- Email: [email protected]
Comparative study of osteoconduction on micromachined and alkali-treated titanium alloy surfaces in vitro and in vivo
ARTICLE IN PRESS
Biomaterials 26 (2005) 1793–1801
Comparative study of osteoconduction on micromachined and alkali-treated titanium alloy surfaces in vitro and in vivo Xiong Lua, Yang Lenga,*, Xingdong Zhangb, Jinrui Xub, Ling Qinc, Chun-wai Chanc a
Department of Mechanical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China b Engineering Research Center in Biomaterials, Sichuan University, Chengdu, China c Department of Orthopaedics and Traumatology, The Chinese University of Hong Kong, Hong Kong, China Received 2 March 2004; accepted 3 June 2004 Available online 17 July 2004
Abstract This study sought to evaluate osteoconduction of Ti-6Al-4 V surfaces under various conditions, including micro-patterned, alkalitreated, micro-patterned plus alkali-treated, and surfaces without any treatment as the control. The through-mask electrochemical micromachining (EMM) was used to fabricate micro-hole arrays on the titanium alloy surface. In vitro calcium phosphate formation on titanium surfaces was in static and dynamic simulated body fluid (SBF). In vivo comparison was conducted in the medullary cavity of dog femur using the implant cages which could provide the same physiological environment for specimens with different surface conditions. In vitro experiments indicate good conduction of calcium phosphate on the alkali-treated surfaces, and also better calcium phosphate deposition on the micro-hole surface than on the flat surfaces in dynamic SBF. In vivo experiments confirm the beneficial effect of alkaline treatment on osteoconduction. The results of in vivo experiments also indicate a synergistic effect of the alkaline treatment and the topographic pattern on osteoconduction. r 2004 Elsevier Ltd. All rights reserved. Keywords: Osteoconduction; Micro-pattern; Alkaline treatment; Simulate body fluid; Animal model
1. Introduction Titanium and titanium alloys are among the most popular materials for orthopedic implants due to their biocompatibility, excellent corrosion resistance, good mechanical properties and lightness [1]. There have been efforts to improve the osseointegration capability of the titanium implants by enhancing osteoconduction on their surfaces using surface morphology and chemistry modifications. Various methods have been used to develop morphological structures of titanium implant surfaces for promoting bone ingrowth and fixation between implants and bone. Among them, powder/fiber/wire mesh metallurgical sintering [2,3], plasma spray processing [4,5] and surface blasting [6] are the widely used methods to modify surface topography of load-bearing titanium *Corresponding author. Tel.: +86-852-2358-7185; fax: +86-8522358-1543. E-mail address: [email protected] (Y. Leng). 0142-9612/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2004.06.009
implant surfaces. These methods may have adverse effects on mechanical properties because of possible residue stress from high temperature process and stress concentrations at the interface between the porous layer and the substrates. Also, these methods generate irregular and random surface geometry which precludes further quantitative analysis of the osseointegration between bone and the implant. Techniques for producing regular surface patterns on titanium implants have been developed, such as reactiveion etching (RIE) [7,8], laser machining [9–12] and the through-mask electrochemical micromachining (EMM) [13–16]. The first two techniques potentially introduce chemical heterogeneity, which might affect osteoblasts behavior. EMM can produce well-defined surface patterns on titanium surfaces without introducing surface chemistry heterogeneity. For surface chemistry modification, various calcium phosphate (Ca–P) coatings on titanium implants have been proposed to improve osseointegration as reviewed in [17]. The adhesive strength between the Ca–P coating
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
Titanium specimens used in this work were made from Ti-6Al-4 V plates with 1 mm in thickness (Baoji Special Iron and Steel Co. LTD., China). A throughmask electrochemical micromachining (EMM) technique was employed to fabricate micropatterns on the Ti specimens as follows. A surface pattern with micro-hole arrays was produced on a photo-mask and transferred to the Ti specimens (20 mm 20 mm) by contact printing photolithography using a positive photoresist. Then, the Ti specimens were attached on a rotating shaft and served as the anode for electrochemical etching. The anode rotated at a speed of 500 rpm in 5 mol/l sodium bromide aqueous solution during electrochemical etching. The dimensions of the micro-holes were measured with a surface profiler (WYKO NT3000, Veeco Instruments Inc., USA). The surface roughness of 0.3870.03 mm was also measured by the surface profiler. To eliminate possible chemical heterogeneity between the etched and non-etched areas, the specimens were placed in diluted sulfate acid (1:5 in volume ratio) for 3 h. 2.2. Alkaline treatment Alkaline treatment was performed by immersing the Ti specimens in 100 ml of 5 mol/l NaOH aqueous solution at 60 C for 24 h. After the alkaline treatment, the specimens were gently washed with distilled water and dried at 40 C for 24 h in an air atmosphere. The
SBF used in this study was prepared following a revised recipe by Kim et al. [38] as follows. Chemicals were dissolved in one liter of distilled water in the sequence of NaCl (5.403 g), NaHCO3 (0.736 g), Na2CO3 (2.036), KCl (0.225 g), K2HPO4 (0.182 g), MgCl2 . 6H2O (0.310 g), HEPES (2-(4-(2-hydroxyethyl)-1-piperazinyl) ethanesulfonic acid) (11.928 g), CaCl2 (0.293 g), Na2SO4 (0.072 g), 1 mol/l NaOH. The fluid was buffered at pH 7.40, 36.5 C with HEPES and 1 mol/l NaOH. The ability of forming Ca–P was evaluated in both static and dynamic SBF environments. The dynamic SBF was realized by pumping the SBF in/out of a 100 ml sample chamber through a two channel peristaltic pump. The flow rate was 3 ml/min, i.e., 100 ml SBF stored in the sample chamber was refreshed at a rate of 3 ml/min by an in/out flow. This flow rate is close to that of body fluid in the human muscle environment [28]. The SBF in the storage tank was replaced with fresh one every three days in order to keep the solution ion concentration stable. The Ca–P deposited on the specimen was examined using an SEM (JEOL 6300, JEOL, Japan) with Energy Dispersive X-ray Spectrometer (EDS). 2.4. In vivo experiments The Ti specimens for the in vivo experiments were cut into 6 mm 8 mm rectangular plates. Seven plates were assembled into a small ultra high molecular weight polyethylene (UHMWPE) cage developed by Ricci et al. [27]. As shown in Fig. 1, the cage with multiple channels enables us to compare the specimens with various surface conditions directly: (a) control without surface treatment, (b) surface patterned, (c) surface patterned
30
1 2
3 4
5 6
7
Unit: mm
8.5
2.1. Electrochemical micromachining
2.3. In vitro experiments
1 1.5
2. Materials and methods
alkali-treated specimens were then placed on an alumina boat heated to 600 C at a rate of 5 C/min for 1 h and cooled to room temperature in the furnace.
