Titanium for osteointegration: Comparison between a novel biomimetic treatment and commercially exploited surfaces

INTRODUCTION

The development of novel surface treatments toimprove osteointegration properties of titanium/ti-tanium alloy based implants is an important area ofresearch in orthopaedic and dental material sci-ence. The goal of osteointegration is a rapid and re-liable achievement of a mechanically stable bone-implant interface and new remodelled bone tissue

in direct contact with implanted material surface,without interposition of low mechanical perform-ing fibrous tissue (1). In the last three decades hydroxylapatite coatingsachieved by plasma spray technique were widely ex-ploited for improving implant osteointegration be-cause of their biological response, resulting in highbone apposition without fibrous tissue interposi-tion at bone-implant interface and high short-term

Journal of Applied Biomaterials & Biomechanics 2004; 2: 35-44

Titanium for osteointegration: Comparison between a novel biomimetic treatment and commercially exploited surfaces

C. GIORDANO1, E. SANDRINI1, B. DEL CURTO1, E. SIGNORELLI2, G. RONDELLI2, L. DI SILVIO3

1Department of Chemistry, Materials and Chemical Engineering “G.Natta”, Politecnico di Milano, Milano - Italy2IENI, Italian National Research Council, Milano Bicocca, Milano - Italy3Guy’s, King’s and St Thomas’ Medical and Dental Institute, King’s College London, Guy’s Hospital, London - UK

ABSTRACT: The objective of this preliminary in vitro biological study was to assess the effect of the surface physicochemical andtopographical properties of a novel bioactive titanium (BSP) obtained by BioSpark™ treatment. A short-term study was per-formed to evaluate the bone cell response to BSP and compare it to two commercially available materials: no treated (TI) andchemically etched (ETC) titanium. Material characterization was carried out using scanning electron microscopy (SEM), en-ergy dispersion spectroscopy (EDS), non-contact laser profilometry (LPM), and Thin Film X-ray Diffraction (TF-XRD). Sur-face analysis showed ETC to have the highest rough surface, followed by TI surface and then BSP being the smoothest materi-al at micro level, but showing a sub micrometer porous structure covered with a “net-like” rough structure. The BSP surfacewas found to consist of a layer of amorphous calcium and phosphorus and crystalline titanium oxide, not detected in the oth-er materials tested. Indirect biological cytotoxicity studies were performed to determine cell viability following incubation withthe eluted extract of the materials. Results indicated no remarkable deterioration in cell viability. In particular, no detectable ef-fect was observed on cellular viability as a result of the chemical interaction between the BSP bioactive surface and the sur-rounding culture medium. Direct cellular studies showed that the material surface resulted in good cell adhesion on BSP sam-ples. This could be related to both the nano-roughness, and also the crystallinity of the superficial layer of titanium oxide cou-pled with bioactive Ca- and P-chemical enrichment. The cellular proliferation analysis demonstrated a remarkably higher ac-tivity for the cells cultured on BSP, with values significantly higher than the other test materials and the control for all timepoints. These findings are highly suggestive that the surface properties of the BioSpark™ treated titanium significantly increasescell proliferation rate. In conclusion, this study has demonstrated that the novel bioactive treatment shows potential as a methodfor improving osteointegration properties of titanium for orthopaedic and dental implants. (Journal of Applied Biomaterials &Biomechanics 2004; 2: 35-44)

KEY WORDS: Titanium, Chemical etching, Anodic spark deposition, Surface treatments, In vitro biological response, Osteointegration

