Metallic Scaffolds for Bone Regeneration

Bone tissue engineering is an emerging interdisciplinary field in Science, combining expertise in medicine, material science and biomechanics. Hard tissue engineering research is focused mainly in two areas, osteo and dental clinical applications. There is a lot of exciting research being performed worldwide in developing novel scaffolds for tissue engineering. Although, nowadays the majority of the research effort is in the development of scaffolds for non-load bearing applications, primarily using soft natural or synthetic polymers or natural scaffolds for soft tissue engineering; metallic scaffolds aimed for hard tissue engineering have been also the subject of in vitro and in vivo research and industrial development. In this article, descriptions of the different manufacturing technologies available to fabricate metallic scaffolds and a compilation of the reported biocompatibility of the currently developed metallic scaffolds have been performed. Finally, we highlight the positive aspects and the remaining problems that will drive future research in metallic constructs aimed for the reconstruction and repair of bone.


Introduction
Human skeletal tissues have complex three-dimensional (3-D) geometries and highly organized internal architectures, which cannot be simply emulated by cells maintained in two-dimensions. Bone is a complex porous composite structure with specific characteristics such as viscoelasticity and anisotropy, both in morphology and mechanical properties [1]. The unique mechanical performance of natural bone is characterized by high toughness, high specific strength, and low stiffness. Porous scaffolds are central to hard tissue engineering strategies because they provide a 3-D framework for delivering reparative cells or regenerative factors in an organized manner to repair or regenerate damaged tissues. Since hard tissues are responsible for the body mechanical stability, materials aimed for repairing, substitution and/or restoration of hard tissues must possess strength, resistance to corrosion/degradation, have a good biocompatibility and exhibit good wear resistance.
The development of successful scaffolds for bone tissue engineering requires a concurrent engineering approach that combines different research fields. During the last three decades, researchers have tailored metallic scaffolds that are useful for a wide variety of medical and dental applications. Surface modification of already proved biocompatible metals is an essential requisite for the utilization to tissue engineering because the metal surface must be controlled to induce the adhesion and proliferation of cells and the adsorption of essential biomolecules.
In this literature review, we will summarize the progress and the state-of-the-art of the metallic scaffolds as well as the reported biocompatibility of each of these metallic structures that has been conceived to be used in specific reconstruction of small or large bone defects. The design of a hard tissue-engineered scaffold logically begins with an intensive characterization of the host tissue properties. The properties of bone and how these apply to the design of a synthetic scaffold are discussed below.

Bone Structure and Properties
Bone is a natural composite material, which by weight contains about 45-60% minerals, 20-30% matrix, and 10-20% water. By including the water fraction in the organic phase, the composition of bone can be simply represented as shown in Figure 1. The matrix is the organic component, which is primarily composed of the protein Type I collagen [2]. Type I collagen is a triple helix that is highly aligned, yielding a very anisotropic structure. The non-collagenous proteins are composed of noncollagenous glycoproteins and bone specific proteoglycans, these proteins include osteocalcin, osteonectin, bone phosphoproteins, bone sialoproteins and small proteoglycans [2]. The noncollagenous proteins have different functions in the regulation of bone mineralization and cell-tomatrix binding interactions with structural proteins. Less than 1% of the non-collagenous proteins contain growth factors influencing the cells but also secreted by them [3]. The cellular component is made of osteoblasts (bone-forming cells), osteoclasts (bone-destroying cells), osteocytes (bonemaintaining cells, which are inactive osteoblasts trapped in the extracelullar matrix) and bone lining cells (inactive cells that are believed to be osteoblasts precursors) [4]. The mineral, inorganic component of bone is a calcium phosphate known as Hydroxyapatite (HA). Hydroxyapatite has a chemical structure represented by the formaula Ca 10 [PO 4 ] 6 [OH] 2 and is present in small crystallites form of approximately 2 × 2 × 40 nm 3 . These crystals undergo important changes in composition with age, thus their biologic functions depend on the amount and the age of the mineral crystals [5]. The inorganic matrix performs two essential functions as an ion reservoir and a structure giving the bone its stiffness and strength. In simple words, the organic matrix provides bone its flexibility and the inorganic material is predominantly responsible for the mechanical properties of bone [6,7].
The human skeleton can be categorized into two types of bone: the cortical bone and the trabecular bone. Although both bone types comprise the same composition, each one contains different proportions of the organic and inorganic materials, degree of porosity and organization. In addition, the combination of cortical and trabecular bone varies according the skeleton regions, which is dependent on the applied mechanical loading. Both, cortical and trabecular bones display timedependent mechanical behavior, as well as damage susceptibility during cyclic loading [8,9].
Despite the multiple functions bone has in the body, its biomechanical role is the most compromised upon injury. Indeed, the other bones in the body can compensate for the injured bone's metabolic function, but if a bone is broken or injured, it can no longer support the load it is meant for, and the body remains handicapped. The mechanical properties of cortical bone have been well documented [10][11][12][13]. They can be measured via traditional testing techniques such as: uniaxial compressive or tensile testing, or three or four-point bending. Cortical bone exhibits a high degree of anisotropy and values of mechanical properties vary between animal species, bone location and testing conditions, age and disease. Testing conditions, for example, may vary between testing dry samples, testing wet samples at 37 C and embedding them in an special resin or not.
Measuring properties of trabecular bone is far more complex than in the case of cortical bone as shown in Table 1. The complexity is due to the small dimensions of the individual trabeculae. When considered mechanically cortical and trabecular bone are not the same material. It is speculated that differences in moduli between cortical and trabecular bone are entirely due to the bone density. The range of cortical bone densities reported for the human proximal femur is 1.5-2 g/cm 3 [14] and the range of apparent trabecular bone density in human proximal femora is 0.2-0.6 g/cm 3 [15]. With either testing technique the mean trabecular Young's modulus is found to be significantly less than that of cortical bone. However, as can be seen in Table 1, some authors have found a value of elastic modulus of trabecular bone as high as those for cortical bone apparently because the test specimens were dried before the mechanical tests [12,16,18,19].
Mechanical properties of human bone depend dramatically on age; 3, 5, and 35-years old femoral specimens had a Young's modulus of 7.0, 12.8 and 16.7 GPa, respectively [20]. Besides age, the nutritional state, physical activity (mechanical loading), bone related diseases, etc., will influence the properties of bone tissue. This fact establishes a major challenge in the design and fabrication of scaffolds aimed to repair specific sites in specific patients. Compiled from references [10][11][12][13]17,19,21].

