M. Araújo E. Linder J. Lindhe Effect of a xenograft on early bone formation in extraction sockets: an experimental study in dog Authors’ affiliations: Mauricio Araújo, Department of Dentistry, State University of Maringá, Paraná, Brazil Elena Linder, Jan Lindhe, Institute of Odontology, The Sahlgrenska Academy at Göteborg University, Gothenburg, Sweden Key words: biomaterial, bone formation, extraction socket, grafting Correspondence to: M. Araújo Rua Silva Jardim 15/sala 03 87013-010 Maringá Paraná Brazil Tel.: þ 55 44 3224 6444 Fax: þ 55 44 3224 6444 e-mail: odomar@hotmail.com Material and methods: Five beagle dogs were used. The distal roots of the third and fourth Abstract Aim: The aim of this study was to study the effect on early bone formation resulting from the placement of a xenograft in the fresh extraction socket in dogs. s mandibular premolars were removed. In one quadrant, a graft consisting of Bio-Oss Collagen was placed in the fresh extraction wound, while the corresponding premolar sites in the contra-lateral jaw quadrant were left non-grafted. After 2 weeks of healing, the dogs were perfused with a fixative, the mandibles removed, the experimental sites dissected, demineralized, sectioned in the mesio-distal plane and stained in hematoxyline–eosine. Results: The central portion of the non-grafted sockets was occupied by a provisional matrix comprised of densely packed connective tissue fibers and mesenchymal cells. Apical and lateral to the provisional matrix, newly formed woven bone was found to occupy most of the sockets. In the apical part of the grafted sockets, no particles of the xenograft could be observed but newly formed bone was present in this portion of the experimental site. In addition, limited numbers of woven bone trabeculae occurred along the lateral socket walls. The central and marginal segments of the grafted sockets, however, were occupied s by a non-mineralized connective tissue that enclosed Bio-Oss particles that frequently were coated by multinucleated cells. s Conclusions: The placement of Bio-Oss Collagen in the fresh extraction wound obviously delayed socket healing. Thus, after 2 weeks of tissue repair, only minute amounts of newly formed bone occurred in the apical and lateral borders of the grafted sockets, while large amounts of woven bone had formed in most parts of the non-grafted sites. Date: Accepted 5 April 2008 To cite this article: Araújo M, Linder E, Lindhe J. Effect of a xenograft on early bone formation in extraction sockets: an experimental study in dog. Clin. Oral Impl. Res. 20, 2009; 1–6. doi: 10.1111/j.1600-0501.2008.01606.x In previous experiments in dogs (Cardaropoli et al. 2003; Araújo & Lindhe 2005), it was observed that healing of an extraction socket included (i) early formation of woven bone (modeling) and (ii) the subsequent replacement of this immature bone with lamellar bone and marrow (remodeling). Healing of the edentulous site was further characterized by (iii) loss of bundle bone in the walls of the socket and (iv) a marked reduction in the height of the buccal bone crest. Similar observations c 2009 The Authors. Journal compilation c 2009 Blackwell Munksgaard were reported from retrospective and prospective studies in man (Pietrokovski & Massler 1967; Schropp et al. 2003; Pietrokovski et al. 2007; for review see Chen et al. 2004). Different approaches were used to prevent the post-extraction ridge reductions. Such methods included the utilization of various graft materials and application techniques (e.g. Becker et al. 1998; Artzi et al. 2000; Froum et al. 2002; Carmagnola et al. 2003; Nevins et al. 2006). 1 Araújo et al . Early bone formation in extraction sockets In a recent study by Araújo et al. (2008), the dog model was used to study healing of an extraction socket that had been grafted s with Bio-Oss Collagen. It was observed that the placement of the xenograft in the fresh extraction socket after a 3-month period of healing appeared to have interfered with the process of bone modeling and remodeling. Thus, while the nongrafted sites underwent marked resorption during the 3-month interval, the dimension of the walls of the grafted sockets as well as the profile of the edentulous ridge remained unchanged. The newly formed hard tissue in the marginal portion of the grafted sites, however, contained a large s number of Bio-Oss particles that were surrounded by immature woven bone. A more detailed analysis of the newly formed tissues in the two differently treated sockets further indicated that the process of tissue remodeling had progressed further in the non-grafted than in the grafted sites. Thus, while about 50% of the tissue volume in the non-grafted