Received 05/12/03; Revised 30/12/03; Accepted 07/01/04

1722-6899/035-10$15.00/0© Società Italiana Biomateriali

mechanical performance. Unfortunately, manystudies have reported delamination and embrittle-ment of HA coatings following long-term implanta-tion, resulting in premature failure of the implant(2-5). As a result, novel surface treatments havebeen explored and developed to improve the os-teointegration and overcome the mechanical draw-backs observed for HA-plasma spray coatings.Material surface topography and chemical charac-teristics play a key role in the osteointegrationprocess by promoting direct bone apposition at theimplant surface. The material bulk properties andmacro-texture are also important for ensuring pri-mary mechanical stability of the implant once insitu (6-9). The chemical composition of any im-plant as well as the bulk and surface treatmentsused, generally determines the characteristics andproperties of the outer surface layer. These charac-teristics include: the nature of the surface electricalcharge; the surface energy; the presence of anygrain boundaries; the chemistry, stoichiometry andcrystallinity of surface salts and oxides (10-15). Theouter surface layer plays a significant role in the fi-nal in vivo response by influencing interfacial reac-tions such as protein adsorption, cell attachmentand proliferation and the subsequent mineralisa-tion. In vitro models using animal and human cells provideuseful information relating to the biological perfor-mance of the material prior to in vivo testing. Nu-merous studies have been reported on selected bonecell response to different textured and chemicallymodified titanium surfaces. However, to date, themode of action of cellular response to different sur-face properties is still an area of great interest (6-9). In the last decade, increasing attention has beenpaid to the development of novel surface treat-ments resulting in modifications that mimic catalyt-ic apatite properties in order to achieve a better os-teointegration of titanium and its alloys (16-20). A novel bioactive surface treatment, namedBioSpark™, has been developed (Nanosurfacess.r.l.), which resulted in a thickened titanium oxidelayer doped with calcium and phosphorus. This lay-er exhibits a high bioactivity resulting in an en-hanced calcium-phosphate nucleation from simu-lated body fluids (21). This has been reported ashaving a positive biological response once in vivo(19, 20).The aim of this study was, therefore, to assess theshort-term cellular response to surface physico-chemical changes and topography of the Bio-Spark™ treated titanium. A human osteosarcomacell line MG63 was used for the in vitro studies (8).The study was also performed on a no treated and

on a chemically etched titanium, both representa-tive of commercially available materials for biomed-ical applications. Surface properties of the tested materials were in-vestigated by scanning electron microscopy (SEM)to assess the morphology of the different surfaces.Thin film X-Ray diffraction (TF-XRD) was carriedout in order to achieve a better understanding ofthe crystalline structure of the differently treatedsurfaces. Also, TF-XRD allows to assess the presenceof crystalline titanium oxides, whose role has beenwidely recognised as crucial for biocompatibilityand, in particular, for the mineralisation processes(22). The effect of surface chemical composition ofthe titanium as a result of the different test treat-ments was analyzed by energy dispersion spec-troscopy (EDS). Micro-roughness has been shown to effect some im-portant cellular responses (6, 31). In this study,non-contact laser profilometry (LPM) was per-formed in order to assess the micro roughness forall the three materials tested. SEM, EDS and TF-XRD for the novel bioactivetreated titanium were also compared to previouslypublished data (21).An indirect cytotoxicity elution test was performedto assess the effect of any toxic leachables on cellu-lar viability by MTT assay. Cell morphology and ad-hesion of the cells seeded directly on material sam-ple surface was assessed using SEM. Cellular prolif-eration was determined using the Alamar Blue™ as-say at specific time points. Statistical analysis was performed using ANOVA testand t-test to compare the data populations and cor-relate cellular response to surface features.

MATERIALS AND METHODS

Samples preparation

All titanium samples were obtained from a titaniumsheet of commercially pure grade 2 titanium(Toresin Titanio, Padova). Test specimens 10x10x1mm3 were cut with a mechanical cutter (Tempe-CNR Milan) and the surface de-greased andcleaned by rinsing in acetone (Sigma Aldrich, Italy)for 5 minutes, followed by washing in distilled waterin an ultrasonic bath (Branson Automatic Cleaner,UK). Three different materials were prepared:• TI: titanium as received by producer without any

further treatment.• BSP: bioactive titanium achieved by BioSpark™

treatment (performed by Nanosurfaces srl).

