Biomaterials 24 (2003) 1309–1316

Detachment of titanium and fluorohydroxyapatite particles inunloaded endosseous implants

D. Martinia,*, M. Finib, M. Franchia, V. De Pasqualea, B. Bacchellia, M. Gamberinia,A. Tintic, P. Taddeic, G. Giavaresib, V. Ottania, M. Raspantid, S. Guizzardie, A. Ruggeria

a Istituto di Anatomia Umana Normale, Via Irnerio 48, 40126 Bologna, Italyb Serv. di Chirurgia Sperim., Ist. di Ric. Codivilla-Putti, IOR Via di Barbiano 1/10, Bologna, Italy

c Dip. di Biochimica ‘‘G. Moruzzi’’, Via Belmeloro 8/2, 40126 Bologna, Italyd Lab.di Morfologia Umana ‘‘Luigi Cattaneo, Via Montegeneroso 71, 21100 Varese, Italy

e Dip.Medicina Sperimentale- Sez. di Istologia, Via Volturno 39, 43100 Parma, Italy

Received 23 April 2002; accepted 28 August 2002

Abstract

The shape, surface composition and morphology of orthopaedic and endosseous dental titanium implants are key factors to

achieve post-surgical and long-term mechanical stability and enhance implant osteointegration.

In this study a comparison was made between 12 titanium screws, plasma-spray-coated with titanium powders (TPS), and 12

screws with an additional coating of fluorohydroxyapatite (FHA-Ti). Screws were implanted in the femoral and tibial diaphyses of

two mongrel sheep and removed with peri-implant tissues 12 weeks after surgery.

The vibrational spectroscopic, ultrastructural and morphological analyses showed good osteointegration for both types of

implants in host cortical bone. The portion of the FHA-Ti implants in contact with the medullary canal showed a wider area of

newly formed peri-implant bone than that of the TPS implants.

Morphological and EDAX analyses demonstrated the presence of small titanium debris in the bone medullary spaces near the

TPS surface, presumably due to the friction between the host bone and the implant during insertion. Few traces of titanium were

detected around FHA-Ti implants, even if smaller FHA debris were present.

The present findings suggest that the FHA coating may act as a barrier against the detachment of titanium debris stored in the

medullary spaces near the implant surface.

r 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Titanium implants; Coating; Fluorohydroxyapatite; Scanning electron microscopy; Vibrational spectroscopy

1. Introduction

Post-surgical and long-term mechanical stability is thefundamental requirement for the osteointegration oforthopaedic and endosseous dental implants. Therefore,the shape, chemical composition and surface morphol-ogy of the implants, as well as the surgical techniques toachieve bone anchoring, are being increasingly studied.

Of the various implant materials, titanium is aparticularly suitable metal for orthopaedic and endoss-eous dental implants on account of its good mechanical

properties and biocompatibility [1]. The host bonetissue, in fact, grows in close contact to the metal, thusensuring implant osteointegration.

The role played by the surface morphology in terms ofimplant stability has been highlighted by many studies[2,3]. A rough implant surface appears to be particularlysuitable for primary implant stability as compared to asmooth implant surface. Moreover, the microcavities onthe titanium implant surface are quickly colonised byblood cells that promote the ingrowth of new bonetissue [4].

To enhance and accelerate osteointegration, titaniumimplants can also be coated with other similar materials,such as bioapatites, which are similar to natural bone interms of mineral content. These coatings have been

*Corresponding author. Tel.: +39-051-242217; fax: +39-051-2091-

659.

E-mail address: martini@biocfarm.unibo.it (D. Martini).

0142-9612/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved.

PII: S 0 1 4 2 - 9 6 1 2 ( 0 2 ) 0 0 5 0 8 - 2

reported to increase bone apposition and implantfixation rates by enhancing chemical bonding at theimplant/new bone interface [3,5–7]. Among bioactiveapatites, fluorinated apatites have shown in vivo the bestchemical stability, good integration with bone andnegligible long-term degradation [8,9].

