Ce-TZP/Al2O3 Nanocomposite as a Bearing Material inTotal Joint Replacement

Kenji Tanaka,1 Jiro Tamura,1 Keiichi Kawanabe,1 Masahiro Nawa,2 Masanori Oka,3 Masaki Uchida,4

Tadashi Kokubo,4 Takashi Nakamura1

1 Department of Orthopedic Surgery, Faculty of Medicine, Kyoto University, Kawahara-cho 54, Shogoin, Sakyo-ku,Kyoto 606-8507, Japan

2 Advanced Technology Research Laboratory, Matsushita Electric Works, Ltd., 1048, Kadoma, Osaka 571-8686, Japan

3 Institute for Frontier Medical Sciences, Kyoto University, Kawahara-cho 53, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan

4 Department of Material Chemistry, Faculty of Engineering, Kyoto University, Yoshida-Honmachi, Sakyo-ku,Kyoto 606-8501, Japan

Received 15 June 2001; revised 13 November 2001; accepted 20 November 2001Published online 00 Month 2002 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.10182

Abstract: The objectives of this study were to investigate the biocompatibility, phase stabil-ity, and wear properties of a newly developed Ce-TZP/Al2O3 nanocomposite, as compared toconventional ceramics, and to determine whether the new composite could be used as abearing material in total joint prostheses. In tests of mechanical properties, this compositeshowed significantly higher toughness than conventional Y-TZP. For biocompatibility tests,cylindrical specimens of both the Ce-TZP/Al2O3 nanocomposite and monolithic alumina wereimplanted into the paraspinal muscles of male Wistar rats. The tissue reactions were almostthe same, and at 24 weeks after implantation, thin fibrous capsules with almost no inflam-mation were observed around both of them. There were no significant differences in mem-brane thickness between the two ceramics. After hydrothermal treatment in 121 °C vapor for18 h, the new composite showed complete resistance to aging degradation, whereas Y-TZPshowed a phase transformation of 25.3 vol% (initial 0.4%) to the monoclinic form. Accordingto the results of pin-on-disk tests, the wear rates of Ce-TZP/Al2O3 nanocomposite and aluminawere 0.55 � 0.04 � 10�7 and 2.12 � 0.37 � 10�7mm3/Nm, respectively. The results of thisstudy suggest that the Ce-TZP/Al2O3 nanocomposite is a promising alternative ceramiccomponent for total joint replacement. © 2002 Wiley Periodicals, Inc. J Biomed Mater Res (ApplBiomater) 63: 262–270, 2002

Keywords: zirconia/alumina composite; total joint prosthesis; mechanical property; biocom-patibility; wear property

INTRODUCTION

Today total joint replacement (TJR) is one of the most suc-cessful options for many patients suffering from end-stagejoint disease. In particular, total hip arthroplasty (THA) andtotal knee arthroplasty (TKA) have allowed those patientswith disabling hip or knee conditions to live independentlywithout pain. But the service life of artificial joints is limited

and it is not unusual for young patients to outlive a total jointimplant.

Aseptic loosening of components associated with osteol-ysis is a very common complication in TJR. Periprostheticosteolysis is thought to be induced chiefly by polyethylene(PE) wear particles.1,2 Since the successful introduction ofmetal-on-PE THA by Charnley, this material combinationhas been used widely as a bearing couple in a variety of totaljoint prostheses.3 However, it has become obvious that, withtime, this combination produces a large number of wearparticles, which might contribute to the inflammatory re-sponse and the resulting osteolysis.4 Therefore, alternativebearing combinations such as metal-on-metal, ceramic-on-ceramic, and ceramic-on-PE have been developed in attemptsto reduce wear debris.

Although these alternative bearing combinations have

Correspondence to: Takashi Nakamura, Department of Orthopedic Surgery, Fac-ulty of Medicine, Kyoto University, Kawahara-cho 54, Shogoin, Sakyo-ku, Kyoto606-8507, Japan (e-mail: ntaka@kuhp.kyoto-u.ac.jp)

Contract grant sponsor: Ministry of Education, Science, Sports, and Culture, Japan;contract grant number: 13680933

No benefit of any kind will be received either directly or indirectly by the author.

© 2002 Wiley Periodicals, Inc.