4
and titanium substrate generate great concern regarding the mechanical integrity of the implants in clinical applications. Kokubo et al. [18–26] introduced an alkaline treatment on titanium surfaces, which might provide strong bone-bonding ability and high bone affinity. Note that the alkaline treatment can be applied to titanium surfaces with any type of morphological features. This study sought to compare the osteoconduction of titanium surfaces that were engineered by surface patterning and alkaline treatment. The osteoconduction of the micropatterned, alkali-treated, micropatterned plus alkali-treated, and control surfaces were evaluated by direct in vitro and in vivo comparison. In vitro ability of bioactive Ca–P formation was examined in static and dynamic simulated body fluid (SBF). The in vivo comparison of different surface conditions was conducted in an implant cages in dog femur, a technique developed by Ricci et al. [27]. The results of this study will guide to future developments of bioactive titanium implants.
1 1
1794
1.5 1
3
1
3
1
3
1 1.5
2
6
Fig. 1. Configuration and arrangement of implant cages for animal model. The dark area marks the cross section of the UMWPE cage and the blank area between the Ti plates is the channels of bone ingrowth. Specimens in the cage are marked by number: (1), Control Ti plate without any treatment; (2) and (3), micro-patterned Ti plates; (4) and (5), micro-patterned plus alkali-treated Ti plates; (6) and (7), alkalitreated Ti plates.
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
and AH-treated, and (d) surface AH-treated. The cage location in dog femur and the short cage compared with the femur length ensured that rational comparisons among the specimens in nearly identical physiological environments could be made (Fig. 2). The opening space (1 6 8 mm3) between the specimen in the cage
Fig. 2. The X-ray radiograph shows the location of cage implanted in dog femur.
1795
ensures the bone ingrowth. The dog was anesthetized with an intramuscular injection of 10% ketamine and 1% xylazine (v : v ¼ 0:25 ml: 0.25 ml per kg body weight) before cage implantation. A sterilized cage was implanted into a rectangle femur cavity of 7-mm deep, 8-mm wide, 18-mm long, and fixed on cortical bone through titanium screws (Fig. 2). The specimens in the cages were placed into the medullary cavity immersed in blood. Two cages were implanted the right and left femurs of same dog for one and two months, respectively. The dog was sacrificed for retrieving the cages. The specimens implanted for one month were examined under SEM while those implanted for two months were fixed for histological observations. The specimens were fixed in 10% formalin in 0.075 mol/l phosphate buffer pH 7.25 for 3 days at room temperature, washed in a buffer solution for 10 min, and dehydrated in serial concentrations of ethanol (70%, 80%, 90%, 99%, 100% and 100% v/v) every 2 days. Then, they were embedded in polymerized methyl methacrylate (PMMA) by immersing the specimens into xylene/100%EtOH, xylene, unpolymerized methyl methacrylate(UMMA)/ xylene (ratio 1:1, v/v), UMMA I, UMMA II, UMMA III, polymerized methyl methacrylate (PMMA) in
Fig. 3. (a) SEM micrograph of micro-hole array obtained from EMM; (b) Single micro-hole; (c) Micro-hole dimensions from surface profiler measurement.
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
1796
Fig. 4. SEM micrographs and EDS spectrum of Ca–P on the Ti-6Al-4V surfaces after 15 days in static SBF. (a) Control specimen; (b) Micropatterned specimen; (c) Alkali-treated specimen; (d) EDS spectrum of c; (e) Surface of micro-patterned plus alkali-treated specimen; (f) Bottom of micro-hole of micro-patterned plus alkali-treated specimen.
sequence with a two-day interval. Thin sections of 250 mm were cut with a saw microtome (SP1600, Leica, Germany). Then, the sections were stuck to the PMMA slides (1 mm thick) and further grinded and polished to 100 mm thick with a grinding and polishing system (RotoPol-21, Struers A/S, Denmark). The polished sections were stained with toluidine-blue. Histomorphology was performed using a light microscope (DM RXA2, Leica, Germany). The bone growth along the surfaces with and without micro-pattern was quantitative analyzed using image analysis software (Qwin Standard, Leica, Germany). The efficiency of osteoconduction was quantified by the bone attachment index (BAI) defined as: BAI ¼
bone area 100%; total tissue area
in which the total tissue area is a rectangle area adjacent to bone/Ti interface in the opening spaces of the cage shown in Fig. 1. The area is 500 mm wide from the interface. The bone area is that occupied by newly formed bone in the total tissue area. Both of the total tissue area and the bone area are determined numerically by the image analysis software.
3. Results and discussion 3.1. Micro-hole characterization Micro-hole arrays on Ti specimen surfaces were successfully produced by the EMM method as shown in the SEM micrographs (Figs. 3a and b). The dimensions of holes measured by the 3D surface profiler
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
Fig. 5. SEM micrographs of Ca–P growth on an alkali-treated microhole by immersing in static SBF. (a) A view of single hole; (b) A view of Ca–P growth inside and outside the hole by tilting view angle of 45 .
indicated a spherical shape with a radius of B300 mm as shown in Fig 3c. The diameter of the holes was sufficiently large to facilitate the ingrowth of blood vessels and the supply of osteocytes [29–31]. These dimensions were similar to the pore size of porouscoated prostheses currently undergoing clinical trials [30]. Apparently, the EMM technique can provide a topographic structure on the titanium surfaces without mechanical property degradation and surface chemical heterogeneity.
1797
not be detected on the micro-patterned surfaces without the alkaline treatment either (Fig. 4b). On the alkalitreated surfaces, a thick layer of Ca–P was formed (Fig. 4c) and confirmed by EDS as shown in Fig. 4d. The pattern of micro-hole arrays seems not to affect the Ca–P formation in static SBF. Ca–P was formed on both the inside and outside surfaces of the micro-holes (Fig. 4e and f). No difference in Ca–P formation capability on the micro-hole surfaces and on the flat surfaces was observed. Growth of globule-like Ca–P crystals could well adapt orientation change of specimen surfaces caused by the geometrical configuration of micro-holes. The Ca–P globules always grew along the direction of surface normal even on the inner wall of the micro-hole (Fig. 5). Apparently, in static SBF the topographic condition of surface has little influence on Ca–P formation, while the chemical treatment of the surface makes a significant difference. The results of experiments in dynamic SBF indicate an effect of surface topography on Ca–P formation in addition to the effect of alkaline treatment. Fig. 6 presents the SEM micrographs of the specimen surfaces after being immersed in dynamic SBF for 15 days. Generally, Ca–P formation on the specimen surfaces in dynamic SBF was less than that in static SBF, and this observation is consistent with prior reports [32–34]. Again, no Ca–P was formed on the specimen surface without any treatment (Fig. 6a). Ca–P was not formed on the micro-patterned surface either (Fig. 6b). On the alkali-treated surfaces (Fig. 6c), Ca–P does not form a thick layer as in static SBF. This difference is also shown in an EDS spectrum (Fig. 6d). In dynamic SBF, Ca–P formation is more pronounced on the bottom of microholes (Fig. 6f) than on flat surfaces (Fig. 6e). Our finding is similar to the results of testing porous Ca–P ceramics in dynamic SBF [33], in which the Ca–P is formed in the inner pores instead of outer surface. Researchers believed that the dynamic SBF tests provide an environment more similar to that in living body than that of static SBF test. Thus, the Ca–P formation in dynamic SBF should be more relevant to Ca–P deposition in vivo [32–37]. Results of our in vivo experiments perhaps clarify any speculation from in vitro experiments. 3.3. In vivo comparisons
3.2. In vitro comparisons In vitro experimental results indicate the differences in the ability to form Ca–P among the specimen surfaces with different treatments. The results of the in vitro experiments in static SBF show the merit of the alkalitreated surfaces. Fig. 4 shows the comparison of the specimen surfaces with different treatment after being immersed in static SBF for 15 days. No deposited Ca–P was found on the control surface (Fig. 4a). Ca–P could
Bone ingrowth was observed in the openings between specimens with the alkali-treated surfaces. X-ray radiographs of the cages retrieved from dog indicate that some openings of the cage implanted for two months were filled with dense substance, but not in the openings of the cage implanted for one month (Fig. 7). The specimens implanted for two months were prepared for histological examination, while the specimens implanted for one-month were examined in SEM.
