Calcium Phosphate-Based Bone Substitutes

Review Article European Journal of Trauma Calcium Phosphate-Based Bone Substitutes Reinhard Schnettler1, Jens Peter Stahl1, Volker Alt1, Theodoros Pavlidis1, Elvira Dingeldein1, Sabine Wenisch2 Abstract Background: The replacement of bone by means of foreign materials was already carried out in prehistoric times. Nowadays autogenous bone grafting is designated as the “golden standard” to fill large osseous defects which result from traumas, tumors, or birth defects. However, its disadvantages such as limited supply of autogenous bone and donor site morbidity have favored the use of bone substitutes. As these materials are characterized by their unlimited availability without bearing the risk of disease transmission, research on improving bone tissue healing by using bone substitutes of synthetic or biological origin is a field of major interest. Focus of Interest: Bone substitutes used clinically in orthopedics, periodontics, oral and maxillofacial surgery as well as in plastic, trauma, and reconstructive surgery comprise a wide variety of materials and have been the focus of interest for the last 80 years. The present review has focused on the frequently used calcium phosphate-based bone substitutes revealing either resorbable or nonresorbable properties. Their excellent biocompatibility due to their close mimicking of the inorganic phase of the natural bone mineral has led to their widespread use in bone reconstructive surgery. Examination Tools: Physicochemical properties of the materials have been shown by X-ray diffraction and scanning electron microscopy, whereas bioreactivity has been investigated by means of comparative histological evaluations and the use of various animal models. Transmission electron microscopy has been suitable for studying cell-mediated degradation at the cellular level. The results are discussed with special regard to the origin, composition, and general characteristics of inorganic bone substitutes. Key Words Bone substitutes • Calcium phosphate • Bone ingrowth • Degradation • Growth factors Eur J Trauma 2004;30:219–29 DOI 10.1007/s00068-004-1393-x Bone Graft Substitutes: Alternatives to Bone Grafting Bone Substitutes are Taken into Account Operative treatment of large bone defects has remained a challenge for orthopedic surgeons. In about 10% of all reconstructive operations caused by traumatic, resectional, or congenital defects, bone transplants and bone substitute materials are necessary. Allogenic bone material carries the potential risk of transmitting tumor cells and a variety of bacterial and viral infections including those that cause AIDS or hepatitis in patients. Additionally, blood group-incompatible bone transplantation can cause the development of antibodies within the AB0 system. This bears the risk of the occurrence of Morbus haemolyticus neonatorum in the case of later pregnancy. Generally, a more or less distinct histoincompatibility, due to the transfer of immunocompetent lymphatic donor cells, exists in allogenic bone transplantation as well as in the transplantation of solid organs [1]. Therefore, and because of its superior osteogenic potential compared to allogenic transplants, autogenous bone grafting continues to be the “golden standard” in reconstructive surgery. However, as the limited availability of autogenous bone is a major problem for the surgeon and his patient, inorganic implants such as calcium phosphate-(CaP-)based bone substi- 1 Department of Trauma Surgery, Justus Liebig University, Gießen, Germany, 2 Experimental Trauma Surgery, Justus Liebig University, Gießen, Germany. Received: December 15, 2003; accepted: June 23, 2004. European Journal of Trauma 2004 · No. 4 © Urban & Vogel 219 Schnettler R, et al. Bone Substitutes tutes of synthetic or biological origin are useful alternatives to autogenous bone grafting [2, 3]. The biocompatibility of HA and the similarities between the crystal structure of HA and bone mineral have led to its widespread use in bone reconstructive Calcium Phosphate-Based Bone Substitutes: an surgery [7–10]. The use of low-density HA with highly Overview interconnected porosity is different to that of dense Bone substitutes should have a good local and systemic HA [11]. The porous structure and its biocompatibilcompatibility, the capability of being substituted by bone ity enable the ingrowth of bone into the implant which and of completely filling any defect. These features repromotes mechanically stable and biologically intequire osteoconductive and/or osteoinductive properties grated repair [7]. Several bone-bonding mechanisms of the implant comparable to those of the natural bone. in different bioactive materials have been discussed