sites was occupied by bone marrow, in the grafted site bone marrow represented only about 25% of the tissue volume. This indicates that the graft may in fact have delayed healing. The objective of the present experiment was to evaluate the effect on early wound healing and bone modeling of the placement of a xenograft in the fresh extraction socket in dogs. Grafted sites In one quadrant of the mandible, a graft s consisting of Bio-Oss Collagen (Geiss tlich , Wolhusen, Switzerland) was placed in the fresh extraction wounds (Fig. 1) in a manner previously described (Araújo et al. 2008). The marginal portion (about 2 mm) of the buccal bone wall of the distal sockets was also covered with the biomaterial. Non-grafted sites The corresponding premolar sites in the contra-lateral jaw quadrant were exposed and managed in the manner described above. No biomaterial was, however, placed in the fresh extraction sockets. The flaps were replaced in the grafted and non-grafted sites and were retained with interrupted sutures. The sutures were removed after 1 week. Mechanical tooth cleaning was performed on days 3, 7 and 10 after surgery. After 2 weeks of healing, the dogs were euthanized with an overdose of Ketamin (Agener União) and perfused, through the carotid arteries, with a fixative containing a mixture of 5% glutaraldehyde and 4% formaldehyde (Karnovsky 1965). The mandibles were removed and placed in the fixative. Each experimental site, including the distal socket area, was diss sected using a diamond saw (Exact Apparatebeau, Norderstedt, Hamburg, Germany), and subsequently demineralized in a 10% solution of EDTA. The Material and methods The ethical committee of the State University of Maringá approved the research protocol. Five beagle dogs about 1 year old and weighing between 10 and 12 kg each were used. During surgical procedures, the animals were anesthetized with intravenously administered Ketamin (10%, 8 mg/kg; Agener União, São Paulo, Brazil). Pocket and crestal incisions were made in the posterior premolar region in both quadrants of the mandible. Buccal and lingual full thickness flaps were elevated to disclose the alveolar crest. The canal of the mesial roots of the third and fourth mandibular premolars was reamed and filled with gutta-percha. The premolars were subsequently hemi-sected with the use of fissure burs. The distal roots were carefully removed with the use of elevators. 2 | Clin. Oral Impl. Res. 20, 2009 / 1–6 blocks were dehydrated in ethanol and isopropanol and embedded in paraffin. Serial sections representing the experimental sites were cut in a mesio-distal plane and parallel to its long axis. The microtome was set at 6 mm. From each site, three sections, representing the central part of the socket and about 20 mm apart, were selected for the histological examination. The sections were stained in hematoxyline–eosine Histological examination The histological examinations were pers formed in a Leitz DM-RBE microscope (Leica, Wetzlar, Germany) equipped with s an image system (Q-500 MC ; Leica). The composition of the newly formed tissue in the extraction socket was determined using a point counting procedure. Thus, a lattice comprising 100 light points (modified from Schroeder & MünzelPedrazzoli 1973) was superimposed over the tissue in the healing socket, and the percentage area occupied by (i) mineralized bone, (ii) bone marrow, (iii) provisional matrix, (iv) granulation tissue and (v) Bios Oss particles was determined (magnification 200). In addition, the number of s Bio-Oss particles in the socket that were in direct contact with multinucleated cells was determined and expressed as percens tage of the total number of Bio-Oss particles identified. Mean values and standard deviations of the different variables were calculated using the dog as the statistical unit. Results Healing was uneventful at all sites. The clinical examination of the experimental sites that was performed after 2 weeks (Fig. 2) disclosed that a slightly invaginated and modestly inflamed soft tissue covered the socket entrance. Gross histological observations Fig. 1. Clinical photographs illustrating the graft of s Bio-Oss Collagen placed in the fresh extraction socket. Note that the graft material also covers the marginal portion of the buccal wall of the socket. The connective tissue of the mucosa covering the grafted as well as the non-grafted sockets was rich in mesenchymal cells, vascular structures and clusters of inflammatory cells. The central portion of the non-grafted sockets (control site; Fig. 3) was occupied by a provisional matrix comprised of c 2009 The Authors. Journal compilation c 2009 Blackwell Munksgaard Araújo et al . Early