ISSN 10683666, Journal of Friction and Wear, 2013, Vol. 34, No. 1, pp. 32–37. © Allerton Press, Inc., 2013. Original Russian Text © L. Lapaj, J. Markuszewski, T. Rybak, A.A. Britsko, V.V. Anosov, M. WieruszKozlowska, 2013, published in Trenie i Iznos, 2013, Vol. 34, No. 1, pp. 43–49.

Wear Analysis of a Ceramic on Ceramic Hip Endoprosthesis L. Lapaja, J. Markuszewskia, T. Rybakb, A. A. Britskoc, *, V. V. Anosovc, and M. WieruszKozlowskaa a

Department of General Orthopaedics, Musculoskeletal Oncology and Trauma Surgery, Poznan University of Medical Sciences, ul. 28 czerwca 1956 nr 135/147, Poznan, 61501 Poland bMetal Forming Institute, ul. Jana Pawla II 14,6, Poznan, 61139 Poland c Department of Orthopaedics, Grodno State Medical University, ul. Gorkogo 80, Grodno, 230009 Belarus *email: [email protected], [email protected] Received September 4, 2012

Abstract—The authors have analyzed the wear of a ceramiconceramic hip joint prosthesis retrieved after 25 years of use. Severe wear and flattening of the femoral head was revealed, while the acetabular component was less worn. Three wear zones can be seen on the femoral head depending on surface topography. The wear patterns of the ceramic components of the prosthetic head indicate the role of microseparation that occurs during implant functioning. Keywords: endoprosthesis, hip joint, wear of ceramics, aseptic loosening DOI: 10.3103/S106836661301008X

cally compatible, undergoes breaking to a lesser degree and its wear debris are biologically inert [8]. First ceramic endoprostheses were brittle, expensive, and the results were unsatisfactory because of the absence of the means for biological fixation (osteointegration in endoprosthesis material) to provide the longterm sta bility of the implant [9]. Improved implants that came into use later had better clinical results, and are now considered as a golden standard in arthroplasty; how ever, data on their longterm operation in vivo, i.e., in the patient’s organism, are limited [10]. The aim of the work is to analyze the wear results of the ceramic–ceramic friction pair of hip endoprosthesis after 25 years of operation in the patient’s organism.

INTRODUCTION Total hip arthroplasty is a surgery performed in patients who suffer from coxarthrosis of degrees 2 and 3 and have severe pain syndrome. During surgical inter vention, the damaged joint surfaces of acetabulum and head are replaced by artificial elements, including ace tabular and femoral components, the latter of which consist of a stem and head. Due to the patient’s daily activities, the endopros thesis undergoes substantial axial loading throughout its entire service life [1]. The degree and rate of wear of the friction pair components in the implant due to constant head movement in the insert of artificial ace tabulum is the most intrinsic characteristic of the arti ficial joint, especially in young patients who lead an active lifestyle. Clinical observations of the people after arthroplasty with the application of metal–poly ethylene friction pairs revealed a high degree of poly ethylene wear. Moreover, microparticles that result from polyethylene wear induced osteolysis around the implant [2]. The problem of aseptic loosening is topi cal in modern orthopedics, despite the improvement of materials [3, 4]. That is why wearresistant friction pairs, such as metal–metal and ceramic–ceramic are preferred. Recent studies have reported the more than 20year survivability of metal–metal friction pairs; however, some other complications occur due to the release of metal ions during their operation [5]. The development and introduction of a “perfect” endoprosthesis is still an open problem [6]. Endoprosthetics of hip joints using a ceramics— ceramics friction pair based on alumina (Al2O3) was first proposed in the 1970s [7]. Alumina ceramics is biologi

MATERIALS AND METHODS A 58yearold man came to the Department of General Ortophaedics, Musculoskeletal Oncology and Trauma Surgery, Poznan University of Medical Sciences (Poland) with complaints of pain in the inguinal region from the right. 25 years ago, his right hip joint had been totally replaced due to avascular necrosis of the femoral head. The implant was the uncemented Mittelmeier (Autophor) endoprosthesis (Osteo AG, Switrzeland), the stem of which was made of CoCr alloy and the attached skirted ceramic femo ral head was 36 mm in diameter. The acetabular com ponent was a conical cap fixed in the pelvis by thread ing. Over 24.6 years, the patient did not consult doc tors until pain appeared. Clinical and Xray examination revealed the aseptic loosening of the endoprosthesis and a revision operation was proposed to the patient (Fig. 1). 32