36

Titanium for osteointegration: Comparison between a novel biomimetic treatment and commercially exploited surfaces

• ETC: rough titanium achieved by acid etching.BSP treatment procedure: the titanium surface was pre-pared in an electrochemical cell by Anodic SparkDeposition technique (23-25). A variable dc powersupply (BVR1200-500-1.5 Belotti Variatori srl, Italy)and two voltmeters were used to supply power tothe circuit and to monitor the voltage gap betweencathode and anode as well as between cathode andpoints of the electrolytic solution in between thetwo electrodes. The solution was constantly stirredusing a magnetic stirrer in a double walled glassbeaker. The temperature of the electrolytic solutionwas monitored and kept at 0 °C (± 2°C) by a con-stant flow of cooled fluid (70% v/v distilled waterand 30% v/v commercial Paraflu-FLSelenia SpA)through the external and the inner wall of thebeaker. Specimens were connected one by one tothe anode. A Teflon sleeve was used to shield thespecimen surface and to avoid sparking at the at-mosphere-sample-electrolyte solution interface.The cathode was formed by a titanium net (c.p.grade 2) cylindrical shaped, whose surface wasabout 60 times bigger than the surface of the ano-dising sample. The BSP material involved two ASD treatments: thefirst was performed in a solution containing calci-um and phosphate ions, the second was in a solu-tion containing only calcium ions. A final treatmentin concentrated potassium hydroxide (60375, Flu-ka, Chemika) water solution at 60 °C was also per-formed. ETC treatment procedure: The titanium specimenswere treated by a double step etching. The first wascarried out in a solution of 1 M NaOH, containing2% v/v H2O2, at 80 °C for 10 minutes (NC-ETC).The second step was carried out in acid water solu-tion at 28 °C for 1 hour. In between the two etchingsteps and following the acid etching, the sampleswere washed three times for 5 minute in distilledwater in ultrasonic bath.

Material surface analysis

All the tests were performed in triplicate.SEM: All samples were sputter-coated with gold(Sputter Coater SC7640, Polaron) viewed by SEM(Cambridge Instruments, STEREOSCAN 360) atan accelerating voltage of 20 kV. EDS: After SEM analysis, samples were examined byelectron dispersion spectroscopy (Oxford, Inca En-ergy). TF-XRD: Sample surface properties were investigat-ed with thin film X-rays diffraction (Philips PW1710) at 40 mA and 40 kV.LPM: The roughness parameters were calculated

on five 5.6 mm-long profiles and measurementswere acquired using a 3D laser profilometer (UBM-Microfocus Compact, NanoFocus AG, Germany).The analysis of the results was performed using twoparameters: the arithmetic mean of the departuresof the roughness profile from the mean line (Ra),and the symmetry of the distribution (Sk) of theprofile height around an ideal average line.

In vitro cellular response

All investigations were performed in triplicate. Sample treatments and controls: The samples were test-ed following sterilisation in ethanol 70% (v/v in wa-ter) followed by UV irradiation (254 nm). Tissueculture plastic (Corning- Costar) was used as nega-tive, non-toxic control for biological response as-sessment.HOS cells: The human osteosarcoma cell line MG63was obtained from the European Collection of Ani-mal Cell Cultures (ECACC 865051601) and cul-tured in Minimum Essential Eagle Medium(EMEM) (Sigma-Aldrich), containing: 10% foetalbovine serum (FBS) (Sigma-Aldrich), 1% non-es-sential amino acids (Sigma-Aldrich), 2mM L-gluta-mine (Sigma-Aldrich), 100 units/ml penicillin (Sig-ma-Aldrich), 0.100 mg/ml streptomycin (Sigma-Aldrich), in an incubator at 37 °C with 5% CO2 anda humidified atmosphere (Forma).Cell seeding: The materials were placed in a 24 welltissue culture plate and seeded with 50 µl of the ap-propriate cell density suspension. Cells were al-lowed to adhere for 1 hour then flooded with 1 mlof culture medium and placed back into the incu-bator.A cell density of 2x106 cells/ml was used for the pro-liferation study and 2x105 cells/ml for SEM mor-phology and adhesion assessment. Elution study: Sterile samples were placed into a 24well tissue culture plate, flooded with 1 ml of cul-ture medium and incubated. At the selected timepoints, the aqueous extracts were collected in ster-ile condition and replaced with 1 ml of fresh medi-um. 100 µl of 1x 105 cell/ml cell suspension wasseeded into a sterile 96 well culture plate and incu-bated to almost 90% confluence. The culture medi-um was then removed and replaced with 100 µl/well of the test extract and the plates incubatedfor a further 72 hours. The extract was then re-moved and replaced with 100 µl of MTT (Sigma-Aldrich) solution (0.5 mg/ml in culture medium)in each well. The plates were incubated for 4 hoursat 37 °C, 5% CO2 in a humidified atmosphere. Fol-lowing this, the medium was then removed and re-placed with 100 µl of DMSO (Merk) to each well

