| The minimization of morbidity in cranio-maxillofacial osseous reconstruction: Bone graft harvesting and coral-derived granules as a bone graft substitute | ||
|---|---|---|
| Prev | Chapter 2. Review of the literature | Next |
The reconstructive options in the osseous reconstruction of the cranio-maxillofacial skeleton include autogenous bone grafts harvested from local or distant sources (Kainulainen et al. 2002a). Allogeneic bone from another individual may also be considered, as might xenogeneic bone from another species. Because the possibilities of immunogenic problems exist, such grafts were first treated with a freezing technique (Herndon & Chase 1954). Later other methods to deal with immunogenicity were developed such as freeze drying, deproteinating and demineralizing techiques (Buchardt 1983). Alloplasts have also been developed to replace bone. In addition a number of surgical procedures have been designed to increase the amount of bone available locally without bone grafting (Dahlin et al. 1988, Gaggi et al. 2000, Oikarinen et al. 2003). Bone reconstruction is best understood if the process of bone healing is first considered (Hollinger & Wong 1996).
Bone is a unique tissue. It can be injured and then can repair itself and return to full function without scarring or deformity (Salter 1983). Embryonic bone development is repeated in the healing of bone. The pattern of bony healing is dictated by the host bed, vascular supply, oxygen tension and the stability of the bone segments (Buckwalter et al. 1995a). Healing can occur either directly as primary bone healing or secondarily, demonstrating an intermediate cartilaginous phase (Hollinger et al. 1994).
Bone healing can be illustrated using the model of the healing bone graft. It is important, when discussing healing in relation to bone reconstruction to differentiate between a graft and an implant. A graft may be defined as a transferable material that contains living cells and can be used for reconstruction. An implant is differentiated from a graft in that it does not contain any living cells. When the stages of graft or implant incorporation are examined, the presence of viable cells that are transferred in a graft will usually differentiate the two (Gray & Elves 1982). A graft of autogenous bone will contain bone-forming cells, fibrin and platelets. The endosteal osteoblasts and hematopoetic cells will survive as long as five days post transplantation due to their ability to absorb nutrients from the surrounding tissues (Marx 1994).
Within hours of placing a graft the initial regenerative process begins (Garge et al. 1998). Entrapped platelets degranulate releasing potent growth factors such as platelet-derived growth factor (PDGF) from their alpha granules and transforming growth factor-beta 1 (TGF-β 1) (Caplan 1995). Endothelial cells initiate capillary ingrowth as they bind PDGF. Next endosteal osteoblasts and hematopoiteic stem cells are stimulated to initiate mitosis increasing their numbers. These cells also commence their production of osteoid (Friedenstein et al. 1996). This is mediated by the binding of TGF-β 1 to cell receptors. After the third day, the influence of the growth factors transplanted with the graft is replaced by the action of locally induced macrophages (Knighton et al. 1983). They efficiently synthesize growth factors and will regulate bone healing from this point. By the end of the second week, the graft will demonstrate complete revascularization. Endosteal osteoblasts from the transplanted bone will begin laying down osteoid and stem cells will begin differentiating into osteoblasts. Resultant islands of bone formation are then seen developing within the graft. Once the graft has become revascularized circulating stem cells, attracted to the wound, may also transform into bone forming units (Marx 1994).
This initial bone formation, which occurs as a result of the transfer of osteocompetent cells contained within the graft, has been referred to as Phase I bone (Axhausen 1956). Complete by six weeks, graft viability is maintained as sufficient quantities of newly mineralized matrix have been deposited. The bone that has formed does so without initial cartilaginous deposition and is referred to as woven bone. This bone is extremely cellular and disorganized but does not demonstrate any independent structural integrity. During the second phase of healing, bone will undergo a remodelling phenomenon referred to as lamellar compaction. The resultant lamellar bone will be less cellular, more mineralized and is highly organized (Buckwalter et al. 1995b). As with all bone, this newly formed matrix will mature as it responds to the physical demands placed upon it. Finally, it will enter into a remodelling phase similar to normal skeletal turn-over (Marx 1994).
Osteoinduction describes a process whereby new bone is produced in an area where there was no bone before, where one tissue or its derivative causes another undifferentiated tissue to differentiate into bone. The phenomenon of osteoinduction was first described in the classic works of Urist (Urist & McLean 1952, Urist 1965, Urist et al. 1977). Bone matrix was shown to induce bone formation within muscle pouches of many species of animals. Later a specific extract from bone, a protein now referred to as Bone Morphogenetic Protein (BMP), was identified as that factor which caused the phenomenon (Urist et al. 1979, Mizutani & Urist 1982). Since then a great deal of research has resulted in the discovery of a variety of entities having different effects on bone (Goldring & Goldring 1996). These compounds may be classified as osteoinducers, osteopromoters or bioactive peptides (Hauschka et al. 1988).