WEAR ANALYSIS OF A CERAMIC ON CERAMIC HIP ENDOPROSTHESIS

During surgical intervention, periprosthetic tis sues were removed and their samples were examined histologically. Rehabilitation was running satisfacto rily and 6 months after the operation, the signs of the osteointegration of the revision endoprosthesis were seen on the Xray. A microscopic examination of the periprosthetic tissue did not reveal any pathology and, according to the Morawietz scale, the tissue was classified as type 1 based on the aseptic loosening cri terion (Fig. 2) [11]. Afterwards, the components of the Mittelmeier endoprosthesis were sent to the Metal Forming Insti tute (Poznan, Poland) for further examination. They were studied using optical microscopy, scanning elec tron microscopy (SEM), and energy dispersive Xray spectroscopy (EDX). A 3D coordinate measuring machine was used to assess the shape of the endopros thetic head and a profilometer was used for friction surface topography.

33

Fig. 1. Preoperation Xray of patient’s hip joint in direct projection with signs of aseptic loosening of the endopros thesis components.

RESULTS OF IMPLANT ANALYSIS It was visually observed that the stem of the retrieved endoprosthesis was polished in the place of contact with the head and was mildly worn, whereas the remaining part showed no wear (Fig. 3). The examination of the ceramic head of the endoprosthesis has shown flattening of its upper pole and dark scratches made by the surgical tool during implant retrieval. The heterogeneous nature of the head fric tion surface should be noted; its peripheral part is the same as the neck and had a structure similar to that of a new article (zone I), while the center of the friction surface in the head pole region appeared to be flat tened (zone II). Tuberous spots appeared on the fric tion surface along the circumference (zone III); the same spots were arranged as rings in the region of the head equator (Fig. 4a). The head skirt did not have any signs of wear. The acetabular component of the endoprosthesis was partially damaged during retrieval: it had two small edge chips and was broken in its main part. Several black marks made by surgical tools during extraction were seen. We did not reveal macroscopic wear scars in the loadbearing cap region; however, the heterogene ity of its structure was noted similarly to that on the head (Fig. 4b). Since the endoprosthetics head was flattened, the wear of both components was measured by the DEA Global Image Clima 7.7.5 tool (Hexagon Metrology, Italy) with an accuracy of up to 1.5 μm. Severe wear occurred in the zone during body loading (Fig. 5), as well as in the polar zone, which resulted in the ellipsoid deformation of the head (Fig. 6). According to calcula tions, the average linear wear rate was about 0.055 mm/year. The wear of the acetabular component was insignificant and did not exceed 200 μm in the loaded zone. JOURNAL OF FRICTION AND WEAR

Vol. 34

No. 1

Fig. 2. Micropreparation of periprosthetic tissues. Black inclusions of ceramic wear paticles in the tissue are seen to the left (×80).

Fig. 3. General view of retrieved endoprosthesis.

Scanning electron microscopy was used to study the head structure in the zones determined above. Zone I (Fig. 7a) has a smooth surface with hollows somewhere that correspond to the granules of the ceramic material. Parallel grooves that arise from head polishing were revealed in zone II (Fig. 7b). The tex 2013

34

LAPAJ et al. (a)

(b)

Zone I Zone II Zone III

Fig. 4. Axial view of implant wear zones: (a) head; (b) acetabular component.

