In Vivo Performance of Two Different Hydroxyapatite Coatings onTitanium Prepared by Discharging in Electrolytes

Hiroki Yamamoto,1 Yo Shibata,1 Tetsuhiko Tachikawa,2 Takashi Miyazaki1

1 Department of Oral Biomaterials and Technology, Showa University School of Dentistry, 1–5-8 Hatanodai, Shinagawa-ku,Tokyo 142-8555, Japan

2 Department of Oral Pathology, Showa University School of Dentistry, 1–5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8555, Japan

Received 7 July 2005; revised 26 August 2005; accepted 9 September 2005Published online 16 December 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.b.30477

Abstract: This study reports a discharging method for bone-like carbonated HA (cHA)-coating (Ca/P 1.71) and stoichiometric HA (sHA)-coating (Ca/P 1.67) with micrometer orderthicknesses on titanium plates, using modified body fluid and acidic calcium phosphatesolutions, respectively. In vivo histological performance of the HA coatings prepared bydischarging in electrolytes was evaluated. Bone-contact indexes of HA coatings were measuredmicroscopically. Additionally, bone-coating interface was analyzed by scanning electron mi-croscopy and the use of an electron probe microanalyzer. Results demonstrated that there wasno significant difference in contact index between HA coatings. However, the cHA coating waspractically replaced by immature bone, and the titanium metal substrate was directly con-nected to the bone structure whereas the sHA coating layer remained and was partiallydetached from the titanium metal substrate. Since detached coating particles are pathogens,and can cause peri-implantitis, the cHA coating was more favorable than the sHA coating evenif contact index was equivalent to that of the sHA coating. It is thought that coating thicknessand chemical composition of coatings are important for biological stability of implants. Inconclusion, since bone-like thin cHA coating showed high osteoconductivity and bone replace-ment, bone-like HA is superior to sHA coating for use in dental implants. © 2005 WileyPeriodicals, Inc. J Biomed Mater Res Part B: Appl Biomater 78B: 211–214, 2006

Keywords: titanium; hydroxyapatite; coating; dental implant

INTRODUCTION

The primary concept of HA-coated titanium implants is basedon the integration of both mechanical strength and high osteo-conductivity, to be used as load-bearing medical devices.

Thermal spraying methods have already been establishedas conventional techniques for the production of HA coatingson titanium implants.1,2 However, bonding between HA coat-ings and the metallic substrate formed primarily throughmechanical interlocking is not strong enough. Additionally,these techniques are unsatisfactory for applying thin, uniformHA coating on implants with various shapes, therefore, re-sulting in easy damage at the interface between the implantand the bone. Detached coating particles are pathogens, andcan cause peri-implantitis.3 Thus, mechanical properties ofthermal-sprayed HA coating on titanium implants have re-mained problematic for long-term clinical applications.

Electrochemical techniques seem attractive for preparingthinner HA coatings on titanium implants, because componentsin electrolytes are uniformly deposited on the surface regardlessof implant shapes.4–6 However, bonding between the coatingand metal substrate is also weak. In addition, because depositsare not well-crystallized, additional thermal processes have beennecessary to improve coating crystallinity.

Given these concerns, new techniques for preparing HAcoatings on titanium implants are under investigation.

We propose that a technique involving discharging inelectrolytes is successful for preparing thin calcium phos-phate coatings on titanium implants.7,8 Coating componentsdissolved in electrolytes are driven onto a titanium surfacethrough a gas layer with a larger electrical field than thatproduced by previous electrochemical techniques. Since coat-ing crystallinity and bonding between the coating and metalsubstrate were much increased, the present dischargingmethod is expected to be a promising technique for preparingHA coatings on titanium implants.

Additionally, our discharging method allowed both bone-likecarbonated HA-coating (Ca/P 1.71) and stoichiometric HA-coating (Ca/P 1.67) with thicknesses of 1–10 �m on titanium

Correspondence to: Y. Shibata (e-mail: yookun@dent.showa-u.ac.jp)

Contract grant sponsors: MEXT, HAITEKU (2005); Japan Society for thePromotion of Science; the Ministry of Education, Culture, Sports, Science andTechnology, Japan

© 2005 Wiley Periodicals, Inc.