ARTICLE IN PRESS 1798
X. Lu et al. / Biomaterials 26 (2005) 1793–1801
Fig. 6. SEM micrographs and EDS spectrum of Ca–P on the Ti-6Al-4V surfaces after 15 days in dynamic SBF. (a) Control specimen; (b) Micropatterned specimen; (c) Alkali-treated specimen; (d) EDS spectrum of c; (e) Flat area of the micro-patterned plus alkali-treated specimen; (f) Microhole bottom area of the micro-patterned plus alkali-treated specimen.
Fig. 7. X-ray Radiograph of retrieved cages after one month (Left) and two months (right) of implantation.
The SEM examinations revealed the formation of calcium phosphate (Ca–P) on the specimen surfaces having alkaline treatment (Fig. 8b), while nothing was found on the control specimen (Fig. 8a). Rod-like depositions indicate Ca–P formation on the alkalitreated surfaces but no formation of new bone. A thin layer of new bone seems to have formed on certain flat surface areas (Fig. 8c) with alkaline treatment and a thick layer of new bone covers the hole surfaces of the same specimen (Fig. 8d). Apparently, there was better osteoconduction on the surfaces with alkaline and topographic treatments. Fig. 9 shows the histological micrographs of specimens after two months of implantation. Such a side-byside comparison of specimens with various surface conditions shows effects of alkaline and topographic
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
1799
Fig. 8. SEM micrographs Ti-6Al-4V surfaces after one-month implantation. (a) Control specimen; (b) Ca–P deposited on alkali-treated surface; (c) Flat area of the micro-patterned plus alkali-treated specimen; (d) Micro-hole bottom area of the micro-patterned plus alkali-treated specimen.
treatments on osteoconduction. No new bone was found on the cage openings where the specimen surfaces do not have alkaline treatment, no matter whether they had surface micro-pattern. There was a large amount of new bone in the cage openings where the specimen surfaces had alkaline treatment and micro-holes. Also, there was new bone in the opening where the specimen surfaces had alkaline treatment but no micro-holes. Image analysis indicates that the average BAI value of the specimen with the hole-pattern and alkali-treatment is around 71% while that of specimen with only alkalitreatment is around 31%. Although more accurate assessment by statistical analysis of a large number of samples is no available, this semi-quantitative estimation already indicates the merit of hole-pattern in guiding bone growth. Table 1 summarizes the in vivo results which indicate that alkaline treatment plays a primary role in enhancing osteoconduction. Although topographic patterning alone did not enhance new bone growth in our histological evaluation, our study shows a synergistic effect of combining topographic and alkaline treatment on osseointegration.
Fig. 9. Histological micrographs of specimens implanted for twomonths after toluidine-blue staining (original magnification: 20). (a) Section near the cage center; (b) Section near the side surface of the cage. (1), Ti plate without any treatment; (2) and (3), micro-patterned Ti plates; (4) and (5), micro-patterned plus alkali-treated Ti plates; (6), alkali-treated Ti plates.
1
2
3
4
5
6
(a)
1
(b)
2
3 4
5
6
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
1800 Table 1 In vivo evaluation of osteoconduction
Control Pattern AH treatment Pattern/AH treatment
One month
Two month
CaP
New Bone
New Bone
x x O O
x x O? O
x x O OO
x-negative, O-positive, O?-uncertain, and OO-very positive.
4. Conclusions The present study confirms the beneficial effect of alkaline treatment on osteoconduction. The study also reveals a synergistic effect of combining topographic and alkaline treatment from both in vitro and in vivo experiments. The experiments in dynamic SBF indicate good conduction of calcium phosphate on the alkali-treated surfaces, and also better calcium phosphate deposition on the hole-patterned surface than on the flat surfaces. A side-by-side comparison of surfaces with various surface conditions in animal model shows the importance of alkaline treatment in bone growth. The bone growth on the alkali-treated surfaces can further be enhanced by topographic patterning.
Acknowledgements This project was financially supported by the Research Grants Council of Hong Kong (HKUST 6037/ 02E) and the Funds for High Impact Areas at The Hong Kong University of Science & Technology. The microfabrication was done in the Microelectronics Fabrication Facility of Hong Kong University of Science and Technology. The authors wish to acknowledge valuable suggestions about electrochemical etching from Prof Changjian Lin and Dr. Haobing Hu of the State Key Laboratory for Physical Chemistry of Solid Surfaces, Xiamen University, China.
References [1] Park JB, Lakes RS. Biomaterials: an introduction, 2nd ed. New York: Plenum Press; 1992. p. 89–92. [2] Pilliar RM. Porous-surface metallic implants for orthopedic applications. J Biomed Mater Res Appl Biomater 1987;21:1–33. [3] Pilliar RM. P/M processing of surgical implants: sintered porous surfaces for tissue-to-implant fixation. Int J Powder Metall 1998; 34:33–45. [4] Simmons CA, Valiquette NV, Pilliar RM. Osseointegration of sintered porous-surfaced and plasma spray-coated implants: an animal model study of early post implantation healing response and mechanical stability. J Biomed Mater Res 1999;47:127–38.