262

been confirmed to produce less wear in vitro5–7 and invivo8–10 than the conventional metal-on-polyethylene com-bination, they have not yet been widely used. With metal-on-metal combination, for example, concerns remain about thelong-term effects (local or systemic) of metal ions.11 On theother hand, alumina and yttria-stabilized tetragonal zirconiapolycrystals (Y-TZP) have been used as bearing componentsin TJR. Alumina components have been used for about 30years,12 although there have been concerns about the risk ofcomponent fractures.13 Y-TZP has been clinically applied inarticulation with PE in TJR in the expectation that it wouldcontribute to the reduction of PE wear,14 but it is still un-known whether the zirconia-on-PE system would really pro-duce less wear in clinical use than metal- or alumina-on-PEsystems.15 In addition, the Y-TZP component might be trou-blesome after long-term in vivo implantation16 because of itstime-dependent phase transformation from tetragonal to mon-oclinic, which could cause degradation of mechanical prop-erties and deterioration of surface quality. Both alumina andY-TZP components have undergone various improvements inorder to solve these serious problems.13,15 However, the trueeffectiveness of these ceramic components in clinical use isnot well understood.

Recently some alumina/zirconia composites have beendeveloped to overcome the drawbacks of conventional ce-ramics.17,18 Although these composites might prove to beviable alternatives as ceramic biomaterials due to their excel-lent mechanical properties and high wear resistance, it re-mains uncertain whether they would help to solve the above-mentioned drawbacks.

Ceria-stabilized tetragonal zirconia polycrystals (Ce-TZP)have been shown to provide very high toughness and com-plete resistance to low-temperature aging degradation(LTAD).19 However, Ce-TZP itself has never been used as aceramic biomaterial, probably because of its lower flexuralstrength compared to Y-TZP. Nawa, Nakamoto, Sekino, and

Nihara20 have developed a Ce-TZP-based nanostructuredzirconia/alumina composite. Ce-TZP/Al2O3 nanocompositepossesses three times the toughness and the equivalentstrength of conventional Y-TZP.20 The objectives of thisstudy were to investigate the biocompatibility, phase stability,and wear property of this new material and to determinewhether it could be used as a bearing material in total jointprostheses.

MATERIALS AND METHODS

Sample Preparation

Ce-TZP/Al2O3 nanocomposites were prepared by sinteringmixed powders of Ce-TZP and Al2O3 at 1440 °C for 4 h. Thedetails of fabrication have been described elsewhere.20 Inbrief, this nanocomposite is composed of 10 mol% CeO2-stabilized TZP doped with 0.05 mol% of TiO2 as a matrixphase and 30 vol% of Al2O3 as a second phase. The signif-icant characteristic of its microstructure is an intragranulartype of nanostructure (Figure 1). This composite is com-monly sintered at 1500 °C for 2 h; however, the sinteringconditions for the test samples in this study were chosen toachieve a finer grain size, because it has been reported thatthe grain size of Y-TZP and alumina is closely related withtheir wear resistance.21 That is, decreasing the grain size islikely to enhance the wear resistance. The grain size wasdetermined by using the linear intercept method; the conver-sion factor used is 1.56. As a result, sintered samples ofCe-TZP/Al2O3 nanocomposite had an average grain size of0.59 �m, which was relatively small compared with that of0.9–1.0 �m when sintered at 1500 °C for 2 h,20 and they hada density of 5.56 g/cm3, which was almost equal to thetheoretical density.

Figure 1. Scanning electron micrographs of Ce-TZP/Al2O3 nanocomposite after thermal etching. (a)�10,000; (b) �40,000. Black grain is alumina and white grain is zirconia. Arrow indicates a nano-sizegrain.

263Ce-TZP/Al2O3 NANOCOMPOSITE AS A BEARING MATERIAL

For reference ceramics, alumina and Y-TZP ceramicswere supplied by Kobe Steel Co. (Kobe, Japan). TZP ceram-ics stabilized with 3 mol% Y2O3 were sintered in air at 1475°C for 2 h. Alumina ceramics were sintered in air at 1500 °Cfor 2 h and underwent hot isostatic pressing. The averagegrain size and density of Y-TZP ceramics were 0.29 �m and6.02 g/cm3, respectively. Those of alumina ceramics were2.15 �m and 3.91 g/cm3.