37

Giordano et al

and the plate mixed until complete dissolution ofthe crystals was obtained. Absorbance was measuredusing a Tecan Genius Plus plate reader (test wave-length: 570 nm; reference wavelength: 630 nm).Cellular adhesion and morphology: For the SEM analysis,the cells were fixed with 1.5% w/w glutaraldehyde(Fluka) buffered in 0.1 M sodium cacodylate (Fluka)at each selected time points. The cells were dehy-drated through a series of ethyl alcohol (BDH) solu-tions (from 20% up to 100% v/v ethyl alcohol in dis-tilled water) followed by a series (from 25% up to100% v/v in ethyl alcohol) of hexamethyldisilazane(Sigma-Aldrich) solutions and finally air dried. Thesamples were sputter coated as described before andexamined by scanning electron microscopy at an ac-celerating voltage of 10 kV. Cellular proliferation assay: Cells were seeded directlyonto the sample surface as previously described. Ateach time point, the culture medium was replacedwith 1 ml Alamar BlueTM (Serotec) solution (10%v/v in culture medium) and the plate incubated for4 hours. Alamar Blue is a REDOX indicator i.e. itresponds to reduction or oxidation of the sur-rounding medium. In this assay it both fluorescesand changes colour in response to the chemical re-duction of culture medium that results from cellgrowth and division. 100 µl (3 replicates for each sample) of the sur-rounding medium was removed from each well,transferred to a 96 well plate and the absorbancemeasured using a Tecan Genius Plus plate reader(test wavelength: 570 nm; reference wavelength:630 nm). The samples were subsequently rinsedwith PBS, 1 ml of culture medium was then added

to each well and the plate was returned to the in-cubator for further incubation.

RESULTS

Material surface analysis

SEM analysis: TI specimen (Fig. 1a-b) showed an ir-regular textured surface with no sharp edges andlightly pronounced grain boundaries.ETC (Fig. 2a-b) surface appeared covered by grainboundaries deeply corroded and a further micro-roughness was observed at a higher magnification.As previously reported (21), BSP texture was regu-lar with many round crest and deep pores clearlyvisible on the surface (Fig. 3a-b). Moreover, thewhole surface was covered by a thin nano-texturedroughness exhibiting a “net-like” structure com-prising of thin nanometric sharp crests. LPM: The surface roughness assessment showed re-markable differences among the three surfacetreatments. ETC presented the highest value of theparameter Ra (Fig. 4a), while BSP and TI exhibitedlower values of this parameter compared to the for-mer surface. ETC, TI and BSP exhibited Sk valuesaround zero, which did not differ significantly fromeach other (Fig. 4b). EDS analysis: Calcium, phosphorous and oxygenpeaks were observed in addition to the peaks of ti-tanium for BSP, in accordance with the literature(21). The TI and ETC surface however, exhibitedonly titanium and no traces of oxygen peaks wereobserved (Fig. 5a-c).

38

Titanium for osteointegration: Comparison between a novel biomimetic treatment and commercially exploited surfaces

a b

Fig. 1 - a-b; Scanning electron microscopy of TI untreated titanium.