Osteoconduction describes bone formation by the process of ingrowth of capillaries and osteoprogenitor cells from the recipient bed into, around and through a graft or bioimplant. Therefore the graft or bioimplant acts as a scaffold for new bone formation (Buchardt 1983). Unlike osteoinduction, this process occurs in an already bone containing environment. Osteoconduction describes the facilitation of bone growth along a scaffold of autogenous, allogenic or alloplastic materials.
There are a number of techniques, which enable the surgeon to maximize the available bone in the cranio-maxillofacial skeleton without harvesting a bone graft. These techniques serve to minimize reconstructive morbidity, as there is no graft donor site. Osteocondensation is one such technique. It can reshape the alveolar bone of the maxilla for example, to more optimally house a dental implant, resulting in better primary stability in areas of poor bone quality. Orthopaedic surgeons have practiced osteocondensation since the early 1960s (Valen & Locante 2000). The major advantage of this technique is that an implant bed is created with either minimal drilling or no bone removal (Syrakas et al. 2000) and with osteotomes, which compress the bone. There are implants, which produce osteocondensation and are called press-fit fixtures (de Wijs & Cune 1997, Valen & Locante 2000). In the cranio-maxillofacial skeleton, osteocondensation is best performed in the maxilla.
The major proponent of osteocondensation in the cranio-maxillofacial skeleton has been Summers who described a method to increase the width of alveolar bone and to facilitate sinus floor elevation, without opening the lateral sinus wall (Summers 1994a,b,c, Summers 1995). The technique was further developed to include the use of D-shaped osteotomes and chisels which produced lateral widening of the alveolar ridge and osteocompression, increasing the density of cancellous bone (Tatum 1986, de Wijs & Cune 1997). This ridge expansion osteotomy is achievable using osteotomes which have concave tips and sharpened edges. The instruments are shaped to allow progressively larger osteotome tips to fit into the opening created by the previous osteotome. Instruments are sensitive to changes in bone texture and density and allow excellent tactile sensation for the surgeon (Summers 1994a). The minimum alveolar width necessary for lateral alveolar widening by compression is 2–3 mm assuming that spongious bone is found between cortical layers (Summers 1994b, Sethi & Kaus 2000).
Alveolar ridges can also be widened using the crestal split technique using osteotomes and chisels to produce a “greenstick fracture” at the base of the alveolus. The remaining periosteum is left intact and attached to the bone. This pedicled buccal cortex is repositioned and a new implant bed is created without any drilling. Lateral widening by completely exposing the labial cortex has also been introduced (Duncan & Westwood 1997). The major benefit of crestal widening is that it allows the thin alveolar bone to be utilized for implantation without grafting (Oikarinen et al. 2003). Esthetics and implant positioning are improved and wider implants can also be used. The bone can be moulded to some extent due to its viscosity (de Wijs & Cune 1997). Bone compression is achieved along with an increase in the density of trabeculations of the adjacent site (Komarnyckyj & London 1998). In addition the resulting gap can, if desired be covered by a nonresorbable membrane (Simion et al. 1992, Engelke 1997) and filled with allogenic material (Engelke 1997). Interpositional autogenous bone grafts have been used to improve bony healing in the gap (Lustman & Lewinstein 1995).
Guided bone regeneration (GBR) has been used for minor augmentation procedures in the cranio-maxillofacial skeleton and prior to dental implant placement (Dahlin et al. 1988, 1989, Borgner et al. 1999, Buser 1990, Simion et al. 2001). GBR is a technique in which bone growth is enhanced by preventing soft tissue ingrowth into the desired area and utilizes either resorbable or nonresorbable membranes. Metallic membranes (von Arx et al. 1996) or membranes supported by a titanium frame (Simion et al. 1994, 1998, 2001) have been tested and have been successful. An acellular dermal matrix has been used as a barrier membrane with demineralized freeze-dried bone allograft (Fowler et al. 2000).
The use of membranes is a controversial issue in implantology and their use is certainly very technique-sensitive (Chiapasco et al. 1999). Nonresorbable membranes need a second operation for their removal (von Arx et al. 1996). Resorbable membranes can be associated with inflammation (Yoshinari et al. 1998). Intact periosteum, a split palatal or gingival flap are regarded by some as natural membranes and their use may obviate the need for a membrane (Ylimaz et al. 1998). Nevertheless, good results with augmentation procedures using membranes have been presented (Buser 1990, Simion et al. 1992, 1994, Lustmann & Lewinstein 1995, Lekovic et al. 1998, Simion et al. 2001). Vertical increase of a narrow alveolar crest has been shown to be possible with membranes (von Arx et al. 1996, Simion et al. 1998).