Mechanical Property

The sintered specimens were cut by a diamond blade saw,and ground with a 600-grit diamond wheel. The rectangularbar specimens, having dimensions of 3 � 4 � 40 mm, weresubjected to the mechanical property tests (n � 6 for eachceramic and in each test). Elastic modulus was determined bythe resonance vibration method. Flexural strength was mea-sured by a three-point bending test at room temperature. Thetensile surfaces of the specimens were polished with a dia-mond liquid suspension. The span length and the cross-headspeed were 30 mm and 0.5 mm/min, respectively. The frac-ture toughness was estimated by the indentation-fracturemethod with the use of the following equation of Marshalland Evans22:

KIC � 0.036E0.4P0.6a�0.7(c/a)�1.5, (1)

where E is the elastic modulus, P is the applied load, a is thehalf length of the Vickers impression, and c is the half lengthof the median crack. Polished surfaces of about 0.01 �mRawere used for the Vickers indentation, with a load of 296 Nfor both alumina and Y-TZP and 490 N for the Ce-TZP/Al2O3. Loading duration was 15 s. To obtain measurablecracks the Ce-TZP/Al2O3 specimens were subjected to largerloads.20

All data were statistically analyzed with the use of aone-way ANOVA with Fisher’s PLSD method as a post hoctest. Differences with p � .05 were considered to be statis-tically significant.

Phase Stability

Plate-shaped specimens of the Ce-TZP/ Al2O3 nanocompos-ite and Y-TZP were prepared by cutting the sintered pellets.One side of each specimen was polished to a mirrorlikesurface finish. Hydrothermal treatments were carried out inan autoclave at 121°C and 0.15 MPa for 18 h. Phase trans-formation caused by the hydrothermal treatment was deter-mined by thin-film x-ray diffractometry with CuK� radiation(50 kV, 300 mA) over an angular range of 25–37°in 2�. Thevolume fraction of the monoclinic phase was determined withthe use of Toraya’s equation.23

Biocompatibility

Ce-TZP/Al2O3 nanocomposite and alumina as a control wereused in the biocompatibility test. Before the test, cylindricalspecimens of 2-mm diameter and 6-mm length were washed

and sterilized in an autoclave for 30 min. Surface roughnessof the test specimens was about 0.1 �mRa for both ceramics.Thirty-six male Wistar rats weighing about 250–300 g wereused. The animals were reared and the experiments werecarried out at the Institute of Laboratory Animals, KyotoUniversity. The Kyoto University guidelines for animal ex-periments were observed and the operations were performedunder general anesthesia (Nembutal at 40 mg/kg bodyweight). A 2-cm midline skin incision was made with a knifein the lumbar region, and the implants were placed into theparavertebral muscles using a cannulated needle. Each ratreceived one alumina specimen on the left side and oneCe-TZP/ Al2O3 specimen on the right. The incision site wasthen sutured.

Four or eight rats were sacrificed at 1, 4, 6, 12, 18, 24weeks after implantation. The implants and the surroundingmuscles were removed en bloc and fixed in 10% bufferedformalin. The tissue specimens with implants were dividedinto two groups for processing. In one group (at 1, 4, 6, 12,24 weeks, each n � 4), the implants were carefully removedfrom the soft tissue masses, and tissue specimens were rou-tinely processed and mounted in paraffin. Several 6-�m sec-tions were cut from each tissue specimen perpendicular to theimplant axis and stained with hematoxylin-eosin. At leastthree sections from each specimen were observed histologi-cally under light microscopy. In the other group (at 6, 12, 18,24 weeks, each n � 4), after fixation, the tissue specimenswith implants were dehydrated in serial concentrations ofethanol, after which they were embedded in polyester resin.Sections 400–500 �m thick were cut with a band saw (BS-3000, EXAKT cutting system, Norderstedt, Germany) per-pendicular to the implant axis, bound to transparent acrylicplates with cyanoacrylic adhesive, and ground to a thicknessof about 60 �m with the use of a grinding–sliding machine(Microgriding, MG-4000, EXAKT). Thereafter, the sectionswere stained with toluidine blue and examined by light mi-croscopy. The thickness of the encapsulating membranearound the implant was measured at six randomly chosensites per section, and three sections taken from each samplewere measured. Regarding the difference of membrane thick-ness, paired tests were made between Ce-TZP/Al2O3 andalumina (n � 4). Differences with p � .05 were considered tobe statistically significant.

Wear Properties

Ceramic-on-ceramic wear tests were performed at room tem-perature using a pin-on-disk machine, in which a stationarypin is loaded by a dead weight on a horizontal, reciprocatingplate. Frictional force was monitored by a strain gauge fixedon a leaf spring attached to the transverse bar holding thewear pin. The ceramic pins were cylindrical, 5 mm in diam-eter, and 15mm long, with truncated conical ends giving flatsurfaces of 1.5 mm in diameter. The ceramic disks were 50mm in diameter and 7 mm thick. A constant force of 20 Nwas applied, resulting in an apparent contact pressure of 11.3MPa. The surface roughness of the disk was about 0.003 �mand that of the pin was �0.01 �m.

