Bone bonding to hydroxyapatite and titanium surfaces on femoral stems retrieved from human subjects at autopsy

Alexandra E. Porter^, , Punam Taak^{a, b}, Linn W. Hobbs^a, Melanie J. Coathup^c, Gordon W. Blunn^c and Myron Spector<sup>d

a</sup> Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge MA 02139, USA
^b Department of Materials Science and Metallurgy, University of Cambridge, New Museums Site, Cambridge CB3 9EU, UK
^c Royal National Orthopaedic Hospital, Stanford, Middlesex, UK
^d Department of Orthopaedic Surgery, Brigham and Women's Hospital, Harvard Medical School and Tissue Engineering, VA Boston Healthcare System, Boston, MA, USA

Received 13 November 2003;  accepted 7 December 2003.  Available online 20 February 2004.
Biomaterials
Volume 25, Issue 21 , September 2004, Pages 5199-5208

1. Abstract

The success of clinical results obtained with many hydroxyapatite (HA)-coated prosthetic designs has deflected attention from the need to extend the life of the HA coating on the device. In the current study the percentages of HA and titanium surfaces to which bone was bonded, on HA-coated and non-coated titanium femoral stems retrieved from human subjects, were evaluated. Plasma-sprayed hydroxyapatite (PSHA)-coated devices demonstrated wide variability in the percentage of the PSHA coating remaining on the stems. The coating was missing from a substantial portion of a stem after only about 6 months of implantation. The percentage of revealed metal to which bone was bonded was significantly less than the percentage of the HA coating demonstrating such bonding. The revealed metal to which bone was bonded was comparable to the same value for a separate group of non-PSHA-coated titanium stems. If HA-coatings degrade over time precipitous decline in performance may occur even after several functional years. Many ultrastructural features of the bone bonded to the HA coatings on these implants from human subjects were comparable to those found on HA-coated devices implanted in a canine model.

Author Keywords: Hydroxyapatite; Degradation; Plasma spraying; Electron microscopy

2. Article Outline

1. Introduction

2. Materials and methods

2.1. Environmental scanning electron microscopy

2.2. Transmission electron microscopy

3. Results

3.1. ESEM observations of bone bonding to HA- and non-HA-coated implant surfaces

3.2. TEM observations

4. Discussion

5. Conclusions

Acknowledgements

References

3. 1. Introduction

Plasma-sprayed hydroxyapatite (PSHA) coated implants were introduced to accelerate the rate of fixation of the implant to the surrounding bone [1]. HA coatings have been shown to be osteoconductive in that light and scanning electron microscopy has demonstrated that bone can form directly onto the PSHA surface. This direct bone apposition [2, 3 and 4] provides for mechanical coupling of the implant to the surrounding bone. The resulting bond to bone provides for significantly higher interfacial attachment strength to osseous tissue than can be achieved with non-coated metallic implants. Moreover, at the nanometer scale, scanning and transmission electron microscopy (SEM and TEM) reveal a continuum of mineral from the PSHA coating to the bone formed on the surface of the PSHA implant. Compositional and structural changes of HA may occur under physiologic conditions, but the extent of these changes and their influence on the integrity of the mechanical coating and implant prognosis are still not completely understood [5, 6 and 7].

Short-term studies comparing the degree of osseointegration (i.e., bone formation up to, but not necessarily on, the surface by light microscopy) to PSHA-coated and non-coated porous implants [1, 8, 9, 10 and 11] have indicated that PSHA coated implants increase the rate of apposition at short time periods. Despite their clinical success in the short term, there has been some concern that in the long term, the PSHA coating dissolves and detaches exposing the metal beneath [12, 13, 14 and 15] and that this may have adverse effects on interfacial bone apposition to the implant and its mechanical stability [16]. An 8-year clinical study found that the prosthesis survival rate was initially higher for HA-coated implants, but decreased significantly below that of titanium plasma-sprayed implants after 4 years [17]. This may be a reflection of the loss of the PSHA coating and an inferior bond of bone to the underlying metallic substrate.