[5] Yang YZ, Tian JM, Tian JT, Chen ZQ, Deng XJ, Zhang DH. Preparation of graded porous titanium coatings on titanium implant materials by plasma spraying. J Biomed Mater Res 2000;52:333–7. [6] Jinno T, Goldberg VM, Davy D, Stevenson S. Osseointegration of surface-blasted implants made of titanium alloy and cobaltchromium alloy in a rabbit intramedullary model. J Biomed Mater Res 1998;42:20–9. [7] Jaeger NAG, Brunette DM. Production of microfabricated surfaces and their effects on cell behavior. In: Brunette DM, Tengvall P, Textor M, Thomsen P, editors. Titanium in medicine. Berlin, Hong Kong: Springer; 2001. p. 343–74. [8] den Braber ET, Jansen HV. Scanning electron microscopic, transmission electron microscopic, and confocal laser scanning microscopic observation of fibroblasts cultured on microgrooved surfaces of bulk titanium substrata. J Biomed Mater Res 1998;40: 425–33. [9] Kasemo B, Gold J. Implant surfaces and interface processes. Adv Dent Res 1999;13:8–20. [10] Ricci JL, Charvet J, Frenkel SR, Chang R, Nadkarni P, Turner J, Alexander H. Bone response to laser microtextured surfaces. In: Davies JE, editor. Bone engineering: based on the proceedings of the bone engineering workshop. Toronto: Em Squared Incorporated; 2000. p. 282–92. [11] Cho SA, Jung SK. A removal torque of the laser-treated titanium implants in rabbit tibia. Biomaterials 2003;24:4859–63. ! Z, Turzo! K, Radnai M, Bor Z, . [12] Bereznai M, Pelsoczi I, Toth Fazekas A. Surface modifications induced by ns and sub-ps excimer laser pulses on titanium implant material. Biomaterials 2003;24:4179–203. [13] Madore C, Landolt D. Electrochemical micromachining of controlled topographies on titanium for biological applications. J Micromech Microeng 1997;7:270–5. [14] Madore C, Piotrowski O, Landolt D. Through-mask electrochemical micromachining of titanium. J Electrochem Soc 1999; 146:2526–32. [15] Chauvy PF, Madore C, Landolt D. Electrochemical micromachining of titanium through a patterned oxide film. Electrochem Solid State 1999;2:123–5. [16] Ferri Y, Piotrowski O, Chauvy PF, Madore C, Landolt D. Twolevel electrochemical micromachining of titanium for device fabrication. J Micromech Microeng 2001;11:522–7. [17] de Groot K, Wolke JGC, Jansen JA. Calcium phosphate coatings for medical implants. Proc Instn Mech Engrs 1998;212: 137–47. [18] Kokubo T, Miyaji F, Kim HM, Nakamura T. Spontaneous formation of bonelike apatite layer on chemically treated titanium metals. J Am Ceram Soc 1996;79:1127–9. [19] Kim HM, Miyaji F, Kokubo T, Nakamura T. Preparation of bioactive Ti and its alloys via simple chemical surface treatment. J Biomed Mater Res 1996;32:409–17. [20] Kim HM, Kokubo T, Fujibayashi S, Nishiguchi S, Nakamura T. Bioactive macroporous titanium surface layer on titanium substrate. J Biomed Mater Res 2000;52:553–7. [21] Fujibayashi S, Nakamura T, Nishiguchi S, Tamura J, Uchida M, Kim HM, Kokubo T. Bioactive titanium: effect of sodium removal on the bone-bonding ability of bioactive titanium prepared by alkali and heat treatment. J Biomed Mater Res 2001;56:562–70. [22] Nishiguchi S, Nakamura T, Kobayashi M, Kim HM, Miyaji F, Kokubo T. The effect of heat treatment on bone-bonding ability of alkali-treated titanium. Biomaterials 1999;20:491–500. [23] Nishiguchi S, Kato H, Fujita H, Kim HM, Miyaji F, Kokubo T, Nakamura T. Enhancement of bone-bonding strengths of titanium alloy implants by alkali and heat treatments. J Biomed Mater Res Appl Biomater 1999;48:689–96.
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801 [24] Nishiguchi S, Kato H, Neo M, Oka M, Kim HM, Kokubo T, Nakamura T. Alkali- and heat-treated porous titanium for orthopedic implants. J Biomed Mater Res 2001;54:198–208. [25] Nishiguchi S, Kato H, Fujita H, Oka M, Kim HM, Kokubo T, Nakamura T. Alkali- and heat-treated porous titanium for orthopedic implants. Biomaterials 2001;22:2525–33. [26] Nishiguchi S, Fujibayashi S, Kim HM, Kokubo T, Nakamura T. Biology of alkali- and heat-treated titanium implants. J Biomed Mater Res 2003;67A:26–35. [27] Frenkel SR, Simon J, Alexander H, Dennis M, Ricci JL. Osseointegration on metallic implant surfaces: effects of microgeometry and growth factor treatment. J Biomed Mater Res Appl Biomater 2002;63:706–13. [28] Qian XJ. The normal and prediction value of human circulation and respiration system Anhui: Anhui Scientific Press, 1989. p. 67. (Chinese) [29] Hench LL. Cementless fixation. In: Pizzoferrato A, editor. Biomaterials and clinical applications: proceeding of the sixth European conference on biomaterials. Amsterdam: Elsevier Science Publisher; 1987. p. 23. [30] Kinenapfel H, Sprey C, Wilke A, Griss P. Implant fixation by bone ingrowth. J Arthroplasty 1999;14:355–68. [31] Windler M, Klabunde R. Titanium for hip and knee prosthesis. In: Brunette DM, Tengvall P, Textor M, Thomsen P, editors.
[32]
[33]
[34]
[35]
[36]
[37] [38]
1801
Titanium in medicine. Berlin; Hong Kong: Springer; 2001. p. 704–6. Siriphannon P, Kameshima Y, Yasumori A, Okada K, Hayashi S. Comparative study of the formation of hydroxyapatite in simulated body fluid under static and flowing system. J Biomed Mater Res 2002;60:175–85. Duan Y, Yao Z, Wang C, Chen JY, Zhang XD. A study of bone-like apatite formation on porous calcium phosphate ceramics in dynamic SBF. J Biomed Eng (Chinese) 2002;19: 365–9. Duan Y, Wang C, Chen JY, Zhang XD. Bone-like apatite formation on calcium phosphate dense ceramics in dynamic SBF. J Inorg Mater (Chinese) 2002; 17: 552–558. Izquierdo-Barba I, Salinas AJ, Vallet-Reg!ı M. Effect of the continuous solution exchange on the in vitro reactivity of a CaO– SiO2 sol–gel glass. J Biomed Mater Res 2000;51:191–9. Vallet-Reg!ı M, Pe´rez-Pariente J, Izquierdo-Barba I, Salinas AJ. Compositional variations in the calcium phosphate layer growth on gel glasses soaked in a simulated body fluid. Chem Mater 2000;12:3770–5. R!amila A, Vallet-Reg!ı M. Static and dynamic in vitro study of a sol–gel glass bioactivity. Biomaterials 2001;22:2301–6. Kim HM, Miyazaki T, Kukubo T, Nakamura T. Revised simulated body fluid. Key Eng Mater 2001;192–195:47–50.