Distraction osteogenesis (DO) of the long bones in growing children has been used for decades to gradually lengthen osteotomized bones without a bone graft. The resulting distraction gap is initially filled with callus, which later matures into bone (Ilizarow 1989). DO has also been adapted to the maxillofacial area and special devices and implants are being developed for that purpose (Gaggi et al. 2000, Watzek G et al. 2000).
The DO technique has also been adapted for limited augmentations of the alveolar crest prior to implantation. Some systems use hardware, which expands the jaw over time, and then is removed at the time of dental implant placement (Watzek et al. 2000). Some have tried to utilize the implant itself as the distraction device (Chin & Toth 1996, Gaggi et al. 1999, 2000). The daily rate of alveolar crest distraction ranges from 0.25–0.5 mm and is initiated from two days to one week after the primary osteotomy. DO is continued up to 30 days and the final gain will be between 4 and 7 mm (Gaggi et al. 2000, Urbani et al. 1999). In some cases overcorrection is recommended (Gaggi et al. 2000). However some local limitations due to the lack of stretching of the palatal tissues, may not allow the distracted segment to move exactly as planned and then only in two dimensions (Chin & Toth 1996). Appliances intended to allow three-dimensional DO have been introduced (Ylimaz et al. 1998, Watzek et al. 2000). The benefits of DO are that donor site morbidity from harvesting of bone grafts and dehiscences of grafted bone are avoided (Chin & Toth 1996). However, a second surgery to remove and perhaps replace hardware is needed if implant-based distraction is not used. While DO could eliminate a donor site and thereby limit morbidity, it is so labour intensive that the patient trades the morbidity of the bone graft donor site for the inconvenience of wearing and tolerating potentially cumbersome hardware for longer periods of time.
At the present time, autogenous bone grafting is the gold standard by which all techniques of osseous reconstruction of the cranio-maxillofacial skeleton must be judged. Autogenous cancellous bone grafts produce the most successful and predictable results (Marx 1994). Free bone grafts act mostly as scaffolds and are thus more osteoconductive than osteoinductive even though osteogenic activity may have remained in the spongious part of the graft (Buchardt 1983). The major disadvantage of autogenous grafts is the need for a second surgical site and the morbidity resulting from harvesting. The source of autograft, however, is not limitless for the patient. A point may be reached in reconstruction where the donor site morbidity may exceed the discomfort of the presenting complaint.
There are essentially two forms of nonvascularized free autogenous bone grafts: cortical and cancellous (Bonutti et al. 1998, Keller et al. 1998, Vinzenz et al. 1998). Buchardt has summarized the three essential differences between the two. Cancellous grafts are revascularized more rapidly and completely than cortical grafts. Creeping substitution of a cancellous graft initially involves an appositional bone formation phase, followed by a resorptive phase, whereas cortical grafts undergo a reverse creeping substitution process. Cancellous grafts tend to repair completely with time whereas cortical grafts remain as an admixture of necrotic and viable bone (Buchardt 1983).
Cortical grafts are able to withstand mechanical forces earlier, however, they take more time to revascularize. Cortical grafts are useful for filling defects where early mechanical loading is required (Boyne 1997). The cortical component can be incorporated into the fixation of the graft and can consequently be used in situations where bone is comminuted or where there are bony voids. In the cranio-maxillofacial skeleton these forms of grafts may also be used to onlay areas such as decreased vertical or horizontal alveolar ridges, to improve facial contours or they can be inlayed within bone to fill bony voids. Common sites for the harvesting of cortical grafts are the cranial vault, ribs and the medial or lateral table of the anterior aspect of the iliac crest, the posterior iliac crest as well as the mandibular symphysis (Kainulainen et al. 2002b).
Cancellous grafts have more widespread applications, are generally easier to manipulate and revascularize more rapidly (Marx 1993). The most abundant source of cancellous bone is the anterior or posterior iliac crest. Cancellous bone imparts no mechanical strength. When cancellous bone is used to reconstruct large continuity defects additional stability and rigid fixation is required, such as that which is afforded by using a titanium mesh system (Tideman et al. 1998). In the cranio-maxillofacial skeleton these grafts are packed into bony defects such as alveolar clefts and maxillary sinus floor augmentations (Boyne & James 1980, Merkx et al. 2003). The corticocancellous graft usually produces the best results by combining the attributes of both graft forms and can be placed easily into an interpositional location (Stoelinga et al. 1978, Egbert et al. 1986). These grafts allow for mechanical stabilization while at the same time providing for good revascularization. Others will particulate corticocancellous bone creating a mixed graft which can be used for the restoration of continuity defects in the jaws (Clokie & Sàndor 2001).