39

Giordano et al

Fig. 4 - a-b; Ra and Sk value investigated by laser beam profiler. Ra analysis: ETC> TI (p=3.67E-7), ETC>BSP (p=6.18E-8) andTI>BSP (p=7.15E-7), n=5.

a b

Fig. 2 - a-b; Scanning electron microscopy of ETC treated titanium.

a b

Fig. 3 - a-b; Scanning electron microscopy of BSP treated titanium.

a b

TF-XRD analysis: The TF-XRD analysis on Ti and ETCsamples showed the absence of crystalline oxide phasesand the presence of titanium as the only detectablecrystalline phase (Fig. 6a). On BSP surface the pres-ence of a crystalline oxide film mainly as anatase phaseand only in a small amount as rutile phase was observedas previously described (21) (Fig. 6b).

In vitro cellular response

Elution study: After 72 hours of elution exposure(Fig. 7), a decrease in cell viability was observedfor the test material extracts and the control atday 2, followed by a comparable increase for allthe extracts at day 3. The results obtained for allthe test materials were comparable to the con-trol samples for all the time points studied.Cell adhesion and morphology: At 2 days post-seed-ing, cells were observed adhering and spreadingon the sample surfaces, surrounded by someshort filopodia for all the test materials (Fig. 8a-c). On the BSP samples, a more flattened cellu-lar morphology was observed with fewer filopo-dia seen anchored on the sample nanostruc-tures (Fig. 9a-c).Cell proliferation: An increase with the time in cellproliferation rate was observed for all the testedmaterials and for the control. The results for allthe selected time points for the tested materialswere comparable to the control, with the excep-tion of cells cultured on the BSP samples, wherea higher proliferation rate than the control wasobserved (Fig. 10).

DISCUSSION

This short-term preliminary study examined theresponse of human osteosarcoma cells to aBioSpark™ treated titanium (BSP), comparedto titanium as received (TI) and etching treatedtitanium (ETC).BSP titanium has been shown to have an en-hanced bioactivity, in particular when soaked in simulated body fluids (21). Following incu-bation, the material showed an enhanced calci-um-phosphate nucleation. The resulting nucle-ated clusters consisted of low crystalline hy-droxylapatite, very similar in structure to the in-organic mineral phase of bone tissue (21). Ac-cording to recent findings reported in the liter-ature (22, 26), the described effect may be dueto the crystalline titanium dioxide covering theBSP surface as indicated by the EDS and TF-XRD analyses.

40

Titanium for osteointegration: Comparison between a novel biomimetic treatment and commercially exploited surfaces

Fig. 5 - a-c; Energy dispersive spectroscopy of: (a) TI un-treated titanium; (b): ETC treated titanium; (c): BSP treat-ed titanium.

a b

c

41

Giordano et al

a

b

Fig. 6 - a-b Thin Film X-ray diffractometry of (a) TI untreat-ed titanium and ETC treated titanium (two etching stepsspectra have been reported); (b): BSP treated titanium.

Fig. 7 - Cellular viability performed by MTT test on cells in-cubated (72 hours) with material elutions (1, 2, 3 days).

a

b

c

Fig. 8 - a-c; SEM micrograph of MG63 cells cultured for twodays on (a) TI untreated titanium; (b): ETC treated titani-um; (c): BSP treated titanium.

Laser profilometry indicated that the ETC hadthe highest roughness value. SEM analysis showedthe presence of deeply corroded grain boundarieson the surface as result of the acid attack. Laser profilometry data and SEM observationswere comparable for the TI surface, which ex-hibited the lower Ra values and revealed a typi-cal texture achieved by low concentrated acidsolution pickling, probably carried out for opti-mising surface purity after sheet rolling. BSP appeared to be the smoothest materialwhen tested by laser profilometer analysis, evenif SEM images showed sub micron features onthe surface, with a porous “net-like” rough struc-ture. This mismatch can be explained by the lat-eral resolution limit of the profilometer given bythe laser spot diameter (1 µm) that did not allowthe evaluation of nano-textured features (27).TF-XRD and the EDS results indicated that inthe case of BSP, the porous film achieved on ti-tanium by ASD treatments was mainly composedof a layer consisting of amorphous calcium andphosphorus and crystalline titanium oxide phas-es, in accordance with the previously publisheddata (21). TF-XRD and EDS results revealed, forboth TI and ETC specimens, that the surfaceswere mainly titanium, and no traces of titaniumoxides were detected. This was probably due tothe fact that the superficial oxide films are toothin and probably amorphous (28-30).It has been reported that micro-roughness mayenhance cells adhesion and specific osteoblastactivity while smooth surfaces might enhancecellular proliferation (6, 31-33). From the biological point of view, the cell viabil-ity tests performed following incubation withthe eluted extract of the samples indicated no

