CORR-10/00-0379103-Clarke 08/23/2000 03:07 PM Page 34

CLINICAL ORTHOPAEDICS AND RELATED RESEARCH Number 379, pp. 3440 2000 Lippincott Williams & Wilkins, Inc.

Clinical and Hip Simulator Comparisons of Ceramic-on-Polyethylene and Metal-on-Polyethylene Wear

Ian C. Clarke, PhD*; and Allen Gustafson, MD**

The benet of reduced polyethylene wear with ceramic in hip replacements does not seem to have been universally appreciated. In this current study, wear predictions from laboratory and clinical studies were compared for ceramic-onpolyethylene and cobalt chrome-on-polyethylene combinations. Many laboratory studies included water-based lubrication and linear-tracking mechanisms. Now it is appreciated that these were inappropriate methods, because of a propensity for very low or virtually no polyethylene wear against ceramics in water. Thus, waterbased studies predicting a 20- to 80-fold advantage for ceramic-on-polyethylene compared with metal-on-polyethylene clearly were in error. However, serum-based simulator studies with high protein-concentrations generally have shown greater wear with alumina-on-polyethyl-

From *Peterson Tribology Lab, and the **Department of Orthopaedics, Loma Linda University Medical Center, Loma Linda, CA. R & D Funding was received over the course of this study: Industrial Ceramics Technology, Inc. Ann Arbor, MI. (funded by the National Institutes of Health); Technology Assessment Transfer Inc., Annapolis, MD (funded by the National Institutes of Health); Biomet Inc., Warsaw, IN; Kyocera Inc., Kyoto, Japan; Sulzer Inc., Berne, Switzerland; Sven Noren Memorial Foundation, Stockholm, Sweden; A. Stark, MD, Department of Orthopaedics, Karolinska University Hospital, Stockholm, Sweden. Reprint requests to Ian C. Clarke, PhD, Peterson Tribology Lab, Loma Linda University Medical Center, 250 East Caroline Street, Loma Linda, CA 92408.

ene than with metal-on-polyethylene. Controversy still remains over the use of such nonphysiologic protein levels. The simulator studies were just beginning to explore the role of serum protein concentrations and the inuence on the various wear models. Polyethylene wear with zirconia systems was particularly affected by serum protein concentrations. In one simulator study, use of proteins in the physiologic range resulted in the alumina-on-polyethylene wear rate decreasing to approximately 50% of that of metalon-polyethylene. In the literature, many hip design and polyethylene variations were reported which confounded the wear analysis. Overall, the clinical data supported the superior performance of ceramic-on-polyethylene systems by a factor of 1.5- to fourfold. However, the amount of supporting data was not large. This summary of laboratory and clinical data indicated that ceramicon-polyethylene hip replacement systems offered on average a 50% wear reduction from metal-onpolyethylene systems.

In 1963, Charnley6 revolutionized the practice of total hip replacement with the introduction of the cemented, stainless steel femoral stem and head combined with cemented polyethylene cups. This metal-on-polyethylene system became the gold standard for future comparison of implants and their clinical results.22 Nevertheless, given the activity levels of the more active and/or younger patients, the rela-

34

CORR-10/00-0379103-Clarke 08/23/2000 03:07 PM Page 35

Number 379 October, 2000

Comparisons Ceramic and Metal-Polyethylene Wear

35

tively poor wear performance of the polyethylene material has proved a challenge for the long-term success because of the release of large volumes of wear debris.1,2,9,40 Thus, hip replacement systems with improved wear resistance are of considerable interest today. An alternate to polyethylene bearings was the concept of the alumina ceramic cup pioneered in France by Pierre Boutin in 1970.3 As reported by Mittelmeyer and Heisel23 approximately 3 years later, designs such as the Autophor and Xenophor were introduced in Europe, giving the surgeon the choice of ceramic femoral heads combined with either polyethylene or alumina caps. This initial experience using ceramic femoral heads did not prove popular in the United States. Given the fact that the major limitation of the polyethylene material lies in its poor wear resistance, one may wonder why alumina ceramic implants have not become more universally accepted. One possibility may be that the perceived benet, that is reduced polyethylene wear with ceramic femoral heads, has not been fully appreciated or satisfactorily documented. In the current report, the wear predictions from laboratory and clinical studies were compared for ceramic-on-polyethylene combinations and CoCr-on-polyethylene combinations. Studies of Wear Screening Studies of wear screening used simple pin-onat or annulus-on-at specimens rather than implant geometries. There are many such historic studies, but many included water-based lubrication and linear tracking.11,13,24,26,31 Although such machines still are used, there has been criticism that wear predictions from such machines frequently were contradicted rst by clinical experience and then by hip simulator studies.5,14,15,19,36,37 The rst admonition about the artifacts inherent with water-lubricated tests came from a study on polyethylene pinon-at wear.18 A second criterion recently was reestablished: there had to be multidirectional wear vectors for successful simulation of polyethylene cup wear.4,37 The orbital hip simulators used in the United States were in-