Therefore research in improving bone tissue healing by [12–14]. biological and mechanical approaches is a field of major By means of scanning electron microscopy the interinterest. connecting pore system of a bone-derived HA ceramic The knowledge of the mechanisms concerning bone can be visualized (Figure 1a). A close-up of the trabecuformation and bone repair and the knowledge of the lae demonstrates the size of the crystals and their tight manner by which bone interacts with bone substitutes connection (Figure 1b). Synthetic ceramics are similar are the basis of the clinical practice in orthopedic surin their crystal configuration, but differ in size und numgery. Nowadays we are in the midst of intensive research ber of pores. which is being focused on the field of cellular and moThe dynamic bone-HA interface is one of the most lecular biology. The majority of data provide informaimportant requirements for the tight junction of cetion about the response of cells to their physiological ramics with living tissues by osteointegration [15]. For environment with special regard to the effects of growth example, HA derived from bovine cancellous bone factors, cytokines, extracellular matrix proteins, and re(Endobon®) offers the dynamic interface between the ceramic and the newly formed bone [16]. It is produced ceptor-cell interactions inducing cellular differentiation by hydrothermal defatting and calcination of bovine and/or transformation. cancellous bone which has an open “foam-like” trabecCurrently available bone substitutes show a variety ular macrostructure [11]. of compositions and properties. Among them inorganic CaP cements consist of particles of different size CaPs are frequently used (cf. Table 1). These materials which are embedded in a cement matrix. It is characinclude CaP ceramics (hydroxyapatite [HA]): Endoteristic of them that they can be handled easily. The bon®, Cerabone®, Ceros 80®, and tricalcium phosphate ® ® [β-TCP]: Ceros 82 ), CaP cements (Biobon [α-BSM], mixing of the CaP powder with the accurate amount of Calcibon®), and nanoparticular HA paste (Ostim®). sterile saline, water or buffer solution results in a paste Depending on the nature of the interfaces between which remains formable for hours at room temperature. them and the host bone, these materials are described It hardens in generally < 20 min at body temperature either as bioinert or bioactive. They undergo processes (37 °C) and then displays limited solubility. Biobon® is of dissolution and precipitation resulting in a strong a fully synthetic microcrystalline hydroxyapatitic bone material-bone interface [4–6]. substitute needing a specific setting reaction which depends upon conversion of precursor CaPs to a poorly crystalline apatite Table 1. Origin and composition of inorganic bone substitutes. HA: hydroxyapatite. with a nanocrystalline structure that Material Origin Composition Porosity Resorption is virtually identical to normal bone [17, 18]. X-ray diffraction is one of Endobon® (Biomet Merck) Bovine bone HA Macroporous Nonresorbable the most important characterization Cerabone® (Coripharm/AAP) Bovine bone HA Macroporous Nonresorbable tools used in solid-state chemistry Ceros 80® (Mathys) Synthetic HA Macroporous Nonresorbable and material science. It is suitable Biobon® (α-BSM, Etex, 26, Synthetic HA In situ curing Resorbable Biomet Merck) paste to determine the composition of difCalcibon® (Biomet Merck) Synthetic HA In situ curing Resorbable ferent minerals and has been used in paste two main areas: the fingerprint charOstim® (Osartis/AAP) Synthetic HA Noncuring paste Resorbable acterization of crystalline materials 220 European Journal of Trauma 2004 · No. 4 © Urban & Vogel Schnettler R, et al. Bone Substitutes Figure 1a. Scanning electron micrograph of an hydroxyapatite ceramic derived from bovine bone. Figure 1b. Higher magnification of hydroxyapatite crystals and their tight connections. Scanning electron microscopy. Figure 1c. X-ray diffraction pattern of Biobon®, natural bone, and hydroxyapatite. Figure 1d. X-ray diffraction pattern of Calcibon®. and the determination of their structure. Each crystalline solid has its unique characteristic X-ray powder pattern which may be used as a “fingerprint” for its identification. Once the material has been identified, X-ray crystallography may be used to determine its structure, crystalline state, interatomic distance and angle. The X-ray diffraction pattern of Biobon® in comparison to natural bone is shown in Figure 1c. Although similar to natural bone HA diffraction peaks are distinctly broader. The X-ray diffraction pattern of Calcibon® in contrast to the pattern of synthetic HA can be seen in Figure 1d. European Journal of