New Strategies in Enhancing Spinal Fusion

Introduction

Spinal fusion surgery can be used to treat multiple conditions including fractures, infections, progressive deformities, and instability with spondylolisthesis [14, 34]. The general premise of this procedure is to surgically stimulate two independent vertebrae to heal together through the use of bone graft and fixation devices. The annual number of spinal fusion surgeries performed in the USA continued to increase from 174,223 in 1998 to 413,171 in 2008 [48]. Unfortunately, the risk of pseudarthrosis with current procedures ranges from 2 to 42% [41]. This presents a large burden to patients, surgeons, and the health care system, as revision surgery for pseudarthrosis has high morbidity rates and costs. Thus, the demand to improve current techniques and the science of bone grafting for spinal fusion is ever increasing. This review will serve to both outline the basics in bone grafting and discuss new advances in the field.

An ideal bone graft contains the properties of osteoinduction, osteoconduction, and osteogenesis. Osteoconduction refers to the physical ability of the graft to act as a scaffold through which capillaries may grow, allowing the integration of host bone–producing cells. Osteoinduction is the chemical process, stimulated by the graft, that promotes differentiation of mesenchymal cells into osteoprogenitor cells with the ability to develop new bone. Osteogenesis involves living osteoprogenitor cells in the graft, including mesenchymal stem cells or osteoblasts, which can drive the production of new bone [49].

Bone graft healing occurs in several stages. After a bone graft is placed, it becomes immersed in hematoma. An inflammatory reaction occurs, such that the hematoma is infiltrated with immunological and fibroblastic cells. A vascular supply is reestablished from bleeding cortical surfaces of host bone and soft tissue over several weeks, and the hematoma is transformed into a fibrovascular stroma. Following this, the reparative phase occurs, wherein necrotic tissue is resorbed and osteoblastic and chondroblastic cells differentiate (weeks 4 to 5). Finally, during the remodeling phase, cartilage is replaced with mineralized woven bone in a centripetal fashion [5].

In order to identify articles on innovations in bone grafting, we searched two commonly used search engines: PubMed and Google Scholar. We searched for articles using key words such as “bone graft biology,” “bone autograft,” “bone allograft,” “demineralized bone matrix,” “ceramics,” “osteoinductive factors,” “tissue engineered scaffolds,” “osteogenic stem cells,” and others. Multiple authors reviewed publications in order to identify appropriate citations. The following provides an overview of our findings.

Bone Graft Materials

Bone Graft Enhancers: Growth Factors and Gene Therapy

Growth factors, including bone morphogenetic protein (BMP), transforming growth factor beta (TGF-β), and platelet-derived growth factor (PDGF), are signaling proteins that induce cellular division/differentiation and bone matrix synthesis [46].

The BMPs, part of the TGF-β superfamily, include BMP-2 through BMP-20 and are related by amino acid sequence homology and dimeric molecular structure [44, 73]. BMPs are multifunctional and clinically interesting for their ability to stimulate osteoinduction rivaling autograft [15]. They were first extracted by Marshall Urist in 1980 from demineralized rabbit bone and were shown to be able to induce bone morphogenesis across species [62]. However, until the advent of molecular techniques allowing for the purification and cloning of each protein, it was unknown which species was responsible for the induction of bone formation (BMP-2–BMP-8) [68, 73]. Molecularly, they function by binding to a cell surface serine–threonine kinase receptor, which then transduces the signal through SMAD and ras/raf proteins to activate the gene expression necessary for bone production [15]. Wang et al. showed that, although these proteins are often present as a group during bone morphogenesis, individual members (such as BMP-2) are necessary and sufficient to drive bone formation [67]. At lower concentrations, BMPs induce endochondral ossification by causing mesenchymal stem cells to differentiate into chondrocytes and produce a cartilage matrix that then becomes calcified [24]. At higher concentrations, this process shifts to direct transmembranous bone formation, whereby bone is formed directly without a cartilage intermediary. BMPs have been shown to be safe and effective promoters of local bone healing in multiple animal studies [6, 11, 12, 56].

There are two preparations of BMPs available for clinical use: recombinant human BMP-2 (rhBMP-2) and rhBMP-7, both manufactured from mammalian cell culture [72]. They are typically delivered to the fusion site via a carrier material, either a collagen sponge or a ceramic such as calcium phosphate. This vehicle both serves as an osteoconductive agent for bone and also improves the tissue retention of BMP. BMP has been shown in humans to increase the rate of fusion, especially in cases of refractory non-union [6]. Given the complex role of RhBMP-2 on osteoclast and osteoblast function, it is important to note that while it does enhance bone growth, it also induces transient bone resorption [55]. Simmonds et al. published the combined results of 11 clinical trials (1302 patients) involving BMP and were able to show an improvement in Oswestry Disability Index (ODI) scores at 24 months compared with iliac crest harvest, as well as a 12% higher fusion rate [57]. Nonetheless, complications have been described with the use of BMP, including a potential higher rate of cancer [10, 57]. Other complications such as retrograde ejaculation, osteolysis, radiculitis, and seroma formation have also been described and must be considered when using rhBMP [10, 42]. The rate and etiology of these complications, however, is likely related in part to the technique and location in which BMP is used [47]. For instance, the use of RhBMP-2 in the anterior cervical spine has largely been abandoned due to complications related to dysphagia, end-plate resorption, and post-operative swelling [64]. RhBMP-2 use has continued for lumbar spine arthrodesis, however, likely due to its association with higher fusion rates and lower rates of BMP-related complications for posterolateral lumbar spinal fusion [65]. New evidence also indicates that the relationship between cancer and RhBMP-2 use may not be substantial [27].