Allogeneic bone is non-vital osseous tissue taken from one individual and transferred to another individual of the same species. There are three forms of allogeneic bone: fresh frozen, freeze-dried and demineralized bone matrix (DBM). Fresh frozen bone is rarely used today for the purposes of bony reconstruction in the cranio-maxillofacial skeleton because of concerns related to the transmission of viral diseases (Buchardt 1983). The risk of transmitting HIV with a properly screened demineralized freeze-dried bone allograft has been calculated to be 1 in 2.8 billion (Russo & Scarborough 1995). Bone harvested from a patient who died from AIDS related disease and was tested for the p24 core protein and reverse transcriptase and has been found to be positive. When this same bone was processed to make DBM, no evidence of either was found (Mellonig et al. 1992). It is therefore assumed that the process to make DBM eliminates or inactivates the p24 core protein and reverse transcriptase.
Freeze-dried allogeneic bone is processed to remove the moisture from the bone. This results in an implant with mechanical strength that can be used to onlay areas or as a crib to retain autogenous bone (Marx 1993). This implant, while osteoconductive, has no osteogenic or osteoinductive capabilities and consequently requires a source of osteocompetent cells. Therefore freeze-dried allogeneic implants are usually placed in conjunction with autogeneic grafts when reconstructing the cranio-maxillofacial skeleton.
By demineralizing the freeze-dried bone to create DBM, the implant loses its mechanical strength but may retain some osteoinductive properties (Urist 1965, Zhang et al. 1997a,b). Removal of the mineral component from the bone matrix may expose native proteins, such as bone morphogenetic protein (BMP). The potential osteoinductive capabilities of DBM make it a valuable tool for the surgeon.
Recent advances have seen DBM incorporated into various carriers such as collagen or selected polymers (Helm et al. 1997, Babush 1998, Morone & Boden 1998). These forms are either sponge-like or gel/putty-like in consistency. Putties are simple to apply and are well retained within the recipient tissue bed. These products could potentially be used in the treatment of periodontal infrabony defects, extraction sites to prevent ridge resorption, alveolar ridge reconstruction, bone reconstruction associated with dental implant placement, bone reconstruction associated with dental implant complications and cysts or bony defects of the jaws (Caplanis et al. 1997, Becker et al. 1998, Campbell 1998, Caplanis et al. 1998, Kim et al. 1998, Kumta et al. 1998, Parashis et al. 1998, Rosenberg & Rose 1998, Wiesen & Kitzis 1998). If larger volumes of bone are required, such as in maxillary sinus floor augmentation prior to dental implant placement, then DBM may be used as a bone graft expander to reduce the volume of bone graft required to fill an osseous defect (Blomqvist et al. 1998, Goldberg & Baer 1998, Stevenson 1998). This reduced graft volume may allow the use of a less morbid intra-oral harvest site. While reducing patient morbidity by potentially avoiding an extra-oral donor site, the major disadvantage of this technique is the cost of the DBM material.
Xenogeneic bone grafts consist of skeletal tissue that is harvested from one species and transferred to the recipient site of another species (van den Bogaerde & White 1997, Hammer et al. 1998). These grafts can be derived from mammalian bones and coral exoskeletons. Bovine derived bone has been commonly used (Block & Posner 1995, Jensen et al. 1996), even though other sources are such as porcine or murine bone are available. Xenogeneic bone was popular in the 1960"s but fell into disfavour due to reports of patients developing autoimmune diseases following bovine bone transplants (Pierson et al. 1968, Buchardt 1983). The re-introduction of these products in the 1990"s comes after the development of methods to deproteinate bone particles (Iwamoto et al. 1997). This processing reduces the antigenicity making these implants more tolerable to host tissues (Basle et al. 1998). The result is that the organic component of bone, referred to at the beginning of this chapter, is almost completely removed.
This inorganic bone matrix then has the structure of bone making it osteoconductive without the osteoinductive abilities imparted by the organic elements. Eventually xenogenic bone should be replaced by host tissue, which would make it useful for defect or extraction site filling in the alveolus prior to dental implant placement or prosthetic rehabilitation (Chappard et al. 1996, Berglundh & Lindhe 1997, Hurzeler et al. 1997, Merkx et al. 1997, Schmitt et al. 1997, Skoglund et al. 1997, Valentini et al. 1998). Resorption of bovine derived bone has been observed in animals studies (Merkx et al. 1997) but not consistently in human clinical trials (Hallman et al. 2001a, Valentin 1998, Skoglund 1997). Since the material is usually a powder it may require some form of retentive structure such as a membrane to keep the xenograft in the desired location (Avera et al. 1997, Zitzman et al. 1997, Hurzuler et al. 1998, Lorenzoni et al. 1998). While bovine xenografts may reduce morbidity by eliminating the donor site, their disadvantage is the concern with the possibility of future bovine spongiform encephalopathy due to potential slow virus transmission in bovine-derived products (Bons et al. 2002, Hunter 2002). Since other alternative biomaterials exist, bovine-derived products should probably be avoided until the concerns regarding potential slow virus transmission are clearly addressed.