42

Titanium for osteointegration: Comparison between a novel biomimetic treatment and commercially exploited surfaces

a

b

c

Fig. 9 - a-c; SEM micrograph of MG63 cells cultured for twodays on (a) TI untreated titanium; (b): ETC treated titani-um; (c): BSP treated titanium.

Fig. 10 - Cell proliferation assessed by Alamar Blue test (1, 2,3 days) of cell culturing on sample surface; BSP: 72 h> 48 h >24 h (p<0.05); TI: 72 h> 48 h > 24 h (p<0.05); ETC: 72 h> 48h > 24 h (p<0.05); TCP: 72 h> 48 h > 24 h (p<0.05); BSP > (TIand ETC and TCP) for each time point: (p<0.05); n = 12.

remarkable deterioration in cell viability fromday 1 to 3. A decrease in cell proliferation how-ever, was observed on day 2, followed by an in-crease for all the materials and control. Generally primary osteoblast growth occurs instages: proliferation, differentiation and extracel-lular maturation and mineralization. Each ofthese stages is reflected in peaks in biochemicaland gene expression levels of specific proteins(34). Usually cell proliferation has to reach a peakbefore the cells can begin to differentiate and passon to the next stage. A drop in proliferation there-fore, usually signals that the cell is in a state of dif-ferentiation. Although purely speculative, as noosteoblast differentiation markers have been mea-sured in this study, the drop in proliferation of theosteosarcoma cell line, might reflect a modifica-tion in cell metabolism with cells advancing intodifferentiation.The cellular proliferation analysis demonstrated amuch higher activity for the cells cultured on BSP,as expected for the smoother material tested (6,31-33). It is interesting to note that the values ob-tained for the BSP are significantly higher thanthe other materials and the control at each timepoint, suggesting an enhancement in the prolifer-ation rate, as a direct result of the surface proper-ties of the bioactive treated titanium.Several studies have been performed on the effectof surface roughness on cell behaviour (6, 31-33,35, 36). In this study, good cell adhesion andspreading was observed on the test materials indi-cating a favourable surface, with cells having nor-mal osteoblast morphology. Interestingly, a moreflattened cellular morphology was observed onthe BSP samples, suggesting a more advanced ad-hesion state and a surface enhancing cell coloni-sation (6). This study has indicated that BSP has the smoothestsurface, while TI and ETC exhibited higher rough-ness values due to deep grain boundaries not ob-served on BSP surface. These results might suggesta more tight and advanced cell adhesion morphol-

ogy on both the micro-rough TI and ETC speci-mens. However, cells on BSP showed a slightly bet-ter “adhesion response”. This may be a direct resultof the nano-roughness of BPS and also the crys-tallinity of the thick titanium oxide surface layer,not shown by the other treated materials. Further-more, the bioactive Ca- and P-chemical enrichmentalso played a role in cell response.

CONCLUSIONS

The preliminary study has shown that BSP bioac-tive treatment provides a potential method for im-proving material surfaces in order to enhance thecellular response and subsequently the osteointe-gration properties of titanium implants. In orderto understand more fully the mechanism of ac-tion of this novel treatment, long-term cellularstudies are necessary in order to determine theeffect on proliferation and differentiation. Fur-thermore, the use of primary human osteoblastcells will provide fundamental basic informationnecessary for biomedical application of this novelmaterial treatment.

ACKNOWLEDGEMENTS

The Authors wish to thank the staff of SAMM – Politecnico diMilano – Italy for their technical support and Nanosurfaces srl,a SAMO Group Company, via Matteotti 37, 40057 Cadrianodi Granarolo Emilia (BO) – Italy, for providing the BioSpark™treated materials.