264 TANAKA ET AL.

Ce-TZP/Al2O3-on-Ce-TZP/Al2O3 and alumina-on-alu-mina combinations (each n � 3) were tested on the pin-on-disk apparatus under distilled-water lubrication at a frequencyof 1 Hz, which resulted in an average sliding speed of 46mm/s, and the sliding distance was about 27 km (correspond-ing to 163 h, or 0.59 million cycles). Before and after thewear tests, the ceramic specimens were cleaned ultrasonicallywith distilled water and ethanol, and thereafter dried in adesiccator. The wear volume of the pin specimen was calcu-lated by the measured weight loss (AT201, Mettler Toledo)and the material density. The wear factor K (mm3/Nm) wasdetermined with the use of the following formula:

K � V/DF (2)

where V (mm3) was the measured wear volume, D (m) wasthe sliding distance, and F (N) was the applied load. The wearon the disk specimen was expressed as the wear track depth,which was measured at three different points with the use ofa profilometer (DEKTAK II, Sloan ). The surface roughnessof the worn disks was measured using the same profilometer.

Wear data were statistically analyzed with the use ofunpaired t tests and Welch’s t test. Differences with p � .05were considered to be statistically significant.

RESULTS

The mechanical properties of the Ce-TZP/ Al2O3 nanocom-posite, Y-TZP and alumina are listed in Table I. The fracturetoughness of the Ce-TZP/Al2O3 nanocomposite was muchhigher than that of either conventional Y-TZP or alumina (p�. 0001), whereas the toughness values measured by theindentation-fracture method might be overestimated becauseof the rising R-curve behavior of Ce-TZP/Al2O3.20 In addi-tion, the flexural strength of the Ce-TZP/Al2O3 nanocompos-ite was 941 � 34 MPa, nearly equal to that of Y-TZP (945 �105 MPa) (p � .93). All mechanical properties investigatedin this study were almost identical to those of the originalstudy reported by Nawa, Nakamoto, Sekino, and Niihara,20 inspite of the treatment with a lower sintering temperature toreduce the grain size.

Figure 2 shows the XRD patterns of the Ce-TZP/ Al2O3

nanocomposite and Y-TZP before and after 18-h autoclavetreatment. A remarkable increase in the monoclinic peak wasobserved in Y-TZP after the treatment, whereas the Ce-TZP/Al2O3 nanocomposite showed no change of the monoclinicphase. The relationship between the volume fraction of themonoclinic phase and aging is shown in Figure 3. Y-TZPshowed an increase in the monoclinic phase from 0.4 to 25.3vol% with time up to 18 h, whereas the Ce-TZP/ Al2O3

nanocomposite showed a slight increase ranging from 5.7 to6.8 vol%. Therefore, it was proven that the Ce-TZP/ Al2O3

nanocomposite had a higher resistance to aging degradationthan Y-TZP.

Macroscopically, there were no tumors, infections, or skinirritations in the biocompatibility test. At 1 week [Figures4(a) and 4(b)] formation of encapsulating membranes, whichwere loose and contained a large number of cells, was ob-served around both the Ce-TZP/ Al2O3 and alumina speci-mens. Some macrophages, a few lymphocytes, and manyactive fibroblasts were seen in all layers of the membrane, butalmost no neutrophils were noted. These cellular reactionsappeared to be influenced mainly by surgical procedures. At4 weeks [Figures 4(c) and 4(d)], the cell density decreasedmarkedly. Flat fibrocytes increased in the superficial andmiddle layers of the membrane, and a few lymphocytes andmacrophages were still present, chiefly in the deep layer,except for one sample in each ceramic group, where moderatemacrophage infiltration was confirmed in the middle and deep

TABLE I. Mechanical Properties of Tested Ceramics

Ce-TZP/Al2O3 Alumina Y-TZP

Density (g/cm3) 5.56 3.91 6.02Grain size (� m) 0.59 2.14 0.29Elastic modulus (GPa) 247 376 203Flexural strength (MPa)a 941 � 34 441 � 42 945 � 105Fracture toughness (MPam1/2)b 20.05 � 0.22 4.35 � 0.31 5.54 � 0.12Hardness (GPa)c 11.71 � 0.03 16.21 � 0.41 11.66 � 0.13

(Mean � S.D.)aCe-TZP/Al2O3 and Y-TZP showed higher values than alumina (p � .0001).bCe-TZP/Al2O3 showed higher values than Y-TZP and alumina (p � .0001).cAlumina showed higher values than Y-TZP and Ce-TZP/Al2O3 (p � .0001).