One interesting xenogeneic transplant, Biocoral, is derived directly from the exoskeletons of corals from the Group Madrepora of the genus acropora (Guillemin et al. 1987). These corals are harvested from the relatively unpolluted waters of the reefs off New Caledonia, a point of importance since corals from contaminated waters can contain petrochemical impurities. Both solid blocks and particulated implants fashioned from this material are composed largely of calcium carbonate and are osteoconductive. When implanted, they are simultaneously incorporated into the human bony skeleton and replaced by human bone. The enzyme carbonic anhydrase, liberated by osteoclasts is responsible for the breakdown of this material. The time for total replacement of this implant by bone in the human craniofacial skeleton is approximately 18 months (Roux et al. 1988b). Since the use of coral-derived granules gives rise to bone with the material’s eventual replacement, it could decrease morbidity by avoiding a bone graft harvest donor site.
Alloplastic bone substitutes are synthetic substances that have been processed for clinical use in osseous regeneration. There are three types of alloplastic substances in clinical use today: hydroxyapatite, other ceramics and polymers.
Hydroxyapatite (HA) is a ceramic. HA can be divided into two groups depending upon its ability to resorb (Jarcho 1986, Alexander et al. 1987, Ricci et al. 1989, Brown & Constanz 1994). Some refer to the internal pore size as a means of differentiating between various types of hydroxyapatite (Holmes 1979, Guillemin et al. 1989, 1995). The porous form of HA allows rapid fibrovascular tissue ingrowth, which may stabilize the graft and help resist micromotion (Kenny et al. 1988, El Deeb & Holmes 1989). HA can be machined to many shapes or consistencies (Schliephake & Neukam 1991, Frayssinet et al. 1992, Marchac 1993). HA has several potential clinical applications including the filling of bony defects, the retention of alveolar ridge form following tooth extraction and as a bone expander when combined with autogenous bone during ridge augmentation and maxillary sinus floor augmentation procedures (Stoelinga et al. 1986, Bifano et al. 1998, Haas et al. 1998a,b, Simion et al. 1998). Although the use of HA can eliminate donor site morbidity, the tendency for granular migration and incomplete resorption has become a long-term problem (Rosen & McFarland 1990, Byrd et al. 1993, Mercier 1996, Prousaefs et al. 2002).
Apart from HA, there are three other types of ceramics: tricalcium phosphate (TCP), bioglasses, and calcium sulphate (Peltier 1961, Shafer & App 1971, Metsger et al. 1982, Hollinger & Batristone 1986, Kim et al. 1998). TCP is a similar to HA being a calcium phosphate with a different stoichiometric profile (Mors & Kaminski 1975, Hollinger et al. 1989). TCP has been formulated into pastes, particles or blocks, which have demonstrated an ability to be biocompatible and biodegradable (Naghara et al. 1992). Clinically the one disadvantage with TCP is its unpredictable rate of bioresorption. Its degradation has not always been associated with concomitant deposition of bone (Ogushi et al. 1991, Buser et al. 1998). Two products (Norian SRS®, Norian Corporation, Cupertino, California, USA and Bone Source®, Leibinger, Dallas, Texas, USA) have been used for the repair of cranial vault defects. Calcium salts are mixed with water to form a paste having an isothermic setting reaction and placed into the defect. Early versions of these materials tended to be easily washed out of the wound by haemorrhage. The materials tend to fracture and are resorbed unevenly in cranial vault defect studies (Clokie et al. 2002).
Bioactive glasses are silico-phosphate chains that been used in dentistry as restorative materials such as glass ionomer cement. These materials have the ability to chemically bond with bone and are supposed to function as small bone regenerative chambers (Ziffe et al. 1991, Merkx et al. 2003). Bioactive glasses may have osteoconductive properties and have been tested in animal trials (Turunen et al. 1997). Bioactive glasses have been used in the treatment of periodontal bony defects (Nasr et al. 2000, Yukna et al. 2001). In order to preserve the form of the alveolar ridge after tooth-loss, bioactive glass root replicates have been introduced (Ylimaz et al. 1998). While these are able to preserve the crestal width and height of the alveolus, they may impair the later placement of dental implants due to incomplete resorption.