Some authors have investigated the biological inter-action between peri-implant tissues and implant surfacein relation to the material used. Attention has beenmainly focussed on the mechanisms of dissolution of animplant material in the host tissues. Regarding titanium,nitric acid passivation has been demonstrated to lead totitanium oxide dissolution and to result in a thinner andless stable oxide layer [10,11]. A porous-like structureprovides more interstitial spaces for higher serum attackand would account for the increased levels of ion releaseover the polished samples [1].

The dissolution of metal ions decreases by 50% uponcoating with hydroxyapatite (HA), thus providing akind of physical barrier to metal dissolution [12].Moreover, an increase in HA coating thickness hasbeen suggested to reduce metal ion release [13].However, HA dissolution has also been observed inaqueous and Ringer’s solutions [14]. HA and fluoroa-patite coating dissolution has been demonstrated tooccur in the presence of bovine serum albuminesolution, although fluoroapatite coating has shown aslightly lower dissolution rate [3].

Titanium screws with different coatings (plasma-sprayed powders of titanium (TPS) and the same TPSwith adjunctive fluorohydroxyapatite coating (FHA-Ti))were implanted in the tibia and femur of mongrel sheepto study the biological interactions between peri-implanttissues and implant surface 12 weeks after surgery.

2. Materials and methods

Twenty-four tapered cylindrical titanium screws,4.5 mm in outer diameter and 28 mm in length (Biocoat-ings, Fornovo Taro, Italy), were coated with differentpowders by plasma spraying: 12 implants were coatedwith TPS (granule size ranging from 290 to 310 mm)(TPS) and another 12 with TPS+FHA powder (granulesize ranging from 3 to 80 mm) (FHA-Ti). All of theprocedures involving the sheep were performed accord-ing to the ethical guidelines on animal experimentationby the University of Bologna.

The implants were implanted bilaterally in thefemoral and tibial diaphyses of two mongrel sheep(3–4 years old) anaesthetised according to a standar-dised protocol: premedication with intramuscular injec-tion of 10 mg/kg b.w. ketamine (Ketavet 100,Farmaceutici Gellini, SpA, Aprilia, Italy), 0.3 mg/kgb.w. xylazine (Rompun, Bayer AG, Leverkusen, Ger-many) and subcutaneous injection of 0.0125 mg/kg b.w.

atropine sulphate; induction with intravenous injectionof 6 mg/kg sodium thiopentone (2.5% solution,Pentothal, Hoechst AG, Germany); maintenance withO2, N2O and 1–2.5% halothane under assisted ventila-tion (Servo Ventilator 900 D, Siemens, Germany). A3.5 mm-diameter drill was used to predrill three holes ineach diaphysis, which were tapped with a 4.5 mm device.Three screws were then inserted into the diaphysealcortex of the tibiae and femora and tightened to the finalinsertion torque of 2.270.1 N m. The TPS and FHA-Tiscrews were implanted on the right and on the left sideof each animal, respectively.

Antibiotics (cefalosporin, 1 g/day for 5 days) andanalgesics (ketoprofen 500 mg/day for 3 days) wereadministered post-operatively. The sheep were main-tained in single boxes for 20 days after each surgery, andwere then returned to external housing conditions. After12 weeks, the animals were euthanised with intravenousadministration of Tanax (Hoechst, Frankfurt am Main,Germany) under general anaesthesia. The implants withthe surrounding peri-implant tissues were then removedand specimens containing implants were sawed forhistological, morphological and spectroscopic analyses.

The diaphyseal bone segments containing the screwswere isolated and fixed in 10% formalin-bufferedsolution (pH 7.2). Some samples were then dehydratedin ethanol and embedded in methyl methacrylate.Thirty–fifty mm thick sections were obtained by sawingand grinding operations (Saw and Grinding, Remet,Bologna, Italy), stained with toluidine blue and acidfuchsin and finally observed with a light microscope.

