would detect differences in the cellular response to wear particles of two titanium ... Wear particles from endoprostheses provoke the release of inflammatory ...
IN VITRO HUMAN MONOCYTE RESPONSE TO WEAR PARTICLES OF TITANIUM ALLOY CONTAINING VANADIUM OR NIOBIUM S. D. ROGERS,
D. W. HOWIE,
S. E. GRAVES,
M. J. PEARCY,
D. R. HAYNES
From the Royal Adelaide Hospital and the University of Adelaide, Australia Our aim was to determine whether in vitro studies would detect differences in the cellular response to wear particles of two titanium alloys commonly used in the manufacture of joint replacement prostheses. Particles were of the order of 1 m in diameter representative of those found adjacent to failed prostheses. Exposure of human monocytes to titanium 6-aluminium 4- vanadium (TiAlV) at concentrations of 7 4 x 10 particles/ml produced a mean prostaglandin E2 release of 2627.6 pM; this was significantly higher than the 317.4 pM induced by titanium 6-aluminium 7-niobium alloy (TiAlNb) particles (p = 0.006). Commercially-pure titanium particles induced a release of 347.8 pM. In addition, TiAlV stimulated significantly more release of the other cell mediators, interleukin-1, tumour necrosis factor and interleukin-6. At lower concentrations of particles there was less mediator release and less obvious differences between materials. None of the materials caused significant toxicity. The levels of inflammatory mediators released by phagocytic cells in response to wear particles may influence the amount of periprosthetic bone loss. Our findings have shown that in vitro studies can detect differences in cellular response induced by particles of similar titanium alloys in common clinical use, although in vivo studies have shown little difference. While in vitro studies should not be used as the only form of assessment, they must be considered when assessing the relative biocompatibility of different implant materials. J Bone Joint Surg [Br] 1997;79-B:311-5. Received 20 August 1996; Accepted after revision 31 October 1996
S. D. Rogers, BSc, Research Assistant D. W. Howie, MB BS, PhD, FRACS, Professor and Head of Department S. E. Graves, MB BS, DPhil, Senior Lecturer Department of Orthopaedics and Trauma, Royal Adelaide Hospital, North Terrace, Adelaide 5000, South Australia. M. J. Pearcy, PhD, Associate Professor School of Engineering, Flinders University of South Australia, Sturt Road, Bedford Park, 5042 South Australia. D. R. Haynes, PhD, Research Officer Department of Pathology, University of Adelaide, Adelaide 5005, South Australia. Correspondence should be sent to Professor D. W. Howie. ©1997 British Editorial Society of Bone and Joint Surgery 0301-620X/97/27192 $2.00 VOL. 79-B, NO. 2, MARCH 1997
Wear particles from endoprostheses provoke the release of inflammatory mediators from macrophages which may differ according to the material used. This may influence the extent of periprosthetic bone loss. The wear of titaniumalloy prostheses, including the articulating surfaces, the surfaces of the modular components, and cemented and cementless femoral stems, results in the accumulation of 1-3 wear particles in the periprosthetic tissues. Despite this, titanium alloys are useful implant materials because of their excellent mechanical properties, their resistance to corro3 sion, and apparent biocompatibility in solid form. There has been concern, however, regarding one of the commonly used alloys, titanium 6-aluminium 4-vanadium alloy (TiAlV), because of the potential adverse effects of vana4,5 dium. An alternative titanium alloy, titanium 6-aluminium 7-niobium (TiAlNb), was therefore developed with comparable mechanical properties to those of high-strength TiAlV 6-8 and with more resistance to corrosion at low pH. TiAlV particles are known to stimulate a macrophage response in vivo and provoke the release of inflammatory 9-12 mediators from these cells in vitro. TiAlNb particles are likely to have similar effects but they may be less marked because of the different metals in TiAlNb or because of its enhanced corrosion resistance. No differences have been reported, however, in the cellular response in vivo to these 13 alloys in solid form and only small differences in the number and types of cells in the tissues after injection of 14 particles of both alloys. This may suggest that the cellular responses are similar, but the in vivo tests are relatively insensitive and important biological differences may not have been detected. Our aim was to determine if in vitro studies could detect differences in the cellular responses to wear particles