troduced with such multidirectional kinematics.7,10 Recent emphasis therefore has included serum lubrication and multidirectional travel for studies regarding polyethylene wear screening. However, there still are no ceramicon-polyethylene and metal-on-polyethylene data for comparison from such machines.5,30 Simulator Wear Comparisons The data from hip simulator studies have been polarized between those studies using water as the joint lubricant11,29,32,34 and those using a biologic lubricant such as bovine serum.20,21, 28,38,41 In a pioneering simulator study of ceramic-on-polyethylene and metal-on-polyethylene with water lubrication, a 20-fold wear improvement was reported with alumina femoral heads.34 A similar study in which 22-, 26- and 32-mm stainless steel and zirconia femoral heads in water lubrication were used also showed a wear advantage for the ceramic-onpolyethylene systems.11 However, it was evident that a brown patchy transfer lm discolored the zirconia femoral heads. This could have been indicative of excessive heat and/or formation of surface transfer lms, both now considered artifacts of an inadequate lubrication regime.1214,16,21,36 One simulator study of implants from ve manufacturers reported almost no polyethylene wear with alumina femoral heads (water lubrication) but much higher wear with stainless steel and CoCr femoral heads.32 Average polyethylene wear ratios varied from 32- to 85-fold in favor of ceramic femoral heads compared with metal femoral heads. In a subsequent study, the same authors ran polyethylene cups of different thickness against zirconia and CoCr femoral heads, again with water lubrication.29 In two sets, the 28-mm and 32mm zirconia heads did not have polyethylene wear. In the remaining three sets, the metalon-polyethylene averaged 2.4- to fourfold more wear than with zirconia heads (7-mm thick cups; 28-mm femoral heads). Thus, overall the water-based simulator studies have shown metal-on-polyethylene wear varying from 2.4- to 85-fold higher than with ceramic-

CORR-10/00-0379103-Clarke 08/23/2000 03:07 PM Page 36

36

Clarke and Gustafson

Clinical Orthopaedics and Related Research

on-polyethylene. There also was a propensity for virtually no polyethylene wear when testing ceramic femoral heads in water, which was an especially prominent feature with zirconia femoral heads. In one of the rst simulator studies using bovine serum as the lubricant,41 a wear ratio of only 1.3-fold was reported for metal-on-polyethylene compared with ceramic-on-polyethylene. One later study of serum simulators reported a tenacious white coating on the femoral heads, particularly evident with zirconia ceramic.21 The authors advocated the use of 20 mmol ethylenediaminetetraacetic acid to prevent precipitation of these calcium phosphate coatings from the serum. In a followup study20 with ethylenediaminetetraacetic acid added to 90% concentration of bovine serum (60 mg/mL proteins), the alumina femoral heads averaged 1.2-fold more polyethylene wear overall than with the metal femoral heads and the CoCr femoral heads averaged 1.8-fold more polyeth-

ylene wear than the zirconia (Fig 1). This appeared to be the rst serum-based study to indicate that in vitro, the alumina could have more polyethylene wear than CoCr femoral heads. The current authors ran a similar study and also reported higher polyethylene wear with alumia femoral heads compared with metal femoral heads (nonpublished data, Clarke IC: Comparisons of M-M and M-PE hip systems at 10 million cycles in hip simulator study. Presented at the 10 Years Metasul Jubilee Meeting. ZurichOerlikon, Switzerland, 1998) (Fig 1). The main differences between the two studies were the 11%8 and 20%20 increases in polyethylene wear with alumina femoral heads compared with metal femoral heads. Polineni et al28 studied the effect of protein concentration and observed a wear trend that was different for ceramic femoral heads compared with metal femoral heads and which also varied with sterilization methods (Fig 2). In addition, all serum concentrations resulted