Some unstained methyl methacrylate-embedded sec-tions were also mounted on stubs with carbon bioadhe-sive film, carbon-coated with an Emitech sputter-coaterand observed with a Philips XL-30 FEG scanningelectron microscope (Philips XL30FEG, Eindhoven,Holland) fitted with secondary electron (SE) and back-scattered electron (BSE) probes, and with X-raydispersive spectroscopy (EDAX), at voltages of10–12 kV.

Vibrational Raman and infrared (IR) spectroscopieswere used for the chemico-physical analysis of theimplanted TPS- and FHA-Ti-coated screws. Micro-Raman spectra were obtained in a non-destructive wayusing a Jasco NRS-2000C instrument with a � 20magnification microscope. All the spectra were recordedin backscattering conditions with 5 cm�1 spectralresolution using the 488 nm line (Innova Coherent 70)with power of ca. 20 mW. The detector was a 160 Kfrozen CCD from Princeton Instruments Inc. IR spectrawere performed using a Jasco Model FTIR 300EFourier transform spectrophotometer with spectralresolution of 4 cm�1. The spectra were obtained fromKBr pellets (about 0.5% w/w) containing finely groundpowders removed from the surface of the implantedscrews.

D. Martini et al. / Biomaterials 24 (2003) 1309–13161310

3. Results

All animals showed no evidence of complicationduring surgery and survived the whole post-surgicalperiod of 12 weeks without any infection. All screwswere used for investigations after explantation.

3.1. Vibrational spectroscopic analysis

The micro-Raman spectra recorded after 12 weeks ina non-destructive way on the screws implanted in thesheep tibia (Fig. 1) were exactly the same as thoserecorded on the corresponding screws in the femur (datanot reported).

The micro-Raman spectrum of the material in contactwith the TPS screw (Fig. 1a) showed the typical bonetissue bands, and therefore bands from organic(1660 cm�1 =Amide I of collagen; 1450 cm�1=dCH2;1250 cm�1=Amide III of collagen; 1003 cm�1=partiallyderived from phosphorylated amino acids) [15] andinorganic (1003 cm�1=partially derived from HPO2�

4

ions; 961, 592, 584, 450 and 434 cm�1=PO3�4 ions; 1070

and 1045 cm�1=PO3�4 and CO2�

3 ions) [16–18] bonecomponents could be observed. The micro-Ramanspectrum of the material in contact with the FHA-Ti-coated screws (Fig. 1b) revealed the typical bandsof the FHA coating before implantation (in particular,the 964 cm�1 band) (Fig. 1c), in addition to thosepreviously mentioned for organic and inorganic bonecomponents.

The IR spectra (Fig. 2) of the material removed fromthe implanted screws showed a trend similar to that ofthe Raman spectra. The spectrum of the material

removed from the TPS-coated screw (Fig. 2a) revealedthe typical bone tissue bands, and therefore bandsfrom organic (1650 cm�1=Amide I of collagen;1540 cm�1=Amide II of collagen; 1455 cm�1=dCH2)[19] and inorganic (1155, 1110, 575 cm�1=HPO2�

4 ions;1035, 962, 604, 564 cm�1=PO3�

4 ions [16,17], 1417, 874,712 cm�1=CO2�

3 ions of a B-type carbonate apatite)[20] bone components could be observed. The IRspectrum of the material removed from the FHA-Ti-coated screw (Fig. 2b) showed the typical coating bands(in particular, those at 1092, 1045 and 570 cm�1)(Fig. 2c), in addition to those previously mentioned fororganic and inorganic bone components.

It is interesting to note that the IR spectra in Figs. 2aand b show a band at 1385 cm�1. Such a band is notattributable to bone components and has been observedin the spectra of carbonate apatites containing impu-rities of the NO�

3 ions [17,21,22]. The presence of the1385 cm�1 band in the spectra of Fig. 2 is thereforeascribable to the NO�

3 component released by thenitrurised microtome saw used for histological samplespreparation.