211

plates, using modified body fluid and acidic calcium phosphatesolutions, respectively.7,8 It is generally accepted that carbonatedapatite appears to degrade bioactive materials, rather than stoi-chiometric HA.9 However, in vivo differences between carbon-ated apatite coating and stoichiometric HA-coating on titaniumimplants have not been fully elucidated yet.

Therefore, in vivo histological performances of HA coat-ings prepared by discharging in electrolytes were also eval-uated. Bone-coating interface was analyzed by scanning elec-tron microscopy and an energy dispersive X-ray spectrometer(EDX).

MATERIALS AND METHODS

Preparation of Specimens

Turned cylindrical titanium (JIS grade II, KS-50, Kobe Steel,Kobe, Japan) with a dimension of � 1.6 � 5.0 mm2 was usedas specimens. Both bone-like carbonated HA (cHA) andstoichiometric HA (sHA) coatings of specimens were pre-pared using a discharging method.7,8 Specimens withoutcoatings were used as controls.

Surgical Technique

The National Institutes of Health (NIH) guidelines for thecare and use of laboratory animals (NIH Pub. No.85–23,Rev.1985) were followed for all experiments involving ani-mals in this study.

Coating cylinder-shaped specimens were sterilized withethylene oxide for 24 h at ambient temperature. Male Wisterrats (10-weeks-old) were anesthetized with ethylether. Theleft femur was chosen as implantation site. The site wasshaved, sterilized, and incised with a scalpel. For the implantcavity, a hole of 1.6 mm in diameter and 5.0 mm deep wasformed perpendicular to the length of the cortical bone, withsimultaneous infusion of sterile physiological saline solution.Cylinder-shaped burrs (� 1.6 � 5.0 mm2, IAT, ISHIFUKU,Tokyo, Japan) were used in turns with a micromotor forimplant surgery, with a rotation speed of 160–180 rpm. Then,each coating specimen was implanted into the bone marrowof the left femur; and in control animals, implantation wasdone into the right femur. Each specimen was implanted intoa rat femur for up to 4 weeks.

Histological Observations

Immediately after euthanasia, femora were dissected fromsurrounding soft tissues, fixed in formaldehyde, and embed-ded in polyester. Undecalcified serial cross sections were cutwith a water-cooled diamond saw, mounted on Plexiglasslides, then ground and polished on a precision grinder(Isomet; Buehler Corporation, Lake Bluff, IL) to a thicknessof 50 �m.

Specimens were stained with toluidine blue for micro-scopic observations, and contact index was defined micro-scopically (E-600, Nikon, Tokyo, Japan) as the new bone

contact to the specimen surface (Fig. 1). Results were ex-pressed as means � standard deviations for six specimens(n � 6). Findings were analyzed statistically by analysis ofvariance (ANOVA).

SEM Analysis

Cross-sectioned specimens after 2 weeks implantation wereprepared by gold-vaporizing with a vacuum evaporation de-vice (IB-2, Eiko Engineering, Ibaraki, Japan). Interface be-tween the specimen and contacted bone was observed bySEM (S-2360N, Hitachi, Tokyo, Japan).

EDX Characterization

Cross-sectioned specimens after 2 weeks implantation weresputter-coated with carbon (CC-40F, Meiwa Shoji Co., To-kyo, Japan). An EDX (Quantum�-IV Type: Kevex, Foster,CA) equipped with an electron microscope (S-2500CX, Hi-tachi Co., Tokyo, Japan) was used to produce back-scatteringelectron images and provide element composition of thecoating and bone.