Figure 2. XRD patterns of (a) Ce-TZP/Al2O3 nanocomposite and (b)Y-TZP before and after 18 h of treatment (T: tetragonal, M, mono-clinic). A remarkable increase in the monoclinic peak was observed inY-TZP after the treatment.

265Ce-TZP/Al2O3 NANOCOMPOSITE AS A BEARING MATERIAL

layers. At 6 weeks [Figures 4(e) and 4(f)], a large number offlat fibrocytes were noted. Although the entire cell densityappeared slightly higher for alumina specimens than for Ce-TZP/Al2O3, the tissue response was decreased for both ce-ramic specimens. A few macrophages, very few lymphocytesand some fibroblasts were seen in the deep layer. At 12 weeks[Figures 4(g) and 4(h)], in three animals, the number ofmacrophages decreased markedly even in the deep layer, andthe fibrocyte density decreased slightly in the superficial andmiddle layers. For both types of ceramic specimens im-planted into one other animal, the cell density was relativelyhigh in all layers and mainly fibroblasts were seen, whichsuggested a delay in wound healing. At 24 weeks [Figures4(i) and 4(j)], the cell density became very low in all mem-brane layers for both ceramic specimens, and most of thecells were fibrocytes. Macrophages and lymphocytes wereseldom seen, although local infiltration of macrophages wasobserved very uncommonly. Regardless of implantation timeand ceramic type, a very small number of eosinophils orforeign body giant cells were observed only rarely. Theforeign-body giant cells were located in the deepest layer ifpresent and they did not appear to respond to tested materials.

Tissue-implant interfaces and encapsulating membranesremained almost completely intact on the polyester sectionsin contrast to the paraffin sections [Figures 5(a) and 5(b)].The membrane thickness was measured on the polyestersections rather than on the paraffin sections because someartificial damage was sometimes seen in the encapsulatingmembranes on the paraffin sections, even though the testimplants had been removed very carefully. The evolution ofthe encapsulating membrane thickness had nearly reached aplateau after 12 weeks, although the thickness increasedslightly for both ceramic types (Figure 6). There were nosignificant differences in membrane thickness between theCe-TZP/Al2O3 nanocomposite and alumina. In conclusion,the soft tissue response to the Ce-TZP/Al2O3 nanocomposite

was almost the same as that to alumina, which has been usedsuccessfully in humans. Some differences that were observedbetween the two ceramic types in the present biocompatibilitytest, especially in early-stage specimens, appeared to be re-lated with the surgical invasion but not with the ceramic type.

Wear and friction data for the Ce-TZP/ Al2O3 nanocom-posite and alumina are shown in Table II. Although theCe-TZP/ Al2O3 nanocomposite showed a higher friction co-efficient (0.27) than that of alumina (0.11), both the wear rateof the pin and the wear track depth of the disk for theCe-TZP/ Al2O3 showed much lower values compared tothose of alumina. The wear rate of the pin for the Ce-TZP/Al2O3 and alumina were 0.55 � 10�7 mm3/Nm and 2.12 �10�7 mm3/Nm, respectively (p � 0.05, Welch’s t test). Thewear-track depths of the disks were 0.46 and 4.59 �m,respectively (p � 0.001, unpaired t test). The surfaces of thewear tracks appeared smooth for both ceramic disk types.Indeed, the roughness of the wear track was almost the samevalue (0.0095 �mRa for the Ce-TZP/Al2O3 nanocompositeand 0.0079 �mRa for alumina).

DISCUSSION

It has been demonstrated that Y-TZP shows excellent wearproperties in vitro when articulated against PE,14,24 andthat it has good biocompatibility and higher mechanicalstrength than alumina.25,26 As a result, Y-TZP was intro-duced into clinical use as a femoral head in THA more than10 years ago. Although the excellent mechanical and wearproperties of Y-TZP are very attractive for TJR, the ma-terial’s drawbacks of LTAD may have a negative effectupon clinical results.

Many aging studies have been conducted on the LTAD ofY-TZP.16,27-29 In contrast to the disappointing results of invitro aging tests reported by Drummond,27 Cales, Stefani, and

Figure 3. Monoclinic fraction (vol%) of Ce-TZP/Al2O3 nanocomposite and Y-TZP as a function ofaging time.