3.2. Histological and ultrastructural analysis

Twelve weeks after surgery, histological and ultra-structural analysis of the methyl methacrylate-em-bedded samples showed evidence of newly formedbone in both coatings.

Both types of implants appeared to be osteointegratedand showed a newly-mineralised tissue in close contactwith almost the entire implant surface, thus filling thespace between the TPS and FHA-Ti implant surfaces

Fig. 1. Micro-Raman spectra of: (a) the material on the implant surface of a TPS-coated screw; (b) the material on the implant surface of a FHA-Ti-

coated screw; and (c) FHA coating before implantation.

D. Martini et al. / Biomaterials 24 (2003) 1309–1316 1311

and the host cortical bone (Figs. 3 and 4). The portionof TPS surface crossing the whole medullary canalshowed a small amount of primary new bone (Fig. 5),while the analogous portion of FHA-Ti surface wascovered by a thicker well-distributed layer of new bone(Fig. 6).

The ultrastructural analysis of the TPS samplesimplanted in the cortical tibia and femur showed thenewly formed peri-implant bone to be in tight contactwith the Ti surface with no spaces at the interface(Fig. 7a). The newly formed bone grew in the coatingmicropores of the FHA-Ti implants and appeared to bein continuity with the implant surface (Fig. 8a). No

evidence of detachment from the bulk titanium screwswas found in either the titanium or FHA coatings.

Detachment of titanium particles was observable atthe TPS surface and inside the newly formed peri-implant bone (Fig. 3). Moreover, small titanium debriswere sometimes detectable in the bone medullary spacesnear the TPS surface. On the contrary, titaniumparticles were not seen in the medullary spaces far fromthe implant (Fig. 3). Small particles of FHA and a smallamount of titanium particles were evident in themedullary spaces near the FHA-Ti implant surface(Fig. 4). In TPS implants, titanium debris were mostlyfound within 200–250 mm from the metal surface,

Fig. 2. IR spectra of: (a) the material removed from the implant surface of a TPS-coated screw; (b) the material removed from the implant surface of

a FHA-Ti-coated screw; and (c) FHA coating before implantation.

Fig. 3. Light microscopy: A TPS implant in the sheep tibia after 12

weeks. A particle of titanium is detectable inside the newly formed

bone (black arrow). Some titanium debris are present in the medullary

spaces near the titanium surface (grey arrows). Scale bar 100mm.

Fig. 4. Light microscopy: An FHA-Ti implant in the sheep tibia after

12 weeks. Some particles of FHA and titanium are visible in the

medullary spaces near the implant surface (arrows). Scale bar 100 mm.

D. Martini et al. / Biomaterials 24 (2003) 1309–13161312

although some debris could occasionally be observed at500 mm and even at greater distances. On the other hand,FHA particles had a wider distribution and could beseen at greater distances in FHA-Ti implants. However,the greater the distance, the smaller the particlesbecame, until being undetectable. They appeared insome way to crumble and disintegrate as their distancefrom the implant surface increased. BSE imaging andEDAX analysis confirmed the presence of titanium andFHA debris in the medullary spaces near the implantsurface (Figs. 7a, b, 8a,b).

No signs of detachment were detectable at thetitanium coating-bulk titanium and titanium coating–FHA coating interfaces, and only some fractures,probably due to sample manipulation, were observedin the FHA coating thickness.