Fig 1. Comparison of polyethylene wear rates in two simulator studies with study durations of two21 and ve million cycles10 and same serum concentration of 60 mg/mL (20 mmol ethylenediametetraacetic acid additive). One study used 32 mm heads with GUR 4150 cups, and the sterilization condition was not specied.21 The second study used 28 mm heads with GUR 4150 cups sterilized at 2.5 Mrad under nitrogen atmosphere.10

CORR-10/00-0379103-Clarke 08/23/2000 03:07 PM Page 37

Number 379 October, 2000

Comparisons Ceramic and Metal-Polyethylene Wear

37

Fig 2. Comparison of volumetric wear rates for polyethylene regarding protein concentration in serum. The human range varied from 20 mg/mL protein in normal synovial uid and increased to more than 35 mg/mL in patients with osteoarthritis and rheumatoid arthritis.28

in some protein degradation. At very low concentrations, there was back transfer of polyethylene onto the ceramic femoral heads and at the higher concentrations there was more precipitation of insoluble proteins. Both conditions resulted in reduced rates of polyethylene wear. The highest rates of polyethylene wear were achieved in the physiologic range of protein concentrations (Fig 2) and in that condition the metal-on-polyethylene wear averaged 1.55fold that of ceramic-on-polyethylene. Thus, the studies of serum-based simulators generally have shown greater wear with alumina-onpolyethylene than with metal-on-polyethylene when nonphysiologic, high protein concentrations were used.20,21 (nonpublished data, Clarke IC: Presented at the 10 Years Metasul Jubilee

Meeting. Zurich-Oerlikon, Switzerland, 1998). When physiological protein concentrations were used, the wear rate of alumina-onpolyethylene seemed to decrease to approximately 50% that of metal-on-polyethylene. Clinical Wear Comparisons It was apparent from radiographic studies comparing ceramic-on-polyethylene and metal-onpolyethylene systems in the same series that the average wear rates with the metal femoral heads invariably were higher than with ceramic femoral heads (Table 1). Overall, the wear ratios varied from 1.7-fold to fourfold higher for metal-on-polyethylene relative to ceramic-on-polyethylene, with an average ratio of 2.7. Of course, there were confounding

CORR-10/00-0379103-Clarke 08/23/2000 03:07 PM Page 38

38

Clarke and Gustafson

Clinical Orthopaedics and Related Research

TABLE 1. Clinical Comparison of Linear Wear Rates for Polyethylene Cups and Ranking of Wear Ratios for Combinations With Metal and Ceramic Heads
Author(s)
Sieber and Weber Schuller and Marti Oonishi et al Wroblewski et al Zichner Ohashi et al Zichner and Willert

Reference
35 33 27 42 43 25 44

Material of Metal Heads

cobalt chrome cobalt chrome stainless steel stainless steel cobalt chrome stainless steel, cobalt chrome cobalt chrome

Wear with Metal

0.2 0.1 0.25 0.07 0.2 0.04 0.2

Wear with Ceramic

0.05 0.03 0.1 0.03 0.1 0.03 0.1

Wear Ratio
4.0 3.7 2.5 2.0 2.0 1.7 2.0

A wear ratio greater than 1.0 indicated higher wear for the metal-on-polyethylene system.

factors. One study25 included data with ball sizes other than 28 mm but these were a minority (22 mm, 32 mm types) and may not have had a great inuence. In another study,27 cups manufactured in the United States (28-mm T28, Zimmer Inc, Warsaw IN) were compared with cups made in Japan (28-mm Bioceram, Kyocera Inc, Kyoto). In addition, some early T-28 cups were molded and some were machined from GUR 4150 extruded bar-stock. With both types of cups being sterilized with 2 to 4 Mrad in air, this may not have had a major effect. The study in which the Charnley system was used not only had ceramic femoral heads but also had a highly cross-linked polyethylene material.42 In the two studies from Europe in which the highest wear ratios were reported33,35 the rotating trunnion type of design, which has not found application in the United States, was used. Excluding the latter two reports, these clinical wear data provided a wear ratio of twofold in favor of ceramic-on-polyethylene with xed femoral heads (Table 1). This 2:1 wear-ratio44 was conrmed recently with the 20-year followup studies reported by Zichner.43 DISCUSSION Ideally, the clinical review only would include studies in which the same design of implant was implanted by the same surgeons, the same head diameter was used, the same polyethyl-