bone formation in extraction sockets Fig. 4. Microphotograph representing the provisional matrix in the non-grafted site. Note the densely packed connective tissue fibers and cells; V, vessel. Hematoxyline–eosine stain; original magnification: 200. Fig. 2. Clinical photographs illustrating a site s grafted with Bio-Oss Collagen after 2 weeks of healing. Note the slightly invaginated crestal surface and the slightly inflamed soft tissue. Fig. 6. Microphotograph representing the woven bone in the non-grafted site. The trabeculae of woven bone were lined with osteoblasts and occasional isolated osteoclasts (arrows). BM, primitive bone marrow; V, vessel; WB, woven bone. Hematoxyline–eosine stain; original magnification: 200. Fig. 5. Microphotograph representing the mineralization front of the woven bone in the non-grafted site. Note that the woven projections had initiated the formation of a primary spongiosa (primary osteons and enclosed vascular structures). PM, provisional matrix; V, vessel; WB, woven bone. Hematoxyline–eosine stain; original magnification: 200. Fig. 3. Microphotograph of a mesio-distal section representing a non-grafted site. Note that most of the alveolar socket is occupied by woven bone and provisional matrix. PM, provisional matrix; GT, granulation tissue; WB, woven bone. Hematoxyline– eosine stain; original magnification: 16. densely packed connective tissue fibers and mesenchymal cells (Fig. 4). Apical and lateral to the provisional matrix, newly formed woven bone was found to occupy most of the socket (Fig. 3). Several of the bone projections formed primary osteons and enclosed vascular structures (Fig. 5). The trabeculae of immature bone were lined with osteoblasts and occasionally isolated osteoclasts, residing in resorption bays, could be observed on the surface of the newly formed bone (Fig. 6). No particles of the xenograft could be observed in the apical part of the grafted sockets (test site; Fig. 7), but newly formed immature bone was found to occupy this portion of the post-extraction wound. In addition, limited numbers of trabeculae of woven bone occurred along the lateral walls of the sockets. The central and marginal segments of the wound were occupied by a connective tissue that enclosed the s Bio-Oss particles (Fig. 7). This connective c 2009 The Authors. Journal compilation c 2009 Blackwell Munksgaard Fig. 7. Microphotograph of a mesio-distal section representing a grafted site. Note that the central and marginal segments of the sockets were occupied s by a connective tissue that enclosed Bio-Oss particles (asterisks). PM, provisional matrix; WB, woven bone. Hematoxyline–eosine stain; original magnification: 16. tissue had the features of a provisional matrix (rich in mesenchymal cells, fibers and vessels) but small amounts of mononuclear leukocytes could consistently also be observed. In addition, the provisional matrix harbored a multitude of large multinucleated cells that were in direct contact with and occupied substantial portions of the periphery of many of the graft particles (Figs 8 and 9). Segments of the graft that 3 | Clin. Oral Impl. Res. 20, 2009 / 1–6 Araújo et al . Early bone formation in extraction sockets Fig. 8. Microphotograph representing the provisional matrix in the grafted site. The provisional matrix harbored large multinucleated cells (arrows) that were in direct contact with and occupied different portions of the periphery of many of the graft s particles. Bp, Bio-Oss particles; PM, provisional matrix. Hematoxyline–eosine stain; original magnification: 200. Fig. 10. Microphotograph representing the provisional matrix in a grafted site. Note that the Bios Oss particle was not only in direct contact with multinucleated cells (arrows) but also with provisional matrix or newly formed woven bone (arrows heads). a, artifact; Bp, Bio-Oss particles; PM, provisional matrix; WB, woven bone. Hematoxyline–eosine stain; original magnification: 400. Fig. 12. Microphotograph representing a higher magnification of a multinucleated cell present in a s shallow invagination on the surface of a Bio-Oss particle in the provisional matrix of a grafted site. s Bp, Bio-Oss particle. Hematoxyline–eosine stain; original magnification: 1000. Table 1. Mean values (SD) of the proportions (%) of the different tissues in the grafted and non-grafted alveolar sockets Grafted site Granulation tissue Provisional matrix Mineralized bone s Bio-Oss Non-grafted site 3.7 (0.8) 3.1 (0.5) 62.7 (12.6) 49.1 (8.2) 14.7 (5.9) 47.8 (15.6) 18.9 (9.8) – Discussion Fig. 9. Microphotograph representing a higher magnification of multinucleated cells (arrows) that s coated the entire periphery of a Bio-Oss particle s in a