of TiAlV and TiAlNb, by measuring the release of prostaglandin E2 (PGE2) from human monocytes. PGE2 was considered important because it is present in periprosthetic tissue retrieved at revision surgery for loosening and has been proposed as a factor in periprosthetic bone resorp15-17 It is also known to be released in vitro in response tion. 11 to TiAlV. Other mediators of bone resorption, such as interleukin-1 (IL-1), tumour necrosis factor (TNF) and interleukin-6 (IL-6), have been identified in periprosthetic tissue and implicated in the inflammatory response to wear 18-22 particles and periprosthetic loosening. We therefore also examined differences in the release of these mediators 311
312
S. D. ROGERS,
D. W. HOWIE,
S. E. GRAVES,
as well as cytotoxicity. Because previous in vitro studies have shown that commercially-pure titanium (cpTi) particles induce the release of fewer bone-resorbing mediators 12 than TiAlV particles, we also compared the response to cpTi particles with that to TiAlV and TiAlNb. MATERIALS AND METHODS Generation of wear particles. Three separate preparations of wear particles were produced: TiAlV (Protasul 64), TiAlNb (Protasul 100), and cpTi. Hemispherical cups with an internal diameter of 44 mm and blocks measuring 4 4 1 mm for each material were provided by Dr W. Frick (Protek AG, Berne, Switzerland) and were prepared using the same manufacturing standards as for human implants. The components were soaked overnight in a 0.1% solution of E-Toxa-Clean (Sigma Chemical Co, St Louis, Missouri) and rinsed three times in endotoxin-free deionised water. They were then soaked for 20 minutes in 20% nitric acid in accordance with ASTM F86-76 standard recommended practice for surface preparation and marking 23 of metallic surgical implants. The rinsing was repeated three times and the components were sterilised by dry heating at 180°C overnight. Wear particles were produced by the shaking of a chamber 24 as has been described previously. The hemispherical cups containing blocks of the same titanium material and a shaking solution were sealed and shaken with phosphate-buffered saline supplemented with 2% fetal calf serum (FCS), 5 g/ ml penicillin and 50 U/ml streptomycin for 72 hours. After discarding the first suspension, subsequent particle suspensions after each 72-hour period were stored at 0°C. Attention was given to producing particles free from endotoxin contamination. All washing and preparation steps were carried out in thoroughly clean, heat-sterilised glassware. Low-endotoxin FCS was purchased in bulk (Commonwealth Serum Laboratories Ltd, Parkville, Australia). Levels of endotoxin were monitored regularly throughout the period of particle generation using the Limulus amoebocyte lysate test. Levels were below 10 pg per ml. The particle suspensions generated for each of the titanium materials were sized separately by differential cen24 trifugation to obtain particles in the range 0.5 to 3.0 m. The size distribution was confirmed using a Coulter counter (Multisizer 2; Coulter Electronics, Luton, UK). The shape of the distribution graphs for each material was similar and the mode values for TiAlV, TiAlNb and cpTi were 0.62, 0.66 and 0.66 m, respectively. The concentration of metal in each suspension was determined by mass spectrometry (VG Plasmaquad PQ2 Plus, VG Isotopes, Winsford, UK) and the concentration as particles per ml was calculated using the total titanium-alloy metal content (g/ml), the density of each alloy, and the mode value for particle 24 diameter. Three suspensions of each particle were prepared by vigorous sonication followed by dilution in RPMI 1640 medium containing 10% FCS, 5 g/ml penicillin and
M. J. PEARCY,
D. R. HAYNES
50 U/ml streptomycin, 2 mM L-glutamine and 50 M 7 -mercaptoethanol. The suspensions contained 4 10 , 7 6 1 10 and 2.5 10 particles per ml and were designated respectively as high, medium and low concentrations. As previous studies have shown a large increase in PGE2 release by macrophages after exposure to TiAlV particles at 7 a concentration of 4 10 particles per ml this was chosen 11 as a high dose and corresponds to 57.6 g/ml. Measurement of particle dimension. We used transmission electron microscopy to measure particle dimensions and to detect shape differences between the titanium pre24 parations. The greatest length and breadth of a particle were measured from photomicrographs. We measured 100 particles in each sample. The different titanium preparations were analysed using Student’s t-test with p 0.025 being taken as significant. The length and breadth of the individual particles were not significantly different (Table I). Table I. Mean size (m ± titanium alloy particles