Another important complication of the breakdown of the PSHA coating is that the resulting particulate material may elicit a phagocytic response by macrophages [18] resulting in a loss of bone apposition to the implant. It has also been proposed that these HA particulate chips contribute to the accelerated wear of metal-on-polyethylene articulation in total joint replacement [12 and 13]. In total hip arthroplasty, failed HA-coated femoral stems displayed areas of coating degradation and separation, and embedded coating particles were found in the polyethylene acetabular inserts. Scratches on the femoral head, produced by particulate HA debris embedded in the acetabular cups [19, 13 and 20], contributed to the accelerated abrasive wear of the polymer.

In light of the recognized loss of coating from PSHA-coated implants, the supposition in support of their continued clinical use has been that as PSHA coatings degrade from the titanium substrate, bone forms on the surface of the exposed metal so as to maintain fixation. In an effort to obtain additional data to support, or reject, this supposition the current study analyzed PSHA-coated and a group of non-coated femoral components from hip arthroplasty retrieved at autopsy. The objective of this work was to evaluate the percentages of the PSHA and exposed metal surfaces to which bone was bonded, using SEM. In order to advance our understanding of the mechanism of bone bonding to PSHA, a second objective was to examine the ultrastructural features adjacent to PSHA coated devices by TEM. These findings from the prostheses implanted in human subjects were compared to observations made in a previous study investigating the ultrastructure of the PSHA-bone interface predisposing to bone bonding in a canine model [21].

4. 2. Materials and methods

The six implants retrieved from human subjects, used in this study, were drawn from a larger population of patients treated for a fractured femoral neck with a Bimetric hemi-arthroplasty (Biomet, UK). Each patient randomly received a plasma-sprayed titanium stem with or without an HA coating on the proximal half of the stem. The PSHA coating had an average crystallinity >85% and an average thickness of 50 
m. The implants and associated femora were collected at autopsy: 3 HA-coated titanium stems (173, 261 and 660 days, post-op) and 3 non-coated titanium stems (40, 650 and 1056 days). On retrieval, the specimens were fixed in 10% formaldehyde solution and excess soft tissue was removed. The proximal region of each femoral component (i.e., the coated area being investigated) was cut using an Exotom cut off machine, into a proximal region (F1), a mid region (F2) and a distal region (F3), each representing a third of the coated area. The specimens were then embedded in polymethyl methacrylate (PMMA). For the purpose of this study, only F3 sections were used.

2.1. Environmental scanning electron microscopy

All samples were machine ground and polished using a Motopol 2000, Bueler Model. Samples were ground on 1200 grit silicon carbide paper and were polished using 5 
m alumina paste then 0.06 
m colloidal white solution. The polished samples were subsequently sputter coated with a thin layer of carbon.

Samples were viewed and energy dispersive X-ray (EDX) microanalysis was performed in the ElectroScan E3 environmental SEM (ESEM) under high vacuum at 20 kV in backscattered electron mode (backscattered electron imaging, BEI). Samples were kept in a desiccator overnight before ESEM observation. At least 10 micrographs for any given sample were taken all at the same magnification, between 122 and 159×. Higher magnifications were used to confirm the apposition of bone on the coating (i.e., no intervening non-mineralized tissue).

To determine the degree of osseointegration to the coatings and percentage of HA coating remaining apposed to the titanium implant surface, for each sample a 50 
m square grid was placed over a randomly selected area of the interface. For 100 squares along the interface, the number of squares in which (a) bone was apposed to the HA coating, (b) was apposed to the porous titanium, or (c) was not apposed to the coating, was counted. Bone apposition to the interface was defined by there being a continuum of mineral from the bone to the interface. In addition, for the HA coated samples, the number of squares in which the HA remained apposed to the titanium was measured. From these measurements, the percentage osseointegration to the HA coating and to the porous titanium coating was calculated, by determining the percentage of counted squares in which bone was opposed to the HA/plain porous coating. Finally, Student's t-test was used to determine: (a) significant differences in osseointegration between the HA-coated and non-HA coated areas of the interface (for both HA-coated and plain porous implants), (b) significant differences in osseointegration to the exposed titanium surface and the HA remaining on the HA-coated implants on the same stems, and (c) significant differences in osseointegration between the exposed titanium on the HA-coated and non-coated implants.

