electron microscope (Quanta FEGSEM with Oxford EDX), used to investigate the worn surface characterization of the retrieved sample. Energy dispersive X-ray (EDX) microanalysis was performed to find out the component of the surface features. 2.3. Oxidation characterization Fourier transform infrared (FTIR) spectroscopy was performed on a Perkin-Elmer Spectrum 100 FTIR Spectrometer. FTIR was used to characterize the extent of oxidative degradation in the polyethylene tibial insert. The sample for FTIR
Fig. 1. Radiographs of the right knee of the patient (a) at presentation 10 years after primary total knee replacement (arrow pointing at decreased joint space at medial joint line with tibial tray in varus position) and (b) anteroposterior and lateral radiographs of the right knee post revision knee replacement.
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
1417
Femoral Component
UHMWPE Tibial Insert
Tibial Tray
Fig. 2. Knee prosthesis components after retrieval from the patient (cleaned).
characterization was prepared by cutting the UHMWPE sample into approximately 1 mm thin section. The FTIR measurements were performed on two distinct regions of the thin section. These are the oxidized surface layer and the oxidized bulk region. Scans of thin sections were acquired at 2 cm 1 intervals from 600 cm 1 to 4000 cm 1. 2.4. Crystallinity measurement The crystallinity of the UHMWPE tibial insert was measured using differential scanning calorimeter (DSC) (DSC 820, Mettler-Toledo Inc., Columbus, OH). The sample for DSC characterization was prepared by using a 5 mm inner diameter serrated coring bit at the weight bearing region. Heating runs were conducted from 30 °C to 200 °C at rate of 10 °C/min on samples weighing about 5.6 mg. The sample was sealed in an aluminium sample pan and used for the measurement in the standard DSC mode. Sample crystallinity was determined by comparing the total heat of melting DH (the area under the endotherm) to the total heat of fusion of fully crystalline UHMWPE (DHf = 293 J/g) [12]. The percentage crystallinity was calculated as 100 DH/DHf. 2.5. Molecular weight measurement Gel-permeation chromatography (GPC) analysis (ALLIANCE GPCV-200) equipment was made in order to measure the polyethylene molecular weight of the tibial insert. Molecular weight analysis was conducted with a 50 mg of polyethylene dissolved in trichlorobenzene at 140 °C. 3. Results The wear damage analysis of the articulating surface of the UHMWPE tibial insert showed different types of features. Fig. 3b and c displays the wear surface of retrieved UHMWPE at 40 magnification under the stereoscope. Light scratching marks, oriented in the vertical direction in the micrographs are noted on both the lateral and medial compartments. Further observation was done in the SEM and the micrographs are shown in Fig. 4. Close observation of the surface at different areas revealed a number of features. Scratches (Fig. 4a), pitting (Fig. 4b), folding (Fig. 4c), and surface delamination (Fig. 4d) can be observed as a result of the wear of the component during service. Such features were seen all over the surface. As can be seen from Fig. 4c, there were some ‘white particles’ scattered on the worn surface. The EDX analysis (Fig. 5) confirmed the presence of Ca and P in these particles. Since hydroxyapatite Ca10(PO4)6(OH)2 is an important component of the human bone [13], the presence of Ca and P in these particles suggests that these particles are bone fragments.
1418
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
(a)
(b)
(c)
1 cm
1 cm
Fig. 3. (a) General view of UHMWPE and optical microscopic images of the surface of tibial insert, (b) lateral compartment and (c) medial compartment.
(a)
(b)
Pit
(c)
(d)
Delamination
Fig. 4. SEM images of UHMWPE tibial plate surface indicating different features of wear damage: (a) scratches, (b) pits, (c) folding and (d) delamination.