Address for correspondence:Enrico Sandrini, MSDip. Chimica, Materiali e Ingegneria Chimica “G. Natta”Politecnico di MilanoVia Mancinelli, 720133 Milano - [email protected]

43

Giordano et al

REFERENCES

1. Brunette DM, Tengvall P, Textor M, Thompsen P. Metal-lurgy and technological properties of titanium and tita-nium alloys. Titanium in Medicine 2002; 3: 25-51.

2. Bloebaum RD, Dupont JA. Osteolysis from a press-fit hy-droxylapatite-coated implant: a case study. J Arthroplasty1993; 8: 195-202.

3. Buma P, Gardeniers JW. Tissue reaction around a hy-

droxylapatite-coated hip prostheses: case report of re-trieved specimen. J Arthroplasty 1995; 10: 389-95.

4. Zhang XS, Revell PA, Evans MA, Tuke MA, Gregson PJ.In vivo biocompatibility and mechanical study of novelbone-bioactive materials for prosthetic implantation. JBiomed Mater Res 1999; 46: 279-86.

5. Bauer TW, Geesink RC, Zimmermann R, Mamahon JT.Hydroxylapatite-coated femoral stems: Histologicalanalysis of components retrieved at autopsy. J Bone Joint

Surg 1991; 73: 1439-52.6. Martin JY, Schwartz Z, Hummert TW, et al. Effect

of titanium surface roughness on proliferation, dif-ferentiation, and protein synthesis of human os-teoblast-like cells (MG63). J Biomed Mater Res1995; 29: 389-401.

7. Deligianni DD, Katsale N, Ladas S, Sotiropolou D,Amebee J, Missirlis YS. Effect of surface roughnessof the titanium alloy Ti-6Al-4V on human bonemarrow cell response and on protein adsorption.Biomaterials 2001; 22: 1241-51.

8. Lincks J, Boyan RD, Blanchard CR, et al. Responseof MG63 osteoblast-like cells to titanium and tita-nium alloy is dependent on surface roughness andcomposition. Biomaterials 1998; 19: 2219-32.

9. Brunette DM, Cherohudi B. The effects of the sur-face topography of micromachined titanium sub-strata on cell behaviour in vitro and in vivo. J Bio-mech Eng 1999; 121: 49-57.

10. Hench LL, Paschall HA. Direct chemical bond ofbioactive glass ceramic materials to bone and mus-cle. J Biomed Mater Res 1973; 7: 25-32.

11. Cheroudi B, Gould TRL, Brunette D M. Titanium-coat-ed micromachined grooves of different dimensions af-fect ephitelial and connective tissue cells differently invivo. J Biomed Mater Res 1990; 24: 1203-19.

12. Kilpadi DV, Lemon JE. Surface energy charachter-isation of unalloyed titanium implants. J BiomedMater Res 1994; 28: 1419-25.

13. Hay DI, Moreno EC. Differential adsorption andchemical affinities of proteins for apatitic surfaces.Journal of Dentristy Research 1979 58: 930-42.

14. Maxiane SH, Di Stefano T, Melican MC, Tiku M L,Zawadsky JP. Bone cell behavior on matrigel-coatedCa/P coatings of varying crystallinities. J BiomedMater Res 1998; 40: 171-9.

15. Hazan R, Brener R, Oron U. Bone growth to metalimplants is regulated by their surface chemicalproperties. Biomaterials 1993; 14: 970-4.

16. Kokubo T, Li P, Ohtsuki C, Nakanishi K, Soga N,De Groot K. The role of hydrated silica, titania andallumina in inducing apatite on implants. J Bio-med Mater Res 1994; 28: 7-15.

17. Kokubo T, Kim H, Miyaji F, Nakamura T. Prepara-tion of Bioactive Ti and its alloys via simple chemi-cal surface treatment. J Biomed Mater Res 1996;32: 409-17.