Biomaterials 26 (2005) 1793–1801
Comparative study of osteoconduction on micromachined and alkali-treated titanium alloy surfaces in vitro and in vivo Xiong Lua, Yang Lenga,*, Xingdong Zhangb, Jinrui Xub, Ling Qinc, Chun-wai Chanc a
Department of Mechanical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China b Engineering Research Center in Biomaterials, Sichuan University, Chengdu, China c Department of Orthopaedics and Traumatology, The Chinese University of Hong Kong, Hong Kong, China Received 2 March 2004; accepted 3 June 2004 Available online 17 July 2004
Abstract This study sought to evaluate osteoconduction of Ti-6Al-4 V surfaces under various conditions, including micro-patterned, alkalitreated, micro-patterned plus alkali-treated, and surfaces without any treatment as the control. The through-mask electrochemical micromachining (EMM) was used to fabricate micro-hole arrays on the titanium alloy surface. In vitro calcium phosphate formation on titanium surfaces was in static and dynamic simulated body fluid (SBF). In vivo comparison was conducted in the medullary cavity of dog femur using the implant cages which could provide the same physiological environment for specimens with different surface conditions. In vitro experiments indicate good conduction of calcium phosphate on the alkali-treated surfaces, and also better calcium phosphate deposition on the micro-hole surface than on the flat surfaces in dynamic SBF. In vivo experiments confirm the beneficial effect of alkaline treatment on osteoconduction. The results of in vivo experiments also indicate a synergistic effect of the alkaline treatment and the topographic pattern on osteoconduction. r 2004 Elsevier Ltd. All rights reserved. Keywords: Osteoconduction; Micro-pattern; Alkaline treatment; Simulate body fluid; Animal model
1. Introduction Titanium and titanium alloys are among the most popular materials for orthopedic implants due to their biocompatibility, excellent corrosion resistance, good mechanical properties and lightness [1]. There have been efforts to improve the osseointegration capability of the titanium implants by enhancing osteoconduction on their surfaces using surface morphology and chemistry modifications. Various methods have been used to develop morphological structures of titanium implant surfaces for promoting bone ingrowth and fixation between implants and bone. Among them, powder/fiber/wire mesh metallurgical sintering [2,3], plasma spray processing [4,5] and surface blasting [6] are the widely used methods to modify surface topography of load-bearing titanium *Corresponding author. Tel.: +86-852-2358-7185; fax: +86-8522358-1543. E-mail address: [email protected] (Y. Leng). 0142-9612/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2004.06.009
implant surfaces. These methods may have adverse effects on mechanical properties because of possible residue stress from high temperature process and stress concentrations at the interface between the porous layer and the substrates. Also, these methods generate irregular and random surface geometry which precludes further quantitative analysis of the osseointegration between bone and the implant. Techniques for producing regular surface patterns on titanium implants have been developed, such as reactiveion etching (RIE) [7,8], laser machining [9–12] and the through-mask electrochemical micromachining (EMM) [13–16]. The first two techniques potentially introduce chemical heterogeneity, which might affect osteoblasts behavior. EMM can produce well-defined surface patterns on titanium surfaces without introducing surface chemistry heterogeneity. For surface chemistry modification, various calcium phosphate (Ca–P) coatings on titanium implants have been proposed to improve osseointegration as reviewed in [17]. The adhesive strength between the Ca–P coating
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
Titanium specimens used in this work were made from Ti-6Al-4 V plates with 1 mm in thickness (Baoji Special Iron and Steel Co. LTD., China). A throughmask electrochemical micromachining (EMM) technique was employed to fabricate micropatterns on the Ti specimens as follows. A surface pattern with micro-hole arrays was produced on a photo-mask and transferred to the Ti specimens (20 mm 20 mm) by contact printing photolithography using a positive photoresist. Then, the Ti specimens were attached on a rotating shaft and served as the anode for electrochemical etching. The anode rotated at a speed of 500 rpm in 5 mol/l sodium bromide aqueous solution during electrochemical etching. The dimensions of the micro-holes were measured with a surface profiler (WYKO NT3000, Veeco Instruments Inc., USA). The surface roughness of 0.3870.03 mm was also measured by the surface profiler. To eliminate possible chemical heterogeneity between the etched and non-etched areas, the specimens were placed in diluted sulfate acid (1:5 in volume ratio) for 3 h. 2.2. Alkaline treatment Alkaline treatment was performed by immersing the Ti specimens in 100 ml of 5 mol/l NaOH aqueous solution at 60 C for 24 h. After the alkaline treatment, the specimens were gently washed with distilled water and dried at 40 C for 24 h in an air atmosphere. The
SBF used in this study was prepared following a revised recipe by Kim et al. [38] as follows. Chemicals were dissolved in one liter of distilled water in the sequence of NaCl (5.403 g), NaHCO3 (0.736 g), Na2CO3 (2.036), KCl (0.225 g), K2HPO4 (0.182 g), MgCl2 . 6H2O (0.310 g), HEPES (2-(4-(2-hydroxyethyl)-1-piperazinyl) ethanesulfonic acid) (11.928 g), CaCl2 (0.293 g), Na2SO4 (0.072 g), 1 mol/l NaOH. The fluid was buffered at pH 7.40, 36.5 C with HEPES and 1 mol/l NaOH. The ability of forming Ca–P was evaluated in both static and dynamic SBF environments. The dynamic SBF was realized by pumping the SBF in/out of a 100 ml sample chamber through a two channel peristaltic pump. The flow rate was 3 ml/min, i.e., 100 ml SBF stored in the sample chamber was refreshed at a rate of 3 ml/min by an in/out flow. This flow rate is close to that of body fluid in the human muscle environment [28]. The SBF in the storage tank was replaced with fresh one every three days in order to keep the solution ion concentration stable. The Ca–P deposited on the specimen was examined using an SEM (JEOL 6300, JEOL, Japan) with Energy Dispersive X-ray Spectrometer (EDS). 2.4. In vivo experiments The Ti specimens for the in vivo experiments were cut into 6 mm 8 mm rectangular plates. Seven plates were assembled into a small ultra high molecular weight polyethylene (UHMWPE) cage developed by Ricci et al. [27]. As shown in Fig. 1, the cage with multiple channels enables us to compare the specimens with various surface conditions directly: (a) control without surface treatment, (b) surface patterned, (c) surface patterned
30
1 2
3 4
5 6
7
Unit: mm
8.5
2.1. Electrochemical micromachining
2.3. In vitro experiments
1 1.5
2. Materials and methods
alkali-treated specimens were then placed on an alumina boat heated to 600 C at a rate of 5 C/min for 1 h and cooled to room temperature in the furnace.