Images of the sample surface were taken by using infinite focus microscope (IFM) (Fig. 6). These images (Fig. 6a and b) revealed the occurrence of the features described above in a more clear way. In addition, IFM also allows the quantification of surface data and provide the dimensional surface profiles (Fig. 6c and d). For example, deep pits were revealed at the bottom right corner of Fig. 6c and upper left corner of Fig. 6d. Pit depths as large as 60 lm and 35 lm were obtained. The surface roughness of UHMWPE tibial component was also measured by IFM. The surface roughness data, which are listed in Table 1, suggest that there is no significant difference between lateral and medial roughness.
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
1419
Fig. 5. (a) SEM image showing white particles scattered on the worn surface of UHMWPE tibial insert, (b) a typical EDX spectrum on the white particles indicating the presence of calcium (Ca) and phosphorus (P).
Fig. 7 shows the FTIR spectra taken on material at the surface and bulk region of the UHMWPE samples. There was no significant difference between the two regions. Prominent peaks due to absorption at 2915 and 2848 cm 1 (CH2 stretching), 1462 and 1471 cm 1 (CH2 and CH3 bending) and 719 and 730 cm 1 (linear chain of at least 4 CH2 groups) are clearly visible in Fig. 7. Additionally, there are a number of small absorption peaks at surface region which can be assigned to oxidation products. These are carboxylic acid (C@O) at 1697.03 and 1300.69 cm 1, ketone (C@O) at 1716 cm 1, ester (C@O) at 1748.26 cm 1 and alcohol (C@O) at 3361.09 cm 1 [14,15]. The band at 1166.54 cm 1 can be mainly assigned to PO 4 [13]. This suggests that the sample contained a small amount of bone fragment which was indicated previously by EDX analysis. In the present study, the crystallinity and molecular weight measurement of the UHMWPE tibial insert were carried out in order to determine whether there was a change after the insert was implanted for 10 years. Fig. 8 shows the melting endotherm for the retrieved UHMWPE tibial insert. The heat of fusion for this sample was 164.85 J/g. It was obtained by integrating the area under the endothermic peak. The value of the average crystallinity obtained from DSC of the UHMWPE tibial insert was 56.6%. Finally, the molecular weight of the retrieved UHMWPE tibial insert was measured by gel-permeation chromatography (GPC). Fig. 9 shows the distribution of molecular weight of UHMWPE tibial insert after the tibial insert was implanted for 10 years. The molecular weight (Mw) of UHMWPE tibial insert was 130,104 g/mol.
4. Discussion Orthopaedic literature contains many articles describing the clinical performance of knee replacement with various designs and materials [4–8]. According to Hood et al. [6], there are seven surface wear damage modes on the articulating surface of retrieved tibial insert. The wear damage modes are pitting, scratching, burnishing, embedded particulate debris, abrasion, permanent deformation and surface delamination. In the present study the most common features on the articulating surface of the retrieved polyethylene tibal insert was delamination. This was followed by scratches, pitting, folding, and third body particles (bone fragment). Ho et al. [16] suggested that high-grade wear consisted of pitting, scratching, and delamination. Thus, the wear damage on the 10 years implanted UHMWPE tibial insert in this current study can be classified as a high-grade wear. Delamination and pitting which are related to fatigue wear were reported by Bradford et al. [17] as the most common feature of damage for total knee replacement. Medel et al. [18] suggested that the fatigue wear mechanism was predominant in the knee implant system due to lower conformity and higher contact stress between the femoral and tibial component. Studies show that the formation of large flakes of wear debris known as delamination was initiated by cracks from the cyclic stress at the contact surface of polyethylene [19]. Similarly, Burnett et al. [20] suggested that the fatigue wear modes (delamination and cracking) are resulted from the combination of higher stresses and reduced mechanical properties. They also reported that the reduction in mechanical properties such as loss of toughness is due to oxidation. It has been observed that the oxidative degradation of polyethylene can lead to a decrease in wear resistance and mechanical properties [15,21,22]. Collier et al. [8] observed that delamination is correlated with zone of high oxidation. They found that subsurface zone which highly oxidized could be related to the delamination. They proposed that the reduction in mechanical properties such as decrease in tensile strength and low elongation, led to the formation of visible subsurface white zone and resulted delamination and cracking. In the present study, the presence of oxidation product such as acid carboxylic, ketone, ester and alcohol from FTIR analysis confirms that oxidation degradation had occurred in the UHMWPE insert. Repetitive cyclic loading on the insert during daily activities over 10 years have resulted in the delamination and pitting in the present retrieved sample.