SD)
of titanium and
Length
Breadth
TiAlV
1.710 ± 0.744
1.102 ± 0.576
TiAlNb
1.507 ± 0.582
0.994 ± 0.457
cpTi
1.615 ± 0.638
0.990 ± 0.393
Isolation and culture of cells. We collected human peripheral blood monocytes from eight healthy volunteer donors and isolated them by Ficoll-hypaque sedimentation and 25 adherence to glass. A suspension of 500 l of 6 4 10 cells/ml from each donor was pipetted onto ethanol-washed glass coverslips in a 24-well Linbro plate. The monocytes were allowed to adhere at 37°C in a humidified 5% CO2 atmosphere. After one hour, the coverslips were washed three times by gentle shaking with Hanks’ balanced salt solution. Non-specific esterase staining established that the remaining adherent cells were greater than 96% pure monocytes. Particle suspensions were added to the cells along with appropriate volumes of media to ensure a final volume of 500 l for each well and each was tested in duplicate. Lipopolysaccharide (LPS) 5 g/ml (Escherichia coli 0111:B4; Sigma Chemical, St Louis, Missouri) was used as a positive control for mediator release. Cells cultured in media alone were used as a negative control (nonstimulated cells). The cells were cultured for 48 hours. The supernatants were removed and were centrifuged to remove metal particles. The resultant aliquoted supernatants were either tested immediately for levels of mediators or stored at –70°C until ready for testing. Analysis of mediator release. We determined levels of 26 PGE2 using a competitive radioimmunoassay. Levels of IL-1, TNF and IL-6 were measured by enzyme-linked immunoassays (Cayman Chemical Co, Ann Arbor, Michigan and MedGenix Diagnostics, Fleurus, Belgium). The THE JOURNAL OF BONE AND JOINT SURGERY
IN VITRO HUMAN MONOCYTE RESPONSE TO WEAR PARTICLES OF TITANIUM ALLOYS
313
We found an increased release of all the mediators after exposure to TiAlV particles compared with TiAlNb and cpTi particles. The mean PGE2 release in response to highconcentration TiAlV particles was 2627.6 ± 846.6 pM and to TiAlNb particles 317.4 ± 140.6 pM (p = 0.006). Highconcentration TiAlV particles stimulated release of PGE2 levels 15 times higher than that from non-stimulated cells (p = 0.005) while high-concentration TiAlNb particles gave
approximately a twofold increase (p = 0.05) as did cpTi particles (p = 0.02) (Fig. 1). Lower concentrations caused little or no increase in PGE2 release. A similar pattern was observed with IL-1 release (Fig. 2). High-concentration TiAlV particles stimulated fivefold higher IL-1 release than that from non-stimulated cells (p = 0.005), while high-concentration TiAlNb particles induced a twofold increase in IL-1 release (p = 0.04) as did cpTi particles (p = 0.01). As with PGE2, IL-1 release was unremarkable after exposure to lower concentrations of any of the titanium preparations. As regards TNF, high-concentration TiAlV particles stimulated release 15 times higher than that from nonstimulated cells (p = 0.002) (Fig. 3) and medium-concentration TiAlV particles induced a smaller threefold increase in release (p = 0.002). High concentrations of TiAlNb particles induced a threefold increase (p = 0.001) as did cpTi (p = 0.001), whereas medium concentrations produced a twofold increase (p = 0.002) as did cpT1 particles (p = 0.02). TNF release induced by the low concentration of any of the particles was unremarkable. High concentrations of TiAlV particles stimulated a fivefold higher release of IL-6 than that from non-stimulated cells (p = 0.002) (Fig. 4). Medium-concentration TiAlV particles induced a twofold increase (p = 0.06). High-concentration TiAlNb particles and cpTi particles produced a twofold increase in IL-6 release (p = 0.06 and p = 0.01, respectively). After exposure to lower concentrations IL-6 release was unremarkable. When the results were analysed according to the percentage change in mediator release relative to non-stimulated cells similar differences in mediator release were seen. Examination of the cultures using phase-contrast microscopy showed no difference in the morphology of the cells, in the phagocytosis of particles of all three materials or in the amount of phagocytosis among the different materials. Particles were relatively non-toxic to the cells. The proportions of viable cells exposed to high concentrations of TiAlV, TiAlNb and cpTi particles were 98 ± 2%, 98 ± 1%
Fig. 1
Fig. 2
The release of PGE2 by human monocytes in response to titanium particles (*p < 0.05; **p < 0.005).