Polymers by their nature can be fashioned in seemingly endless configurations (Barrows 1986, Shalaby 1988, Haas et al. 1998a). Combinations of polyglycolic acid (PGA) and polylactic acid (PLA) have been successfully used in the form of bioresorbable sutures for many years (Aderriotis & Sàndor 1999) and more recently as bioresorbable fixation materials (Suuronen et al. 1999, 2000). Giant cell reactions presented as a problem with earlier combinations of this material (Brekke 1995). As with bioglasses, root replicates have been introduced to preserve the form of the alveolar ridge after tooth-loss. These are made of PLA (Suhonen & Meyer 1996). The ability of PLA implants to preserve the crestal width and height is an advantage. Unfortunately because of incomplete resorption they may impair the later placement of dental implants (Suhonen & Meyer 1996). The future of bone regeneration could lie with this class of synthetic materials (Clokie & Sàndor 2001). These materials could be better utilized once their ability to resorb at variable rates, over set periods of time is better understood and an appreciation for their compatibility with the emerging bioactive agents is developed. The ideal would be a completely synthetic bioimplant, which is predictably degradable and is innately osteocompetent (Clokie & Sàndor 2001). Such synthetic materials could also play a very important role in tissue engineering (Vesala et al. 2002), serving as bioactive scaffolds.
One important advantage related to all xenogenic and allogenic materials is that they could potentially be used as bone graft expanders by mixing them with autogenous bone chips. This mixing could decrease the volume of autogenous bone graft needed, which in turn could convert an extra-oral harvesting procedure to an intra-oral harvesting procedure, potentially reducing donor site morbidity (Hallman et al. 2001a, Kainulainen et al. 2002a). However, data from clinical histology indicates that not all xenogenic and allogenic materials will be resorbed and replaced by autogenous bone with time (Hallman et al. 2001b, Merkx et al. 2003). This may leave the augmented bone with a composite rather than a homogenous structure, which could influence future dental implant survival (Merkx et al. 2003).
In fact Merkx et al. found after an extensive review of clinical reports, that autogenous bone without anorganic additives seemed to result in the greatest amount of bone in sinus floor augmentation after a four to six month healing period. Bovine bone material and HA seemed to result in the lowest amount of bone formed (Merkx et al. 2003).
An osteoactive agent is any material which has the ability to stimulate the deposition of bone (Clokie & Sàndor 2001). The phenomenon of osteoinduction was first described in the works of Urist and co-workers in (Urist & McLean 1952, Urist 1965, Kale & Di Cesare 1995). Bone matrix was shown to induce bone formation when implanted within muscle pouches of a number of different species of animals. Urist’s group identified a specific extract from bone, a protein now referred to as Bone Morphogenetic Protein (BMP), as that factor which caused the phenomenon (Urist et al. 1977, 1979, Mizutani & Urist 1982). Since then, many other entities have been found with a variety of effects on bone (Goldring & Goldring 1996). These may be classified as osteoinducers, osteopromotors or bioactive peptides (Hauschka et al. 1988).
The compounds in the first two categories are growth factors, a group of complex proteins of approximately 6 to 45 kilo Daltons which function to regulate normal physiological processes and biological activities such as receptor signalling, DNA synthesis, and cell proliferation (Wozney et al. 1988, Schliephake 2002). Growth factors that are referred to as cytokines have a lymphocytic origin, being nonantibody proteins released by one cell population on contact with a specific antigen and act as intracellular mediators. Other growth factors are described as morphogens. These are diffusible substances in embryonic tissues that influence the evolution and development of form, shape or growth. Still other growth factors are mitogens. They induce blast transformation by regulating DNA, RNA and protein synthesis (Kawamura & Urist 1988).
An example of the importance of such factors in cranial growth is the effect of fibroblast growth factors (FGF) and their receptors. Normal growth and morphogenesis of the cranial vault reflect a delicate balance between cell proliferation in the sutures of membranous bones and osteogenesis of the cranial bones (Moore et al. 2002). The discovery that mutations in FGF receptors cause the major craniosynostisis syndromes implicates FGF-mediated signalling in the skeletogenic differentiation of the cranial neural crest (Sarkar et al. 2001, Sándor et al. 2001). In fact blocking of endogenous FGF-2 activity prevents cranial vault osteogenesis (Moore et al. 2002), whereas mutant FGF receptors can induce chondrogenesis in neural crest cells, potentially perturbing this complex process of skeletogenesis (Petiot et al. 2002).