266 TANAKA ET AL.

Lilley28 and Shimizu et al.29 concluded from the good resultsof long-term in vivo and in vitro aging tests that well-manufactured Y-TZP ceramics show almost no phase trans-formation even in long-term clinical use. Such contradictoryresults are probably because the LTAD of Y-TZP is depen-dent on its microstructure. As described by Cales, the qualityof Y-TZP has been improved continually by manufacturersand there should be few concerns about the degradationassociated with long-term implantation in the human body. Inparticular, hot isostatic pressing and/or slight alumina dopingcould provide much higher resistance to aging degradation.30

However, bearing components in TJR might actually be ex-posed to more severe conditions than other orthopedic im-plants because of frictional heating.31 Although autoclaveaging might be too extreme for simulating body environ-ments, the present results suggest that Ce-TZP/Al2O3 nano-composites would maintain much higher resistance to LTADthan experimental Y-TZP.

Another notable advantage of Ce-TZP/Al2O3 nanocom-posites is that they exhibit much higher fracture toughness.Generally, it is quite difficult to improve both toughness andstrength in oxide ceramics. Y-TZP, having higher flexuralstrength, was thought to reduce the risk of catastrophic com-ponent fracture before its clinical introduction. Indeed, Y-TZP component fractures are now very rare, but they stilloccur.15 Ce-TZP/Al2O3 nanocomposites, possessing ex-tremely high toughness, may eliminate such catastrophicfractures.

There have been numerous in vivo studies on the biocom-patibility of various biomaterials.16,25,26,32,33 Therin, Christel,and Munier investigated the soft-tissue response to metalsand ceramics quantitatively and concluded that, in the shortterm, a high fibrocyte density and few macrophages might bean indicator for biocompatibility.32 The present study foundmany fibroblasts and a decreased number of macrophagesaround both the Ce-TZP/Al2O3 nanocomposite and alumina

Figure 4. Tissue response around tested materials (hematoxylin-eosin; original magnification �400).(a) One week, Ce-TZP/Al2O3. (b) One week, alumina. (c) Four weeks, Ce-TZP/Al2O3. (d) Four weeks,alumina. (e) Six weeks, Ce-TZP/Al2O3. (f) Six weeks, alumina. (g) Twelve weeks, Ce-TZP/Al2O3. (h)Twelve weeks, alumina. (i) Twenty-four weeks, Ce-TZP/Al2O3. (j) Twenty-four weeks, alumina.

267Ce-TZP/Al2O3 NANOCOMPOSITE AS A BEARING MATERIAL

specimens at 4 week. A disadvantage of in vivo biocompat-ibility testing is that it is often influenced by multiple factorssuch as surgical procedures, implant shape, and surfaceroughness. In practice there was an attempt to minimize thesurgical invasion, but the tissue responses observed in theshort term seemed to be mainly operative artifacts. In addi-tion, as the surface roughness of the two ceramic types couldnot be controlled, that difference may have influenced theoutcome.

It is well known that some metals release elements bycorrosion and that these elements can cause adverse tissue

reactions when implanted in vivo.33 In the animal experi-ments conducted by Laing, Ferguson, and Hodge, for exam-ple, a pure nickel implant produced a severe cellular inflam-matory reaction 6 months after implantation.33 Such adversetissue reactions or reactions against the materials themselveswere not seen with any of the Ce-TZP/Al2O3 nanocompositeor alumina specimens.

In considering whether a biomaterial would be adaptableas a bearing component of TJR, wear properties are as im-portant as biocompatibility or mechanical properties; there-fore, wear properties should be investigated for any devel-

Figure 4. (continued)

Figure 5. Tissue response around tested materials (toluidine blue; original magnification �400). (a)Eighteen weeks, Ce-TZP/Al2O3. (b) Eighteen weeks, alumina. Tissue–implant interfaces and encap-sulating membranes remained almost completely intact. I � implanted ceramic; E � encapsulatingmembrane; M � muscle; bar � 50 �m.

268 TANAKA ET AL.

oped biomaterial. Burger et al. developed a new alumina/zirconia composite that combined the excellent mechanicalproperties of Y-TZP with the excellent wear properties ofalumina.17 Affatato, Goldoni, Testoni, and Toni investigatedthe wear behavior of experimental Y-TZP/Al2O3 compositesby a hip simulator using serum and reported no significantdifferences between the experimental composites and com-mercial alumina.18 The results of the ceramic-on-ceramicwear tests in this study suggest that the Ce-TZP/Al2O3 nano-composite would have higher wear resistance than alumina,at least under certain test conditions. Although Ce-TZP/Al2O3 showed lower wear rate despite relatively higher fric-tion coefficient than alumina, such a contradictory result isnot uncommon in ceramic-on-ceramic wear tests.34,35 How-ever, the wear test conducted in this study was small scaleand not physiological. The pin-on-disk test was conductedbecause it is considered convenient and suitable for firstevaluation of new ceramics. Ceramic-on-ceramic combina-tions have been occasionally examined by pin-on-disk tests,whereas ISO 6474 has recommended ring-on-disk tests. Asthe results of laboratory wear tests generally depend on testconfigurations and conditions, sufficient evaluation of wearresistance of the Ce-TZP/Al2O3 nanocomposite cannot nec-essarily be made on the basis of the present results alone. Amore practical wear test, such as with a hip simulator, will benecessary for further investigation of its wear properties.