grafted site. Bp, Bio-Oss particle. Hematoxyline–eosine stain; original magnification: 1000. were not occupied by the multinucleated cells were found to be in direct contact with provisional matrix or newly formed mineralized bone (Fig. 10). In the lateral portions of the socket, several adjacent Bios Oss particles were frequently found to be ‘connected’ (bridged) by zones of immature bone (Fig. 11). There were no obvious signs of active degradation of the bovine mineral and only at a few sites could multinucleated cells be observed in shallow invaginations on the surface of the graft (Fig. 12). Morphometric measurements The percentage area occupied by the various tissues in the alveolar socket is presented in Table 1. The proportion of 4 | Clin. Oral Impl. Res. 20, 2009 / 1–6 Fig. 11. Microphotograph representing the lateral portions of the socket in a grafted site. Note few s multinucleated cells at the surface of the Bio-Oss s particles and that several adjacent Bio-Oss particles were bridged by newly formed woven bone. Bp, Bios Oss particles; V, vessel; WB, woven bone. Hematoxyline–eosine stain; original magnification: 400. granulation tissue was 3.7 0.8% (SD) in the grafted and 3.1 0.5% in nongrafted sites. The proportions of provisional matrix were 62.7 12.6% (grafted sites) and 49.1 8.2% (non-grafted sites), and the corresponding proportions of mineralized bone were 14.7 5.9% and s 47.8 15.6%. The Bio-Oss particles occupied 18.9 9.8% of the tissue present in the grafted sockets. The percentage of s Bio-Oss particles that were in direct contact with multinucleated cells was 53 12.2%. The present experiment indicated that the early phase of hard tissue formation was retarded in extraction sockets that, immediately after tooth removal, had been s grafted with Bio-Oss Collagen. It appeared that this delayed wound healing and bone modeling may have been influenced by the multinucleated cells that occurred in tissues harboring the xenograft. Thus, in the grafted sites, substantial amounts of newly formed bone could only be detected in the apical portion of the socket where the graft material was absent. In the remaining portions of the grafted sockets, a mildly inflamed provisional matrix surrounded s the majority of the Bio-Oss particles, the surface of which was frequently, but not always, coated with multinucleated cells. In isolated areas of the grafted sites, multinucleated cells were absent and the foreign material occasionally surrounded by the woven bone that bridged adjacent granules of the bovine bone. c 2009 The Authors. Journal compilation c 2009 Blackwell Munksgaard Araújo et al . Early bone formation in extraction sockets In the non-grafted control sites, large amounts of woven bone had formed in most compartments of the socket. This finding is in agreement with observations from similar experiments that investigated tissue modeling and remodeling in extraction sockets as well as in mechanically produced defects in the alveolar ridge in dogs (Cardaropoli et al. 2003, 2005; Araújo & Lindhe 2005). Similar amounts of new bone had formed in the apical segments of the test and control sites, while in the remaining portions of the grafted sockets, only minute amounts of immature bone could be found. There are reasons to assume that following root extraction and graft placement, the ensuing flow of blood into the cone-shaped s socket forced the Bio-Oss Collagen material in marginal direction. A void was hereby established in the apical zone of the socket that could house a coagulum that apparently was unaffected by the translocated foreign material. Hence, in this apical portion of the test socket as well as in the entire non-grafted socket, tissue modeling evidently followed a normal patter, i.e. the clot was replaced with granulation tissue, provisional matrix and primary spongiosa (Cardaropoli et al. 2003). In the non-mineralized portions of the newly formed tissue in the grafted sockets, large multinucleated cells were found to surround a substantial number of the Bios Oss granules. Thus, 450% of the granules were more or less completely coated by multinucleated cells. In the remaining s portions of the graft, the Bio-Oss particles were surrounded either by a leukocytecontaining provisional matrix or occasionally newly formed woven bone. In such locations, multinucleated cells were conspicuous by their absence. The occurrence of multinucleated cells s on the surface of Bio-Oss material was previously described (e.g. Piatelli et al. 1999; Merkx et al. 2000; Tadjoedin et al. 2003; Houshmand et al. 2007; Simion et al. 2007), and the cells were identified as osteoclasts and their presence related to the resorption and gradual