ene material was used, and the same xation modes were used, with the experimental protocol alternating only in the choice of either metal or ceramic femoral heads. There were only three studies, which appeared to satisfy these criteria: two in which the rotatingtrunnion design was used33,35 and one in which the more common modular but nonrotating head was used.43 In these studies, the metal-on-polyethylene wear rates averaged two to four times higher than ceramic-onpolyethylene. In particular, the long followup times in the latter study were particularly impressive, conrming the twofold wear reduction with ceramics.44 Thus, the simulator studies with water lubrication predicting 20- to 85-fold advantage for ceramic-on-polyethylene were clearly an artifact of the test methods used. Even in the simulator studies with serum lubrication, artifacts were encountered attributable to the formation of calcium phosphate lms and also from the precipitation of insoluble proteins.16,17,38,39 In particular, the zirconia femoral heads were particularly at risk for protein degradation artifacts leading to lower polyethylene wear.16,17,20,21 This may be attributable to the fact that zirconia has a lower thermal conductivity than either CoCr or alumina, runs at a higher temperature and therefore was more prone to promote serum degradation problems. The one simulator study in which physiologic protein concentrations were used successfully showed the 50% wear

CORR-10/00-0379103-Clarke 08/23/2000 03:07 PM Page 39

Number 379 October, 2000

Comparisons Ceramic and Metal-Polyethylene Wear

39

reduction for ceramic-on-polyethylene over metal-on-polyethylene.28 From the current review, it was clear that the amount of data showing superiority of the ceramic-on-polyethylene systems is not large. Implant variations and use of different polyethylene processing confounded the clinical literature. In addition, the simulator studies were just beginning to explore the role of serum test parameters in the wear of metal-onpolyethylene systems, including serum type, protein concentrations, volume of test chambers, operating temperatures, and cup mounting positions. As more success in showing clinical wear rates of metal-on-polyethylene systems is gained by using simulator models, future wear studies of the ceramic-on-polyethylene combination will benet. Acknowledgments
The authors thank the orthopaedic faculty of Loma Linda University Medical Center and the staff of the Peterson Tribology Lab, Loma Linda University Medical Center for their support in these studies. The authors also thank Biomet Inc, (Warsaw, IN); CeramTec Inc, (Plochingen, Germany); Industrial Ceramics Technology Inc, (Ann Arbor, MI); Kyocera Inc, (Kyoto, Japan); Sulzer Inc, (Berne, Switzerland); Technology Assessment Transfer Inc, Annapolis MD; A. Stark, MD, Department of Orthopaedics, Karolinska University Hospital; and the Sven Noren Memorial Foundation, Stockholm, Sweden, for implant donations and research support.

6.

7. 8.

9.

10. 11.

12.

13.

14.

15. 16.

References
1. Amstutz HC, Campbell P, Kossovsky N, et al: Mechanism and clinical signicance of wear debris-induced osteolysis. Clin Orthop 276:718, 1992. 2. Bono JV, Sanford L, Toussaint JT: Severe polyethylene wear in total hip arthroplastyObservations from retrieved AML PLUS hip implants with an ACS polyethylene liner. J Arthroplasty 9:119125, 1994. 3. Boutin P: Arthroplastie totale de la hanche par prothese en alumine frittee. Rev Chir Orthop Reparatrice Appar Mot 58: 229246, 1972. 4. Bragdon CR, OConnor DO, Lowenstein JD, et al: The importance of multidirectional motion on the wear of polyethylene. Proc Inst Mech Eng 210: 157165, 1996. 5. Bragdon CR, OConnor DO, Lowenstein JD, et al: Development of a new pin on disk testing machine for evaluation of polyethylene wear. Proceedings of 17. 18.

19.