CONCLUSION

The mechanical properties, phase stability, biocompatibility,and wear properties of a newly developed Ce-TZP/Al2O3

nanocomposite were investigated in comparison with aluminaand Y-TZP. From the present results it can be predicted thatthis new composite would be strongly resistant to cata-strophic fracture and to degradation associated with long-term implantation. Therefore, the Ce-TZP/Al2O3 nanocom-posite is a promising alternative ceramic bearing material.We are now conducting further studies on this new ceramic

for clinical application using combinations of both ceramic-on-ceramic and ceramic-on-PE.

REFERENCES

1. Schmalzried TP, Jasty MJ, Harris WJ. Periprothestic bone lossin total hip arthroplasty. J Bone Joint Surg 1992;74A:849-863.

2. Shanbhag AS, Jacobs JJ, Glant TT, Gilbert JL, Black J, GalanteJO. Composition and morphology of wear debris in faileduncemented total hip replacement arthroplasty. J Bone JointSurg 1994;76B:60–67.

3. Charnley J. Low friction arthroplasty of the hip: Theory andpractice. New York: Springer; 1979.

4. Jiranek WA, Machado M, Jasty MJ, Jevsevar D, Wolfe HJ,Goldring SR, Goldberg MJ, Harris WJ. Production of cytokinesaround loosened cemented acetabular components. Analysiswith immunohistochemical techniques and in situ hybridization.J Bone Joint Surg 1993;75A:863–879.

5. Saikko V, Pfaff HG. Low wear and friction in alumina/aluminatotal hip joints: A hip simulator study. Acta Orthop Scand1998;69:443–448.

6. Goldsmith AA, Dowson D, Isaac GH, Lancaster JG. A com-parative joint simulator study of the wear of metal-on-metal andalternative material combinations in hip replacements. Proc InstMech Eng 2000;H214:39–47.

7. Clarke IC, Good V, Williams P, Schroeder D, Anissian L, StarkA, Oonishi H, Schuldies J, Gustafson G. Ultra-low wear ratesfor rigid-on-rigid bearings in total hip replacements. Proc InstMech Eng 2000;H214:331–347.

8. Zichner LP, Willert HG. Comparison of alumina–polyethyleneand metal–polyethylene in clinical trials. Clin Orthop 1992;282:86–94.

9. Walter A. On the material and the tribology of alumina–aluminacouplings for hip joint prostheses. Clin Orthop 1992;282:31–46.

10. Willert HG, Buchhorn GHH, Gruebel D, Kruester G, SchaffnerS, Schenk R, Semlitsch M. Wear behavior and histopathologyof classic cemented metal on metal hip endoprostheses. ClinOrthop 1996;329S:S160–186.

11. Jacobs JJ, Skipor AK, Doorn PF. Cobalt and chromium con-centrations in patients with metal on metal total hip replace-ments. Clin Orthop 1996;329S:256–263.

12. Boutin P. Arthroplastie totale de la hanche par prosthese enalumine fritte. Rev Chir Orthop 1972;58:229–246.

13. Willmann G. Ceramic femoral head retrieval data. Clin Orthop2000;379:22–28.

14. Kumar P, Oka M, Ikeuchi K, Shimizu K, Yamamuro T, Oku-mura H, Kotoura Y. Low wear rate of UHMWPE againstzirconia ceramic (Y-PSZ) in comparison to alumina ceramicand SUS 316L alloy. J Biomed Mater Res 1991;25:813–828.

15. Cales B. Zirconia as a sliding material. Clin Orthop 2000;379:94–112.

TABLE II. Wear and Friction Properties of Tested Ceramics

Ce-TZP/Al2O3 Alumina

Wear rate of pin(� 10�7 mm3/Nm) 0.55 � 0.04 2.12 � 0.37

Wear track depth(� m) 0.46 � 0.16 4.59 � 0.64

Steady statefriction coefficient 0.27 � 0.03 0.11 � 0.02

Wear track roughnessafter test(� m Ra) 0.0095 � 0.006 0.0079 � 0.004

(mean � S.D.)