elimination of the graft material (Berglundh & Lindhe 1997; Araújo et al. 2001; Cardaropoli et al. 2005). It has not been made clear, however, whether the multinucleated cells found on the surface of this xenograft are indeed active osteoclasts as suggested by, e.g., Piatelli et al. (1999), Tadjoedin et al. (2003), Tapety et al. (2004) or non-active, impaired osteoclasts cells as suggested by Taylor et al. (2002). It is documented that multinucleated cells originate from mononuclear phagocytes that belong to the hematopoietic line of stem cells (Bredan & Xing 2007). Such mononuclear phagocytes (macrophages) may fuse and differentiate into osteoclasts or giant cells as reported by Vignery (2000) who claimed that while osteoclasts develop on bone surfaces, other multinucleated cells such as giant cells differentiate ‘primarily in chronic inflammatory sites in response to bacterial invasion and foreign bodies such as implants . . ..’ In the tissue sampled from the grafted sites of the current experiment, the multinucleated cells on the graft surface were almost never found to reside in resorption bays (Fig. 13), while morphologically similar cells present on adjacent surfaces of host bone were almost consistently located in characteristic Howship’s lacunae and were classified as active osteoclasts. It has been documented that a foreign body reaction can be elicited to a xenograft that is clinically non-immunogenic, nontoxic and chemically inert (for review see Luttikhuizen et al. 2006). Furthermore, it was proposed (Ratner 2001) that such a reaction may occur during early phases of wound healing. The presence of multinucleated cells in the granulation tissueprovisional matrix in the grafted sockets of the present experiment indicated that the xenograft particles may have been recognized as being foreign to the host, and that a foreign body reaction may have been elicited. The early phase of such a reaction Fig. 13. Microphotograph representing a higher magnification of multinicleated cells (arrows) on the graft surface in the provisional matrix of a grafted site. Note the multinucleated cells were present on a s smooth non-invaginated surface. Bp, Bio-Oss particle. Hematoxyline–eosine stain; original magnification: 1000. is associated with inflammatory processes (Vignery 2000; Ratner 2001; Luttikhuizen et al. 2006) that may impair hard tissue deposition. This may explain the delayed formation of bone that was observed in the graft-containing areas of the extraction socket, and that the multinucleated cells found on the surface of the xenograft were giant cells. Most of the graft particles present in the test sites were surrounded either by a dense provisional matrix or newly formed woven bone. In such locations, multinucleated cells were conspicuous by their absence. In a previous experiment in the dog (Araújo et al. 2008), socket grafting was performed and healing evaluated 3 months after tooth extraction. In sections from this longer term study, multinucleated cells were never observed at the surface of the xenos graft, but most of the Bio-Oss granules were in direct contact with immature woven bone. Hence, in the interval between 2 weeks and 3 months of healing, the multinuclear cells may have completed their function, undergone apoptosis and disappeared. References Araújo, M., Linder, H., Wennström, J. & Lindhe, J. s (2008) The influence of Bio-Oss collagen on healing of an extraction socket: an experimental study in the dog. The International Journal of Periodontics & Restorative Dentistry 28: 123–135. c 2009 The Authors. Journal compilation c 2009 Blackwell Munksgaard Araújo, M.G., Carmagnola, D., Berglundh, T., Thilander, B. & Lindhe, J. (2001) Orthodontic movement in bone defects augmented 5 | Clin. Oral Impl. Res. 20, 2009 / 1–6 Araújo et al . Early bone formation in extraction sockets s with Bio-Oss . An experimental study in dogs. Journal of Clinical Periodontology 28: 73–80. Araújo, M.G. & Lindhe, J. (2005) Dimensional ridge alterations following tooth extraction. An experimental study in the dog. Journal of Clinical Periodontology 32: 212–218. Artzi, Z., Tal, H. & Dayan, D. (2000) Porous bovine bone mineral in healing of human extraction sockets. Part 1: histomorphometric evaluations at 9 months. Journal of Periodontology 71: 1015– 1023. Becker, W., Cameron, C., Sennerby, L., Urist, M. & Becker, B. (1998) Histologic findings after implantation and evaluation of different grafting materials and titanium micro screws into extraction sockets: case reports. Journal of Periodontology 69: 414–421. Berglundh, T. & Lindhe, J. (1997) Healing around implants placed in bone defects treated with Bios Oss . An experimental study in the