Bone Tissue Engineering
The goal of bone tissue engineering is to repair bone defects, which are difficult or even impossible to treat by conventional methods. This usually involves the use of 3-D bone graft substitutes to treat bone losses due to traumatic injury or revision surgery to augment the natural regenerative capacity of the body [22,23]. Bone tissue engineering employs a multidisciplinary approach, drawing on the principles of cell biology, molecular development biology, materials science and biomechanics, to aid in the repair of tissues damaged beyond the natural healing capacity of the bone. There are several approaches to bone engineering, ranging from inorganic bone fillers (in common clinical use) [24] to in situ bone induction by bone-inductive growth factors (in limited use) [25,26] to laboratory cultured bone cells and gene therapy (in experimental phase) [27][28][29]. All these methods, however, have two common requirements: a physical continuity across the damage site that has to be provided to guide the bone growth, and the avoiding of scar formation.
In general, three essential elements are needed to successfully engineer a biological tissue or organ: 1) Tissue forming cells (osteogenic cells) and/or signaling biomolecules 2) Biocompatible scaffolds conducive to normal cell functions, and 3) Quantitative measures of tissue's regenerative outcome.
An ideal strategy for the tissue engineering of bone is the harvesting of osteogenic cells from the patient, which are then expanded in culture and seeded on a scaffold or graft that act as a guide and stimulus for tissue ingrowth in 3-D ( Figure 2). Ideally, the need to regenerate tissue can be forecasted in advance, and cells taken from the patient can be seeded onto a scaffold, grown in vitro, and then reimplanted back into the patient, resulting in a healing of the damaged tissue. In a tissue-engineered scaffold, mesenchymal stem cells (MSCs) are usually included to give rise to bone cells. These stem cells can be readily extracted from the bone marrow of adult mammals (including humans), and can be induced to differentiate into natural tissue. The scaffold material can be preseeded in vitro with osteogenic cells to promote bone formation. At the implant site these cell/scaffold constructs contribute to bone formation. The role of the scaffold is to act as a carrier that restricts the movement of these MSCs cells away from the implantation site and to provide support for new bone formation. Figure 3 schematically illustrates the cell-based strategy for tissue regeneration. The osteogenic cells lay down bone extracellular matrix in the surface of the scaffold as woven immature bone. Over time, a mature bone structure will form inside and in the exterior part of the scaffold allowing the regeneration of the tissue. Growth factors such as basic fibroblasts growth factors (FGFs), platelet-derived growth factor (PDGF), insulin-like growth factor (IGF), epidermal growth factor (EPG), transforming growth factorbeta (TGF-), bone morphogenetic proteins (BMPs), etc., also would be applied in the tissue engineered scaffolds to promote bone formation. When the scaffold material is loaded with specific bone-inductive growth factors, these exogenous growth factors are then released at the implantation site, where they can act upon locally resident cells as well as recruiting other more distant cells to form new bone tissue. Bone morphogenetic proteins are active bone-inducing factors that act on immature mesenchymal cells, including osteoblasts, resulting in osteogenesis [30]. To date, molecular cloning has isolated several types of BMPs, and recombinant BMP molecules have been synthesized [31]. BMPs 2, 4, 6, and 7 are generally considered to be the most osteoconductive of the bone morphogenetic proteins. BMP-2, specifically promotes undifferentiated mesenchymal cells into osteoblasts, leading to bone formation [32].
While these factors place special demands on all aspects of tissue engineering, scaffold design takes on a role of particular importance. We will discuss this topic separately in the following section.
Finally, postoperatively high quality image examinations are required to investigate the effectiveness of the implantation such as the position of the scaffold and evaluate the developing status of surrounding anatomic structures. For the clinical determination of the bone ingrowth inside the scaffold recently advances of the X-ray micro-computed tomography (CT) imaging have shown sufficient resolution for the accurate identification of the bone ingrowth within the metallic porous structure. However, the complex process of bone remodeling inside a tissue-engineered construct, made up of scaffold material, host bone, mineralized bone and soft tissue, makes the partitioning of the tomogram into discrete phases non-trivial [33]. In the past, CT was not suitable for metallic scaffolds as the metal heavily attenuates X-rays. The presence of metal resulted in dark and bright grainy artifacts, which obscure important details of the scan images [34]. However, improved algorithms for metal artifact reduction has been developed recently [35,36], and the combination of 2 mm thick aluminum filter and a 10 mm thick polymethylmethacrylate filter has been employed improving the signal-to-noise ratio in the images. By doing this it can be reduced the streak artifacts caused by the metallic material [37].