Trauma 2004 · No. 4 © Urban & Vogel 221 Schnettler R, et al. Bone Substitutes Comparative in Vivo Studies – Material and Methods In order to evaluate the characteristics such as osteointegration, osteoinduction, biocompatibility, and degradation of CaP-based implants in vivo, their biological reactivity has been investigated in various animal models. Cylindrical bone defects were created by using a diamond-coated drill (diamond bone cutting system [DBCS]) with water irrigation. The sizes of the defects were 9.6 mm in diameter and 10 mm in depth in miniature pigs, 11 mm in diameter and 20 mm in depth in sheep, and 5.6 mm in diameter and 8 mm in depth in rabbits, respectively. The animals were left in vivo for different periods (cf. Table 2). Using the method of Donath & Breuner [19] undecalcified samples were embedded in methyl methacrylate and sawed into 20-µm sections for histological examinations in light microscopy. Subsequently, the sections were stained with toluidine blue. Enzymatic detection of tartrate-resistant acidic phosphatase (TRAP) activity was performed by incubation of the slices in a solution of naphthol AS-BI phosphate (Sigma Chemical) and fast red violet LB salt (Sigma Chemical) in 0.2 M acetate buffer (pH 5.0) containing 50 mM (+) tartaric acid for 20 min at 37 °C. Then, the slices were counterstained with hematoxylin. For ultrastructural examinations small samples were postfixed at 4 °C for 24 h in Yellow Fix (4% paraformaldehyde, 2% glutaraldehyde, 0.04% picric acid). After several washes in 0.1 M phosphate buffer (pH 7.2) the non-decalcified specimens were fixed for 2 h in 1% osmium tetroxide (OsO4), washed carefully and repeatedly in 0.1 M phosphate buffer (pH 7.2), and deTable 2. Inorganic bone substitutes and animal model. Material Species Model Endobon® (Biomet Merck) Miniature pigs Cerabone® (Coripharm/AAP) Rabbits Ceros 80® (Mathys) Miniature pigs Biobon® (α-BSM, Etex, 26, Biomet Merck) Calcibon® (Biomet Merck) Sheep Femur condylus, press fit Femur condylus, press fit Femur condylus, press fit Tibia head Sheep Tibia head Ostim® (Osartis/AAP) Ostim® (Osartis/AAP) Sheep Rabbit Tibia head Femur condylus, press fit 222 hydrated in series in graded ethanol. Subsequently, the specimens were embedded in Epon (Serva, Heidelberg, Germany). Polymerization was performed at 60 °C for 20 h. Thin sections were cut with a diamond knife (45°, Diatome, Switzerland) on an ultracut (Reichert-Jung, Germany). Semithin sections (1 µm) were stained with Richardson (1% methylene blue, 1% borax, 1% azure II). Ultrathin sections (80 nm) were counterstained with uranyl acetate and lead citrate (Reichert Ultrostainer, Leica, Germany) and examined in a Zeiss EM 109 transmission electron microscope. In order to study the nature of new bone formation at defined time intervals, fluorochrome labeling was performed by means of tetracyline, Alizarin complexon, Calcein green, and Calcein blue during the implantation period. Results and Discussion HA Ceramics The histological response to implants of different chemistry and structure was similar in HA ceramics but differed considerably when resorbable implants were used. In sintered HA ceramics (Figures 2a and 2b), active areas of bone deposition as well as resorption and remodeling were present. Bone ingrowth started (Figure 2c) from the 5th week onward and was complete after 10 weeks. At this time the bone around the completely integrated implant underwent remodeling caused by osteoclastic (Figure 2d) and osteoblastic acivities. Abundant evidence of osteoblastic activity could be seen during the early stages of reconstitution, but particularly at the time when appositional activity was present. Bone ingrowth tended to proceed from the bottom of the defect as well as from the walls. There were no signs of fibrous encapsulation, but in some histological secFollow-up tions aggregations of HA crystallites surrounded by macrophages could 42 days, 84 days be observed (Figure 2e). This phe42 days, 84 days nomenon might have been caused by the implantation procedure. 