The release of IL-1 by human monocytes in response to titanium particles (*p < 0.05; **p < 0.005).
results of the duplicate tests were averaged. The difference in mean PGE2 release induced by the high concentration of TiAlV particles was compared with that induced by the same concentration of TiAlNb particles. We took an increase in PGE2 release of 30% or more as an important difference because of laboratory variation and because we felt that a difference of less than 30% may not be relevant in the human situation. A p value of 0.05 was taken as significant. To determine the patterns of release of all the mediators in response to the three titanium materials, the level of mediator release was compared with that from non-stimulated cells. Statistical significance was calculated using the paired Student’s t-test. Because there was considerable variation between the monocyte response to each donor, the results were also expressed as a percentage of the nonstimulated cells; the level of mediator release by nonstimulated cells was taken to be 100% and the mean percentage change for each treatment was calculated. To determine the variation in mediator release due to errors in measurement techniques, IL-6 release from a single batch of cells from one donor was measured ten times. The mean IL-6 release after exposure of the cells to high concentrations of TiAlV, TiAlNb, cpTi particles and of non-stimulated cells was 996 ± 99, 401 ± 62, 399 ± 32 and 208 ± 45 pg/ml, respectively. Cell viability. Cell viability was assessed by trypan blue 27 exclusion. RESULTS
VOL. 79-B, NO. 2, MARCH 1997
314
S. D. ROGERS,
D. W. HOWIE,
S. E. GRAVES,
M. J. PEARCY,
D. R. HAYNES
Fig. 3
Fig. 4
The release of TNF by human monocytes in response to titanium particles (*p < 0.05; **p < 0.005).
The release of IL-6 by human monocytes in response to titanium particles (*p < 0.05; **p < 0.005).
and 96 ± 3%, respectively compared with 97 ± 3% from non-stimulated cells.
These results differ from in vivo studies which showed 13 14 either no or only slight differences in the cellular responses to TiAlV, TiAlNb and cpTi. These studies, however, may have lacked the sensitivity necessary to detect differences in response to these materials since it is often difficult to make quantitative assessments of the response to biomaterials in vivo. In our study the wear particles of each material were of 28,30-33 Studies have comparable size to those found in vivo. shown that tissues retrieved during revision surgery for aseptic loosening have had an average of 818 g titanium 1,3,34 per gram of tissue. Assuming a particle size of 1 m 8 this corresponds to an average 4 10 particles per gram of tissue of TiAlV. This means that the increased stimula7 tion with 4 10 particles/ml of TiAlV in vitro is likely to be relevant to the human situation. Recent studies on how wear particles induce a macrophage response in vitro have highlighted the effect of low 35 pH within the phagosome. This was shown to be directly related to stimulation of mediator release by the macrophage. TiAlNb is as corrosion-resistant as commercially 36 pure titanium and is more corrosion-resistant than TiAlV. TiAlNb alloy is less susceptible to corrosion at low pH than 6,8 TiAlV alloy which may explain the significant differences in cellular response observed between them. Other wear particles, such as those of cobalt-chrome alloy, are susceptible to intracellular corrosion, but produce more 35,37 while causing less release of mediators toxic effects 11 than TiAlV. Our study has shown that human monocytes in vitro release significantly more bone-resorbing mediators after exposure to particles of TiAlV compared with TiAlNb. While the results of in vitro studies should not be directly transposed to the human situation, we postulate that if there are differences in release of bone-resorbing mediators in human periprosthetic tissues then TiAlV wear particles may cause more osteolysis than TiAlNb particles.
DISCUSSION Our study has shown in vitro differences in cellular responses to wear particles generated from different titanium materials. The large increase in PGE2 release in response to TiAlV compared with TiAlNb confirmed our hypothesis, and the finding of a much greater release of all the other inflammatory cytokines in response to TiAlV compared with TiAlNb and cpTi supports the view that TiAlV causes greater stimulation of phagocytic cells than the other two materials. Previous in vitro studies have shown that TiAlV particles stimulate the release of PGE2, IL-1, TNF and IL-6 from human peripheral blood monocytes and rodent peritoneal 9-12 This stimulatory effect was shown to be macrophages. slightly less when human monocytes were exposed to cpTi 12 particles. Our data support these findings and in addition we have shown that release of these mediators was less with TiAlNb than with TiAlV. We selected human peripheral blood monocytes as the target cells since they are precursors of tissue macrophages which have consistently been identified in association with 28 wear particles in periprosthetic tissues and are known to produce potent inflammatory and bone-resorbing mediators. Previous studies have shown an inflammatory 11 response to TiAlV by rodent peritoneal macrophages, but we felt that it was important to use human cells because the pattern of response to particulate may vary according to the 29 type of phagocytic cell and because we were specifically addressing the differences in response to biomaterials in clinical use. The levels of mediator release varied considerably among the eight donors. Some of this variation was shown to be due to errors in measurement techniques, but most was attributed to differences in response between donors. The pattern of response, however, was similar for all eight donors.
