Titanium surface modification and its effect on the

Amoroso Pier-Francesco Robert J. Adams Mark G. J. Waters David W. Williams

Authors’ affiliations: Amoroso Pier-Francesco, Robert J. Adams, David W. Williams, Department of Oral Surgery, Medicine & Pathology, Dental School, Cardiff University Cardiff, UK Mark G. J. Waters, Department of Matrix Biology and Tissue Repair, Dental School, Cardiff University, Cardiff, UK Correspondence to: David W. Williams Department of Oral Surgery, Medicine & Pathology Dental School Cardiff University Heath Park, Cardiff CF14 4XY UK Tel.:/Fax: þ 44 (0)29 20742442 e-mail: [email protected]

Titanium surface modification and its effect on the adherence of Porphyromonas gingivalis: an in vitro study

Key words: bacterial adherence, Porphyromonas gingivalis, titanium Abstract Aim: Titanium dental implants are an important treatment option in the replacement of missing teeth. Implant failures can, however, occur and may be promoted by the loss of tissue as a result of local bacterial infection (peri-implantitis). Objectives: Bacterial adherence to implant surfaces is believed to be influenced by material surface roughness and surface-free energy parameters. Consequently, the aim of this study was to modify these properties of titanium and identify what effect these modifications had on subsequent bacterial adherence. Materials and methods: In this study, 16 titanium samples of different roughness (Ra 34.57– 449.42 nm) were prepared using specific polishing procedures. A further six samples were chemically altered by argon plasma discharge treatment and immersion in silane solutions to produce different surface hydrophobicities. An in vitro adhesion assay using Porphyromonas gingivalis was used to assess the effect of modification on bacterial adherence. Results: A significant reduction in adhesion to materials categorised as being ‘very smooth’ (Ra 34.57 5.79 nm) was evident. This reduction did not occur with ‘smooth’ (Ra 155.00 33.36 nm), ‘rough’ (Ra 223.24 9.86 nm) or ‘very rough’ (Ra 449.42 32.97 nm) surfaces. Changing material surface hydrophobicity was also not found to effect bacterial adhesion. Conclusions: Adhesion of P. gingivalis to titanium was inhibited at surface roughness levels below those generally encountered for implant collars/abutments (Ra 350 nm). Considerations of these findings may be beneficial in the production of titanium implants in order to reduce bacterial colonisation.

Date: Accepted 30 May 2006 To cite this article: Pier-Francesco A, Adams RJ, Waters MGJ, Williams DW. Titanium surface modification and its effect on the adherence of Porphyromonas gingivalis: an in vitro study. Clin. Oral Impl. Res. 17, 2006; 633–637 doi: 10.1111/j.1600-0501.2006.01274.x

Copyright r Blackwell Munksgaard 2006

Dental implants are root analogues that are surgically placed into the jaw bone and used to support crowns, bridges and dentures. The use of dental implants is now widespread (van Steenberghe et al. 1999) and is an increasingly important alternative to certain conventional dental treatments. The most common dental implant material is ‘commercially pure titanium’ which has good strength, is resistant to corrosion and has a modulus of elasticity similar to that

of bone (Parr et al. 1985). A direct structural and functional connection between bone and the surface of a titanium implant can develop in a process termed as osseointegration (Bra˚nemark et al. 1977). In recent years, modifications have been made to machined titanium implants with the aim of optimising osseointegration. Modifications to increase the speed and success rate of osseointegration have included the use of hydroxyapatite coatings, increasing surface