Figure 6. Encapsulating membrane thickness (mean�SD).

269Ce-TZP/Al2O3 NANOCOMPOSITE AS A BEARING MATERIAL

16. Piconi C, Maccauro G. Zirconia as a ceramic biomaterial.Biomaterials 1999;20:1–25.

17. Burger W, Richter HG. High strength and toughness aluminamatrix composites by transformation toughening and ‘in situ’platelet reinforcement (ZPTA). The new generation of bioce-ramics. In: Giannini S, Moroni A, editors, Bioceramics 13.Bologna: Trans Tech; 2000. p 545–548.

18. Affatato S, Goldoni M, Testoni M, Toni A. Mixed-oxidesprosthetic ceramic ball heads. Part 3: Effect of the ZrO2 fractionon the wear of ceramic on ceramic hip joint prostheses. Along-term in vitro wear study. Biomaterials 2001;22:717–723.

19. Tsukuma K. Mechanical properties and thermal stability ofCeO2 containing tetragonal zirconia polycrystals. Am CeramSoc Bull 1986;65:1386–1389.

20. Nawa M, Nakamoto S, Sekino T, Niihara K. Tough and strongCe-TZP/alumina nanocomposites doped with titania. Ceram Int1992;18:11–17.

21. Zum Gahr KH, Bundschuh W, Zimmerlin B. Effect of grain sizeon friction and sliding wear of oxide ceramics. Wear 1993;162-164:269–279.

22. Marshall DB, Evans AG. Reply to comment on elastic/plasticindentation damage in ceramics: The median/radial crack sys-tem. J Am Ceram Soc 1981;64:C182.

23. Toraya H, Yoshimura M, Somiya S. Calibration curve forquantitative analysis of the monoclinic-tetragonal ZrO2 systemby x-ray diffraction. J Am Ceram Soc 1984;67:C119–C121.

24. Derbyshire B, Fisher J, Dowson D, Hardaker C, Brummitt K.Comparative study of the wear of UHMWPE with zirconiaceramic and stainless steel femoral heads in artificial hip joints.Med Eng Phys. 1994;16:229–236.

25. Christel P, Meunier A, Heller M, Torre JP, Peille CN. Mechan-ical properties and short-term in-vivo evaluation of yttrium-oxide-partially-stabilized zirconia. J Biomed Mater Res 1989;23:45–61.

26. Ichikawa Y, Akagawa Y, Nikai H, Tsuru H. Tissu compatibilityand stability of a new zirconia ceramic in vivo. J Prosthet Dent1992;68:327–330.

27. Drummond JL. In vitro aging of yttria-stabilized zirconia. J AmCeram Soc 1989;72:675–676.

28. Cales B, Stefani Y, Lilley E. Long-term in vivo and in vitroaging of a zirconia ceramic used in orthopaedy. J Biomed MaterRes 1994 ;28:619–624.

29. Shimizu K, Oka M, Kumar P, Kotoura Y, Yamamuro T, Maki-nouchi K, Nakamura T. Time-dependent changes in the me-chanical properties of zirconia ceramic. J Biomed Mater Res1993;27:729–734.

30. Blaise L, Villermaux F, Cales B. Ageing of zirconia: Everythingyou always wanted to know. In: Giannini S, Moroni A, editors.Bioceramics 13. Bologna: Trans Tech; 2000 p 553–556.

31. Lu Z, McKellop H.Frictional heating of bearing materials testedin a hip joint wear simulator. Proc Inst Mech Eng H 1997;211:101–108.

32. Therin M, Christel P, Munier A. Analysis of the general featuresof the soft tissue response to some metals and ceramics usingquantitative histomorphometry. J Biomed Mater Res 1994;28:1267–1276.

33. Laing PG, Ferguson AB, Hodge ES. Tissue reaction in rabbitmuscle exposed to metallic implants. J Biomed Mater Res1967;1:135–149.

34. Chevalier J, Cales B, Drouin JM, Stefani Y. Ceramic–ceramicbearing systems compared on different testing configurations.In: Sedel L, Rey C, editors. Bioceramics 10. Paris: Elsevier;1997. p 271–274.

35. Sawae Y, Murakami T, Tashima S, Shimotoso T. Improvementin microstructure and wear property of alumina ceramics forjoint prostheses. In: Ohgushi H, Hastings GW, Yoshikawa T,editors. Bioceramics 12. Singapore: World Scientific 1999. p99–102.

270 TANAKA ET AL.