18. Kokubo T, Kim H, Oka M, et al. Alkali and heat-treated porous titanium for orthopaedic implants.J Biomed Mater Res 2001; 54: 198 -208

19. Kokubo T, Kim H, Tamura J, et al. Bioactive titani-um: effect of sodium removal on the bone bondingability of bioactive titanium prepared by alkali andheat treatment. J Biomed Mater Res 2001 56: 562-70.

20. Fujibayashi S, Neo M, Kim HM, Kokubo T, Naka-mura T. Osteinduction of porous bioactive titani-um metal. Biomaterials 2004; 25: 443- 50.

21. Sandrini E, Chiesa R, Rondelli G, Santin M, CigadaA. A novel biomimetic treatment for an improvedosteintegration of titanium. Journal of Applied

Biomaterials & Biomechanics 2003; 1: 33-41.22. Yang B, Uchida M, Kim HM, Zhang X, Kokubo T.

Preparation of bioactive titanium metal via anodicoxidation treatment. Biomaterials 2004; 25: 1003-10.

23. Chiesa R, Sandrini E, Rondelli G, Santin M, CigadaA. Osteointegration of titanium and its alloys by an-odic spark deposition and other electrochemicaltechniques. Journal of Applied Biomaterials & Bio-mechanics 2003; 1: 91-107.

24. Kurze P, Kryssman W, Dittrich K-H, Schneider H-G.Process charachteristics and parameters of anodicoxidation by spark deposition (ANOF). Crystal Re-search and Technology 1984; 19: 973-9.

25. Kurze P, Kryssman W, Dittrich K-H, Schneider H-G.Structure and properties of ANOF layers. CrystalResearch and Technology 1984; 19: 93-9.

26. Wei M, Uchida M, Kim HM, Kokubo T, NakamuraT. Apatite forming ability of Ca-O-containing tita-nia. Biomaterials 2002; 23: 167-72

27. Brunette DM, Tengvall P, Textor M, Thompsen P.Surface characterisation techniques. Titanium inMedicine 2002; 5: 105-7.

28. Sitting C, Textor M, Spencer ND, Wieland M, Val-loton PH. Surface charachterisation of implantmateials CP Ti, Ti-6Al-7Nb, and Ti-6Al-4V with dif-ferent pre-treatments. J Mater Sci Mater Med 1999;10: 35-46

29. McCafferty E, Wightman JP. An X-ray photoelec-tron spectroscopy sputter profile study of the na-tive air-formed oxide film on titanium. AppliedSurface Science 1999; 14: 92-100.

30. Poilleau J, Devilliers D, Garrido F, Durrand-Vidal S,Mahe E. Structure and composition of passive tita-nium oxide films. Material Science Engineering1997; 47: 235-43.

31. Boyan BD, Hummert TW, Kieswetter K, SchraubDM, Dean DD, Schwartz Z. Effect of titanium sur-face characteristics on chondrocytes and os-teoblasts in vitro. Scanning Electron Microscopy(Cells & Materials) 1995; 5: 323-35.

32. Davies JE, Lowenberg B, Shiga A. The bone-titani-um interface in vitro. J Biomed Mater Res 1990; 24:1289-306.

33. Cochran DL, Simpson J, Weber H, Buser D. At-tachment and growth of peridontal cells onsmooth and rough titanium. Int J Oral MaxillofacImplants 1995; 9: 289-97.

34. Di Silvio L, Gurav N. Osteoblasts. In: Koller MR,Palsson BO, Masters JRW. Human cell culture.Kluwer Academic Publishers 2001; 221-41.

35. Keller JC, Draughn RA, Wightman JP, DoughertyWJ, Meletiou SD. Charachterisation of sterilizedCP titanium implant surfaces. Int J Oral MaxillofacImplants 1990; 5: 360-7

36. Schwartz Z, Martin JY, Dean DD, Simpson J,Cochran DL, Boyan BD. Effect of titanium surfaceroughness on chondrocyte proliferation, matrixproduction, and differentiation depends on thestate of cell maturation. J Biomed Mater Res 1996;30: 145-55.

44

Titanium for osteointegration: Comparison between a novel biomimetic treatment and commercially exploited surfaces