4
and titanium substrate generate great concern regarding the mechanical integrity of the implants in clinical applications. Kokubo et al. [18–26] introduced an alkaline treatment on titanium surfaces, which might provide strong bone-bonding ability and high bone affinity. Note that the alkaline treatment can be applied to titanium surfaces with any type of morphological features. This study sought to compare the osteoconduction of titanium surfaces that were engineered by surface patterning and alkaline treatment. The osteoconduction of the micropatterned, alkali-treated, micropatterned plus alkali-treated, and control surfaces were evaluated by direct in vitro and in vivo comparison. In vitro ability of bioactive Ca–P formation was examined in static and dynamic simulated body fluid (SBF). The in vivo comparison of different surface conditions was conducted in an implant cages in dog femur, a technique developed by Ricci et al. [27]. The results of this study will guide to future developments of bioactive titanium implants.
1 1
1794
1.5 1
3
1
3
1
3
1 1.5
2
6
Fig. 1. Configuration and arrangement of implant cages for animal model. The dark area marks the cross section of the UMWPE cage and the blank area between the Ti plates is the channels of bone ingrowth. Specimens in the cage are marked by number: (1), Control Ti plate without any treatment; (2) and (3), micro-patterned Ti plates; (4) and (5), micro-patterned plus alkali-treated Ti plates; (6) and (7), alkalitreated Ti plates.
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
and AH-treated, and (d) surface AH-treated. The cage location in dog femur and the short cage compared with the femur length ensured that rational comparisons among the specimens in nearly identical physiological environments could be made (Fig. 2). The opening space (1 6 8 mm3) between the specimen in the cage
Fig. 2. The X-ray radiograph shows the location of cage implanted in dog femur.
1795
ensures the bone ingrowth. The dog was anesthetized with an intramuscular injection of 10% ketamine and 1% xylazine (v : v ¼ 0:25 ml: 0.25 ml per kg body weight) before cage implantation. A sterilized cage was implanted into a rectangle femur cavity of 7-mm deep, 8-mm wide, 18-mm long, and fixed on cortical bone through titanium screws (Fig. 2). The specimens in the cages were placed into the medullary cavity immersed in blood. Two cages were implanted the right and left femurs of same dog for one and two months, respectively. The dog was sacrificed for retrieving the cages. The specimens implanted for one month were examined under SEM while those implanted for two months were fixed for histological observations. The specimens were fixed in 10% formalin in 0.075 mol/l phosphate buffer pH 7.25 for 3 days at room temperature, washed in a buffer solution for 10 min, and dehydrated in serial concentrations of ethanol (70%, 80%, 90%, 99%, 100% and 100% v/v) every 2 days. Then, they were embedded in polymerized methyl methacrylate (PMMA) by immersing the specimens into xylene/100%EtOH, xylene, unpolymerized methyl methacrylate(UMMA)/ xylene (ratio 1:1, v/v), UMMA I, UMMA II, UMMA III, polymerized methyl methacrylate (PMMA) in
Fig. 3. (a) SEM micrograph of micro-hole array obtained from EMM; (b) Single micro-hole; (c) Micro-hole dimensions from surface profiler measurement.
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
1796
Fig. 4. SEM micrographs and EDS spectrum of Ca–P on the Ti-6Al-4V surfaces after 15 days in static SBF. (a) Control specimen; (b) Micropatterned specimen; (c) Alkali-treated specimen; (d) EDS spectrum of c; (e) Surface of micro-patterned plus alkali-treated specimen; (f) Bottom of micro-hole of micro-patterned plus alkali-treated specimen.
sequence with a two-day interval. Thin sections of 250 mm were cut with a saw microtome (SP1600, Leica, Germany). Then, the sections were stuck to the PMMA slides (1 mm thick) and further grinded and polished to 100 mm thick with a grinding and polishing system (RotoPol-21, Struers A/S, Denmark). The polished sections were stained with toluidine-blue. Histomorphology was performed using a light microscope (DM RXA2, Leica, Germany). The bone growth along the surfaces with and without micro-pattern was quantitative analyzed using image analysis software (Qwin Standard, Leica, Germany). The efficiency of osteoconduction was quantified by the bone attachment index (BAI) defined as: BAI ¼
bone area 100%; total tissue area
in which the total tissue area is a rectangle area adjacent to bone/Ti interface in the opening spaces of the cage shown in Fig. 1. The area is 500 mm wide from the interface. The bone area is that occupied by newly formed bone in the total tissue area. Both of the total tissue area and the bone area are determined numerically by the image analysis software.
3. Results and discussion 3.1. Micro-hole characterization Micro-hole arrays on Ti specimen surfaces were successfully produced by the EMM method as shown in the SEM micrographs (Figs. 3a and b). The dimensions of holes measured by the 3D surface profiler
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
Fig. 5. SEM micrographs of Ca–P growth on an alkali-treated microhole by immersing in static SBF. (a) A view of single hole; (b) A view of Ca–P growth inside and outside the hole by tilting view angle of 45 .
indicated a spherical shape with a radius of B300 mm as shown in Fig 3c. The diameter of the holes was sufficiently large to facilitate the ingrowth of blood vessels and the supply of osteocytes [29–31]. These dimensions were similar to the pore size of porouscoated prostheses currently undergoing clinical trials [30]. Apparently, the EMM technique can provide a topographic structure on the titanium surfaces without mechanical property degradation and surface chemical heterogeneity.
1797
not be detected on the micro-patterned surfaces without the alkaline treatment either (Fig. 4b). On the alkalitreated surfaces, a thick layer of Ca–P was formed (Fig. 4c) and confirmed by EDS as shown in Fig. 4d. The pattern of micro-hole arrays seems not to affect the Ca–P formation in static SBF. Ca–P was formed on both the inside and outside surfaces of the micro-holes (Fig. 4e and f). No difference in Ca–P formation capability on the micro-hole surfaces and on the flat surfaces was observed. Growth of globule-like Ca–P crystals could well adapt orientation change of specimen surfaces caused by the geometrical configuration of micro-holes. The Ca–P globules always grew along the direction of surface normal even on the inner wall of the micro-hole (Fig. 5). Apparently, in static SBF the topographic condition of surface has little influence on Ca–P formation, while the chemical treatment of the surface makes a significant difference. The results of experiments in dynamic SBF indicate an effect of surface topography on Ca–P formation in addition to the effect of alkaline treatment. Fig. 6 presents the SEM micrographs of the specimen surfaces after being immersed in dynamic SBF for 15 days. Generally, Ca–P formation on the specimen surfaces in dynamic SBF was less than that in static SBF, and this observation is consistent with prior reports [32–34]. Again, no Ca–P was formed on the specimen surface without any treatment (Fig. 6a). Ca–P was not formed on the micro-patterned surface either (Fig. 6b). On the alkali-treated surfaces (Fig. 6c), Ca–P does not form a thick layer as in static SBF. This difference is also shown in an EDS spectrum (Fig. 6d). In dynamic SBF, Ca–P formation is more pronounced on the bottom of microholes (Fig. 6f) than on flat surfaces (Fig. 6e). Our finding is similar to the results of testing porous Ca–P ceramics in dynamic SBF [33], in which the Ca–P is formed in the inner pores instead of outer surface. Researchers believed that the dynamic SBF tests provide an environment more similar to that in living body than that of static SBF test. Thus, the Ca–P formation in dynamic SBF should be more relevant to Ca–P deposition in vivo [32–37]. Results of our in vivo experiments perhaps clarify any speculation from in vitro experiments. 3.3. In vivo comparisons
3.2. In vitro comparisons In vitro experimental results indicate the differences in the ability to form Ca–P among the specimen surfaces with different treatments. The results of the in vitro experiments in static SBF show the merit of the alkalitreated surfaces. Fig. 4 shows the comparison of the specimen surfaces with different treatment after being immersed in static SBF for 15 days. No deposited Ca–P was found on the control surface (Fig. 4a). Ca–P could
Bone ingrowth was observed in the openings between specimens with the alkali-treated surfaces. X-ray radiographs of the cages retrieved from dog indicate that some openings of the cage implanted for two months were filled with dense substance, but not in the openings of the cage implanted for one month (Fig. 7). The specimens implanted for two months were prepared for histological examination, while the specimens implanted for one-month were examined in SEM.