dog. Clinical Oral Implants Research 8: 117–124. Bredan, B.F. & Xing, L. (2007) Biology of RANK, RANKL, and osteoprotegerin. Arthritis Research and Therapy 9: 1–7. Cardaropoli, G., Araújo, M., Hayacibara, R., Sukekava, F. & Lindhe, J. (2005) Healing of extraction sockets and surgically produced – augmented and non-augmented – defects in the alveolar ridge. An experimental study in the dog. Journal of Clinical Periodontology 32: 435–440. Cardaropoli, G., Araújo, M. & Lindhe, J. (2003) Dynamics of bone tissue formation in tooth extraction sites. An experimental study in dogs. Journal of Clinical Periodontology 30: 809–818. Carmagnola, D., Adriaens, P. & Berglundh, T. (2003) Healing of human extraction sockets filled s with Bio-Oss . Clinical Oral Implants Research 14: 137–143. Chen, S.T., Wilson, T.G. Jr. & Hämmerle, C.H. (2004) Immediate or early placement of implants following tooth extraction: review of biologic basis, clinical procedures, and outcomes. Inter- 6 | Clin. Oral Impl. Res. 20, 2009 / 1–6 national Journal of Oral & Maxillofacial Implants 19 (Suppl.): 12–25. Froum, S., Cho, S.C., Rosenberg, F., Rohrer, M. & Tarnow, D. (2002) Histological comparison of healing extraction sockets implanted with bioactive glass or demineralized freeze-dried bone allograft: a pilot study. Journal of Periodontology 73: 94–102. Houshmand, B., Rahimi, H., Ghanavati, F., Alisadr, A. & Eslami, B. (2007) Boosting effect of bisphosphonates on osteoconductive materials: a histologic in vivo evaluation. Journal of Periodontal Research 42: 119–123. Karnovsky, M.J. (1965) A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy. Journal of Cell Biology 27: 137A– 138A. Luttikhuizen, D.T., Harmsen, M.C. & Van Luyn, M.J.A. (2006) Cellular and molecular dynamics in the foreign body reaction. Tissue Engineering 12: 1955–1970. Merkx, M.A., Maltha, J.C. & Freihofer, H.P. (2000) Incorporation of composite bone implants in the facial skeleton. Clinical Oral Implants Research 11: 422–429. Nevins, M., Camelo, M., De Paoli, S., Friedland, B., Schenk, R.K., Parma-Benfenati, S., Simion, M., Tinti, C. & Wagenberg, B. (2006) A study of the fate of the buccal wall of extraction sockets of teeth with prominent roots. The International Journal of Periodontics and Restorative Dentistry 26: 19–29. Piatelli, M., Favero, G.A., Scarano, A., Orsini, G. & Piatelli, A. (1999) Bone reactions to anorganic bovine bone (Bio-Oss) used in sinus augmentation procedures: a histologic long-term report of 20 cases in humans. International Journal of Oral & Maxillofacial Implants 14: 835–840. Pietrokovski, J. & Massler, M. (1967) Alveolar ridge resorption following tooth extraction. Journal of Prosthetic Dentistry 17: 21–27. Pietrokovski, J., Starinsky, R., Arensburg, B. & Kaffe, I. (2007) Morphologic characteristics of bony edentulous jaws. Journal of Prosthodontics 16: 141–147. Ratner, B.D. (2001) Replacing and renewing: synthetic materials, biomimetics, and tissue engineering in implant dentistry. Journal of Dental Education 65: 1340–1347. Schroeder, H.E. & Münzel-Pedrazzoli, S. (1973) Correlated morphometric and biochemical analysis of gingival tissue. Journal of Microscopy 99: 301–329. Schropp, L., Wenzel, A., Kostopoulos, L. & Karring, T. (2003) Bone healing and soft tissue contour changes following single-tooth extraction: a clinical and radiographic 12-month prospective study. The International Journal of Periodontics and Restorative Dentistry 23: 313–323. Simion, M., Fontana, F., Rasperini, G. & Maiorana, C. (2007) Vertical ridge augmentation by expanded-polytetrafluoroethylene membrane and a combination of intraoral autogenous bone graft and deproteinized anorganic bovine bone (Bio Oss). Clinical Oral Implants Research 18: 620–629. Tadjoedin, E.S., de Lange, G.L., Bronckers, A.L., Lyaruu, D.M. & Burger, E.H. (2003) Deproteinized cancellous bovine bone (Bio-Oss) as bone substitute for sinus floor elevation. A retrospective, histomorphometrical study of five cases. Journal of Periodontology 30: 261–270. Tapety, F.I., Amizuka, N., Uoshima, K., Nomura, S. & Maeda, T. (2004) A histological evaluation of the involvement of Bio-Oss in osteoblastic differentiation and matrix synthesis. Clinical Oral Implants Research 15: 315–324. Taylor, J.C., Cuff, S.E., Leger, J.P.L., Morra, A. & Anderson, G.I. (2002) In vitro osteoclast resorption of bone substitute biomaterials used for implant site augmentation: a pilot study. International Journal of Oral & Maxillofacial Implants 17: 321–330. Vignery, A. (2000) Osteoclasts and giant cells: macrophage-macrophage fusion mechanism. International Journal of Experimental Pathology 81: 291–304. c 2009 The Authors. Journal compilation c 2009 Blackwell Munksgaard