633

Pier-Francesco et al . Surface modified titanium and adherence of Porphyromonas gingivalis

roughness and surface-free energy. Increasing surface roughness has been achieved by blasting the surface with titanium oxide, analytical etching, sand blasting and acid etching. Failure of dental implants does occur, although the reported rate of failure is variable. Esposito et al. (1998) reported a failure rate of 2812 implants as being 7.7% over 5 years, although failure rates as high as 35% have been described elsewhere (Jaffin & Berman 1991). Implant failures fall broadly into two categories; the failure to achieve osseointegration (early failure) and the failure to maintain osseointegration (late failure). The aetiology of early failure is multi-factorial and includes hostrelated factors, the primary stability of an implant, micro-motion and surgical experience (Sennerby & Roos 1998). Smoking has also been reported as being associated with a higher rate of implant failure (Bain & Moy 1993). Heydenrijk et al. (2002) suggested that late failures occurring within 1 year of loading were caused by host factors and overload while later failures were proposed to be due to overload and peri-implantitis. Peri-implantits is an inflammatory process affecting the tissues around an osseointegrated implant, resulting in the loss of supporting bone (Albrektsson & Isidor 1994). The bacteria implicated in peri-implantitis are primarily those associated with periodontal disease and include strictly anaerobic bacteria such as Porphyromonas gingivalis, Actinobacillus actinomycetemcomitans, Prevotella intermedia and spirochaetes (Mombelli 1997). The prevalence of peri-implantitis is again not clear but has been reported to account for approximately 10% of late implant failures (Mombelli & Lang 1998; Esposito et al. 1999). There are many reported risk factors for peri-implantits including poor oral hygiene, the depth of peri-implant pocket and the surface roughness of the transmucosal component of an implant (Quirynen et al. 2002). The risk factors associated with peri-implantitis appear to be related to the composition of the bacterial environment around an implant and the ability of bacteria to adhere to the implant material (Quirynen et al. 2002). While several investigations have studied the effect of material modification on bacterial adhesion, there are relatively few studies examining modifications aimed at

634 |

Clin. Oral Impl. Res. 17, 2006 / 633–637

promoting osseointegration (Steigenga et al. 2003). Microbial adhesion to biomaterials has previously been related to factors such as surface-free energy (Yoshinari et al. 2000), chemical composition and physical attributes such as material surface irregularities and roughness (Nakazato et al. 1989; Esposito et al. 1997, 1998). Current opinion appears to indicate that low surface-free energy materials, with reduced surface roughness limits plaque accumulation in vivo and that the influence of surface roughness on plaque accumulation appears to be more important than surface-free energy or electrical charge (Quirynen et al. 1999). It is generally accepted that a threshold value of 200 nm represents the average roughness (Ra) below which bacterial adhesion cannot be further reduced (Bollen et al. 1997). In the present study, titanium samples were prepared at different surface roughness and different hydrophobicity. The aim of the study was to assess the effect of such modifications on the adhesion of the peri-implant pathogenic bacterium, P. gingivalis.

Materials and methods Modification of titanium surface roughness

Sixteen samples of commercially pure titanium (Goodfellow Cambridge Limited, Cambridgeshire, UK; 10 10 1 mm) were prepared. This preparation involved polishing the samples for 24 h using an Eco mini dry rotating machine (Otec Pra¨zisionsfinish Gmbh, Straubenhardt-Feldrennach, Germany) to produce materials with a standardised smooth surface. Twelve of these samples were subjected to additional polishing regimes using different grades of polishing discs, or sandblasting with either glass or aluminium oxide beads. From these 16 samples, four groups of materials (four specimens in each) were obtained of different surface roughness. The groups were categorised as being ‘very smooth’ (hand polished with rotary brushes for a mirror finishing process), ‘smooth’ (polished only with the Eco mini dry machine), ‘rough’ (sandblasted with 50 mm glass beads; 5 bars pressure) and ‘very rough’ (sandblasted with 50 mm aluminium oxide beads; 6.5 bars pressure). The roughness of these samples was measured using

an atomic force microscope (AFM; Digital Instruments, Santa Barbara, CA, USA) in tapping mode and scan sizes were 400 mm2 with a scan rate o1 Hz. Additional AFM parameters such as integral grain, proportional gain and amplitude were modified during scanning to generate the best resolution of the image (Bowen et al. 2001). The Ra value (Verran & Boyd 2001) and the Zrange (Rz) were measured at four sites on each specimen. Modification of titanium surface hydrophobicity