ARTICLE IN PRESS 1798
X. Lu et al. / Biomaterials 26 (2005) 1793–1801
Fig. 6. SEM micrographs and EDS spectrum of Ca–P on the Ti-6Al-4V surfaces after 15 days in dynamic SBF. (a) Control specimen; (b) Micropatterned specimen; (c) Alkali-treated specimen; (d) EDS spectrum of c; (e) Flat area of the micro-patterned plus alkali-treated specimen; (f) Microhole bottom area of the micro-patterned plus alkali-treated specimen.
Fig. 7. X-ray Radiograph of retrieved cages after one month (Left) and two months (right) of implantation.
The SEM examinations revealed the formation of calcium phosphate (Ca–P) on the specimen surfaces having alkaline treatment (Fig. 8b), while nothing was found on the control specimen (Fig. 8a). Rod-like depositions indicate Ca–P formation on the alkalitreated surfaces but no formation of new bone. A thin layer of new bone seems to have formed on certain flat surface areas (Fig. 8c) with alkaline treatment and a thick layer of new bone covers the hole surfaces of the same specimen (Fig. 8d). Apparently, there was better osteoconduction on the surfaces with alkaline and topographic treatments. Fig. 9 shows the histological micrographs of specimens after two months of implantation. Such a side-byside comparison of specimens with various surface conditions shows effects of alkaline and topographic
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
1799
Fig. 8. SEM micrographs Ti-6Al-4V surfaces after one-month implantation. (a) Control specimen; (b) Ca–P deposited on alkali-treated surface; (c) Flat area of the micro-patterned plus alkali-treated specimen; (d) Micro-hole bottom area of the micro-patterned plus alkali-treated specimen.
treatments on osteoconduction. No new bone was found on the cage openings where the specimen surfaces do not have alkaline treatment, no matter whether they had surface micro-pattern. There was a large amount of new bone in the cage openings where the specimen surfaces had alkaline treatment and micro-holes. Also, there was new bone in the opening where the specimen surfaces had alkaline treatment but no micro-holes. Image analysis indicates that the average BAI value of the specimen with the hole-pattern and alkali-treatment is around 71% while that of specimen with only alkalitreatment is around 31%. Although more accurate assessment by statistical analysis of a large number of samples is no available, this semi-quantitative estimation already indicates the merit of hole-pattern in guiding bone growth. Table 1 summarizes the in vivo results which indicate that alkaline treatment plays a primary role in enhancing osteoconduction. Although topographic patterning alone did not enhance new bone growth in our histological evaluation, our study shows a synergistic effect of combining topographic and alkaline treatment on osseointegration.
Fig. 9. Histological micrographs of specimens implanted for twomonths after toluidine-blue staining (original magnification: 20). (a) Section near the cage center; (b) Section near the side surface of the cage. (1), Ti plate without any treatment; (2) and (3), micro-patterned Ti plates; (4) and (5), micro-patterned plus alkali-treated Ti plates; (6), alkali-treated Ti plates.
1
2
3
4
5
6
(a)
1
(b)
2
3 4
5
6
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801
1800 Table 1 In vivo evaluation of osteoconduction
Control Pattern AH treatment Pattern/AH treatment
One month
Two month
CaP
New Bone
New Bone
x x O O
x x O? O
x x O OO
x-negative, O-positive, O?-uncertain, and OO-very positive.
4. Conclusions The present study confirms the beneficial effect of alkaline treatment on osteoconduction. The study also reveals a synergistic effect of combining topographic and alkaline treatment from both in vitro and in vivo experiments. The experiments in dynamic SBF indicate good conduction of calcium phosphate on the alkali-treated surfaces, and also better calcium phosphate deposition on the hole-patterned surface than on the flat surfaces. A side-by-side comparison of surfaces with various surface conditions in animal model shows the importance of alkaline treatment in bone growth. The bone growth on the alkali-treated surfaces can further be enhanced by topographic patterning.
Acknowledgements This project was financially supported by the Research Grants Council of Hong Kong (HKUST 6037/ 02E) and the Funds for High Impact Areas at The Hong Kong University of Science & Technology. The microfabrication was done in the Microelectronics Fabrication Facility of Hong Kong University of Science and Technology. The authors wish to acknowledge valuable suggestions about electrochemical etching from Prof Changjian Lin and Dr. Haobing Hu of the State Key Laboratory for Physical Chemistry of Solid Surfaces, Xiamen University, China.
References [1] Park JB, Lakes RS. Biomaterials: an introduction, 2nd ed. New York: Plenum Press; 1992. p. 89–92. [2] Pilliar RM. Porous-surface metallic implants for orthopedic applications. J Biomed Mater Res Appl Biomater 1987;21:1–33. [3] Pilliar RM. P/M processing of surgical implants: sintered porous surfaces for tissue-to-implant fixation. Int J Powder Metall 1998; 34:33–45. [4] Simmons CA, Valiquette NV, Pilliar RM. Osseointegration of sintered porous-surfaced and plasma spray-coated implants: an animal model study of early post implantation healing response and mechanical stability. J Biomed Mater Res 1999;47:127–38.