4. Discussion

Osteointegration is the fundamental requirement forthe long-term success of orthopaedic or dental implants,and primary stability is a key factor to achieveosteointegration. An extensive and close contact be-tween the implant and the host bone surfaces is thecondition that maintains primary stability and avoidsexcessive interfacial micromotion during bone healingthat may be detrimental to the osteointegration process[23]. Some authors, however, have observed that thepresence of adequate spaces to allow bone remodelling isalso useful to enhance and accelerate osteointegration[24], provided that implant mobilisation is avoided.Therefore, clinicians may obtain satisfactory primarystability through the friction occurring at the implantsurface–host bone interface, thus avoiding implantmicromotion. Such friction, however, could increase

Fig. 6. Light microscopy: An FHA-Ti implant in the sheep tibia after

12 weeks. The portion of FHA-Ti surface is covered by a well-

distributed layer of new bone. Scale bar 500mm.

Fig. 7. (a) TPS implants in the sheep tibia after 12 weeks. BSE

analysis. Titanium debris are observable in the medullary spaces near

the implant surface (arrows). Scale bar 500mm. (b) The EDAX analysis

of (a) (Ti=light grey; Ca=dark grey).

Fig. 5. Light microscopy: A TPS implant in the sheep tibia after 12

weeks. The portion of TPS surface crossing the whole medullary canal

shows a small amount of primary new bone. Scale bar 500mm.

D. Martini et al. / Biomaterials 24 (2003) 1309–1316 1313

stress concentrations on the implant coating and alter itsmorphology and integrity, resulting in the detachmentof metal particles. Some studies have reported the long-term presence of metallic wear particles from endoss-eous implants in the liver, spleen, small aggregates ofmacrophages and even in para-aortic lymph nodes [25].Metal ions released from implants may arise fromdissolution, fretting and wear, and may be a source ofconcern due to their potentially harmful local andsystemic carcinogenic effects [1,26]. However, local andsystemic adverse effects of titanium ion release are notuniversally recognised. Some authors have in factobserved that titanium-alloy dental implants with andwithout plasma-sprayed HA coating did not show anytoxic effects on cells in dogs [27]. In any case, areduction in metal ion release is also preferableon account of the adverse effect that they may haveon the sensitive differentiation processes necessary for

normal bone formation [1]: an excessive metal ionrelease can inhibit cell function and apatite formation[28,29].

In this study TPS- and FHA-Ti-coated screws wereimplanted in the femoral and tibial diaphyses of twomongrel sheep. Histological observations showed goodosteointegration of both implants in host bone 12 weeksafter surgery. No significant differences in osteointegra-tion were found between the two implants in terms ofdiaphyseal compact bone. However, a different responseof the host tissue to TPS and FHA-Ti coatings wasobserved in the implant portion crossing the medullarycanal. On the TPS implant surfaces, a small amount ofprimary new bone was observed around some nuclea-tion sites, while on the analogous portion of the FHA-Ti-coated implants there was a thicker and morehomogeneously distributed layer of new bone near theimplant. These observations confirm the stimulatingeffect of apatites on osteogenesis. The new bone appearsas a fairly continuous layer overlapped with a well-preserved FHA-Ti coating, whose in vivo level ofstability is implicitly confirmed [8,9].

The spectroscopic analysis of the material on theimplant surface of TPS and FHA-Ti screws showed thepresence of bone tissue, as revealed by the typical bandsof its organic and inorganic components. The char-acteristic bands of HPO2�

4 ions (at 1003 cm�1 in theRaman spectra and at 1155, 1110 and 575 cm�1 in theIR spectra) indicated the presence of newly formedbone, since they have been reported to decrease duringthe aging and maturation of the mineral phase[15,30,31]. In vivo and in vitro [32] bone depositionoccurs by means of amorphous precursors, such astricalcium phosphate and octacalcium phosphate, whichcontain HPO2�

4 ions. Successively, these amorphousmetastable phases are transformed into a crystallineoctacalcium phosphate-like phase which, in turn, istransformed into a crystalline apatitic phase by takingup OH� and HPO2�

4 ions. With maturation, this apatiticphase is transformed into non-stoichiometric HA.Termine et al. [30] have observed that the mineralcomponent of young rat deproteinated bone showsanother band, in addition to the typical bands of PO3�

4

ion, which, seen at B1100 cm�1, is absent in thespectrum of mature bone and is typical of the HPO2�

4

ion.It is interesting to note that in the FHA-Ti-coated

screws both IR and Raman spectra showed the presenceof bands attributable to bone in addition to the coatingbands. These results suggest that bone grows in closecontact with the FHA coating and confirm a high levelof osteointegration.