General Desirable Properties of the Bone Scaffolds
A scaffold for hard tissue reconstruction is a three dimensional construct, which is used as a support structure allowing the tissues/cells to adhere, proliferate and differentiate to form a healthy bone/tissue for restoring the functionality. In almost all the clinical cases, scaffolds for hard tissue repair in a load-bearing area are not temporary, but permanent. They most retain their shape, strength and biological integrity through the process of regeneration/repair of the damaged bone tissue. Bone replacement constructs for bone defects reconstructions would need to be biocompatible with surrounding tissue, radiolucent, easily shaped or molded to fit perfectly into the bone defect, nonallergic and non-carcinogenic, strong enough to endure trauma, stable over time, able to maintain its volume and osteoconductive (able to support bone growth and encourage the ingrowth of surrounding bone) [38][39][40][41][42].
Apart from the above-mentioned material requirements, the structural requirements expected for the possible candidate for bone scaffold are numerous, ranging from the maximum feasible porosity to the porous architecture itself. Pore size and interconnectivity are important in that they can affect how much cells can penetrate and grow into the scaffold and what quantity of materials and nutrients can be transported into and out of the scaffold. In other words, pore size distributions, porosity and the interconnectivity of the scaffold should be sufficient for cell seeding, cell migration, matrix deposition, vascularization and mass transport of nutrients from and to the cells. Physiologically, previous research has shown that the optimum pore size for promoting bone ingrowth is in the range of 100-500 m [43][44][45]. However, the scientific community has not reached yet a consensus regarding the optimal pore size for bone ingrowth.
From a mechanical perspective, scaffold materials aimed for the repair of structural tissues should provide mechanical support in order to preserve tissue volume and ultimately to facilitate tissue regeneration. The most critical mechanical properties to be matched by the scaffold are bone loading stiffness, strength and fatigue strength. When the scaffold's stiffness exceeds that of natural bone, stress concentration in the surrounding bone can cause bone failure. Conversely, when the scaffold's stiffness is less than that of natural bone, stress concentration in the scaffold can cause implant failure as well as bone atrophy. This effect of stiffness mismatch, which gives rise to uneven load sharing between bone and implant, is known as stress shielding [46]. Stress shielding affects the bone remodeling and healing process. The underloaded bone adapts to the low stress environment and becomes less dense and consequently weak. In addition to matching bone stiffness, the scaffold should also match or exceed the strength of natural bone. The scaffold must resist physiological forces within the implantation site and should have sufficient strength and stiffness to function for a period until in vivo tissue ingrowth has filled the scaffold matrix. An equal or excess strength ensures that the scaffold has equivalent or better load bearing capabilities than natural bone. For last, for a nonresorbable scaffold, it is very important to consider the fatigue strength, since the scaffold will be exposed to cyclic loading during the rest of the patient's life. Complete design of the scaffold must take into account both the mechanical considerations and the biological requirements to produce a globally optimized structure with an adequate chemical composition able to allow the subsequent ingrowth of bone.
In the scaffold design, surface properties including: topography, surface energy, chemical composition, surface wettability, surface bioactivity, etc., must all be considered, taking into account that in a complex porous 3-D scaffold the surface is not just the outside surface, but also the internal 3-D surfaces. For example, the modification of scaffolds materials with bioactive molecules is a technique to tailor the scaffold bioactivity. In addition, reduction of micromotion can be obtained by appropriately tailoring the material surface of the scaffold. The development of the required interface is not only highly influenced by surface chemistry, but also more specifically by nanometer and micrometer scale topographies. The surface roughness is found to influence the cell morphology and growth. It has been proved that alteration in surface topography by physical placement of grooves and depressions changes the cell orientation and attachment [47][48]. In general, smooth surfaces exhibit less cell adhesion than rough surfaces. On the other hand, surface porosity is another important factor in bone replacement [49][50]. It has been reported that materials coated with a porous surface exhibit less fibrous capsule formation than bulk or non-porous materials [51].
Surface modifications, such as, immobilization of biofunctional polymers and biopolymers, calcium phosphate ceramic coatings, hybridization with biocompatible and essential biomolecules are needed to achieve the required tissue induction properties. Countless procedures have been developed to modify the surface of biomaterials. Table 2 shows an overview of the surface modification methods available for titanium and its alloys. It has been widely demonstrated that surface treatment of titanium and its alloys has a critical influence on biocompatibility.  On the other hand, to program scaffolds with biological structures, cells and growth factors need to be integrated into the scaffold fabrication for bone tissue engineering, so that the bioactive molecules can be released from the scaffold in order to stimulate or modulate new tissue formation. Through surface modifications the metallic scaffold surface can be tailored to improve the adhesion of cells and adsorption of biomolecules in order to stimulate the bone formation and to facilitate faster healing [53]. Currently, a significant research effort is aimed at the biochemical modification of metallic surfaces. Table 3 presents some of the biochemical methods for surface modification of scaffolds introduced in recent years. The goal of the biochemical surface modification is to immobilize proteins, enzymes or peptides on biomaterials for the purpose of inducing specific cell and tissue responses, or in other words, to control the tissue-scaffold interface with molecules delivered directly to the interface. Nowadays, for the regeneration of structural tissues the bone tissue engineering is focusing in the development of a common framework for designing and building porous structures having both materials and biological components. In summary, the hybridization with active biofactors (cells, genes and/or proteins), the chemical composition, and the topography (structure, morphology) of the scaffold surface are known to be extremely important in bone replacement, since they regulate the type and degree of the interactions that take place at the interface: adsorption of ions and biomolecules such as proteins; formation of calcium phosphate layers; or interaction with different type of cells (macrophages, bone marrow cells, osteoblasts, etc.) [54,55]. For this reason, in recent years there has been a lot of research effort aimed at optimizing and controlling surface properties of the bone scaffolds with a view to customizing a certain material for the required application. Equally important for the success of the loading-bearing scaffold is the postoperative stability. The scaffold and the surrounding bone must be tightly apposed to ensure osteointegration. In order to achieve stability over the time the scaffold must fulfill each patient anatomical requirement. Anatomically, the external geometry and size of the scaffold should be the same as those of the tissue defect in order for the scaffold to fit and anchor into the defect site. Computed-aided tissue engineering enables the application of advanced computer aided technologies and biomechanical engineering principles to derive systematic solutions for the designing and fabrication of patientspecific scaffold [56].
Finally, the scaffold for bone repair should be easy to manufacture with highly consistent pore sizes, pore distribution, pore density and interconnectivity with a narrow size distribution range of the structural parameters over the entire volume of the scaffold [57]. And for last, the scaffold must withstand sterilization procedures without loss of properties and have an acceptable shelf-life.