While BMPs potentiate cell differentiation (causing mesenchymal cells to become osteoblasts), PDGF and TGF-β are growth factors that cause cells to divide, as well as increase production of matrix proteins. Preparation of platelet rich plasma, which is a source of PDGF and TGF-β, may also be used to stimulate bone growth, although it has not yet shown definitive osteoinductive properties in spinal fusion [46].

Gene therapy, wherein the BMP gene is delivered to a host cell by a vector, has been shown in animal models to induce fusion in nonosteoid tissue [2, 40]. Compared with locally delivered BMP, gene therapy could increase the bioavailability of BMP by producing ongoing osteogenic expression locally [7]. Vectors range from viral (using a virus such as adenovirus or herpes virus to transfect a cell) to nonviral (liposomes, electroporation). Delivery to a host may be either ex vivo (implantation of transfected cells into the host) or in vivo (injection of genes directly into host cells) [39, 40]. Riew et al. were able to show that ex vivo BMP-transfected bone marrow cells, replanted in rabbits, were able to produce spinal fusion [50]. Currently, gene therapy is limited by the massive immune response against viral vectors [2].

Systemic Therapy to Enhance Spinal Fusion

Many patients treated with spinal fusion have a metabolic bone disease that increases the risk of complications during and after the procedure [4]. Pre-clinical models support the use of intermittent parathyroid hormone to enhance spinal fusion in rabbits [42]. The pre-operative use of teriparatide has also been shown to both increase the insertional torque of a pedicle screw and decrease post-operative screw loosening [26, 31]. While the use of parathyroid hormone analogs is currently approved for treatment of osteoporosis in patients at risk for fracture, there is significant research indicating that it could also enhance bone growth during spinal fusion [36, 60]. Further investigations are warranted to ensure the safe and appropriate use of this novel treatment to enhance clinical outcomes for spinal fusion.

Tissue Engineering and Scaffolds

Biomaterial scaffold is a generic term used for materials that function by imitating the physical and chemical properties of native bone tissue, such that osteoblasts may easily incorporate into and replace them [21]. An ideal scaffold should have a three-dimensional structure strong enough to mimic the mechanical properties of bone, allow for osteoinduction by containing surface proteins necessary for osteoblast attachment, and contain cells and signaling factors to promote osteogenesis [25]. The material must also be immunologically inert to prevent the formation of fibrous tissue, which can lead to aseptic loosening [43]. The multiple classes of scaffolds include allograft, autograft, metals, ceramics, polymers, and composites. They may either be synthetic vs. natural or be resorbable vs. non-resorbable. Each class of scaffold has varying mechanical, chemical, and immunological properties that must be considered.

Currently, there is no single scaffold material that contains all the ideal properties for a bone graft substitute [43]. Autografts are considered the gold standard, but they require the cost of donor site morbidity [1]. Allografts have limited osteoinductive and osteogenic properties and also have the potential for disease transmission [22]. Metals are strong in both tension and compression; however, they do not provide a natural substrate for cell adhesion and can induce a non-specific immune response [46]. With the advent of additive manufacturing (where three-dimensional structures are created in a layered fashion and controlled by computer design), new surface properties are being developed in metals to allow for improved osteointegration [25]. Ceramics, including bioactive glass, calcium phosphates, and corals, as described above, are useful in promoting osteoconduction but have limited osteoinduction potential [25, 46]. In addition, they do not have ideal bone mechanical properties and tend to be weak in tension and shear. Polymers include a vast array of materials, ranging from natural (collagen, chitosan, silk, hyaluronic acid, and peptides) to synthetic compounds (polyglycolic acid and polylactic acid). Naturally derived polymer scaffolds, such as collagen or chitosan, have the ability to resorb and also contain signaling motifs for cell migration. Still, they too lack the ideal mechanical properties of bone [46].

Peptide hydrogels are a new scaffolding material that have shown promise in the regeneration of tissues, with limited reparative potential of cartilaginous, neuronal, and cardiac tissues [23]. Hydrogels are synthesized from the molecular self-assembly of amphiphilic peptides into an entangled nanofiber structure, which is similar to the extracellular matrix of native tissues [22]. Moreover, they can be engineered to contain epitopes such as the α5β1 integrin receptor that promote cell migration and adhesion [37, 46]. Hydrogel materials may be combined with osteogenic cells and assembled into a matrix that allows osteoid formation and can be tuned to degrade at an appropriate time [46, 66]. Thus, they could represent a new horizon in bone grafting, offering the panacea of osteoinductive, osteoconductive, and osteogenic properties.

In conclusion, there is a vast array of bone graft choices with a variety of strengths and weaknesses. The science behind bone grafting is growing and driving improved clinical outcomes in spine surgery. With the advent of new technology, bone grafting will become faster, cheaper, and more efficacious, all the while sparing the patient the morbidity of local graft harvest.