2.2. Transmission electron microscopy

Specimens that were 3 mm×1 mm×1 mm in size, containing the bone-HA-titanium interface, were sawn from the segments of the prostheses. Samples were rinsed in neat ethanol for 5 min and then twice in propylene oxide (Ted Pella, Reading, CA) for 5 min each. They were then infiltrated with Spurr's resin (Polysciences and Ted Pella) over several days, as follows. The samples were agitated in 1:3, 1:2, 1:1, and 2:1 solutions in propylene oxide for 23 h each and then vacuum infiltrated with neat Spurr's resin for 23 h. They were cured in fresh Spurr's resin for 23 h at 60°C. Spurr's resin was prepared with 10 g VCD (vinyl cyclohexene dioxide), 4.5 g DER-736 (diglycidyl ether of polypropylene glycol), 26 g nonenyl succinic anhydride (NSA) and 0.4 g dimethylaminoethanol (DMAE). Extra resin was stored and covered in the fume-cupboard for no more than 48 h.

Silver to gold sections (70-90 nm thickness) cut by an ultramicrotome were collected on Parlodion coated grids (Ted Pella) and coated with a thin layer of evaporated carbon. For ultramicrotomy, a 45° diamond knife was used. Sections were cut so that the titanium was cut first and then the bone. It was found that a higher success rate was achieved by cutting away as much titanium as possible with a razor blade, before cutting with the diamond knife.

Additional ultramicrotomed sections were stained with uranyl acetate and lead citrate. The grids were placed sample side down on individual drops of 4% uranyl acetate in 30% ethanol in water for 20 min in a Petrie dish in a light-tight box. They were then rinsed 10 times each in each of three beakers of distilled water and allowed to dry. The grids were then placed sample side down on individual drops of 2% lead citrate in water for 7 min and rinsed in a similar fashion. To prevent precipitation of the lead from solution, a few NaOH pellets were placed in the Petrie dish to neutralize carbonic acid (formed from carbon dioxide). In order to achieve a higher success rate, care was taken to ensure that the slices were immersed in the lead citrate solution for 7 min only.

TEM and selected electron diffraction (SAED) were performed in the Jeol 2000 FX operated at 200 kV.

5. 3. Results

3.1. ESEM observations of bone bonding to HA- and non-HA-coated implant surfaces

Titanium, HA coating and bone could clearly be distinguished by BEI (Fig. 1 and Fig. 2) and confirmed by EDX. Bone appeared to grow along the surface and into the pores of the HA coating (e.g., Fig. 1a) at all time periods. Of interest was that a large percentage of the titanium surface of the PSHA-coated stems did not contain HA coating ( Fig. 1b). In some regions of these devices bone could be seen apposed to the titanium surface.

(31K)

Fig. 1. (a) ESEM section displaying bone bonding to the PSHA coating of a prosthesis obtained after 173 days in vivo. (b) Higher magnification ESEM micrograph showing bone growing contiguous to the HA coating.

(57K)

Fig. 2. (a) ESEM micrograph illustrating that along the periphery of the HA-coated stems the bone bonded to the HA did not display trabecular bridges to the surrounding cancellous bone 261 days in vivo. (b) ESEM image, illustrating remodelling trabecular bone apposing an HA coated titanium hip replacement at 261 days in vivo. Bone apposition defined as a continuum of mineral from the synthetic HA to the surrounding bone mineral. Images taken in backscattered electron mode.

It was of interest that along the periphery of the HA-coated stems, the bone bonded to the HA did not display trabecular bridges to the surrounding cancellous bone (Fig. 2a). Regions of the coating to which bone was bonded displayed a continuum of mineral from the PSHA coating into the mineralized bone matrix ( Fig. 2b). These images guided the selection of specimens for TEM. In regions where bone was bonded to the surface, the lacunae of osteocytes were observed about 3 
m from the coating. These osteocytes were derived from the precursor osteoblasts that initially attached to the biological layer that formed on the HA coatings and were responsible for bone bonding. Bone matrix could also be found in small voids within the PSHA coating (Fig. 2b).

For the titanium porous coating, after 40 and 650 days, there were large regions of fibrous tissue interface present. The bone surrounding the implant retrieved after 40 days displayed fragmented trabecular bone that was left over likely from surgery (Fig. 3). In the region adjacent to the fragmented trabecular bone, the bone has a scalloped surface typical of that seen in areas that had recently undergone osteoclastic resorption. After 1056 days, the morphology of bone apposed to the porous titanium (Fig. 4a) was similar to that apposed to the HA-coated titanium implant and bone trabeculae appeared to grow along the titanium implant surface (Fig. 4b). In many regions, lamellae could be seen in the trabecular bone ( Fig. 4b).