Currently Used Metallic Scaffolds Materials and Their Limitations
To date there are several biocompatible metallic materials that are frequently used as implanting materials in dental and orthopedic surgery to replace damaged bone or to provide support for healing bones or bone defects. Standard surgical implant materials include stainless steel 316 L (ASTM F138), Co based alloys (mainly ASTM F75, and ASTM F799) and titanium alloys; where Ti-6Al-4V (ASTM F67 and F136) are the most employed. However, the main disadvantage of metallic biomaterials is their lack of biological recognition on the material surface. To overcome this restraint, surface coating or surface modification presents a way to preserve the mechanical properties of established biocompatible metals improving the surface biocompatibility. Moreover, in order to enhance communication between cells, facilitating their organization within the porous scaffold; it is desired to integrate cell-recognizable ligands and signaling growth factors on the surface of the scaffolds. Indeed, biofactors that influence cell proliferation, differentiation, migration, morphologies and gene expression can be incorporated in the scaffold design and fabrication to enhance cell growth rate and direct cell functions [58]. Another limitation of the current metallic biomaterials is the possible release of toxic metallic ions and/or particles through corrosion or wear possible that lead to inflammatory cascades and allergic reactions, which reduce the biocompatibility and cause tissue loss [59]. A proper treatment of the material surface may help to avoid this problem and create a direct bonding with the tissue.
On the other hand, depending on the materials properties, some metallic materials are too weak to be arranged into the desired architecture with a controlled porous structure and some metals are too stiff and would fracture when arranged into certain architectures. Each metallic material possesses different processing requirements and the degree of processability of each metal to form a scaffold is variable also.

Tantalum
Porous tantalum is a biomaterial with a unique set of physical and mechanical properties. It has a high-volume porosity (>80%) with fully interconnected pores to allow secure and rapid bone ingrowth [60]. In addition, it has a modulus of elasticity similar to that of bone, which minimizes stressshielding. Porous tantalum is a structural material and has sufficient strength to allow physiological load-carrying applications and represents an alternative metal for primary and revision total knee arthroplasty (TKA) with several unique properties. Bobyn and coworkers [60,61] presented basic scientific data that lend support for the use of this material, which is a trabecular metal composed of a carbon substrate that has elemental tantalum deposited on the surface. This trabecular metal has been shown to be highly biocompatible in several animal models [60,61,133]. Studies have demonstrated substantial cortical bone ingrowth between the trabecular network as well as high levels of bone growth onto the scaffold itself. Initial stability of the trabecular metal itself is also higher than that of standard materials, such as cobalt chrome. Furthermore, this new material offers better osteoconduction than other technologies used for biological fixation. Although porous tantalum is in its early stages of evolution, the initial clinical data [135][136][137][138] and preclinical studies [178][179][180][181][182][183][184][185] support its use as an alternative to traditional orthopedic implant materials.

Magnesium
The use of magnesium and its alloys for surgical applications is of particular interest. These alloys have great potential, and it has been shown that they are fully bioresorbable, have mechanical properties aligned to bone, induce no inflammatory or systemic response, are osteoconductive, encourage bone growth, and have a role in cell attachment [62]. Furthermore, because of its biodegradability, the second surgery for the removal of the scaffold might be avoided. All these facts suggest that Mg has significant potential as a load-bearing biomaterial. Indeed, there is a renewed interest in the use of this material in biomedical applications, e.g. for coronary stents [63,64], and more recently, researchers have concentrated on the application of magnesium-rare-earth alloys with new elemental contributions of cerium, neodymium and praseodymium for bone fixation devices [65,66] for osteo-applications. Recently, Mg-Ca alloys have been also produced and evaluated in vitro and in vivo as biodegradable biomaterials for orthopedic applications [67]. However, concerns over the toxicity of dissolved Mg have been raised, but it has been shown that the excess of magnesium is efficiently excreted from the body in urine [68]. In addition, concern does remain over the use of pure Mg as the dissolution rate in physiological conditions is rapid, potentially leading to hyper-magnesia, although a number of potential routes to controlling the corrosion rate have been proposed; especially providing it with a ceramic coating [69], titanium coating [70] or through the use of Mg alloys, including AZ31, AZ91, WE43, LAE442 and Mg-Mn-Zn alloys [65,66,71]. Although limited long-term survival data is available for Mg or Mg alloys porous scaffolds, the material seems promising for certain bone ingrowth applications such as trabecular bone regeneration.