340°350° 330° 320° 310° 300°

0°

10°20°

30° 40° 50° 60° 70°

290°

6 mm 6a 9 mm

280°

80°

270°

90°

260°

100° 110°

250° 240° 230° 220° 210° 200°

3 mm 3a

190° 180° 170°

120° 130° 140° 150° 160°

9a 12 mm 12a 15 mm 15a 18 mm 18a

Fig. 5. Loss of head spherical shape of endoprosthesis axial view. Solid lines correspond to the perimeters of head section by par allel planes with 3 mm between them. Dashed lines designate the section perimeters with the same planes of a spherical head.

mm 18 2 16 14 1 12 10 8 6 4 2 0 –18 –14 –10 –6 –2 2 6 10 14 mm –16 –12 –8 –4 0 4 8 12 16 Fig. 6. Loss of spherical shape by the endoprosthesis head, side view. (1) surface contour of retrieved head; (2) sphere 36 mm in diameter.

ture of zone III is heterogeneous with hollows that evolved due to grains of material falling out during wear, and contributed to surface flattening. The same microscopic zones were revealed in the acetabular component. Energy dispersive Xray spectroscopy of both com ponents has shown them to consist of alumina ceram ics, while surface scratches contain mainly iron parti cles (Fe). The absence of Cr, Co, and Mo indicates the absence of metallic fragments of the stem in the fric tion zone. The profilogram of zone I consists of peaks and depressions (Fig. 8) that correspond to the surface changes as a result of polishing (Rt = 2.960). Zone II has a smooth flattened surface (Rt = 2.226) with an insignificant number of hollows that correspond to grain boundaries in the material (Fig. 8). The highest average roughness was determined in zone III (Rt = 3.548) with the largest number of peaks and depres

JOURNAL OF FRICTION AND WEAR

Vol. 34

No. 1

2013

WEAR ANALYSIS OF A CERAMIC ON CERAMIC HIP ENDOPROSTHESIS

(a)

30 μm

(c)

30 μm

35

50 μm

(b)

Fig. 7. Scanning electron microscopy of the head of removed implant: (a) zone I, (b) zone II, (c) zone III.

sions on the profilograms that corresponded to the protruding and fallen out grains of the ceramic mate rial (Fig. 8). The surface topography of the respective zone of the acetabular component was identical to the endoprosthetic head with the same Rt. DISCUSSION The study results indicate that a ceramic–ceramic friction pair is characterized by mild wear and wear parti cles are biologically inert [1, 8]. Despite the good biolog ical properties of the Mittelemeier endoprosthesis tribo joint, frequent cases of its aseptic loosening are reported, however some papers, albeit limited in number, involve the results of its longterm observation [12]. In our opin ion, the present work represents a unique case of the longterm operation of this implant. The high frequency of the loosening of the Mittelmeier endoprosthesis is apparently due to the fact that its design ignored the prin ciple of biological fixation [8]. Smooth contact surfaces of the stem and the threaded cup ensure primary stability, JOURNAL OF FRICTION AND WEAR

Vol. 34

No. 1

but not osteointegration. Therefore, when the patient begins loading the limb, micromovements between the bone and the endoprosthesis surface arise, which lead to osteolysis, loosening, and implant migration. This agrees with clinical observations of patients with implanted modified Autophor 900S endoprostheses. The modified stem was partially coated with metal beads to provide osteointegration and clinical observations showed a longer life of this endoprosthesis [9]. Findings from pre vious studies and experimental results indicate that short comings in endoprosthesis design, rather than its wear, are responsible for aseptic loosening. The wear of the endoprosthesis head in the ceramic–ceramic friction pair has been previously described in work [13]. The authors believe that the wear of the head and the cap of the Mittelmeier endoprosthesis is caused by the low quality of the fric tion materials and unstable initial position of the ace tabular component; however, the latter fact was not supported by the observations made by other authors. 2013

36

LAPAJ et al. R–Profil wypoziomowany Filtr ISO 11562(M1) Lc = 0.800 mm 5.0 [μm] 0

–5.0 Koncówka pomirowa TKU300 Lt = 4.80 mm Vt = 0.50 mm/s R–Profil wypoziomowany Filtr ISO 11562(M1) Lc = 0.800 mm 2.0

4.8

[μm] 0

–2.0 Koncówka pomirowa TKU300 Lt = 4.80 mm Vt = 0.50 mm/s R–Profil wypoziomowany Filtr ISO 11562(M1) Lc = 0.800 mm 2.5

4.8

[μm] 0

–2.5 Koncówka pomirowa TKU300 Lt = 4.80 mm Vt = 0.50 mm/s

4.8

Fig. 8. Surface profilograms of zones I, II, and III of the endoprosthesis head of the hip joint removed after 25 years of functioning.