RESULTS

Histological Observations

Contact indexes of cHA, sHA, and control specimens, 2weeks after implantation, were 64.1, 66.1, and 31.0, respec-tively (Fig. 2). Contact indexes of both HA-coated specimenssignificantly increased (p � 0.01) compared with those ofcontrol specimens even after 2 weeks. There was no signif-icant difference in contact index between HA coatings (p �0.01). Contact indexes of cHA, sHA, and control specimens,4 weeks after implantation, were 70.8, 75.7, and 64.1, respec-tively. There was no significant difference in contact indexbetween the specimens (p � 0.01).

Figure 1. Schematic illustration of contact index calculation. Contactindex was defined as the total length of surface Ti connected withnewly formed immature bone (1)/length of surface Ti in bone marrow(2) � 100 (%).

Journal of Biomedical Materials Research Part B: Applied BiomaterialsDOI 10.1002/jbmb

212 YAMAMOTO ET AL.

SEM Observations

The cHA coating was observed to be practically replaced byimmature bone, and the titanium metal substrate was directlyconnected to the bone structure [Fig. 3(A)]. On the contrary,the coating layer observed to be remained on sHA-coatedimplants [Fig. 3(B)]. The newly formed bone was not closelyconnected to the control specimen [Fig. 3(C)].

EDX Characterization

Two weeks after implantation, EDX observations also re-vealed that the cHA coating was absorbed, and completelyreplaced by immature bone [Fig. 4(A)]. The newly formedbone was directly connected to titanium substrate. Mean-while, the newly formed bone was connected to the remain-ing coating layer on the sHA specimen. Additionally, thesHA coating layer was partially detached from the titaniummetal substrate [Fig. 4(B)]. The newly formed bone was notdirectly connected to the control specimen [Fig. 4(C)].

DISCUSSION

Because of the higher contact index of HA coatings comparedwith that of control specimens 2 weeks after implantation,both HA coatings prepared by discharging were effective inearly stage osteogenesis. Since no significant differences be-tween HA coatings were detected, it seemed that chemicalcomposition of coatings did not affect, at least, the histolog-ical contact index.

However, SEM images showed that cHA coatings werereplaced, and the titanium metal substrate connected directlyto immature bone. Additionally, EDX also revealed that cHAcoatings were practically absorbed, and were replaced byimmature bone. Since residual coating layers may detachunder load-bearing conditions, cHA coatings were more fa-vorable than sHA coatings even if contact index was equiv-alent with sHA coatings.

Figure 2. Bone-contact indexes of specimens. Results are expressed as means � standard devia-tions for six specimens (n � 6). Findings were analyzed statistically by analysis of variance (ANOVA).

Figure 3. SEM photographs of cross-sections of specimens. Bone-coating interface of cHA coating (A), sHA coating (B), and controlspecimen (C) 2 weeks after implantation. Bar � 10 �m.

Journal of Biomedical Materials Research Part B: Applied BiomaterialsDOI 10.1002/jbmb

213TWO DIFFERENT HYDROXYAPATITE COATINGS ON TITANIUM

Materials used for orthopedic implants in bone replacementsshould be gradually resorbed and replaced by new bone.9 Con-tinuous balanced osseous remodeling is essential for successfulosteogenesis of implant devices.10 A recent report indicated thatcarbonate apatite could be resorbed by osteoclasts, and was notsusceptible to phagocytosis by macrophages. Osteoclast adhe-sion and resorption on carbonated HA are much greater thanthose on stoichiometric HA.9 However, since synthesis of car-bonate apatite is generally associated with low crystallinity,carbonate apatite coatings without thermal treatments degradeprior to osteogenesis during implantation.11 Since bone-likecHA coatings prepared by discharging in modified body fluidacquire high crystallinity and a thinner coating layer is obtainedwithout such treatments,7,8 cHA coatings are favorable bone-contact in this study.

On the contrary, sHA coatings were connected to immaturebone through the residual coating layer. Because thickness ofcurrent thermal-sprayed HA coatings is 50–100 �m, these coat-ings are easily damaged at the interface between the titaniumand the bone. Thus, techniques resulting in thinner HA coatingshave been explored.8,12 However, since detachment of the coat-ing layer was detected even in the case of sHA coatings withmicrometer order thickness, it was reasonable to say that coatingthickness and chemical composition of coatings are importantfor biological stability of implants.