An additional six titanium samples (dimensions 20 10 1 mm) were polished with the Eco mini dry machine using a cotton rotating disc. The Ra value of these samples was later determined to be 86 15.54 nm. Two specimens were not treated further and acted as controls. Four specimens were placed in an Emscope sputter coater (SC500A provided by Surpluseq Com. Inc., Phoenix, AZ, USA) and exposed to argon plasma discharge on all sides for 4 min on three occasions. The plasma-treated samples were then paired and immersed in a specific silane (ABCR GmbH & Co. KG, Karlsruhe, Germany) solution (2% in ethanol) for 1 h. The two silanes were 2-methoxy-[(polyethyleneoxy) propyl]-trimethoxysilane and 1H,1H,2H, 2H-perfluorooctytriethoxysilane, with the former used to generate a hydrophilic surface and the latter a more hydrophobic surface. Changes in hydrophobicity of the surfaces were assessed using a dynamic contact angle analyser (Cahn Instruments Inc., Thermo Electron Corporation, Basingstoke, UK). P. gingivalis adherence assay

P. gingivalis ATCC 33277 was used for all adhesion experiments. The bacterium was cultured for 72 h on fastidious anaerobe agar (Lab M, Bury, UK) and colonies resuspended in 0.14 M phosphate-buffered saline (PBS), pH 7.2 to a turbidity equivalent to a McFarland 0.5 index. A portion (50 ml) of this suspension was used to inoculate fastidious anaerobe broth (Lab M), which was incubated anaerobically at 371C for 60 h. The broths were centrifuged (4000 g for 10 min) and the bacterial pellet was washed twice using PBS. P. gingivalis was finally diluted in PBS to a standard concentration (0.001 absorbance at


Table 1. Average roughness (Ra) of titanium and adhesion of Porphyromonas gingivalis Specimen description

Ra (nm) SD

Very smooth Smooth Rough Very rough

34.57 155 223.24 449.42

5.79 33.36 9.86 32.97

Adhesion of P. gingivalis (median values of % coverage)

Adherence range

0.0087 1.26 0.57 0.67

0.0009–0.1 0.09–16.3 0.06–12.15 0.15–6.68

The Kruskal–Wallis statistic test demonstrated a significant difference in roughness between each of these specimen groups of samples (P ¼ 0.019).

Table 2. Surface hydrophobicity of titanium and advancing and receding angle Silane treatment

Hydrophilic

Mean SD Hydrophobic

Mean SD Untreated

Mean SD

Advancing contact angle

Receding contact angle

Pre-treatment

Post-treatment

Pre-treatment

Post-treatment

85.29 75.45 85.26 78.95 76.01 91.62 78.58 73.78 80.62 6.17

44.671 44.66 53.5 54 46.17 53.73 54.29 46.21 49.65 4.56

32.73 32.73 26.13 22.8 37.6 38.33 24.28 23.51 29.76 6.36

39.1 38.03 38.05 39.1 42.97 43.4 42.94 43.17 40.85 2.47

76.561 76.831 74.491 83.431 87.071 93.591 85.471 93.241 83.84 7.43

99.34 99.44 107.1 104.72 101.3 90.89 90.95 101.3 99.38 5.84

26.82 26.77 26.21 26.86 39.59 39.54 39.92 39.68 33.17 6.96

55.78 55.1 64.69 65.46 39.52 57.05 39.27 57.05 54.24 9/96

88.09 75.97 86.76 88.2 90.39 86.28 89 80.4 85.64 4.91

N/A N/A N/A N/A N/A N/A N/A N/A

42.62 42.37 42.29 44.7 38.21 42.59 43.12 38.39 41.79 2.28

N/A N/A N/A N/A N/A N/A N/A N/A

The difference in the advancing and angle before the plasma treatment was not statistically significant (P ¼ 0.08). The difference in values of the advancing and angle after the plasma treatment was statistically significant (P ¼ 0.001).