[5] Yang YZ, Tian JM, Tian JT, Chen ZQ, Deng XJ, Zhang DH. Preparation of graded porous titanium coatings on titanium implant materials by plasma spraying. J Biomed Mater Res 2000;52:333–7. [6] Jinno T, Goldberg VM, Davy D, Stevenson S. Osseointegration of surface-blasted implants made of titanium alloy and cobaltchromium alloy in a rabbit intramedullary model. J Biomed Mater Res 1998;42:20–9. [7] Jaeger NAG, Brunette DM. Production of microfabricated surfaces and their effects on cell behavior. In: Brunette DM, Tengvall P, Textor M, Thomsen P, editors. Titanium in medicine. Berlin, Hong Kong: Springer; 2001. p. 343–74. [8] den Braber ET, Jansen HV. Scanning electron microscopic, transmission electron microscopic, and confocal laser scanning microscopic observation of fibroblasts cultured on microgrooved surfaces of bulk titanium substrata. J Biomed Mater Res 1998;40: 425–33. [9] Kasemo B, Gold J. Implant surfaces and interface processes. Adv Dent Res 1999;13:8–20. [10] Ricci JL, Charvet J, Frenkel SR, Chang R, Nadkarni P, Turner J, Alexander H. Bone response to laser microtextured surfaces. In: Davies JE, editor. Bone engineering: based on the proceedings of the bone engineering workshop. Toronto: Em Squared Incorporated; 2000. p. 282–92. [11] Cho SA, Jung SK. A removal torque of the laser-treated titanium implants in rabbit tibia. Biomaterials 2003;24:4859–63. ! Z, Turzo! K, Radnai M, Bor Z, . [12] Bereznai M, Pelsoczi I, Toth Fazekas A. Surface modifications induced by ns and sub-ps excimer laser pulses on titanium implant material. Biomaterials 2003;24:4179–203. [13] Madore C, Landolt D. Electrochemical micromachining of controlled topographies on titanium for biological applications. J Micromech Microeng 1997;7:270–5. [14] Madore C, Piotrowski O, Landolt D. Through-mask electrochemical micromachining of titanium. J Electrochem Soc 1999; 146:2526–32. [15] Chauvy PF, Madore C, Landolt D. Electrochemical micromachining of titanium through a patterned oxide film. Electrochem Solid State 1999;2:123–5. [16] Ferri Y, Piotrowski O, Chauvy PF, Madore C, Landolt D. Twolevel electrochemical micromachining of titanium for device fabrication. J Micromech Microeng 2001;11:522–7. [17] de Groot K, Wolke JGC, Jansen JA. Calcium phosphate coatings for medical implants. Proc Instn Mech Engrs 1998;212: 137–47. [18] Kokubo T, Miyaji F, Kim HM, Nakamura T. Spontaneous formation of bonelike apatite layer on chemically treated titanium metals. J Am Ceram Soc 1996;79:1127–9. [19] Kim HM, Miyaji F, Kokubo T, Nakamura T. Preparation of bioactive Ti and its alloys via simple chemical surface treatment. J Biomed Mater Res 1996;32:409–17. [20] Kim HM, Kokubo T, Fujibayashi S, Nishiguchi S, Nakamura T. Bioactive macroporous titanium surface layer on titanium substrate. J Biomed Mater Res 2000;52:553–7. [21] Fujibayashi S, Nakamura T, Nishiguchi S, Tamura J, Uchida M, Kim HM, Kokubo T. Bioactive titanium: effect of sodium removal on the bone-bonding ability of bioactive titanium prepared by alkali and heat treatment. J Biomed Mater Res 2001;56:562–70. [22] Nishiguchi S, Nakamura T, Kobayashi M, Kim HM, Miyaji F, Kokubo T. The effect of heat treatment on bone-bonding ability of alkali-treated titanium. Biomaterials 1999;20:491–500. [23] Nishiguchi S, Kato H, Fujita H, Kim HM, Miyaji F, Kokubo T, Nakamura T. Enhancement of bone-bonding strengths of titanium alloy implants by alkali and heat treatments. J Biomed Mater Res Appl Biomater 1999;48:689–96.
ARTICLE IN PRESS X. Lu et al. / Biomaterials 26 (2005) 1793–1801 [24] Nishiguchi S, Kato H, Neo M, Oka M, Kim HM, Kokubo T, Nakamura T. Alkali- and heat-treated porous titanium for orthopedic implants. J Biomed Mater Res 2001;54:198–208. [25] Nishiguchi S, Kato H, Fujita H, Oka M, Kim HM, Kokubo T, Nakamura T. Alkali- and heat-treated porous titanium for orthopedic implants. Biomaterials 2001;22:2525–33. [26] Nishiguchi S, Fujibayashi S, Kim HM, Kokubo T, Nakamura T. Biology of alkali- and heat-treated titanium implants. J Biomed Mater Res 2003;67A:26–35. [27] Frenkel SR, Simon J, Alexander H, Dennis M, Ricci JL. Osseointegration on metallic implant surfaces: effects of microgeometry and growth factor treatment. J Biomed Mater Res Appl Biomater 2002;63:706–13. [28] Qian XJ. The normal and prediction value of human circulation and respiration system Anhui: Anhui Scientific Press, 1989. p. 67. (Chinese) [29] Hench LL. Cementless fixation. In: Pizzoferrato A, editor. Biomaterials and clinical applications: proceeding of the sixth European conference on biomaterials. Amsterdam: Elsevier Science Publisher; 1987. p. 23. [30] Kinenapfel H, Sprey C, Wilke A, Griss P. Implant fixation by bone ingrowth. J Arthroplasty 1999;14:355–68. [31] Windler M, Klabunde R. Titanium for hip and knee prosthesis. In: Brunette DM, Tengvall P, Textor M, Thomsen P, editors.
[32]
[33]
[34]
[35]
[36]
[37] [38]
1801
Titanium in medicine. Berlin; Hong Kong: Springer; 2001. p. 704–6. Siriphannon P, Kameshima Y, Yasumori A, Okada K, Hayashi S. Comparative study of the formation of hydroxyapatite in simulated body fluid under static and flowing system. J Biomed Mater Res 2002;60:175–85. Duan Y, Yao Z, Wang C, Chen JY, Zhang XD. A study of bone-like apatite formation on porous calcium phosphate ceramics in dynamic SBF. J Biomed Eng (Chinese) 2002;19: 365–9. Duan Y, Wang C, Chen JY, Zhang XD. Bone-like apatite formation on calcium phosphate dense ceramics in dynamic SBF. J Inorg Mater (Chinese) 2002; 17: 552–558. Izquierdo-Barba I, Salinas AJ, Vallet-Reg!ı M. Effect of the continuous solution exchange on the in vitro reactivity of a CaO– SiO2 sol–gel glass. J Biomed Mater Res 2000;51:191–9. Vallet-Reg!ı M, Pe´rez-Pariente J, Izquierdo-Barba I, Salinas AJ. Compositional variations in the calcium phosphate layer growth on gel glasses soaked in a simulated body fluid. Chem Mater 2000;12:3770–5. R!amila A, Vallet-Reg!ı M. Static and dynamic in vitro study of a sol–gel glass bioactivity. Biomaterials 2001;22:2301–6. Kim HM, Miyazaki T, Kukubo T, Nakamura T. Revised simulated body fluid. Key Eng Mater 2001;192–195:47–50.