We wish to thank Professor Barrie Vernon-Roberts for his advice and support, and Zophia Krzemionka for her advice on preparation of the THE JOURNAL OF BONE AND JOINT SURGERY
IN VITRO HUMAN MONOCYTE RESPONSE TO WEAR PARTICLES OF TITANIUM ALLOYS
manuscript. This work was supported in part by research grants from Protek AG, Berne, Switzerland, and the National Health and Medical Research Council of Australia. One or more of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition benefits have also been or will be directed to a research fund, foundation, educational institution, or other non-profit institution with which one or more of the authors is associated. REFERENCES 1. Agins HJ, Alcock NW, Bansal M, et al. Metallic wear in failed titanium-alloy total hip replacements: a histological and quantitative analysis. J Bone Joint Surg [Am] 1988;70-A:347-56. 2. Black J, Sherk H, Bonino J, et al. Metallosis associated with a stable titanium-alloy femoral component in total hip replacement: a case report. J Bone Joint Surg [Am] 1990;72-A:126-30. 3. Lombardi AV, Mallory TH, Vaughan BK, Drouillard P. Aseptic loosening in total hip arthroplasty secondary to osteolysis induced by wear debris from titanium-alloy modular femoral heads. J Bone Joint Surg [Am] 1989;71-A:1337-42. 4. Rae T. The toxicity of metals used in orthopaedic prostheses: an experimental study using cultured human synovial fibroblasts. J Bone Joint Surg [Br] 1981;63-B:435-40. 5. Wapner K. Implications of metallic corrosion in total knee arthroplasty. Clin Orthop 1991;271:12-20. 6. Semlitsch M. Classic and new titanium alloys for production of artificial hip joints. In: International Conference on Titanium Products and Applications. Dayton, Ohio: Titanium Development Association, 1986. 7. Semlitsch M. Titanium alloys for hip joint replacements. Clin Mater 1987;2:1-13. 8. Simpson JP. The electrochemical behaviour of titanium and titanium alloys with respect to their use as surgical implant materials. In: Christel P, Meunier A, Lee AJC, eds. Biological and biomechanical performance of biomaterials. Amsterdam: Elsevier Science Publishers BV, 1986:63-8. 9. Bennett NE, Wang JT, Manning CA, Goldring SR. Activation of human monocyte/macrophages and fibroblasts by metal particles; release of products with bone resorbing activities. Trans 37th Ann Meet Orthop Res Soc 1991;16:188. 10. Goldring SR, Kroop SF, Petrison KK, et al. Metal particles stimulate prostglandin E2 (PGE2) release and collagen synthesis in cultured cells. Trans 36th Ann Meet Orthop Res Soc 1990;15:444. 11. Haynes DR, Rogers SD, Hay S, Pearcy MJ, Howie DW. The differences in toxicity and release of bone-resorbing mediators induced by titanium and cobalt-chromium-alloy wear particles. J Bone Joint Surg [Am] 1993;75-A:825-34. 12. Shanbhag AS, Black J, Jacobs JJ, Galante JO, Glant TT. Human monocyte response to submicron fabricated and retrieved polyethylene, Ti-6Al-4V and Ti particles. Trans 40th Ann Meet Orthop Res Soc 1994;19:849. 13. Perren SM, Geret V, Tepic M, Rahn BA. Quantitative evaluation of biocompatability of vanadium free titanium alloys. In: Christal P, Meunier A, Lee AJC, eds. Biological and biomechanical performance of biomaterails. Amsterdam: Elsevier Science Publishers BV, 1986:397. 14. Thoma E. Animal testing of the biocompatability of different titanium alloy implants. Unpublished doctoral thesis. University of Gottingen, 1990. 15. Cila E, Alpaslan AM, Melli M, Tokgozoglu AM. Prostaglandin E2 activity in the synovial-like membrane. J Arthroplasty 1994;9:67-71. 16. Dorr LD, Bloebaum R, Emmanual J, Meldrum R. Histologic, biochemical and ion analysis of tissue and fluids retrieved during total hip arthroplasty. Clin Orthop 1990;261:82-95. 17. Goldring SR, Schiller AL, Roelke M, et al. The synovial-like membrane at the bone-cement interface in loose total hip replacements and its proposed role in bone lysis. J Bone Joint Surg [Am] 1983;65-A:575-84.