Titanium and Titanium Alloys
Titanium is found to be well tolerated and nearly an inert material in the human body environment. In an optimal situation titanium is capable of osseointegration with bone [72]. In addition, titanium forms a very stable passive layer of TiO 2 on its surface and provides superior biocompatibility. Even if the passive layer is damaged, the layer is immediately rebuilt. In the case of titanium, the nature of the oxide film that protects the metal substrate from corrosion is of particular importance and its physicochemical properties such as crystallinity, impurity segregation, etc., have been found to be quite relevant. Titanium alloys show superior biocompatibility when compared to the stainless steels and Cr-Co alloys. Titanium-aluminum-vanadium alloys (ASTM F136, ASTM F1108 and ASTM F1472) have better mechanical properties than commercially pure titanium (cp Ti) (ASTM F67) and are used more widely in total joint implants. However, concerns have been expressed about the presence of long-term Ti-6Al-4V implants, because elements such as vanadium are toxic in the elemental state. These concerns have led to the development of new beta titanium alloys with nontoxic alloying elements like Ta, Nb, Zr [73]. Other currently available titanium alloys include ASTM F1295 (wrought Ti-6Al-7Nb alloy), ASTM F1713 (wrought Ti-13Nb-13Zr alloy), ASTM F1813 (wrought Ti-12Mo-6Zr-2Fe alloy) and ASTM F2066 (wrought Ti-15Mo alloy) and Ti-5Al-2.5Fe (ISO 5832-10). Further biocompatibility enhancement and lower modulus has been achieved through the introduction of second generation titanium orthopedic alloys including Ti-15Mo-5Zr-3Al, Ti-15Zr-4Nb-2Ta-0.2Pd, Ti-12Mo-6Zr-2Fe, Ti-15Mo-3Nb-3O and Ti-29Nb-13Ta-4.6Zr. This new generation of Ti alloys is at present under development and investigation, and it does not seem to be commercialized yet. In general, porous titanium and titanium alloys exhibit good biocompatibility. Bioactive titanium meshes have been successfully used in spine fusion surgery for the past two decades [74]. The titanium mesh cage contoured into cylindrical shape has been used successfully for anterior lumbar interbody fusion (ALIF) for more than 15 years in surgery. Titanium mesh cages were also used with autografts for bone grafting in spinal fusion. This is restricted by factors such as complications and second site morbidity [74]. One method to overcome this problem is the use of hydroxyapatite to provide the necessary bioactivity to the titanium mesh cage with a porous network to facilitate osteoconduction [196,199]. Moreover, despite the great advances in complete tissue engineered oral and maxillofacial structures, the current gold standard for load bearing defect sites such as mandible, maxilla and craniofacial reconstruction remains titanium meshes and titanium 3-D scaffolds. On the other hand, Ti and its alloys are not ferromagnetic and do not cause harm to the patient in magnetic resonance imaging (MRI) units. Titanium osseointegration can be potentially improved by loading the scaffold with specific growth factors. In applications where there are existing gaps, such as craniofacial reconstruction or augmentation of bone or peri-implant defects, increased regeneration of bone, often has been accomplished with delivery of TGF- and BMP-2 via titanium scaffold [30,75]. The latter growth factors are capable to elicit specific cellular responses leading to rapid new tissue formation. Stem cells have also been cultured in vitro onto titanium scaffolds [76] to induce the formation of calcified nodules in order to increase the production of mineralized extracellular matrix (ECM) onto the cells/scaffold constructs.

Nickel-Titanium Alloy (Nitinol)
Nitinol is one of the most promising titanium implants that find various applications as it possesses a mixture of novel properties, even in a porous state, such as shape memory effect (SME), enhanced biocompatibility, superplasticity, and high damping properties [77,78]. Since the elastic modulus of the Nitinol foams (~2.3 GPa) and the compressive strength (~ 208 MPa) are close to that of the bone and due its good biocompatibility porous NiTi have been used in making intramedullary nails and spinal intervertebral spacers used in the treatment of scoliosis [79]. Extensive in vivo testing and preclinical experience indicates that Nitinol is highly biocompatible, more than stainless steels [79,80]. Moreover, good biocompatibility on surface modified NiTi has been reported [81][82][83][84]. The demonstrated biocompatibility of Nitinol, its physical properties and SME, suggest that this alloy may offer substantial gains in the orthopedic field. These gains revolve around creating scaffolds that change shape after implantation due to the SME of Nitinol that can be initiated at the temperature of the human body. However, there is a problem of allergy and toxicity for NiTi alloys associated with the release of Ni ions. The concern of Ni toxicity and potential carcinogenicity has limited the use of NiTi alloys in Europe and the USA. In order to overcome this problem, surface modifications such as oxidation treatment of NiTi to obtain a Ni-free surface [85] and several alternative Ni-free shape memory alloys, mainly Nb-based, are currently under development although their long-term biological performance will have to be assessed in the future [86].