20.

the Fifth World Biomaterials Congress, Toronto 78, 1996. Charnley J: Total Prosthetic Replacement of the Hip Joint Using a Socket of High Density Polyethylene. Publication No. 1. Wigan, Lancashire, England, Centre for Hip Surgery, Wrightington Hospital 1966. Clarke IC: Wear-screening and joint simulation studies versus materials selection and prosthesis design. Crit Rev Bioeng J 8: 2991, 1982. Clarke IC, Anissian L, Stark A, et al: Comparisons of M-M and M-PE Hip Systems at 10 Million Cycles in Hip Simulator Study. In Rieker C, Windler M, Wyss U (eds). METASUL A Metal-on-Metal BearingA 10 year Anniversary. Berne, Hans Huber 93109, 1999. Clarke IC, Campbell P: Interface Failure Dynamics: Osteoclastic and Macrophagic Bone Loss. In Kenedi RM, Paul JP (eds). Progress in Bioengineering. Glasgow, Scotland, Strathclyde University 104115, 1989. Clarke IC, McKellop HA, Okuda R, et al: Materials and prosthesis design criteriaHip simulator studies. Trans Orthop Res Soc: 7: 292, 1982. Cooper JR, Dowson D, Fisher J: The effect of transfer lm and surface roughness on the wear of lubricated ultra high molecular weight polyethylene. Clin Mat 14: 295302, 1993. Derbyshire BJ, Fisher D, Dowson et al: Comparative study of the wear of UHMWPE with zirconia ceramic and stainless steel femoral heads in articial hip joints. Med Eng Phys 16: 229236, 1994. Dowson D, Harding RT: The wear characteristics of ultra-high-molecular weight polyethylene against a high density alumina ceramic under wet (distilled water) and dry conditions. Wear 75:313331, 1982. Good VD, Clarke IC, Anissian L: Water and bovine serum lubrication compared in simulator PTFE/ CoCr wear model. J Biomed Mater Res 33:275283, 1996. Hall RM, Unsworth A, Siney P, Wroblewski BM: Wear in retrieved Charnley acetabular sockets. Proc Inst Mech Eng 210: 197207, 1996. Liao YS, Benya PD, McKellop H: Stability of serum as a lubricant in wear simulator tests of prosthetic joints. Proceedings of the Fifth World Biomaterials Congress Toronto, Canada 871, 1996. Lu Z, McKellop H: Frictional heating of bearing materials tested in a hip joint wear simulator. Proc Inst Mech Eng 211: 101108, 1997. McKellop H, Clarke I, Markolf K, et al: Wear characterization of UHMWpolyethylene: A method for accurately measuring extremely low wear rates. J Biomed Mat Res 12: 895927, 1978. McKellop H, Clarke I, Markolf K, et al: Friction and wear properties of polymer, metal, and ceramic prosthetic joint materials evaluated on a multichannel screening device. J Biomed Mater Res 15: 619653, 1981. McKellop H, Lu B: Friction, Lubrication and Wear of Polyethylene/Metal and Polyethylene/Ceramic Hip Prostheses on a Joint Simulator. Proceedings of the Fourth World Biomaterials Congress, Berlin 110, 1992.