Bone morphogenetic protein (BMP) has been shown to have osteoinductive properties (Wozney 1989, Wozney et al. 1990). BMP is recognized to be part of a larger family of growth factors referred to as the TGF-β superfamily (Sampath et al. 1990) with a 30–40% homology in amino acid sequence with other members in the family. BMP acts as an extracellular molecule that can be classified as a morphogen as its action recapitulates embryonic bone formation. The identifying pattern of the BMP subfamily is their seven conserved cysteine residues in the carboxy-terminal portion of the protein and this is where the unique activity of BMP’s is thought to reside (Sampath et al. 1990).
Bovine & porcine sources were used in much of the original work attempting to purify the BMP molecule, a protein less than 50 kilo Daltons in size (Sampath & Reddi 1981, Besho et al. 1989, Rosen et al. 1989, Ko et al. 1990, Wang et al. 1990) and a number of recombinant human forms of BMP (rhBMP) have been derived. Interestingly the amount of human rhBMP necessary to produce bone induction in vivo is more than ten times higher than that of highly purified native bone extracted BMP (Tuominen 2001). This difference was also demonstrated between human BMP derived from human bone matrix and human rhBMP (Bessho et al. 1999), suggesting that native BMP is a combination of different BMP’s or represents a synergy between them (Wang et al. 1990). This has revived interest in xenogenic derived native BMP’s (Viljanen et al. 1997). Although concern regarding the immunigenicity of interspecies BMP has been raised in the literature, moose-derived BMP showed strong osteoinductive capacity and weak immunogenicity in a sheep study (Viljanen et al. 1996).
Large and small animals have been used to study the influence of BMP on bone regeneration (Nilsson et al. 1986, Yamazaki et al. 1988, Johnson et al. 1989, Nakahara et al. 1989). Critical sized osseous defects are defined as bony defects of a specific size, which will not heal spontaneously with bone tissue alone but with fibrous scar (Lindholm et al. 1988, Hollinger & Kleinschmidt 1990, Lindholm 1995). Bone lesions above a critical size become scarred rather than regenerated, leading to nonunion (Petite et al. 2000). BMP has demonstrated the ability to heal many different varieties of critical sized defects including cranial vault defects, long bone defects and mandibular continuity defects (Lindholm et al. 1988, Covey & Albright 1989, Johnson et al. 1990, Lindholm 1995) without the addition of a bone graft.
One of the challenges in the use of BMP is in its delivery to a site of action. As a morphogen BMP is rapidly absorbed into the surrounding tissues dissipating its effectiveness. Many different carrier vehicles have been used to deliver BMP including other noncollagenous proteins, DBM, collagen, HA, PLA and or PGA combinations, calcium carbonate, calcium sulphates and fibrin glue (Harakas 1984, Urist et al. 1984, Damien et al. 1993, Urist 1995, Ono et al. 1995, Davis & Sàndor, 1998, McCallister et al. 1998, Si et al. 1998, Lindholm 2002b). More recently biodegradable gels, collagen sponges impregnated with BMP and silica glass have been used as carriers (Boyne 1996, Howell et al. 1997a, Bostrom & Camacho 1998, Johnson & Urist 1998, Lindholm 2002a). DBM has been shown to contain BMP and may be used as a bone graft substitute with predictable healing in critical sized rabbit calvarial defects (Clokie et al. 2002) and has been used successfully in a human mandibular defect in vivo with native human BMP, a poloxamer carrier and bank bone (Moghadam et al. 2001).
The proteins in the family of transforming growth factor β (TGF-β ) should be considered as osteopromotors, agents, which enhance bone healing. TGF-β is found in the same supergene family as BMP. TGF-β has been shown to participate in all phases of bone healing (Celeste et al. 1990). During the initial inflammatory phase TGF-β is released from platelets and stimulates mesenchymal cell proliferation. It is chemotactic for bone forming cells, stimulating angiogenesis and limiting osteoclastic activity at the revascularization phase. Once bone healing enters osteogenesis then TGF-β increases osteoblast mitoses, regulating osteoblast function and increasing bone matrix synthesis, inhibiting type II collagen but promoting type I collagen. Finally, during remodelling it assists in bone cell turn-over (Mohan & Baylink 1991, Roberts & Sporn 1993, Miyazono et al. 1994, Cunningham et al. 1995). TGF-β has a biphasic effect, which suppresses proliferation and osteoblastic differentation at high concentrations (Schliephake 2002).
While less work has been undertaken to explore the applications of TGF-β than with BMP’s as an adjunct to bone healing, TGF-β may be more effective than BMP in those situations where enhanced bone healing is preferred to bone induction (Clokie & Sàndor 2001). Moreover, combinations of BMP and TGF-β , may enhance the osteoinductivity of an implant while, at the same time, making it osteopromotive. As with BMP, carrier vehicles for the delivery of TGF-β are under development.