Hybrid Materials
Hybrid materials are those in which more than one class of material is employed in the scaffold. Today there are many different types of materials combinations principally used in artificial joints and bone implants. Many combinations of materials and surface modifications are aimed to stimulate specific responses at the molecular level. The synergistic combination of two types of materials may produce new structures that possess novel properties.
Common material combinations are synthetic polymer with bio-ceramic and synthetic/natural polymers with metals. Novel metal-ceramic-polymer hybrid materials have also been proposed for the fabrication of load-bearing scaffolds. In many clinical cases, composite scaffolds may prove necessary for reconstruction of structural diseases and bone defects. Nevertheless, the mechanical property requirements for hard tissue repair are difficult to satisfy using porous polymer/ceramic composites. Particularly, scaffolds based on HA or tricalcium phosphates (TCP) are very stiff, maybe brittle and may have different viscoelastic properties from bone [87]. To assure the mechanical integrity, hybrid constructs of porous Ti/TCP ceramic and cells have been tried and have demonstrated better osteogenic properties compared with Ti scaffold alone after implantation in goats [150]. Porous Ti is usually combined with bone inductive materials or cells, which endow the osteoinductive property leading to a rapid bone healing.

3-D Metallic Scaffolds Fabrication Technologies
Numerous fabrication techniques have been developed for the production of 3-D metallic scaffolds of high porosity and surface area for load-bearing applications. The basic goal of the available manufacturing techniques is to produce a micro-architecture in a scaffold that is highly porous to allow for cell adhesion, vascularization and nutrient flow. Mechanical considerations however, limit the range of porosities at the optimum pore size that can be employed to produce functional structures. Strength and ductility of porous structures are very sensitive to final density, pore size, material type, and fabrication parameters. Metallic scaffolds can be produced in a variety of ways, using conventional techniques or advanced processing methods. The choice of the technique depends on the requirements of the final application. Selection of the scaffold material and design, the method by which to construct them, and the possible additional surface modification are important to the success of using the scaffold to regenerate new bone.

Conventional Fabrication Methods
Conventional methods for manufacturing metallic scaffolds include sintered metal powders [88], sintered metal fibres [89], space-holder method [90], replication of polymeric sponge [91], fiber meshes and fiber bonding [92], self propagating high temperature synthesis (SHS) [93], spark plasma sintering (SPS) [94] or field assisted consolidation technique (FAST) [95], gas injection into the metal melt [96], decomposition of foaming agents [97][98][99], templated vapor deposition [60] and solid-state foaming by expansion of argon-filled pores [100]. However, there are inherent limitations in these processing methods, which offer little capability to control precisely pore size, pore geometry, pore interconnectivity, spatial distribution of pores, porosity, etc. As a result, there are really few manufacturing technologies capable of producing porous structures that possess the majority of the desired requirements. Moreover, the manufacturing of porous titanium and its alloys is associated with some difficulties; most notably the extreme chemical affinity of liquid titanium to atmospheric gases such as oxygen, hydrogen, and nitrogen, which eventually leads to strongly reduced ductility [101]. Table 4 shows a comparison between the different conventional fabrication methods that have been applied to produce metallic porous structures. ( =20-90%) Fiber meshes sintering [92] (  90%) Fiber bonding [173] (  70%)
The porous structure of the closed-cell structures is equiaxed and pores are surrounded by a metallic wall. In contrast, open-cell structures incorporate interconnected pores. Porous metals with elongated pores aligned in one direction-lotus structures-have recently been described also [96,98]. Scaffolds fabricated using conventional technologies have been employed clinically. Sintered bead coatings have been developed commercially using cobalt chrome and titanium alloys and have been shown to produce a durable biological bond that may last over ten years post implantation [103][104][105][106]. Diffusion bonded fiber-mesh porous structures have also been shown to successfully promote long-term implant fixation [107][108][109][110]. However, the maximum porosities attainable using these technologies is less than 50% at the required 100-700 m pore sizes [111].

Rapid Prototyping (RP) Technology
In early 1980s, rapid prototyping technology emerged in the hi-tech manufacture industry [112]. Since this technique can fabricate products with complex structure and individuation at a small batch, it can be realized in one design-manufacture process with high flexibility. Products with different shapes can be obtained by only modifying the computer-aided design (CAD) model using 3-D tomography data or magnetic resonance imaging (MRI) data, shortening the production cycle. The digital information is then converted to a machine specific cross-sectional format, expressing the model as a series of layers. The file is then implemented on the RP machine, which builds customer designed 3D objects by layered manufacturing strategy. Each layer represents the shape of the crosssection of the model at a specific level. Conventional manufacturing methods [111] are either difficult to employ or are unsuccessful in producing such porous devices with complex structure with the tight constraints of porosity, optimum pore size, or mechanical strength that are required. The drawbacks of the traditional methodologies for producing porous constructs include long fabrication periods, laborintensive processes, incomplete removal of residual chemicals or volatile porogenic elements, poor repeatability, irregularly shaped pores, insufficient interconnectivity of pores and thin wall structures, etc. RP techniques, also variously called solid free-form fabrication (SFF) or rapid manufacturing (RM), are considered a viable alternative for achieving extensive and detailed control over the scaffold architecture, shape and interconnectivity [113].
RP systems can also be utilized to produce a sacrificial mould to fabricate scaffolds. The multistep method involves casting of material in a mold and then removing or sacrificing the mold to obtain the final scaffold. Another important biological requirement is the surface properties of the fabricated scaffold. The topography of rapid prototyped surfaces can be further modified by sandblasting, shotpeening, vibratory deburring, spark anodization, electropolishing, acid etching, etc. Taking advantage of the possibilities of RP techniques load-bearing scaffolds with any predesigned structure and mechanical properties can be produced; so that they mimic the properties of the native bone and possess suitable strength for the intended application. Furthermore, CAD enables computational modeling and finite element analysis (FEA) prior to fabrication. Fluid flow analysis or stress distribution profiles can be obtained from computational models, thus allowing for re-design and scaffold optimization with minimal effort.
Until now RP developments mainly focused on polymer and ceramic materials [114]. However, recently several investigations have been carried out in order to produce 3-D porous metallic scaffolds using the RP route from 3-D solid models produced in CAD. For example, Li et al. [115] used a RP technology called 3D fiber deposition (3DF) for the fabrication of porous Ti-6Al-4V scaffolds with fully interconnected porous network and highly controllable porosity and pore size. Curodeau et al. [116] produced porous CoCr scaffolds manufactured by sacrificial wax template or investment casting. Murr et al. [117] reported the direct metal fabrication of non-stochastic titanium structures by electronbeam melting (EBM). Mullen et al. [118] produced porous titanium constructs by selective laser melting (SLM). This group also demonstrated that optimized structures can be produced with ideal qualities for bone ingrowth applications.  Table 5 summarizes the key features of several RP techniques commonly used for the fabrication of porous metallic scaffolds. In general, the dimensional accuracy, mechanical properties, and applicable materials are restricted by each particular technology. Future development of porous constructs is mainly concerned with improving the RP techniques for creating specialized, low costs structures, which give long-term mechanical reliability to the engineered porous metal-bone interface.