The three zones of the modified endoprosthesis friction surface identified by SEM were reported else where [8, 14, 15]. Shishido et al. introduced a wear classification that consisted of five degrees depending on SEM data [1, 12]. Nevelos et al. distinguished three types of wear [13]. The authors of [1, 12, 14, 16] believe that regions of wear with a smooth surface are obtained by polishing. Apparently, this mechanism also occurred in the investigated implant. On the con trary, the surface texture in zone III points to microc racks on the friction surface and falling out separate particles of the material. Similar effects were noted in the areas of the least implant wear degree [1, 12–14]. Moreover, Zeng et al., using microscopy by focused ion beam revealed microcracks beneath the friction surface, which can be responsible for particle separa tion as described by our profilometry results [14]. Clinical and simulation studies have made it possi ble to attribute the increase in the endoprosthesis wear of the ceramic–ceramic friction pair to microsepara tion of friction surfaces during walking [15, 17]. It has been found that at each step, when the axial load to the implant is minimal, the endoprosthetic head is ex shifted, especially in patients with increased laxity of the connective tissue. As a result, in the takeoff phase, the hip head takes an incorrect position and tangential load leads to the increased local wear of the at foot strike phase [15]. Such behavior of the implant is

called microseparation and results in specific wear scars which were found by many authors in the region of the head equator and the edge of acetabular compo nent [1, 8, 12, 15, 16]. Longterm microseparation (microchipping) leads to increasing friction, which causes a local increase in peak loads, microcracks to appear on the friction surface, and ceramic material particles to fall out. Free particles are involved in pol ishing the friction surface [14]. Our study has revealed a characteristic wear zone on the friction surface shaped as a strip, which con firms the role of microseparation in its formation. The flat part of the head also had signs of stress fractures of ceramic material grains. This cannot be attributed to wear and microseparation, as well wear caused by chipped particles of the material in third body mecha nism. We assume that the loss of the spherical shape due to wear is aggravated by walking kinematics. This results in increased contact stress in the weight bearing part of the head, which encourages the separation of material grains and increased wear of the friction pair. The shape of the rough worn surface of the head cor responds to the load distribution during a standard step cycle, which reinforces our hypothesis. Further studies by Grushko et al. indicate the increased degree of wear with increasing gap between head and cap, which, as our studies have shown, also

JOURNAL OF FRICTION AND WEAR

Vol. 34

No. 1

2013

WEAR ANALYSIS OF A CERAMIC ON CERAMIC HIP ENDOPROSTHESIS

contributes to increasing local axial pressure and leads to the accelerated damage of the friction surface [18]. CONCLUSIONS An analysis of the implant retrieved during a revi sion operation has shown a severe wear of the ceramic hip endoprosthesis after its 25year use. Despite the complex investigation, we were unable to explain the mechanisms responsible for such high wear rates of the ceramic–ceramic friction pair. This indicates the need for further investigations in this field in order to fully understand the tribological behavior in vivo of ortho pedic implants. ACKNOWLEDGMENTS The study was supported by the Polish National Cen ter for Research and Development, grant no. 030081 10/2011. REFERENCES 1. Shishido, T., Clarke, I.C., Williams, P., et al., Clinical and Simulator Wear Study of Alumina Ceramic THR to 17 Years and beyond, J. Biomed. Mater. Res. Part B: Appl. Biomater., 2003, vol. 67, pp. 638–647. 2. Atkins, G.J., Haynes, D.R., Howie, D.W., and Findlay, S.M., Role of Polyethylene Particles in Peri Prosthetic Osteolysis: A Review, World J. Orthop., 2011, vol. 2, no. 10, pp. 93–101. 3. Wendland, J., GierzynskaDolna, M., Rybak, T., et al., Investigation for a New Biomaterial for the Hip Endoprostheses Elements, Obr. Plast. Met., 2009, vol. 20, no. 2, pp. 3–19. 4. Johanson, P.E., Digas, G., Herberts, P., et al., Highly Crosslinked Polyethylene Does not Reduce Aseptic Loosening in Cemented THA 10Year Findings of a Randomized Study, Clin. Orthop. Relat. Res., 2012, vol. 470, no. 11, pp. 3083–3093. 5. Saito, S., Ishii, T., Mori, S., et al., LongTerm Results of Metasul MetalonMetal Total Hip Arthroplasty, Orthopedics, 2010, vol. 33, no. 8. 6. Bernthal, N.M., Celestre, P.C., Stavrakis, A.I., et al. Disappointing ShortTerm Results with the DePuy ASR XL MetalonMetal Total Hip Arthroplasty, J. Arthroplasty, 2012, vol. 27, pp. 539–544.