660 nm) as determined by spectrophotometric measurement. Titanium samples were completely immersed in the prepared bacterial suspensions and incubated by rotation on a mixer at 371C for 1 h. The titanium was removed and washed with PBS. Adherent bacteria were stained with 0.1% w/v acridine orange for 60 s and observed by fluorescent microscopy (Olympus Optical Co. Ltd., Southall, UK). Five digital images were obtained for each sample. Four fields of views (each of 0.0316 mm2) were located 3 mm away from the two edges of the sample corners. The fifth field of view (0.0316 mm2) was at the centre of the titanium sample. Enumeration of adherent bacteria utilised

specific imaging software (Openlab; Improvision, Coventry, UK) and this quantified the total number of pixels corresponding to bacteria. Using a graticule, it was then possible to calculate the total percentage area coverage by adherent bacteria.

Statistical analysis

The Kruskal–Wallis test was used to determine whether there was a statistically significant difference in the surface roughness and contact angle parameters. In the case of bacterial adherence analysis both the Kruskal–Wallis and Mann–Whitney test were used.

Results Analysis of surface hydrophobicity

roughness

and

The Ra ( SD) data obtained from AFM scans for the titanium sample groups are presented in Table 1. The range of roughness recorded was 34.57–449.42 nm and statistical analysis demonstrated a significant difference between each group of samples (P ¼ 0.019). The mean ( SD) advancing and receding contact angles in water of six titanium samples prior and post-silane treatment are presented in Table 2. After treatment with the ‘hydrophobic’ silane a significant (P ¼ 0.001) increase in both the advancing and receding contact angles was noted. This was in contrast with the ‘hydrophilic’ silane-treated materials where a significant (P ¼ 0.001) decrease in the advancing contact angle was evident. Effect of surface roughness on adhesion of P. gingivalis

Significant differences in the bacterial adhesion to the four surface roughness groups were evident (P>0.001). A highly significant difference between the very smooth and other sample groups (P ¼ 0.001; Fig. 1; Table 1) was detected. There were no differences in bacterial adherence evident between these other groups of materials. Effect of hydrophobic changes on the adhesion of P. gingivalis

The percentage area covered by adherent bacteria to titanium surfaces of different hydrophobicity is presented in Table 3. There was no significant difference in adhesion to these different surfaces (P ¼ 0.273).

Discussion Dental implants generally have good longterm success rates and offer an increasingly important treatment modality in the replacement of missing teeth. Dental implants can however fail and in recent years there has been much focus on determining the risk factors associated with failure (Esposito et al. 1999). One such risk factor is peri-implantitis caused by a range of bacteria including P. gingivalis, P. intermedia, A. actinomycetemocomitans, Fusobacterium nucleatum and Bacteroides spp. (Mombelli 1997). In order to cause infec-

635 |



tion the bacteria must first colonise and be retained at the implant site. In practice it has been suggested that the use of an implant surface inhibitory to bacterial adherence may contribute to reducing periimplantitis. The aim of this study was, therefore, to chemically and physically modify titanium surfaces and assess the effect of such modification on the adherence of the bacterium P. gingivalis. It is generally believed that roughened surfaces influence microbial colonisation by enhancing microbial retention within surface irregularities. Verran & Boyd (2001) have proposed three categories of surface roughness, termed as macro-

Fig. 1. Porphyromonas gingivalis adherence to (a) smooth and (b) very smooth titanium. Original magnification 600.