1420
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
Fig. 6. IFM images taken on wear surface of UHMWPE tibial inset; (a and b) surface images and (c and d) 3D surface images corresponding dimensional surface profiles.
Table 1 Surface roughness of wear UHMWPE component surface.
Lateral Medial
Top (lm)
Left (lm)
Bottom (lm)
Right (lm)
Average
1.931 1.843
1.753 2.204
1.242 1.343
1.722 1.454
1.662 1.711
1421
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
-1
3361.29 cm -1
3361.09 cm
-1
1697.24 cm -1 1716.2 cm -1 1748.67 cm
-1
1697.03 cm -1 1716 cm -1 1748.26 cm
1300.51 cm -1
1300.69 cm
-1
1166.85 cm -1
1166.54 cm
-1
Bulk region
Surface region
Fig. 7. IR spectrum recorded on the surface and bulk region of UHMWPE samples.
Fig. 8. A plot of heat flow versus temperature for UHMWPE tibial insert.
Fig. 9. The distribution of molecular weight of UHMWPE tibial insert after being implated for 10 years.
1422
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
The gamma sterilization method has significant effect on delamination of the polyethylene surface. Severe delamination was found on the retrieved polyethylene surface with gamma radiation in air and in nitrogen [8,23]. No delamination was found on the gas plasma sterilized polyethylene insert [4]. The degradation does not occur immediately after sterilization but over time. It has been reported that degradation occurs over a period prior to implantation (shelf ageing) and in vivo services [21,24]. Sterilization by gamma irradiation in air has been shown to have the potential to accelerate the oxidation of polyethylene components [25]. It is suggested that gamma sterilization have played a role in the degradation of the retrieved insert. A correlation between implantation time and presence of delamination was found by Collier et al. [8]. Retrieved samples with short implantation duration, less than 4 years, displayed signs of delamination or cracking on 17–27% of cases. In contrast, 65% of the retrieved tibial components displayed a signs of delamination or cracking for durations greater than 4 years. Apart from delamination, pitting, scratching and folding were also observed in the present sample. Garcia et al. [5] have found that pitting and scratching were the most predominant damage modes on the surface damage of 40 rotating-platform TKAs. Third body particles have been reported by Crowninshield et al. [7] and Muratoglu et al. [26] as a main cause of formation multiple surfaces folding, pitting and scratching. Crowninshield et al. [7] found that the third body particles were embedded in and dislodged from the polyethylene surface, causing the formation of pitting during the motion and loading at knee prosthesis. In addition, scratches were formed when third body particles were pushed along the articular surface and leaving tracks of deformation within the polyethylene surface. Hirakawa et al. [27] suggested that small debris particles attributed to the formation of scratches and larger debris particles which produce by polyethylene fatigue attributed to the delamination formation. These findings clearly showed the formation of surface damage modes were related to each other. The absence of surface burnishing on the UHMWPE surface distinguishes this work from the previous reports. The absence of burnishing wear can be caused by the design of knee implant. Engh et al. [4] have found that the knee implant with fixed bearing system have small percentage of the surface being burnished when compared with the mobile bearing system due to small contact area. In this present study, the ApolloÒ Total Knee system which is fixed bearing system did not exhibit burnishing on the polyethylene surface. In the present study, the crystallinity and molecular weight measurement of the UHMWPE tibial insert were carried out in order to determine whether there is a change after the insert was implanted for 10 years. Kurtz [28] suggested that the degree of crystallinity for medical grade polyethylene should be within 39–75%. Thus in this study, the crystallinity falls within the suggested range. Since the crystallinity data of UHMWPE tibial insert prior to implantation was also not available, no comment can be made on the role of crystallinity on the performance of the present implant. However, it is expected that there is a change in the molecular weight of UHMWPE after implanted for 10 years. In the present case, the molecular weight of UHMWPE was measured as 130, 104 g/mol. This result is approximately ten times lower than the actual average molecular weight of a medical grade polyethylene which is 1.5 106 g/mol [28]. Diabb [11] also reported on the reduction of molecular weight after retrieval. They found that the molecular weight of the polyethylene tibial insert just after 7 months was 166,855 g/mol. The degradation in molecular weight of retrieved UHMWPE tibial insert in the present case suggest that polyethylene properties have changed. This is believed to have contributed to the wear damage. 5. Conclusions The following conclusions can be drawn from the present study. High-grade wear and oxidation degradation are associated with the failure of the 10 years implanted UHMWPE tibial insert. The most prominent wear damage mode identified in the retrieved tibial insert is delamination. This is followed by pitting, folding and scratching. The pit depth as large as 60 lm was measured on the surface. Third body particles also occur on the insert surface. The retrieved UHMWPE tibial insert showed low molecular weight, indicating that polyethylene properties have changed in vivo which contributed to wear damage.