Ultrastructural observations confirmed a high level ofbiointegration since the newly formed bone can growinside the FHA coating porosity in the same way as inother HA coatings. The strong adhesion between the

Fig. 8. (a) FHA-Ti implants in the sheep tibia after 12 weeks.

Titanium (white arrow) and FHA (grey arrow) debris are detectable

in the medullary spaces. Scale bar 500mm. (b) The EDAX analysis

reveals FHA and titanium debris (Ti=light grey; Ca=dark grey).

D. Martini et al. / Biomaterials 24 (2003) 1309–13161314

coating and new bone was confirmed by some artificialfractures due to the mechanical stress developed duringsawing (Saw and Grinding System) that developed onthe coating thickness but not at the bone–coatinginterface.

The morphological analysis showed the presence ofmedullary spaces near the TPS surface. Small titaniumdebris were detectable inside these areas and in thenewly formed peri-implant bone. On the contrary,titanium particles were never found in the medullaryspaces located far from the implant surface. The size andposition of these particles suggest that they may be dueto the friction between the implant and the host boneduring surgical implantation. The screws were in factimplanted in diaphyseal compact bone and should beconsidered unloaded. Consequently, no fretting or wearmay have led to the detachment of metal particles fromthe implant coatings after surgery.

The EDAX analysis demonstrated that the chemicalnature of the released debris depends on the composi-tion of the implant surface. In FHA-Ti implants only afew traces of titanium were seen together with someFHA granules dispersed in the bone medullary spacesnear the implant surface. The lower amount of titaniumdebris in the present samples may be due to the FHAcoating that could protect titanium during the insertionor healing process. The FHA debris were in any casealways present in the medullary spaces of these samplesbut represent biocompatible material [8]. Moreover,FHA resorption may be expected to occur according tothe results of a study on more stable fluoroapatitecoatings by Overgaard et al. [33], who have de-monstrated that osteoclast-like cells, osteocytes, macro-phage-like cells and fibroblasts phagocytize fluoroapa-tite fragments, thus indicating cell-mediated coatingresorption.

On the contrary, evident titanium dispersion insidethe medullary spaces was observed when TPS wasimplanted. These titanium particles decreased in size atgreat distances from the implant surface. This reductionin size of the particles may be due to mechanical effects,but also to a gradual and passive dissolution. Such amechanism has been reported to always occur on anymetal surface interacting with surrounding fluids andtissues [34].

In conclusion, the current findings suggest that:

(1) the medullary spaces not only represent areaspromoting bone turnover [35], but can also act asa sort of ‘‘door’’ to connect the implant surface tothe systemic compartment,

(2) among the porous surfaces of titanium implants,the FHA coating not only offers the positive factorof providing a substrate to enhance new boneformation, but also represents a protective coatingto limit detachment of titanium particles.

Acknowledgements

The authors thank Biocoatings for providing thesamples. This research project was supported by grantsfrom CNR (Progetto finalizzato CNR/MSTA II,Sottoprogetto Biomateriali) and MIUR (60%).

References

[1] Browne M, Gregson PJ. Effect of mechanical surface pretreat-

ment on metal ion release. Biomaterials 2000;21:385–92.

[2] Ungersbock A, Pohler OEM, Perren SM. Evaluation of soft tissue

reactions at the interface of titanium limited contact dynamic

compression plate implants with different surface treatments: an

experimental sheep study. Biomaterials 1996;17:797–806.