Tantalum
The efficacy of tissue-engineered tantalum constructs has been tested extensively in preclinical and clinical trials. Tables 6 and 7 show respectively the results of some preclinical and clinical trials using porous tantalum scaffolds.    Valuable preclinical results in laboratory animal experiments using commercially available tantalum constructs have led to the development of further applications of porous tantalum. For example, in total hip arthroplasty, spinal fusions, structural support of osteonecrosis and tumor related lesions, hand surgery lesions, maxillofacial surgery, etc. Data gained from these experiments have been invaluable leading to the advances of clinical trials in a controlled fashion. The majority of these short-term clinical studies exhibited promising favorable results, but long-term studies are needed. Nowadays, porous tantalum (Trabecular Metal™) in vivo testing is undergoing phase III and phase IV clinical trials.

Magnesium
Magnesium and magnesium alloys have similar mechanical properties with natural bone, but their high susceptibility to corrosion has limited their application in orthopedics. In the case of biodegradable scaffolds, it is desirable for the scaffold materials to be biodegraded completely after an appropriate period in a human body. An important method to slow down the degradation rate of magnesium is surface modification. Some surface modifications have been developed for porous Mg constructs to control the degradation rate as well as to improve the biocompatibility [126,127].  Nowadays only in vitro [126,127] and preclinical studies using animal models have proposed the usage of Mg scaffolds as degradable scaffolds for bone substitute applications. Indeed, works dealing with the in vivo behavior of porous magnesium at the preclinical level are still very scarce. Table 8 lists the data derived from some preclinical studies using magnesium or magnesium alloys constructs.

Titanium
Porous titanium and titanium alloys have been shown to possess excellent mechanical properties as permanent orthopedic implants under load-bearing conditions [128]. Many basic scientific preclinical and clinical studies support the utility of Ti scaffolds. For marginal bone defects and bone augmentation Ti foams allow for bone ingrowth through interconnected porous [155]. On the other hand, titanium fiber-mesh is a useful scaffold material that warrants further investigation as a clinical tool for bone reconstructive surgery. In vitro, titanium fiber-mesh acts as a scaffold for the adhesion and the osteoblastic differentiation of progenitor cells [129]. In vivo, the material reveals itself to be osteoconductive, demonstrating encouraging results [182]. The studies described in Table 9, performed in clinically relevant large animal models, provide a wealth data demonstrating the safety and feasibility of the use of titanium scaffolds in the healing of bone defects.        Results of these preclinical studies confirm that healing of bone is possible using biochemicallymodified Ti scaffolds, specifically by the use of growth factors and osteoprogenitor cells. However, the assessment of the potential for the use of these biochemically-modified Ti scaffolds for clinical applications in the future lies on the capability of the researches to show excellent long-term results. Clinically, cylindrical titanium meshes have been used with consistently good results for large anterior column defect reconstructions. Implantation of synthetic cages into the anterior column seems to offer immediately effective segmental stability, correction of the sagittal plane deformity, and restoration of the anterior vertebral support from a biomechanical standpoint. These anterior interbody cages provide a satisfactory axial load-bearing capacity, and morcellized autograft can be used to fill the inside of the cage [158]. Table 10 lists the results of some clinical studies that employed porous Ti scaffolds for hard tissue repairing and reconstruction. Although there is a paucity of literature regarding the clinical outcomes and result of porous titanium scaffolds, longer follow-up periods and a larger sample group of patients are required in order to obtain reliable clinical success rates.

Nickel-Titanium Alloy (Nitinol)
Porous Nitinol (PNT) has been used in maxillofacial and some orthopaedic surgeries in Russia and China for approximately 15 years [130]. PNT has aroused interest also in intervertebral disc pathologies as an interbody fusion bone scaffold [131]. However, until now few preclinical trials using animal models and very scarce clinical trials have been carried out and more research may be required to better understand the biological performance of PNT. Table 11 shows some preclinical studies carried out with porous Nitinol as scaffold material. The majority of the clinical studies using Nitinol meshes are limited to the thoracic and cardiovascular surgery field where Nitinol finds application especially in self-expanded metallic stents.   Table 12 shows the results of two specific clinical trials using porous Nitinol. The first trial employed a porous Nitinol superelastic expandable cage that can be twisted into any shape and return to the original shape when the compression force is lifted [171]. The second trial corresponds to the midface endoprosthetic area. PNT constructs were used in the midface reconstruction in 129 patients with encouraging results [172]. However, continued follow-up is necessary to determine the clinical long-term success of these PNT constructs.