Acknowledgements The authors would like to thank University Malaya Medical Centre (UMMC) for supplying the sample. This research was supported by Postgraduate Research Grant (PPP), Grant No: PS079-2009B from University of Malaya. References [1] Santavirta S, Konttinen YT, Lappalainen R, Anttila A, Goodman SB, Lind M, et al. Materials in total joint replacement. Curr Orthop 1998;12:51–7. [2] O’Rourke MR, Callaghan JJ, Goetz DD, Sullivan PM, Johnston RC. Osteolysis associated with a cemented modular posteriorcruciate-substituting total knee design: five to eight years follow up. J Bone Joint Surg Am 2002;84:1362–71.
S. Liza et al. / Engineering Failure Analysis 18 (2011) 1415–1423
1423
[3] Whaley AL, Trousdale RT, Rand JA, Hanssen AD. Cemented long-stem revision total knee arthroplasty. J Arthroplast 2003;18:592–9. [4] Engh GA, Zimmerman RL, Parks NL, Engh CA. Analysis of wear in retrieved mobile and fixed bearing knee inserts. J Arthroplast 2009;24(6):28–32. [5] Garcia RM, Kraay MJ, Messerschmitt PJ, Golberg VM, Rimnac CM. Analysis of retrieved ultra high molecular polyethylene tibial components from rotating-platform total knee arthroplasty. J Arthroplast 2009;24:131–8. [6] Hood RW, Wright TM, Burstein AH. Retrieval analysis of total knee protheses: a method and its application to 48 total condylar protheses. J Biomed Mater Res 1983;17:829–42. [7] Crowninshield RD, Wimmer MA, Jacobs JJ, Rosenberg AG. Clinical performance of contemporary tibial polyethylene components. J Arthroplast 2008;21(5):754–61. [8] Collier JP, Sperling DK, Currier JH, Sutula LC, Saum KA, Mayor MB. Impact of gamma sterilization on clinical performance of polyethylene in the knee. J Arthroplast 1996;11(4):377–89. [9] William IR, Mayor MB, Collier JP. The impact of sterilization method on wear in knee arthroplasty. Clin Orthop 1998;356:170–80. [10] Ashraf T, Newman JH, Desai VV, Beard D, Nevelos JE. Polyethylene wear in a non congruous unicompartmental knee replacement: a retrieval analysis. Knee 2004;11:177–81. [11] Diabb J, Juarez-Hernandez A, Reyes A, Gonzalez-Rivera C, Hernandez-Rodriguez MAL. Failure analysis for degradation of a polyethylene knee prosthesis component. Eng Fail Anal 2009;16:1770–3. [12] Lopéz JM, Gómez-Barrenata E. Estructura y morfología de las regiones degradadas del polietileno de ultra alto peso molecular en prótesis articulares. Biomecánica 1999;7(13):2–11. [13] Slouf M, Sloufora I, Horak Z, Stepanek P, Entlicher G, Krejcik M, et al. New fast method for determination of number of UHMWPE wear particles. J Mater Sci Mater Med 2004;15:1267–78. [14] Costa L, Luda MP, Trossarelli L, Brach del Prever EM, Crova M, Gallinaro P. Oxidation in orthopaedic UHMWPE sterilized by gamma-radiation and ethylene oxide. Biomaterials 1998;19:659–68. [15] Goldman M, Lee M, Gronsky