(Ra 10 mm), micro- (Ra 1 mm) and nano-roughness (Ra 0.2 mm). Microroughness has been suggested to be appropriate for dental implants (Bollen et al. 1997) and it has been postulated that a surface with an Ra below 0.2 mm is unlikely to promote microbial adherence due to the larger size of most bacteria (NB P. gingivalis is approximately 1.5 1 mm in size). In the present study 14 (four very smooth, four smooth and the six chemically coated) of the 22 titanium samples had Ra values below 0.2 mm. We would, therefore, have expected these 14 samples to have the same percentage coverage by P. gingivalis. However, the percentage coverage of P. gingivalis on very smooth (Ra 34 nm) titanium surfaces was significantly lower than that detected on smooth surfaces (Ra 155 nm) even though this is lower than the Ra value of 200 nm. There was also no significant difference in percentage coverage of P. gingivalis between ‘smooth’ samples (Ra 155 nm) and ‘very rough’ samples (Ra 449 nm). Therefore, while our results largely concur with those of Bollen et al. (1997), that smooth surfaces are less likely to be associated with bacterial adhesion; this present study suggests a lower surface roughness cutoff value (between 34 and 155 nm) for reduced adhesion of P. gingivalis (Bollen et al. 1997). Interestingly, increasing the surface roughness above 155 nm did not enhance the adherence of P. gingivalis and it could be postulated the increased size of surface irregularities was then too large to offer increased bacterial retention. Importantly, commercially available Bra˚nemark-type dental implants display a range of surface roughness (350– 2500 nm) exceeding the value determined in this study to reduce bacterial adhesion.

Table 3. Surface hydrophobicity of titanium and adhesion of Porphyromonas gingivalis Specimen group

Description

Adhesion of P. gingivalis (median % coverage)

Range of adherence

1a 1b 2a 2b 3a 3b

Hydrophilic

0.0253 0.0464 0.0389 0.0835 0.0434 0.153

0.00345–0.285 0.0062–5.12 0.00585–2 0.00186–4.18 0.0134–2.38 0.00892–6.2

Hydrophobic Untreated Control

No significant difference in adhesion to these specimens was evident (P ¼ 0.273). Ra values of 86 15.54 nm were obtained for the samples used for chemical alteration and the Kruskal–Wallis test did not demonstrate significant difference in roughness between each of these groups of samples (P ¼ 0.416). Ra, average roughness.

636 |


In addition to the physical effects of titanium surfaces, the influence of chemical modification on P. gingivalis adherence was also investigated. In these experiments, the relative bacterial adherence to the controls and silane-treated materials was similar and indicated that under these in vitro conditions physical parameters of the titanium were more influential in bacterial retention than chemical factors. This is in agreement with Quirynen et al. (1999) who stated that surface roughness was more important in plaque accumulation around an implant than surface-free energy. Microbial adherence to surfaces are obviously more complex in the presence of host conditioning films where proteins can act as receptors, potentially masking the underlying surface characteristics of the material. Such an issue was not addressed in this study and represents an area for future research. Furthermore, co-adherence between bacteria often facilitates the attachment of organisms normally incapable of binding to host surfaces. The result of such co-adherence can ultimately lead to the development of a biofilm community. Biofilm bacteria behave differently in terms of their physiology and resistance to antimicrobials compared with their planktonic (free living) counterparts and therefore once established can represent a significant challenge for removal. In such instances prevention of initial adherence of the pioneer coloniser would be preferable. Although this study only compares the effects of two titanium surface parameters on the adhesion of one bacterial species (albeit a significant periodontal pathogen), considerations of these findings could offer a starting point in future design of titanium implants/abutments aimed at reducing bacterial colonisation. However, extensive in vivo assessment would need to be undertaken to account for additional host variables (e.g. pellicle proteins and additional bacterial groups) that may be present at the implant insertion site.

Acknowledgements: This work was partly supported by a grant from the Universita` Cattolica S. Cuore, Rome. We are grateful to Mrs Pat Bishop and Mrs Gillian Fellows (Dental School, Cardiff) for their assistance with the microbiological aspects of this study.