CORR-10/00-0379103-Clarke 08/23/2000 03:07 PM Page 40

40

Clarke and Gustafson

Clinical Orthopaedics and Related Research

21. McKellop H, Lu B, Benya P: Friction lubrication and wear of cobalt-chromium, alumina and zirconia hip prostheses compared on a joint simulator. Trans Orthop Res Soc 17:402, 1992. 22. Malchau H, Herberts P, Ahnfelt L: Prognosis of total hip replacement in Sweden. Follow-up of 92,675 operations performed 19781990. Acta Orthop Scand 64:497506, 1993. 23. Mittelmeier H, Heisel J: Sixteen years experience with ceramic hip prostheses. Clin Orthop 282:6472, 1992. 24. Mosleh M, Arinez J, Suh NP: Delamination wear of polyethylene enhanced by gamma irradiation. Trans Orthop Res Soc 21:480, 1996. 25. Ohashi T, Inoue S, Kajikawa K, et al: Clinical Wear Rate of Acetabular Component Accompanied With Alumina Ceramic Head. In Oonishi H, Ooi Y (eds). Orthopaedic Ceramic Implants. Vol 8. Japanese Society Orthopaedics 223232, 1988. 26. Oonishi H, Kuno M, Tsuji E, Fujisawa A: The optimum dose of gamma radiationHeavy doses to low wear polyethylene in total hip prostheses. J Mater Sci Mater Med 8:1118, 1997. 27. Oonishi H, Takayaka Y, Clarke IC, et al: Comparative wear studies of 28 mm ceramic and stainless steel total hip joints Over 2 to 7 year period. J Long Term Eff Med Implants 2:3747, 1992. 28. Polineni VK, Wang A, Essner A, et al: Effect of lubricant protein concentration on the wear of UHMWPE acetabular cups against cobalt-chrome and alumina femoral heads. Twenty-third Annual Proceedings of the Biomaterials Society. New Orleans 154, 1997. 29. Saikko VO: Wear of polyethylene acetabular cups against zirconia femoral heads studied with a hip joint simulator. Wear 176:207212, 1994. 30. Saikko VO, Ahlroos T: Wear simulation of UHMWPE for total hip replacement with a multidirectional motion pin-on-disk device: Effects of counterface material, contact area, and lubricant. J Biomed Mater Res 49:147154, 1999. 31. Saikko V, Kleimola M: A reciprocating pin-on-at study of the long-term behavior of ultra-high molecular weight polyethylene sliding against Co-Cr-Mo alloy. Proceedings of the Sixth Nordic Symposium on Tribology. Uppsala, Sweden 227234, 1994. 32. Saikko VO, Paavolainen PO, Slatis P: Wear of the polyethylene acetabular cup. Acta Orthop Scand 64:391402, 1993. 33. Schuller HM, Marti RK: Ten-year socket wear in 66

34.

35.

36.

37.

38.

39.

40.

41.

42.

43. 44.

hip arthroplasties. Ceramic versus metal heads. Acta Orthop Scand 61:240243, 1990. Semlitsch M, Lehmann M, Weber H et al: New prospects for a prolonged functional life-span of articial hip joints by using the material combination polyethylene/aluminum oxide ceramic/metal. J Biomed Mater Res 11:537552, 1977. Sieber HP, Weber BG: HDPE Ceramic Total Hip Replacement: The St. Gallen. In Vivo and In Vitro Experience. In Vincenzini P (ed). Ceramics in Surgery. Amsterdam, Elsevier Scientic Publishing Co 287294, 1983. Wang A, Essner A, Stark C, et al: Comparison of the size and morphology of UHMWPE wear debris produced by a hip joint simulator under serum and water lubricated conditions. Biomaterials 17:865871, 1995. Wang A, Polineni VK, Essner A, et al: The signicance of nonlinear motion in the wear screening of orthopaedic implant materials. J Testing Eval 25:239245, 1997. Wang A, Polineni VK, Essner A, et al: Role of proteins and hyaluronic acid in the lubrication and wear of UHMWPE acetabular cups. Trans Orthop Res Soc 23:218, 1998. Wang A, Polineni VK, Essner A, Stark C, Dumbleton JH: The impact of lubricant protein concentration on the outcome of hip joint simulator testing. Trans Orthop Res Soc 24:52, 1999. Willert HG, Buchhorn G Bucchorn U: Tissue Response to Wear Debris in Artical Joints. In Weinstein A, Gibbons D, Brown S, Ruff W. (eds). Implant Retrieval: Material and Biological Analysis. Washington, DC, United States Department of Commerce 239267, 1981. Wright KWJ, Scales JT: Use of Hip Joint Simulators for the Evaluation of Wear of Total Hip Prostheses. In Winter GD, Leray JL, de Groot K (eds). Evaluation of Biomaterials. New York, John Wiley and Sons 135146, 1980. Wroblewski BM, Siney PD, Dowson D, et al: Prospective clinical and joint simulator studies of a new total hip arthroplasty using alumina ceramic heads and cross-linked polyethylene cups. J Bone Joint Surg 78B:280285, 1996. Zichner LP: Two decades experience of alumina/ polyethylene in Germany. In Press. Zichner LP, Willert HG: Comparison of aluminapolyethylene and metal-polyethylene in clinical trials. Clin Orthop 282:8694, 1992.