[3] Zeng H, Chittur KK, Lacefield WR. Dissolution/reprecipitation

of calcium phosphate thin films produced by ion beam sputter

deposition technique. Biomaterials 1999;20:443–51.

[4] Hayakawa T, Yoshinari M, Nemoto K, Wolke JGC, Jansen JA.

Effect of surface roughness and calcium phosphate coating on the

implant/bone response. Clin Oral Implant Res 2000;11:296–304.

[5] Moroni A, Orienti L, Stea S, Visentin M. Improvement of the

bone pin interface with hydroxyapatite coating: an in vivo long-

term experimental study. J Orthop Trauma 1996;10:236–42.

[6] Kitsugi T, Yamamuro T, Nakamuro T, Kotani S, Kokubo T,

Takeuchi H. Four calcium phosphate ceramics as substitutes for

not-weight-bearing. Biomaterials 1993;14:216–24.

[7] Gessink RG. Hydroxyapatite-coated total hip prostheses: two-

year clinical and roentgenographic results of 100 cases. Clin

Orthop 1990;261:539–85.

[8] Savarino L, Stea S, Ciapetti G, Granchi D, Donati ME, Cervellati

M, Visentin M, Moroni A, Pizzoferrato A. The interface of bone

microstructure and an innovative coating: an X-ray diffraction

study. J Biomed Mater Res 1998;40:86–91.

[9] Gineste L, Gineste M, Ranz X, Ellefterion A, Guilhem A,

Rouquet N, Frayssinet P. Degradation of hydroxyapatite,

fluoroapatite and fluorhydroxyapatite coatings of dental implants

in dogs. J Biomed Mater Res 1999;48:224–34.

[10] Browne M, Gregson PJ. Surface modification of titanium alloy

implants. Biomaterials 1994;15:894–8.

[11] Badawy WA, Alkharafi FM. Stability of titanium and zirconium

anodic films in nitric-acid solutions-EIS comparative investiga-

tion. Bull Electrochem 1996;12:505–10.

[12] Ducheyne P, Bianco PD, Kim C. Growth enhancement by

calcium-phosphate coatings on porous titanium-alloys—the effect

of shielding metal dissolution product. Biomaterials 1992;13:

617–24.

[13] Sousa SR, Barbosa MA. Effect of hydroxyapatite thickness on

metal ion release from Ti-6Al-4 V substrates. Biomaterials

1996;17:397–404.

[14] Raynaud S, Champion E, Bernache-Assolant D, Tetard D.

Dynamic fatigue and degradation in solution of hydroxyapatite

ceramics. J Mater Sci Mater Med 1998;9(4):221–7.

[15] Sauer GR, Zunic WB, Durig JR, Wuthier RE. Fourier transform

Raman spectroscopy of synthetic and biological calcium phos-

phates. Calcif Tissue Int 1994;54:414–20.

[16] Casciani FS, Condrate RA. The infrared and Raman spectra of

several calcium hydrogen phosphates. In: Ean C, editor. 2nd

International Congress on Phosphorus Compounds, Proceedings.

Paris: Phosphorus Publications. 1980. p. 175–90.

[17] Nelson DGA, Featherstone JDB. Preparation, analysis and

characterization of carbonated apatites. Calcif Tissue Int 1982;

34:S69.

D. Martini et al. / Biomaterials 24 (2003) 1309–1316 1315

[18] Nishino M, Yamashita S, Aoba T, Okazaki M, Moriwaki Y. The

laser-Raman spectroscopic studies on human enamel and

precipitated carbonate-containing apatites. J Dent Res 1981;60:

751–5.

[19] Mendelsohn R, Hassakhani A, Di Carlo E, Boskey A. FT-IR

microscopy of endochondral ossification at 20mm spatial resolu-

tion. Calcif Tissue Int 1989;44:20–4.

[20] Bonel G. Contribution "a l’!etude de la carbonatation des apatites.