42 days, 84 days However, due to their porosity, osteoconductive HA implants can un42 days dergo a significant degree of degra42 days dation and resorption. Apposition of lamellar bone along the implant 30 days, 60 days surfaces could be seen 4–6 weeks 20 days, 40 days after HA ceramic implantation (Figures 2f and 3a). European Journal of Trauma 2004 · No. 4 © Urban & Vogel Schnettler R, et al. Bone Substitutes Figure 2a. Bovine bone-derived hydroxyapatite. Bone formation within the interface region. Microradiography. Figure 2b. Bovine bone-derived hydroxyapatite with areas of bone deposition. Microradiography. Figure 2c. Hydroxyapatite ceramic 42 days after implantation in miniature pigs. Bone ingrowth has already started. Figure 2d. Active area of bone resorption mediated by osteoclasts. Histological section, toluidine blue, 125×. Figure 2e. Loose hydroxyapatite crystals localized at the ceramic surface are surrounded by macrophages and multinucleated giant cells. Histological section, toluidine blue, 640×. Figure 2f. Lamellar bone formation along the external ceramic surface 84 days after implantation in the rabbit. Histological section, toluidine blue, 250×. European Journal of Trauma 2004 · No. 4 © Urban & Vogel 223 Schnettler R, et al. Bone Substitutes Figure 4a. Semithin section revealing cellular events after 42 days of implantation of Calcibon® into the tibia head of sheep. Toluidine blue, 115×. Figure 4c. Ultrastructure of an osteoclast localized at the implant (Biobon®) surface. 5,763×. Considering their properties, porous and dense HA implants cause different tissue responses. Ultimately, porosity and interconnectivity of a ceramic are the determining factors for the amount and the type of newly formed bone. All in all, HA ceramics with an interconnective pore system are highly osteoconductive and act as a scaffold for appositional bone formation. When the newly formed bone has grown through all the pores of an osteoconductive material, the implant-bone composite changes its original biomechanical properties. Figure 4b. Multinucleated cells localized at the implant (Biobon®) surface express high levels of tartrate-resistant acidic phosphatase. Semithin section counterstained with hematoxylin. 640×. 224 Calcium Phosphate Cements Using the sheep tibia head defect model, we could demonstrate that resorption of CaP cements is mediated by European Journal of Trauma 2004 · No. 4 © Urban & Vogel Schnettler R, et al. Bone Substitutes osteoclasts. Degradation of the implant spatially and temporally coincides with the formation of new bone. Comparative analysis revealed accelerated degradation and formation in the Biobon® group in contrast to the Calcibon® group. However, as Figure 3b shows, in both groups the close association between the bone-forming cells and the implant surfaces could be observed even after 6 weeks. In the case of Biobon® the semithin section (Figure 3d) shows long slender projections of bone marrow – including numerous blood vessels – forming a circumscribed labyrinthine system within the cement. The outer surface of the bone marrow localized in close vicinity to the cement is covered by a cellular band of cuboid-shaped osteoblasts. Adjacent areas of newly formed bone are also closely associated with the implant. The transmission electron micrograph (Figure 3c) shows two adjacent osteoblasts responsible for the synthesis of the irregularly arranged bundles of collagen fibers. They cover the cement surface and also occupy some of the small spaces within the implant. Bone formation occurred already 6 weeks after implantation of the CaP cement Calcibon® into the tibia head defect of sheep. Note the expanded area of host bone in the lower half of the semithin section (Figure 4a). The bone is in close contact with a neighboring bone marrow projection. The bone marrow areas facing the cement surface are covered with multinucleated osteoclasts as well as with osteoblasts. The former are visible at the upper margin of the bone marrow projection. The osteoclastic ruffled borders appear as lightly stained cellular regions closely attached to the cement surface. Osteoblasts can be seen in the lower margin of the bone marrow projection. They are in close apposition to each other and have roundish or oval profiles. The bone marrow is well vascularized, and in vicinity to the endothelial tubes osteoblastic cells can be observed. The osteoblasts reveal a wide range of shapes and cytoplasmic densities caused by their different maturational stages. Bone healing requires the subsequent substitution of the implanted CaP cement by natural bone. Substitution depends upon numerous events [20], especially upon a resorption rate similar to the rate of bone formation [2]. Although the mechanisms of osteoclast-induced bone resorption have been studied extensively [21–23], controversial data exist concerning degradation of CaP cements in vitro and in vivo [24–33]. Up to now there has been intense debate whether multinucleated giant cells involved in biomaterial degradation are actually European Journal of Trauma 2004 · No. 4 © Urban & Vogel osteoclasts, because previous investigations suggested that multinucleated cells at the implantation site do not possess osteoclast features but only the specificity of macrophage polykaryons [34–36]. Up to now there has been accumulating evidence that osteoclasts are able to resorb CaP ceramics [24, 29, 30, 32], and recent in vitro experiments have revealed additional information regarding the physiological functions of osteoclasts during degradation of CaP surfaces [27, 28, 32]. As these studies ascribe a role to osteoclasts in phagocytosis without abolishing their resorption capacity, the functions of this cell population seem more complex than initially thought. Therefore, we have investigated the in vivo mechanism of CaP cement (Biobon®) degradation by means of transmission electron microscopy [37]. The results have revealed osteoclast-mediated degradation of HA implanted into sheep bone caused by simultaneous resorption and phagocytosis. After 6 weeks of implantation osteoclasts were localized immediately beneath the ceramic surface. Enzymatic staining demonstrated that the multinucleated cells attached to the implant surface expressed high levels of TRAP (Figure 4b). They formed resorption lacunae and exhibited typical ultrastructural features such as the ruffled border, the clear zone, and the dorsal microvilli (Figure 4c). Their resorption capacity also became evident by alterations of the electron density and the shape of the CaP crystals localized within the acidic microenvironment of the ruffled border. Moreover, the osteoclasts were simultaneously capable of phagocytosing the resorbed CaP crystals. The formation of endophagosomes was performed (1) by the uptake of particles into large intracellular vacuoles generated by deep invagination of the membranes of the osteoclastic ruffled border, and (2) by the encircling of particles due to the development of pseudopodia-like plasma protrusions of the ruffled border. The formation of endophagosomes was followed by the in situ fragmentation of the inclusion material, which was subsequently released into the extracellular space, and phagocytosed by macrophages. In accordance with these results a recent in vitro study has documented ultrastructurally osteoclast-induced degradation of CaP ceramics by simultaneous resorption and phagocytosis [32]. Successful incorporation of an implant, its degradation, and its subsequent replacement by autologous tissue are dependent upon the cellular responses during the initial and chronic inflammatory phases of healing [38]. The steps toward successful osteointegration and bone healing are mediated by highly interrelated im- 225 Schnettler R, et al. Bone Substitutes Figure 3a. Fluorochrome labeling revealing bone apposition in close association with the hydroxyapatite surface 4 weeks after implantation into miniature pigs. 1,600×. Figure 3b. Osteoconductive properties of Biobon® after 6 weeks of implantation into the tibia head of sheep. Histological section, toluidine blue, 57×. Figure 3c. Ultrastructure of osteoid-producing osteoblasts localized in close vicinity to the electron-dense Biobon® implant. 6,348×. 226 Figure 3d. Semithin section of the Biobon® implant which is infiltrated with long, slender areas of osteoid and cuboid-shaped osteoblasts. Toluidine blue, 115×. European Journal of Trauma 2004 · No. 4 © Urban & Vogel Schnettler R, et al. Bone Substitutes lular matrix proteins [38]. Thus, the protein adsorption capability of the implant surface has profound effects on the subsequent attachment of cells [38]. Adhesion of osteoclasts to the implant surface by means of the αvβ3-integrin depends upon adsorption of extracellular matrix proteins containing an arginine-glycine-aspartatic acid (RGD) binding site sequence such as virtonectin. In addition, the capability of osteoclasts to resorb CaP implants may also be related to the solubility of the implant. In contrast to the CaP used in the present study – displaying minimal solubility [17, 18] – CaP disks in vitro, showing high solubility, were not resorbed by osteoclasts [40]. This decrease in osteoclastic activity Figure 5a. Histological overview of the tibia head of the sheep: in contrast to the empty defect (left defect) revealing sparse bone formation along the border of the host bone, the defect might be caused by environmental treated with Ostim® (right defect) was completely filled with new bone after an implantation changes leading to high extracelperiod of 60 days. Histological section, toluidine blue, 57×. lular calcium levels, which in turn induce increased intracellular levels in osteoclasts responsible for the decreased podosome assembly and the resorption of osteoclasts [40]. Furthermore, osteoclastic resorption of apatite cements is totally different from sintered apatite possibly because of differences in crystallinity. Although the data documenting osFigure 5b. Defect bridging within the Figure 5c. Bridging of the critical-size defect in the teoclastic resorption capacity of sintibia head can be