R, Pruitt L. Oxidation of ultrahigh molecular weight polyethylene characterized by Fourier transformation infrared spectrometry. J Biomed Mater Res 1997;37(1):43–50. [16] Ho FY, Ma HMJJ, Liau, Yeh CR, Huang CH. Mobile-bearing knees reduce rotational asymmetric wear. Clin Orthop Relat Res 2007;462:143–9. [17] Bradford L, Baker DA, Graham J, Chawan A, Ries MD, Pruitt LA. Wear and surface cracking in early retrieved highly cross-linked polyethylene acetabular liners. J Bone Joint Surg Am 2004;86:1271–82. [18] Medel FJ, Kurtz SM, Parvizi J, Klein GR, Kraay MJ, Rimnac CM. In vivo oxidation contributes to delamination but not pitting in polyethylene components for total knee arthroplasty. J Arthroplast in press. [19] Blunn GW, Joshi AB, Lilley PA, Engelbrecht E, Ryd L, Lidgren L, et al. Polyethylene wear in unicondylar knee prostheses. 106 Retrieved Marmor, PCA, and St. Georg tibial components compared. Acta Orthop Scand 1992;63(3):247–55. [20] Burnett RSJ, Biggerstaff S, Currier BH, Collier JP, Barrack RL. Unilateral tibial polyethylene liner failure in bilateral total knee arthroplasty–bilateral retrieval analysis at 8 years. J Arthroplast 2007;22(5):753–8. [21] Kurtz SM, Hozack W, Marcolongo M, Turner J, Rimnac C, Edidin A. Degradation of mechanical properties of UHMWPE acetabular liners following longterm implantation. J Arthroplast 2003;18(1):68–78. [22] Wannomae KK, Christensen SD, Freiberg AA, Bhattacharyya S, Harris WH, Muratoglu OK. The effect of real-time aging on the oxidation and wear of highly cross-linked UHMWPE acetabular liner. Biomaterials 2006;27:1980–7. [23] Brandt JM, Medley JB, MacDonald SJ, Bourne RB. Delamination wear on two retrieved polyethylene inserts after gamma sterilization in nitrogen. Knee 2011;18(2):125–9. [24] H. Oonishi, S.C. Kim, M. Kyomoto, M. Iwamoto, M. Ueno, Comparison of in-vivo wear between polyethylene inserts articulating against ceramic and cobalt–chrome femoral components in total knee prostheses, Bioceramics and Alternative Bearings in Joint Arthroplasty 12th BIOLOXÒ Symposium Seoul, Republic of Korea, 2007, pp. 149–159. [25] Costa L, Jacobson K, Bracco P, Brach del Prever EM. Oxidation of orthopaedic UHMWPE. Biomaterials 2002;23:1613–24. [26] Muratoglu OK, Ruberti J, Melotti S, Spiegelberg SH, Greenbaum ES, Harris WH. Optical analysis of surface changes on early retrievals of highly crosslinked and conventional polyethylene tibial inserts. J Arthroplast 2003;18(7):42–7. [27] Hirakawa K, Bauer TW, Yamaguchi M, Stulberg BN, Wilde AH. Relationship between wear debris particles and polyethylene surface damage in primary total knee arthroplasty. J Arthroplast 1999;14(2):165–71. [28] Kurtz S. The UHMWPE Handbook: Ultra-High Molecular Weight Polyethylene in Total Joint Replacement. first ed. New York: Elsevier Inc.; 2004.