(48K)

Fig. 3. ESEM micrograph of a plain porous coated specimen retrieved 40 days post implantation. Illustrating fragmented trabecular bone remaining from surgery.

(36K)

Fig. 4. (a) Higher magnification ESEM micrograph showing bone growing along the plain porous implant interface, 1056 days post implantation. In many regions, lamellae could be seen in the trabecular bone.

Quantitative analysis of coating degradation showed the percentage of HA coating remaining adherent to the HA-coated implant after different time periods of implantation (Fig. 5). While there were only 3 data points, it appeared unlikely that there was any clear correlation between the percentage of remaining HA coating and implantation time (Fig. 5).

(9K)

Fig. 5. Percentage of HA coating remaining apposed to the titanium implant surface after various time periods.

Evaluation of the percentage of bone apposed to the implant revealed differences in the regions with and without the HA coating (Fig. 6a). There was a clear indication of more (2-fold) bone bonded to the HA coating versus the exposed metal (Fig. 6b). A paired Student's t-test showed the difference in bonding in the 2 groups to be statistically different (p=0.0103).

(26K)

Fig. 6. (a) Comparison of the degree of osseo-integration to HA-coated regions and regions where the titanium was exposed on HA coated stems after various time periods. (b) Comparison of osseointegration to HA-coated regions and regions where the titanium was exposed, n=3. Error bars indicate the standard error.

The percentage of the surface of the non-HA coated implants to which bone was found to be apposed ranged from less than 20% to 35% (Fig. 7). The bone bonding to the titanium stems (i.e., not the HA-coated stems) was comparable to the bonding to the titanium regions exposed on the HA-coated stems: 24±5%, mean±standard error of the mean, for the apposition to the metal on the HA-coated stem versus 21±14% for the non-HA-coated devices (Student's unpaired t-test, p=0.44). Although for the porous implant it appeared that there was a trend toward an increase in bone apposition between 650 and 1056 days (Fig. 7).

(11K)

Fig. 7. Percentage of titanium implant surface apposed by bone after various time periods post-implantation.

3.2. TEM observations

TEM revealed numerous crystallites in the region adjacent to the HA coating and in the interstices among the HA particles forming the coating after 660 days (Fig. 8a-c). The presence of the HA grains was confirmed by spots characteristic of single crystals seen in SAED mode. In some regions the crystallites were plate-like in morphology, in other areas they are more needle-like in morphology. The plate-like morphology was indicated by the lower contrast images of the features when observed en face (ef), and the sharper, higher contrast image of the crystallites when viewed on edge (oe) (Fig. 8a). The size of the crystallites was approximately 3-7×30-40×60-100 
m. In the interstices among the HA particles forming the coating the crystallites seemed to comprise a nanocrystalline phase (nc) (Fig. 8a and b) and in other areas they appeared to comprise fused globular aggregates of apatite crystallites (ga) ( Fig. 8b). The crystallites appeared to be denser on the HA grains, whereas, in contrast, some areas in the surrounding matrix were devoid of HA ( Fig. 8a and b). Selected area diffraction from the interface region containing the nanocrystalline phase exhibited polycrystalline diffraction rings with spacings consistent of apatite.

(58K)

Fig. 8. (a) TEM micrograph of interface between the HA coating and embedded biological material opposing the coating 660 days post-implantation. Showing globular aggregates (ga) of apatite crystallites and apatite crystallites viewed en face (ef) and on edge (oe). (b) TEM micrograph of tissue opposing the HA coating 660 days post-implantation, showing the nanocrystalline phase (nc). (c) TEM micrograph of tissue opposing the HA coating 660 days post-implantation stained with uranyl acetate and lead citrate 660 days post implantation. Showing collagen fibers (cf) with a characteristic 64 nm banding pattern and an intervening apatite layer opposing the coating.

Staining of the HA-coated sample confirmed the presence of collagen fibres (cf) next to the implant surface, with a characteristic 64 nm banding pattern (Fig. 8c). There appeared to be an intervening layer of fused apatite crystallites between the implant surface and the collagen fibres. Aligned HA platelets were observed to be directly associated with these collagen fibrils.