seen 60 days after rabbit 4 weeks after implantation of Ostim®. Histotered apatite are still contradictory, implantation of Ostim® in contrast to logical section, toluidine blue, 115×. the empty defect. Histological section, it is becoming clear that osteoclastic toluidine blue, 250×. resorption activity of sintered apatite depends upon numerous parameters plant characteristics such as surface composition, sursuch as the method of sintering, sintering temperature, face roughness, surface energy, and surface topography sintering period, porosity, and surface roughness [40]. [38, 39]. Especially in connection with the factors that govern osteoclastic resorption the surface characterisNanoparticular HA Paste tics play an important role by influencing the types of In the case of Ostim® the stoichiometry of the product is cells that attach to the implant [38, 39]. Rough apatatic of importance. According to HA the CaP ratio of Ostim® surfaces appear to be more suitable for osteoclastic atis 1.67. Its surface is expanded (100 m2/g) because of the tachment than smooth surfaces [26]. Macrophages are nanosized crystals, and comparable to the surface area supposed to be involved in “cleaning” the surface of of bone mineral. The biomaterial Ostim® is an injectable dense HA implants from loose implant particles via paste which was well tolerated in sheep, minipigs, and phagocytosis [39] whereas attachment of osteoclasts rabbits. It binds to bone and stimulates bone healing. As to the implant surface is mediated largely by extracelthe paste does not harden in situ, cell migration into the European Journal of Trauma 2004 · No. 4 © Urban & Vogel 227 Schnettler R, et al. Bone Substitutes implantation area coincides with revascularization. Irrespective of the species used for the experimental procedures, fragmentation of the paste could be observed immediately after implantation. This fragmentation into round particles of different size must be regarded as an essential step toward successful reconstitution of tissue integrity. It enables cellular infiltration of the implantation site and enables osteoblasts to perform bone formation in an osteoconductive manner. After implantation of Ostim® into tibia head defects of sheep, the defects were completely filled with new bone even after 60 days (Figure 5a). According to their role as guiding structures for bone formation, the implant particles were covered by newly formed bone (Figure 5b). Due to proceeding bone healing, ramifications of trabecular bone could be seen between the implant particles (Figure 5c). This spatial ingrowth pattern, taken together with the stimulation of osteoblast activity, supports complete reconstitution (e.g., bridging) of critical-size defects of rabbits, sheep, and minipigs within 4 weeks. Conclusion and Outlook Autogenous bone grafting used to fill volumetric defects of bone is associated with the risk of morbidity of the patients, while the use of bone graft substitutes fills bony voids and reduces morbidity. Moreover, in contrast to allograft bone these materials bear no risk of disease transmission. Depending upon the defect location and a determinate capacity of bone formation, the appropriate material must be selected from a number of commercially available bone substitutes. Therefore, the knowledge of the characteristic features of bone substitutes is essential. Comparative animal studies are helpful in this field, but, the validation of bone graft substitutes in the scope of clinical use provides essential data for approaches to the operative treatment of large bone defects. Current strategies for bone tissue regeneration focus on rapid cell growth rates and high cell differentiation for the development of implantable matrices that mimic biological tissues [41]. One important approach to bone healing and bone ingrowth is the use of growth factors. Some of these factors, e.g., bone morphogenic proteins (BMPs) [42, 43], transforming growth factor-β (TGF-β) [44–46], insulin-like growth factor (IGF) [44, 46], and fibroblast growth factor (FGF) [47–53], act as local regulators of cellular activity and seem to have osteoinductive and angiogenetic potential. Ikade et al. [54] reported the regeneration of several tissue types after the use of growth factors and different 228 carriers. In a current study [55] bone formation and bone ingrowth in miniature pigs in response to bFGF-coated HA ceramic cylinders have been investigated. The experimental results prove that composite implants provide angiogenesis, bone formation and bone ingrowth that is comparable to the results of autogenous grafts and emphasize the importance of further research in the field of new carrier and release systems for growth factors. Acknowledgments We would like to thank Anne Hild und Maren Cassens for their excellent technical assistance, and the companies Biomet Merck and Coripharm for financial support. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Schnettler R, Dingeldein D, Tausch W, et al. Untersuchungen zur knöchernen Intergration einer Hydroxylapatit-Keramik und bFGF im Vergleich zu autogenen Spongiosazylindern. Osteosynthese Int 1994;2:118–26. Tay BK, Patel VV, Bradford DS. Calcium sulfate- and calcium phospate-based bone substitutes. Mimicry of the mineral phase of bone. Orthop Clin North Am 1999;30:615–23. Burstein FD. Bone substitutes. Cleft Palate Craniofac J 2000;37:1–4. LeGeros RZ, Daculsi G. In vivo transformation of biphasic calcium phosphate ceramics: ultrastructural and physicochemical characterizations. In: Yamauro, Hench L, Wilson-Hench J, eds. Handbook of bioactive ceramics, vol II. Boca Raton: CRC Press, 1990:17–28. LeGeros RZ, LeGeros JP. Bone substitute materials and their properties. In: Schnettler R, Markgraf E, Hrsg. Knochenersatzmaterialien und Wachstumsfaktoren. Stuttgart–New York: Thieme, 1997:12–8. De Bruijn JD, Klein CPAT, De Groot K, et al. The ultrastructure of the bone-hydroxyapatite interface in vitro. J Biomed Mater Res 1992;26:1365–82. Revell PA, Hing KA, Tanner KE, et al. Osseointegration of porous hydroxyapatite. In: Schnettler R, Markgraf E, Hrsg. Knochenersatzmaterialien und Wachstumsfaktoren. Stuttgart–New York: Thieme, 1997:28–30. Hende LL. Bioceramics and the origin of life. J Biomed Mater Res 1989;23:685–703. Denissen HW, Van Dijk HJA, De Groot K, et al. Mechanical properties of biomaterials. In: Hastings GW, Williams DF, eds. Mechanical properties of biomaterials. Chichester: Wiley & Sons, 1980:55–61. Eggli PS, Muller W, Schenk RK. Porous hydroxyapatite and tricalcium phosphate cylinders with two different pore size ranges implanted in the cancellous bone of rabbits. A comparative histomorphometric and histologic study of bony ingrowth and implant substitution. Clin Orthop 1988;232:127–38. Bauer HJ, Bauer M, Dingeldein E. Zusammensetzung und Aufbau einer porösen Hydroxylapatitkeramik biologischer Herkunft. Osteosynthese Int 1994;2:106–10. Daculsi G, LeGeros RZ, Mitre D. Crystal dissolution of biological and ceramic apatites. Calcif Tissue Int 1989;45:95–103. Nordström EG, Hero H, Jorgensen RB. Mechanical properties of hydroxyapatite/mica composite. Biomed Mater Eng 1994;4:309–15. Nordström DG, Karson KH. Carbonate-doped hydroxyapatite. J Mater Sci Mater Med 1990;1:182–4. Daculsi G, Bouler JM. Bone bioactive ceramic interface: a dynamic process. In: Schnettler R, Markgraf E, Hrsg. Knochenersatzmaterialien und Wachstumsfaktoren. Stuttgart–New York: Thieme, 1997:23–7. European Journal of Trauma 2004 · No. 4 © Urban & Vogel Schnettler R, et al. Bone Substitutes 16. Daculsi G, Dard M. Bone calcium-phosphate ceramic interface. Osteosynthese Int 1994;2:153–6. 17. Knaack D, Goad MEP, Aiolova M, et al. Resorbable calcium phosphate bone substitute. J Biomed Mater Res 1998;43:399–409. 18. Knaack D, Aiolova M, Tofighi A, et al. A fully resorbable calcium phosphate bone substitute. In: Vincenzini P, ed. Materials in clinical applications. Techna Srl, 1999:55–75. 19. Donath K, Breuner G. A method for the study of undecalcified bones and teeth with attached soft tissue. J Oral Pathol 1982;11:318–26. 20. Heymann D, Pradal G, Benahmad M. Cellular mechanisms of calcium phosphate degradation. Histol Histopathol 1999;14:871–7. 21. Mostov K, Werb Z. Journey across the osteoclast. Science 1997;276:219–20. 22. Heymann D, Guicheux J, Gouin F, et al. Cytokines, growth factors and osteoclasts. Cytokine 1998;10:155–68. 23. Teitelbaum SL. Bone resorption by osteoclasts. Science 2000;289:1504–8. 24. Baslé MF, Chappard D, Grizon F, et al. Osteoclastic resorption of Ca-P biomaterials implanted in rabbit bone. Calcif Tissue Int 1993;53:348–56. 25. Davies JE, Shapiro G, Lowenberg BF. Osteoclastic resorption of calcium phosphate ceramic thin film. Cells Mater 1993;3:245–56. 26. Gomi K, Lowenberg B, Shapiro G, Davies JE. Resorption of synthetic hydroxyapatite by osteoclasts in vitro. Biomaterials 1993;14: 91–6. 27. Wang W, Ferguson DPJ, Quinn JMW, et al. Osteoclasts are capable of particle phagocytosis and bone resorption. J Pathol 1997;182:92–8. 28. Wang W, Ferguson DPJ, Quinn JMW, et al. Biomaterial particle phagocytosis by bone-resorbing osteoclasts. J Bone Joint Surg Br 1997;79:849–56. 29. Yamada S, Heymann DM, Bouler JM, et al. Osteoclastic resorption of biphasic calcium phosphate ceramic in vitro. J Biomed Mater Res 1997;37:346–52. 30. Yamada S, Heymann D, Bouler JM, et al. Osteoclastic resorption of calcium phosphate ceramics with different hydroxyapatite/ β-tricalcium phosphate ratios. Biomaterials 1997;18:1037–41. 