In conclusion, since bone-like thin cHA coating showedhigh osteoconductivity and bone replacement abilities, bone-like HA is superior to sHA coating for use in dental implants.

This study was conducted by the Department of Oral Biomate-rials and Technology, Showa University School of Dentistry andOral Pathology, Showa University School of Dentistry. The authorsgratefully acknowledge financial supports received from a MEXT,HAITEKU (2005), a Grant-in-Aid for Scientific Research (B) fromthe Japan Society for the promotion of Science and a Grant-in-Aidfor Encouragement of Young Scientists (B) from the Ministry ofEducation, Culture, Sports, Science and Technology of Japan.

REFERENCES

1. Groot K, de Geesink R, Klein CPAT, Serekian P. Plasmasprayed coating of hydroxyapatite. J Biomed Mater Res 1987;21:1375–1381.

2. Oguchi H, Ishikawa K, Ojima S, Hirayama Y, Seto K, EguchiG. Evaluation of a high-velocity flame-spraying technique forhydroxyapatite. Biomaterials 1992;13:471–477.

3. Wheeler SL. Eight-year clinical retrospective study of titaniumplasma-sprayed and hydroxyapatite-coated cylinder implants.Int J Oral Maxillofac Implants 1996;11:340–350.

4. Ban S, Maruno S. Hydrothermal-electrochemical deposition ofhydroxyapatite. J Biomed Mater Res 1998;42:387–395.

5. Ban S, Maruno S. Morphology and microstructure of electro-chemically deposited calcium phosphates in a modified simu-lated body fluid. Biomaterials 1998;19:1245–1253.

6. Ban S, Hasegawa J. Modification of electrochemically depos-ited apatite using supercritical water. Dent Mater J 2001;20:247–256.

7. Takashima H, Shibata Y, Kim TY, Miyazaki T. Hydroxyapatitecoating on a titanium metal substrate by a discharging methodin modified artificial body fluid. Int J Oral Maxillofac Implants2004;19:66–72.

8. Shibata Y, Takashima H, Yamamoto H, Miyazaki T. Function-ally gradient bonelike hydroxyapatite coating on a titaniummetal substrate created by a discharging method in HBSS with-out organic molecules. Int J Oral Maxillofac Implants 2004;19:177–183.

9. Redey SA, Razzouk S, Rey C, Bernache-Assollant D, Leroy G,Nardin M, Cournot G. Osteoclast adhesion and activity onsynthetic hydroxyapatite, carbonated hydroxyapatite, and natu-ral calcium carbonate: Relationship to surface energies. Bioma-terials 1992;13:471–477.

10. Kawai H, Shibata Y, Miyazaki T. Glow discharge plasmapretreatment enhances osteoclast differentiation and survival ontitanium plates. Biomaterials 2004;25:1805–1811.

11. Landi E, Tampieri A, Celotti G, Vichi L, Sandri M. Influence ofsynthesis and sintering parameters on the characteristics ofcarbonate apatite. Biomaterials 2004;25:1763–1770.

12. Thian ES, Huang J, Best SM, Barber ZH, Bonfield W. Magne-tron co-sputtered silicon-containing hydroxyapatite thinfilms—An in vitro study. Biomaterials 2005;26:2947–2956.

Figure 4. Upper panel indicates BSE images of cross-sections of (A) cHA coating, (B) sHA coating,and control specimens (C) 2 weeks after implantation. Lower panel indicates EDX analysis ofcross-sections of (D) cHA coating, (E) sHA coating, and control specimens (F) 2 weeks after implan-tation. Bar � 20 �m.

Journal of Biomedical Materials Research Part B: Applied BiomaterialsDOI 10.1002/jbmb

214 YAMAMOTO ET AL.