Ann Chim 1972;7:127–44.

[21] Vallet-Reg"ı M, Izquierdo-Barba I, Salinas AJ. Influence of P2O5

on crystallinity of apatite formed in vitro on surface of bioactive

glasses. J Biomed Mater Res 1999;46:560–5.

[22] Izquierdo-Barba I, Salinas AJ, Vallet-Reg"ı M. In vitro calcium

phosphate layer formation on sol-gel glasses of the CaO–SiO2

system. J Biomed Mater Res 1999;47:243–50.

[23] Pilliar RM. Modern metal processing for improved load-bearing

surgical implants. Biomaterials 1991;12:95–100.

[24] Futami T, Fujii N, Ohnishi H, Taguchi N, Kusakari H,

Ohshima H, Maeda TJ. Tissue response to titanium

implants in the rat maxilla: ultrastructural and histochemical

observations of the bone-titanium interface. Periodontology 2000;

71:287–9.

[25] Urban RM, Jacobs JJ, Tomlinson MJ, Gavrilovic J, Black J,

Peoch M. Dissemination of wear particles to the liver, spleen and

abdominal lymph nodes of patients with hip or knee replacement.

J Bone Jt Surg Am 2000;82:457–77.

[26] Ferguson Jr AB, Akahoshi Y, Laing PJ, Hodge ES. Character-

istics of trace ions released from embedded metal implants in the

rabbit. J Bone Jt Surg Am 1962;44A:323–6.

[27] Lavos-Valerato IC, Konig Jr B, Rossa Jr C, Marcantonio Jr E,

Zavaglia AC. A study of histological responses from Ti-6Al-7Nb

alloy dental implants with and without plasma-sprayed hydro-

xyapatite coating in dogs. J Mater Sci Mater Med 2001;12:273–6.

[28] Blumenthal NC, Cosma V. Inhibition of apatite formation by

titanium and vanadium ions. J Biomed Mater Res 1989;23:13–22.

[29] Blumenthal NC, Postner AS. In vitro model of alluminium-

induced osteomalacia: inhibition of hydroxyapatite formation and

growth. Calcif Tissue Int 1984;36:439–41.

[30] Termine JD, Eanes ED, Greenfield DJ, Nylen MU, Harper RA.

Hydrazine-deproteinated bone mineral. Physical and chemical

properties. Calcif Tiss Res 1973;12:73–90.

[31] Sauer GR, Wuthier RE. Fourier Transform Infrared character-

ization of mineral phases formed during induction of mineraliza-

tion by collagenase-released matrix vescicles in vitro. J Biol Chem

1988;263:13718–24.

[32] Rey C, Fr"eche M, Heughebaert M, Heughebaert JC, Lacout A,

Lebugle A, Szilagyi J, Vignoles M. Apatite chemistry in

biomaterial preparation, shaping and biological behaviour. In:

Bonfield W Hastings GW, Tanner KE, editors. Bioceramics,

vol. 4. Oxford: Butterworth-Heinemann, 1991. p. 57–63.

[33] Overgaard S, Lind M, Josephsen K, Maunbach AB, B .unger C,

S�balle K. Resorption of hydroxyapatite and fluoroapatite

ceramic coatings on weight-bearing implants: a quantitative and

morphological study in dogs. J Biomed Mater Res 1998;39:

141–152.

[34] Bianco PD, Ducheyne P, Cuckeer JM. Titanium serum and urine

levels in rabbits with a titanium implant in the absence of wear.

Biomaterials 1996;20:1937–42.

[35] Ruggeri Jr A, Strocchi R, Franchi M, Martini D, Raspanti M,

Congiu T, Ruggeri A. Peri-implant medullary cisternae at the

interface of bone-smooth surface endosseous implant. Bull Group

Int. Rech Sci Stomatol Odontol 1999;41:104–11.

D. Martini et al. / Biomaterials 24 (2003) 1309–13161316