References Albrektsson, T. & Isidor, F. (1994) Consensus report of session IV. In: Lang, N.P. & Karring, T., eds. Proceedings of the 1st European Workshop on Periodontology, 365–369. London: Quintessence Publishing Co. Ltd. Bain, C.A. & Moy, P.K. (1993) The association between the failure of dental implants and cigarette smoking. International Journal of Oral & Maxillofacial Implants 8: 609–615. Bollen, C.M.L., Lambrecht, P. & Quirynen, M. (1997) Comparison of surface roughness of oral hard materials to the threshold surface roughness for bacterial plaque retention: a review of the literature. Dental Materials 13: 258–269. Bowen, W.R., Lovitt, R.W. & Wright, C.J. (2001) Atomic force microscope studies of stainless steel: surface morphology and colloidal particle adhesion. Journal of Materials Science. Materials in Medicine 36: 623–629. Bra˚nemark, P.I., Hansson, B.O., Adell, R., Breine, U., Lindstro¨m, J., Halle´n, O. & Ohman, A. (1977) Osseointegrated implants in the treatment of the edentulous jaw. Experience from a 10-year period. Scandinavian Journal of Plastic and Reconstructive Surgery 16: 1–132. Esposito, M., Hirsch, J.M., Lekholm, U. & Thomsen, P. (1997) Failure patterns of four osseointegrated oral implant systems. Journal of Materials Science. Materials in Medicine 8: 843–847. Esposito, M., Hirsch, J.M., Lekholm, U. & Thomsen, P. (1998) Biological factors contributing to failures of osseointegrated oral implants (I). Suc-

cess criteria and epidemiology. European Journal of Oral Sciences 106: 527–551. Esposito, M., Hirsch, J.M., Lekholm, U. & Thomsen, P. (1999) Differential diagnosis and treatment strategies for biological complications and failing implants: a review of the literature. The International Journal of Oral & Maxillofacial Implants 14: 473–490. Heydenrijk, K., Meijer, H.J., van der Reijden, W.A., Raghoebar, G.M., Vissink, A. & Stegenga, B. (2002) Microbiota around root-form endosseous implants: a review of the literature. The International Journal of Oral & Maxillofacial Implants 17: 829–838. Jaffin, R.A. & Berman, C.L. (1991) The excessive loss bra˚nemark fixtures in type IV bone: a 5-year analysis. Journal of Periodontology 62: 2–4. Mombelli, A. (1997) Etiology, diagnosis, and treatment considerations in peri-implantitis. Current Opinion in Periodontology 4: 127–136. Mombelli, A. & Lang, N.P. (1998) The diagnosis and treatment of peri-implantitis. Periodontology 2000 17: 63–76. Nakazato, G., Tsuchiya, H., Sato, M. & Yamauchi, M. (1989) In vivo plaque formation on implant materials. The International Journal of Oral & Maxillofacial Implants 4: 321–326. Parr, G.R., Gardner, L.K. & Toth, R.W. (1985) Titanium: the mystery metal of implant dentistry. Dental materials aspect. The Journal of Prosthetic Dentistry 54: 410–414. Quirynen, M., De Soete, M. & van Steenberghe, D. (1999) Intra-oral plaque formation on artificial

surfaces. In: Lang, N.P., Karring, T. & Lindhe, J., eds. Proceedings of the 3rd European Workshop on Periodontology, 102–129. Berlin: Quintessence Books. Quirynen, M., De Soete, M. & van Steenberghe, D. (2002) Infectious risks for oral implants: a review of the literature. Clinical Oral Implants Research 13: 1–19. Sennerby, L. & Roos, J. (1998) Surgical determinants of clinical success of osseointegrated oral implants: a review of the literature. International Journal of Prosthodontics 11: 408–420. Steigenga, J.T., Al-Shammari, K.F., Nociti, F.H., Misch, C.E. & Wang, H.L. (2003) Dental implant design and its relationship to long-term implant success. Implant Dentistry 12: 318–324. van Steenberghe, D., Quirynen, M. & Naert, I. (1999) Survival and success rates with oral endosseous implants. In: Lang, N.P., Karring, T. & Lindhe, J., eds. Proceedings of the 3rd European Workshop on Periodontology, 242–254. Berlin: Quintessence Books. Verran, J. & Boyd, R.D. (2001) The relationship between substratum surface roughness and microbiological and organic soiling: a review. Biofouling 17: 59–71. Yoshinari, M., Oda, Y., Kato, T., Okuda, K. & Hirayama, A. (2000) Influence of surface modifications to titanium on oral bacterial adhesion in vitro. Journal of Biomedical Material Research 52: 388–394.

637 |