VOL. 79-B, NO. 2, MARCH 1997
315
18. Andrew SM, Andrew JG, Hoyland JA, Wroblewski BM, Freemont AJ. Interleukin-6 in membrane from late loosening of hip arthroplasties. Trans 39th Ann Meet Orthop Res Soc 1993;18:243. 19. Chiba J, Rubash HE, Kim KJ, Iwaki Y. The characterisation of cytokines in the interface tissue obtained from failed cementless total hip arthroplasty with and without femoral osteolysis. Clin Orthop 1994;300:304-12. 20. Frondoza C, Jones LC, Opishinski D, et al. Periprosthetic membranes from cemented and non-cemented aseptically loose total joint replacements produce TNF and IL-1. Trans 40th Ann Meet Orthop Res Soc 1994;19:154. 21. Jiranek WA, Machado M, Jasty M, et al. Production of cytokines around loosened cemented acetabular components: analysis with immunohistochemical techniques and in situ hybridization. J Bone Joint Surg [Am] 1993;75-A:863-79. 22. Kim KJ, Chiba J, Rubash HE. In vivo and in vitro analysis of membranes from hip prostheses inserted without cement. J Bone Joint Surg [Am] 1994;76-A:172-80. 23. American Society for Testing and Materials. Standard recommended practice for surface preparation and marking of metallic surgical implants. In: Annual Book of ASTM Standards. Part 46. Philadelphia: American Society for Testing and Materials, ASTM F86-76, 1980. 24. Rogers SD, Pearcy MJ, Hay SJ, et al. A method for production and characterisation of metal prosthesis wear particles. J Orthop Res 1993;11:856-64. 25. Haynes DR, Whitehouse MW, Vernon-Roberts B. The prostaglandin E1 analogue, misoprostol, regulates inflammatory cytokines and immune functions in vitro like the natural prostaglandins E1, E2 and E3. Immunology 1992;76:251-7. 26. Kelly RW, Deam S, Cameron MJ, Seamark RF. Measurement by radioimmunoassay of prostaglandins as their methyl oximes. Prostaglandins Leukot Med 1986;24:1-14. 27. Hanks JH, Wallace JH. Determination of cell viability. Proc Soc Exper Biol and Med 1958;98:188-92. 28. Howie DW. Tissue response in relation to type of wear particles around failed hip arthroplasties. J Arthroplasty 1990;5:337-48. 29. Glant TT, Jacobs JJ. Response of three murine macrophage populations to particulate debris: bone resorption in organ cultures. J Orthop Res 1994;12:720-31. 30. Kossovsky N, Liao K, Gelman A, et al. Photon correlation spectroscopy analysis of the submicrometre particulate fraction in human synovial tissues recovered at arthroplasty or revision. In: St John KR, ed. Particulate debris from medical implants: mechanisms of formation and biological consequences. ASTM STP 1144. Philadelphia: American Society for Testing and Materials, 1992:68-74. 31. Lee J-M, Salvati EA, Betts F, et al. Size of metallic and polyethylene debris particles in failed cemented total hip replacements. J Bone Joint Surg [Br] 1992;74-B:380-4. 32. Margevicius KJ, Bauer TW, McMahon JT, Brown SA, Merritt K. Isolation and characterisation of debris in membranes around total joint prostheses. J Bone Joint Surg [Am] 1994;76-A:1664-75. 33. Shanbhag AS, Jacobs JJ, Glant TT, et al. Composition and morphology of wear debris in failed uncemented total hip replacement. J Bone Joint Surg [Br] 1994;76-B:60-7. 34. Huo MH, Salvati EA, Lieberman JR, Betts F, Bansal M. Metallic debris in femoral endosteolysis in failed cemented total hip arthroplasties. Clin Orthop 1992;276:157-68. 35. Haynes DR, Rogers SD, Howie DW, Pearcy MJ, Vernon-Roberts B. Drug inhibition of the macrophage response to metal wear particles in vitro. Clin Orthop 1996;323:316-26. 36. Scales JT. Black staining around titanium alloy prostheses: an orthopaedic enigma. J Bone Joint Surg [Br] 1991;73-B:534-6. 37. Howie DW, Vernon-Roberts B. The synovial response to intraarticular cobalt-chrome wear particles. Clin Orthop 1988;232: 244-54.