Platelet-derived growth factor (PDGF) is angiogenic and is known to stimulate the reproduction and chemotaxis of connective tissue cells, matrix deposition (Singh et al. 1982, Antonaides & Williams 1983, Bowen-Pope et al. 1984, Ross et al. 1986). These properties are all crucial to bone healing.
Insulin-like growth factor (IGF) has demonstrated a capacity to increase bone cell mitoses and increase the deposition of matrix. PDGF and IGF have shown an ability to work together during the reparative stages of bone healing. PGDF-IGF impregnated devices have proven to increase bone healing in defects associated with dental implants and teeth (Giannobile et al. 1996, 1997, Howell et al. 1997b).
Platelets are known to contain a number of different growth factors of which TGF-β , and PDGF are two. As platelets degranulate they release these factors which may play a role in initiating graft healing. Platelet rich plasma (PRP) is one potential source of concentrated platelets that could be used in bone regeneration (Landesberg et al. 1998, Marx et al. 1998, Whitman & Berry 1998). A single unit of freshly harvested autologous blood is centrifuged at 5,600 rpm to separate the platelet poor plasma from the erythrocytes and the buffy coat (platelets and leukocytes). Once platelet poor plasma is removed, the specimen is further centrifuged at 2,400 rpm to separate the packed red blood cells from the PRP. The remaining PRP contains 500,000 to 1,000,000 platelets, which are mixed with a thrombin/calcium chloride (1,000units/10%) solution to form a gel (Marx et al. 1998). This gel can then be used in conjunction with bone regeneration materials such as HA or DBM as a source of autogeneic growth factors (Landesberg et al. 1998). When used in combination with autogenous bone, PRP is reported to increase the maturation rate of a bone graft up to 2 fold and also increase the bone density of the graft (Marx et al. 1998, 2002).
The last category of bioactive molecules is the polypeptide group. They may act as osteoinducers or osteoenhancers. Two short amino acids chain peptides that have demonstrated a bone activity are known as P-15 and OSA-117MV. The P-15 polypeptide was designed to take advantage of a conformational arrangement known as the "beta bend", which was found to have an influence on bone induction and growth when utilized in some in vitro studies (Qian & Bhatnager 1996, Yukna et al. 1998). The OSA molecule is even smaller than P-15 and was discovered in relation to the treatment of osteoporosis where OSA"s effect is concentrated in areas of high stress. Researches have started to explore the local effects of this peptide and initial reports (Clokie & Sàndor 2001) suggest that it may enhance the osteoinductive effect of demineralized bone matrix.
The area of tissue engineering has brought to the forefront, the possibilities of hybrids of biomaterials seeded with osteocompetent cells to be used as an implant. The “hybrid graft” could consist of a porous matrix, on which bone marrow cells could grow (Petite et al. 1995).
The use of bone marrow as the source of cells is logical as bone marrow contains stem cells which have the potential to differentiate along various pathways and lines, including the direction of bone producing osteocompetent cells (Friedenstein 1976, Owen 1985, Triffitt 1987, Beresford 1989, Friedenstein et al. 1996). Seeding a porous matrix with bone marrow cells could enhance the osteogenic potential of the matrix as a hybrid. Another possibility is the tissue culturing of bone marrow cells to further expand their numbers (Petite et al. 1995). Bone marrow derived cells are responsive to the influence of dexamethasone and 1, 25 dihydroxycholecalciferol (Leboy et al. 1991, Petite et al. 1995) and can be influenced to differentiate in the direction of bone cells. Human bone marrow cells have been reported to adhere to porous coral matrices (Petite et al. 1995, 2000) and to matrices made of HA and TCP (Ohgushi et al. 1989a,b, 1991, Bernard & Picha 1991). A coral scaffold together, with in vitro-expanded marrow stromal cells have been used as tissue-engineered artificial bone. This artificial bone has been used to treat a large segmental long bone defects in the murine model with morphogenesis leading to complete recorticalization and the formation of a medullary canal (Petite et al. 2000). Osseous cells could also be combined with such matrices, making hybrid grafts. The source of bone cells could be suction trap harvested bone (Kainulainen et al. 2002c, Lindholm et al. 2002). In the case of suction trap harvested bone cells, future hybrid grafts for the same individual could be made at the time of harvesting, or from the same harvested, but stored frozen cells, at a later date (Lindholm et al. 2002). The development of such hybrids, the culturing of bone cells and improvements in cell storage methods may be the way of the future and could also diminish donor site morbidity by the elimination of the donor site.