JOURNAL OF FRICTION AND WEAR

Vol. 34

No. 1

37

7. Stuchin, S.A. and Frankel, V.H., Early Experience with the Autophor Prosthesis Using a Posterior Approach, Yale J. Biol. Med., 1984, vol. 57, pp. 843–850. 8. Magnissalis, E.A., Xenakis, T.A., and Zacharis, C., Wear of Retrieved Ceramic THA Components⎯Four Matched Pairs Retrieved after 5–13 Years in Service, J. Biomed. Mater. Res., 2001, vol. 58, pp. 593–598. 9. Yoon, T.R., Rowe, S.M., Kim, M.S., and Cho, S.G., Fifteen to 20Year Results of Uncemented Tapered Fully PorousCoated CobaltChrome Stems, Int. Orthop., 2008, vol. 32, pp. 317–323. 10. Aldrian, S., Nau, T., Gillesberger, F., Petras, N., and Ehall, R., MediumTerm Analysis of Modern Ceramic onCeramic Bearing in THA, Hip Int., 2009, vol. 19, pp. 36–40. 11. Morawietz, L., Classen, R.A., Schröder, J.H., et al., Proposal for a Histopathological Consensus Classifica tion of the Periprosthetic Interface Membrane, J. Clin. Pathol., 2006, vol. 59, pp. 591–597. 12. Shishido, T., Yamamoto, K., Tanaka, S., et al., A Study for a Retrieved Implant of CeramiconCeramic Total Hip Arthroplasty, J. Arthroplasty, 2006, vol. 21, pp. 294–298. 13. Nevelos, J.E., Prudhommeaux, F., Hamadouche, M., et al., Comparative Analysis of Two Different Types of Alumina–Alumina Hip Prosthesis Retrieved for Asep tic Loosening, J. Bone Joint Surg. Br., 2001, vol. 83, pp. 598–603. 14. Zeng, P., Rainforth, W.M., Inkson, B.J., and Stewart, T.D., Characterization of Worn Alumina Hip Replacement Prostheses, J. Biomed. Mater. Res. Part B: Appl. Biomater., 2012, vol. 100, pp. 121–132. 15. AlHajjar, M., Leslie, I.J., Tipper, J., et al., Effect of Cup Inclination Angle during Microseparation and Rim Loading on the Wear of BIOLOX(R) Delta CeramiconCeramic Total Hip Replacement, J. Biomed. Mater. Res. Part B: Appl. Biomater., 2010, vol. 95, pp. 263–268. 16. Tateiwa, T., Clarke, J.C., Pezzotti, G., et al., Surface MicroAnalyses of LongTerm Worn Retrieved ‘Osteal’ Alumina Ceramic Total Hip Replacement, J. Biomed. Mater. Res. Part B: Appl. Biomater., 2007, vol. 83, pp. 562–570. 17. Stewart, T., Tipper, J., Streicher, R., et al., LongTerm Wear of HIPed Alumina on Alumina Bearings for THR under Microseparation Conditions, J. Mater. Sci. Mater. Med., 2001, no. 12, pp. 1053–1056. 18. Grushko, A.V., Sheykin, S.E., and Rostotskiy, I.Y., Contact Pressure in Hip Endoprostethic Swivel Joints, J. Friction Wear, 2012, vol. 33, pp. 124–129.

2013