6. 4. Discussion

Uncertainty remains about the longevity of PSHA-coated prostheses. Concerns have arisen that in the long term the PSHA coating dissolves and detaches exposing the metal beneath [12, 13, 14 and 19] and that this may have adverse effects on interfacial bone apposition to the implant and on its mechanical stability. In the current study, PSHA-coated devices retrieved from human subjects at autopsy demonstrated wide variability in the percentage of the PSHA coating remaining on the stems. The coating was missing from a substantial portion (more than 70% of the perimeter) of a stem after only about 6 months of implantation. All of the stems demonstrated a loss of the PSHA from more than 50% of the surface. It was not possible in this investigation to determine the chronology of the loss of the PSHA coating or to determine what factors are determinants of this loss. It is of some interest that these stems were likely providing adequate function despite the loss of the PSHA because none of the devices were removed due to loosening. While this finding appears to support the supposition that adequate stability of the PSHA-coated prosthesis can be maintained despite the loss of the coating, the longer-term prognosis for the devices may have been affected by the loss. Clearly, additional studies will be required to determine how the loss of the PSHA affects the long-term performance of the implant.

As expected, a high percentage of the surface of the PSHA coating remaining on the stems displayed bone bonding. A notable finding of this study was that the percentage of the underlying metal surface (revealed due to the loss of the PSHA) to which bone was bonded was significantly less than the percentage of the HA coating demonstrating such bonding. That so much less of the metal was bonded to bone would question the long-term stability that could be expected of the stem after the PSHA coating is lost. It was of interest that the percentage of the revealed metal to which bone was bonded was comparable to the same value for a separate group of non-PSHA-coated titanium stems. These findings underscore the importance of determining the factors leading to HA coating loss from the implant.

The current study employed the use of samples selected from a large inventory of human autopsy specimens. The autopsy samples investigated in this study were all chosen from the same site to keep consistency of the region of the hip studied and the six samples chosen were the only samples for which material was left remaining. It would have been useful to have more data about the plasma-spraying technique, the physical and chemical characteristics of the coatings and also the percentage of titanium implant surface with HA after plasma spraying, but this information was not available. The compelling findings of the current study can serve as rationale for continuing this line of investigation, despite the challenge in obtaining PSHA-coated devices from autopsy.

It is well established that improved osseointegration arises due to the osteoconductive nature of the HA coating [1, 8, 9, 10 and 11]. The current study demonstrates the tenacity of bone bonding to HA-coated implant surfaces in human subjects, several of whom were of advanced age. Extensive bone growth was observed along the surface and even the pores and voids of the HA coating. In comparison, large regions of fibrous tissue interface were found along the exposed titanium coating. All these observations suggest that the HA coating augments processes of bone bonding to the implant enabling faster rehabilitation of the patient.

That the clinical results obtained with many HA-coated prosthetic designs continue to be so good has deflected attention from the need to extend the life of the HA coating on the device. There has been some concern that the HA resorbs over time, exposing the metal beneath and that this may have an adverse effect on bone bonding to the interface [16]. In all implants studied, the percentage bone bonding to the exposed metal on the HA-coated stems was significantly lower than that to the HA coating. Furthermore, there was no enhancement of bone-bonding to the underlying titanium on the HA stems compared to the uncoated titanium stems. There are several reports of PSHA coating dissolution and degradation over time [12, 13, 14 and 19]. Low percentage bone bonding to the exposed metal on the HA-coated stems questions the durability of attachment of HA if lost. If the fixation of bone to exposed titanium is inferior to the bone bonding to the HA coating, precipitous decline in performance may occur even after several functional years. An additional concern with HA coatings is that although HA enhances bone bonding, trabecular bridging to the surface of cancellous bone was found to be sparse. This observation further questions the mechanical integrity of bone bonding to the HA and the long-term performance of the implant.

A notable finding in the present study was the presence of bony fragments, probably remaining from surgery, that were observed remaining in the region adjacent to the porous titanium coating after 40 days implantation. These fragments appear to provide a scaffold for new bone formation. Interestingly, after 1094 days there was an increase in bone apposition to the plain porous-coated implant and the morphology of the bone opposed to the implant was similar to that observed opposed to the HA-coated implants at all time periods. The bone fragments left over from surgery have undergone osteoclastic resorption and the new remodelled bone grown up to and along the interface.