31. Zheng Q, Du J, Xia Z, et al. Biodegradation of tricalcium phosphate ceramics by osteoclasts. J Tongji Med Univ 1998;18: 257–61. 32. Heymann D, Guicheux J, Rouselle AV. Ultrastructural evidence in vitro of osteoclastic-induced degradation of calcium phosphate ceramic by simultaneous resorption and phagocytosis mechanisms. Histol Histopathol 2001;16:37–44. 33. Kawaguchi H, Ogawa T, Shirakawa M, et al. Ultrastructural and ultracytochemical characteristics of multinucleated cells after hydroxyapatite implantation into rat periodontal tissue. J Periodontal Res 1992;27:48–54. 34. Wada T, Hara K, Quian HY, et al. Ultrastructural and histochemical study of β-tricalcium phosphate resorbing cells in periodontium of dog. J Periodontal Res 1989;24:391–401. 35. Dersot JM, Colombier ML, Lafont J, et al. Multinucleated giant cells elicited around hydroxyapatite particles implanted in craniotomy defects are not osteoclasts. Anat Rec 1995;242:166–76. 36. Willert HG, Bertram HB, Buchhorn GH. Osteolysis in alloarthroplasty of the hip: the role of ultra-high molecule weight polyethylene wear particles. Clin Orthop 1990;258:95–107. 37. Wenisch S, Stahl JP, Horas U, et al. In vivo mechanisms of hydroxyapatite ceramic degradation by osteoclasts: fine structural microscopy. J Biomed Mater Res 2003;67A:713–8. 38. Kieswetter K, Schwartz Z, Dean DD, et al. The role of implant surface characteristics in the healing of bone. Crit Rev Oral Biol Med 1996;7:329–45. European Journal of Trauma 2004 · No. 4 © Urban & Vogel 39. Müller-Mai CM, Voigt C, Gross U. Incorporation and degradation of hydroxyapatite implants of different surface roughness and surface structure in bone. Scanning Microsc 1991;4:501–11. 40. Doi Y, Iwanaga, H, Shibutani T, et al. Osteoclastic responses to various calcium phosphates in cultures. J Biomed Mater Res 1999;47:424–33. 41. Liebmann-Vinson A, Hemperly JJ, Guarino RD, et al. Bioactive extracellular matrices: biological and biochemical evaluation. In: Lewandrowski KU, Wise DL, Trantolo DJ, et al., eds. Tissue engineering and biodegradable equivalents. Scientific and clinical applications. New York–Basel: Marcel Dekker Inc., 2002:709–64. 42. Urist MR. Bone: formation by autoinduction. Science 1965;150:853–99. 43. Urist MR, Lietze A, Dawsen E. Tricalcium phosphate delivery system for bone morphogenetic protein. Clin Orthop 1984;187:277–84. 44. Schmidmaier G, Wildemann B, Heeger J, et al. Improvement of fracture healing by systemic administration of growth hormone and local application of insulin-like growth factor-1 and transforming growth factor-beta1. Bone 2002;31:165–72. 45. Yu Y, Yang JL, Chapman-Sheath PJ, et al. TGF-beta, BMPS, and their signal transducing mediators, Smads, in rat fracture healing. J Biomed Mater Res 2002;60:392–7. 46. Blumenfeld I, Srouji S, Lanir Y, et al. Enhancement of bone defect healing in old rats by TGF-beta and IGF-1. Exp Gerontol 2002;37:553–65. 47. Tabata Y, Ikada Y. Vascularization effect of basic fibroblast growth factor released from gelatin hydrogels with different biodegradabilities. Biomaterials 1999;20:2169–75. 48. Rifkin DB, Moscatlli D. Structural characterization and biological functions of basic fibroblast factor. J Cell Biol 1989;109:1–6. 49. Abraham JA, Wlang JL, Tunolo A, et al. Human basic fibroblast growth factor: nucleotide sequence and geromic organization. EMBO J 1986;5:2523–8. 50. Aspenberg P, Wans JS. Basic fibroblast growth factor. Dose- and time dependance in rats. Trans Orthop Res Soc 1994;19:18111. 51. Globus RK, Petterson-Bückendahl P, Gospodarowicz D. Regulation of bovine bone cell proliferation by fibroblast growth factor and transforming growth factors. Endocrinology 1988;123:98–105. 52. Gospodarowicz D, Ferrara N, Schweigerer G, et al. Structural characterization and biological functions of fibroblast growth factor. Endocrinol Rev 1987;8:95–114. 53. Jingnski S, Hedemann A, Kana SK, et al. Acid fibroblast growth factor injections stimulates cartilage enlargement and inhibits cartilage gene expressions in rat fracture. J Orthop Res 1990;8:364–71. 54. Ikade Y, Tabata Y, Lewandrowski KU, et al. Significance of drugs delivery in tissue engineering. In: Lewandrowski KU, Wise DL, Trantolo DJ, et al., eds. Tissue engineering and biodegradable equivalents. Scientific and clinical applications. New York–Basel: Marcel Dekker Inc., 2002:145–63. 55. Schnettler R, Alt V, Dingeldein E, et al. Bone ingrowth in bFGF coated hydroxyapatite ceramic implants. Biomaterials 2003;24:4603–8. Address for Correspondence PD Dr. Sabine Wenisch Experimental Trauma Surgery Kerkrader Straße 9 35394 Gießen Germany Phone (+49/641) 4994-160, Fax -161 e-mail: Sabine.Wenisch@chiru.med.uni-giessen.de 229