Hybrid Constructs
Upon the placement of a scaffold into a surgical site, there is a cascade of molecular and cellular processes that provides for new bone growth and differentiation along the biomaterial surface. The goal of a number of current strategies is to provide an enhanced osseous stability through microsurface mediated events. These strategies can be divided into those that attempt to enhance the immigration of new bone (e.g., osteoconduction) through changes in surface topography (e.g., surface roughness, porous surface, etc.), biological means to manipulate the type of cells that grow onto the surface and strategies to utilize the scaffold as a vehicle for local delivery of a bioactive coating (adhesion matrix or growth factor such as BMPs) that may achieve osteoinduction of new bone differentiation within the scaffold surface. Calcium phosphate (Ca-P) ceramics have been successfully proposed as bone substitutes because of their chemical similarities with bone mineral. Hybrid constructs of Ti or Ta/osteogenic cells, Ti or Ta/Ca-P ceramics, and Ti or Ta/growth factors have demonstrated very good osteogenic properties compared with non-modified Ti or non-modified Ta, suggesting that the surface modification of Ti and Ta plays an important role in bone tissue engineering. In the case of Ca-P coated scaffolds physicochemical and crystallographic continuity have been observed in vivo between the calcium phosphate coated external surface and the newly mineralized layer [179]. This mineralized interface ensures a physicochemical and mechanical cohesion between the scaffold and the host bone. Tables 13 and 14 show some preclinical trials made in different animal models using hybrid Ti and Ta constructs, respectively. In the field of tissue engineering, the grafting of arginine-glycine-aspartic acid (RGD) peptides has been the focus of much attention [176,181,185]. Their presence at biomaterials surfaces improved cell adhesion. Indeed, this peptide sequence is present in various extracellular matrix and plasma proteins, and it constitutes a major recognition site of a large number of adhesive extracellular matrix, blood and cell surface proteins [185]. The immobilization of bioactive molecules such as BMP-2 into the metallic biomaterial surface leads to the differentiation of the cells towards osteogenic lineage improving the osseointegration [174]. Based upon the promising results obtained from the preclinical studies, carefully selected and controlled clinical trials with rhBMP-2 have begun [201]. Ti scaffolds have also found application as delivery systems for transforming growth factors -1 (TGF-1) [183]. On the other hand, adjusting the underlying micro-and nanotopography is also a smart way to trigger and modulate specific cellular functions. This approach was attempted by Wu et al. [170] using porous NiTi, and Takemoto et al. [154] using porous Ti. The combined effect of topography and biochemical cues using bone-stimulating agents is indeed an interesting path to improve the biocompatibility. Table 13. Preclinical studies using titanium-ceramic, titanium-polymer, or cell loaded titanium scaffolds.

Author Animal model & implantation site Clinical results Demonstrated properties
Zhang, E. et al.       Table 15 lists the data derived from some clinical studies using hybrid Ti constructs. Although the various approaches used in bone tissue engineering result in increased bone formation, there is a lack of long-term data able to elucidate how long this de novo bone formation can be maintained. Formal examination of these clinical cases is pending. Moreover, there are a number of challenges to be overcome in the transition from preclinical studies in experimental animals to clinical trials in humans. In order to allow comparisons between different preclinical studies and their outcomes, it is essential that animal models and methods to evaluate the achieved results become standardized to accomplish the accumulation of reliable data leading to the development of intelligent constructs. Furthermore, it should be kept in mind that most of the cell-loaded scaffolds studies were performed using young adult or even fetal animal cells and not with cells from elderly patients. Therefore, extensive research will be needed to determine if results can be extended to the human situation and used in a clinical situation for treating human bone defects.

Summary
Porous metallic scaffolds are used in tissue engineering to replace damaged hard tissues in order to restore its functionality. These structural scaffolds possess an imposed pore structure and interconnectivity and are designed to maintain their shape and strength through the process of repair of the injured bone. For the long-term replacement of bone defects porous metallic scaffolds offer the advantage of interfacial porosity as well as permanent structural framework. They can be made by a number of processes (e.g. powder metallurgy, decomposition of foaming agents, replication, rapid prototyping technologies, among many others). Enormous progress has been made in the development of metallic scaffolds by rapid prototyping techniques and many researchers and surgeons believe that instead of biodegradable scaffolds, biochemically-modified porous metallic scaffolds are more suitable for the development of implants for load-bearing applications. To date, there are many in vivo and in vitro tissue-culturing approaches for bone repair using metallic scaffolds with macro-porous structure.
Porous metallic structures have been tested as a bone-engineered construct using the cell-based and the growth-factor-based strategies. It has been also demonstrated that coating the metallic scaffolds with various proteins such as collagen, RGD-peptide, vibronectin and fibronectin leads to accelerated osseointegration and enhanced bone formation in vivo. Future directions of research in this field will probably focus on the efficient combinations of osteoinductive materials, osteoinductive growth factors and cell-based tissue regeneration approach using composite constructs carriers to reconstruct and repair hard tissues. The goal is to obtain a functional replacement of the injured hard tissue in a procedure that avoids the step of bone harvesting. Therefore, a perfectly controlled hybrid scaffold still remains to be developed.