There has been some debate as to whether the ultrastructural features in animal model systems mimic those observed in humans. A previous short term (6 h-10 days) study imaging the interface between bone and a PSHA coating in a canine model, revealed clusters of plate-like apatite crystallites at 6 h and nodular aggregates of apatite crystallites after 3 h in vivo [21]. Staining revealed invasion of the implant interface by aligned collagen fibrils after 10 days in vivo. Similar ultrastructural features were observed in a longer-term (6 and 12 week) ovine study investigating osseointegration to sintered granules of pure HA and silicon-substituted HA implants [22]. Many ultrastructural features of the bone bonded to the HA coatings on these implants from human subjects were comparable to those found on HA-coated devices implanted in canine models [21]. TEM revealed numerous plate-like and needle-like apatite crystallites nucleated on the PSHA coating and in the interstitial pore spaces within the coating after 660 days. The voids in which a calcium and phosphate ion concentration is allowed to increase to saturation, thereby initiating the nucleation of apatite. Staining revealed collagen fibrils around the PSHA implant. These findings confirm the validity of a canine animal model for investigating ultrastructural features of bone bonding to PSHA coatings.

7. 5. Conclusions

The success of clinical results obtained with many HA-coated prosthetic designs has deflected attention from the need to extend the life of the HA coating on the device. The current results demonstrate that fixation of bone to exposed titanium is inferior to the bone bonding to the HA coating. If HA coatings degrade over time precipitous decline in performance may occur even after several functional years. Many ultrastructural features of the bone bonded to the HA coatings on these implants from human subjects were comparable to those found on HA-coated devices implanted in a canine model and sintered HA granules implanted in an ovine model.

8. Acknowledgements

The US National Science Foundation, Division of Bioengineering and Environmental Systems, through Grant 9904046, as the principal sources of funding; and as secondary sources the Armourers and Brasiers for support of AEP; The BioKinetics Foundation; Biomet Merck; and Darwin College Cambridge for supplementary support of AEP.

9. References

1. H. Oonishi, M. Yamamoto, H. Ishimaru, E. Tsuji, S. Kushitani, M. Aono and Y. Ukon, The effect of hydroxyapatite coating on bone growth into porous titanium alloy implants. J Bone Surg Br 71 2 (1989), pp. 213-216. Abstract-MEDLINE | Abstract-EMBASE  

2. T.S. Golec and J.T. Krauser, Long-term retrospective studies on hydroxyapatite-coated endosteal and subperiosteal implants. Dent Clin North Am 36 (1992), pp. 39-65. Abstract-MEDLINE  

3. R.S. Corpe, D.E. Steflik, R.Y. Whitehead, M.D. Wilson, M.D. Young, T.R. Young and C. Jaramillo, Correlative experimental animal and human clinical retrieval evaluations. Crit Rev Biomed Eng 28 3-4 (2000), pp. 395-398. Abstract-MEDLINE | Abstract-Elsevier BIOBASE | Abstract-EMBASE  

4. C.K. Chang, J.S. Wu, D.L. Mao and C.X. Ding, Mechanical and histological evaluations of hydroxyapatite-coated and noncoated Ti6Al4 V implants in tibia bone. J Biomed Mater Res 56 1 (2001), pp. 17-23. Abstract-MEDLINE | Abstract-EMBASE   | Full Text via CrossRef

5. R.Y. Whitehead, L.C. Lucas and W.R. Lacefield, The effect of dissolution on plasma sprayed hydroxyapatite coatings on titanium. Clin Mater 12 (1993), pp. 31-39. Abstract-MEDLINE | Abstract-Compendex | Abstract-EMBASE  

6. R.D. Bloebaum, G.E. Lundeen, K.N. Bachus, I. Ison and A.A. Hoffmann, Dissolution of particulate hydroxyapatite in a macrophage organelle model. J Biomed Mater Res 40 (1991), pp. 104-114. Abstract-Compendex  

7. D.E. Macdonald, F. Betts, M. Stranick, S. Doty and A.L. Boskey, Physicochemical study of plasma-sprayed hydroxyapatite-coated implants in humans. J Biomed Mater Res 54 (2001), pp. 480-490. Abstract-MEDLINE | Abstract-EMBASE   | Full Text via CrossRef

8. A. Moroni, V.L. Caja, E.L. Egger, L. Trinchese and E.Y. Chao, Histomorphometry of hydroxyapatite coated and uncoated porous titanium bone implants. Biomaterials 15 11 (1994), pp. 926-930. Abstract

9. K.A. Thomas, Hydroxyapatite coatings. Orthopedics 17 3 (1994), pp. 267-278. Abstract-MEDLINE | Abstract-EMBASE  

10. M. Svehla, P. Morberg, B. Zicate, W. Bruce, D. Sonnabend and W.R. Walsh, Morphometric and mechanical evaluation of titanium implant integration: comparison of five surface structures. J Biomed Mater Res 51 1 (2000), pp. 15-22. Abstract-MEDLINE | Abstract-Compendex   | Full Text via CrossRef

11. P. Ducheyne, L.L. Hench, A. Kagan, M. Martens, A. Bursens and J.C. Mulier, Effect of hydroxyapatite impregnation on skeletal bonding of porous coated implants. J Biomed Mater Res 14 (1988), pp. 225-237.

12. R.D. Bloebaum, D. Beeks, L.D. Loor, C.G. Savory, J.A. DuPont and A.A. Hoffmann, Complications with hydroxyapatite particulate separation in total hip arthroplasty. Clin Orthop 298 (1994), pp. 19-26. Abstract-MEDLINE | Abstract-EMBASE  

13. R.D. Bloebaum and J.A. Dupont, Osteolysis from a press-fit hydroxyapatite-coated implant. A case study. J Arthroplasty 8 2 (1993), pp. 195-202. Abstract-EMBASE | Abstract-MEDLINE  

14. I. Baltag, K. Watanabe, H. Kusakari, N. Taguchi, O. Miyakawa, M. Kobayashi and N. Ito, Long-term changes of hydroxyapatite-coated dental implants. J Biomed Mater Res (Appl Biomat) 53 (2000), pp. 76-85. Abstract-MEDLINE | Abstract-Compendex | Abstract-EMBASE   | Full Text via CrossRef

15. T.W. Bauer, R.C.T. Geesink, R. Zimmerman and J.T. McMahon, Hydroxyapatite-coated femoral stems: histological analysis of components retrieved at autopsy. J Bone Jt Surg 73-A (1991), pp. 1439-1452. Abstract-EMBASE | Abstract-MEDLINE  

16. F.B. Bagambisa, U. Joos and W. Schilli, Mechanisms and structure of the bone between bone and hydroxyapatite ceramics. J Biomed Mater Res 27 (1993), pp. 1047-1055. Abstract-MEDLINE | Abstract-Compendex | Abstract-EMBASE  

17. S.L. Wheeler, Eight year clinical retrospective study of titanium plasma-sprayed and hydroxyapatite-coated cylinder implants. Int J Oral Maxillofac Implants 11 (1996), pp. 340-350. Abstract-MEDLINE  

18. Y. Harada, J.T. Wang, V.A. Doppalapudi, A.A. Willis, M. Jasty, W.H. Harris, M. Nagase and S.R. Goldring, Differential effects of different forms of hydroxyapatite and hydroxyapatite/tricalcium phosphate particles on human monocyte/macrophages in vitro. J Biomed Mater Res 31 1 (1996), pp. 19-26. Abstract-EMBASE | Abstract-MEDLINE | Abstract-Compendex   | Full Text via CrossRef

19. T.W. Bauer, B.N. Stulberg, J. Ming and R.G.T. Geesink, Uncemented acetabular components: histologic analysis of retrieved hydroxyapatite-coated and porous implants. J Arthroplasty 8 (1993), pp. 167-177. Abstract-EMBASE | Abstract-MEDLINE  

20. J.P. Collier, V.A. Surprenant, M.B. Mayor, M. Wrona, R.E. Jensen and H.P. Surprenant, Loss of hydroxyapatite coating on retrieved, total hip components. J Arthroplasty 8 (1993), pp. 195-202.

21. A.E. Porter, L.W. Hobbs, V. Benezra Rosen and M. Spector, The ultrastructure of the plasma-sprayed hydroxyapatite-bone interface predisposing to bone bonding. Biomaterials 23 (2002), pp. 725-733. SummaryPlus | Full Text + Links | PDF (557 K)

22. Porter AE, Patel N, Skepper JN, Best SM, Bonfield W. Effect of silicate substitution in sintered hydroxyapatite on remodelling processes at the bone-implant interface. Biomaterials, in press.

Corresponding author. Tel.: [phone]; fax: [phone]

---