Keywords

Introduction

Bone grafts and graft substitutes are materials that are used to rapidly induce or support biologic bone remodeling after surgical procedures to reconstruct bony structures and correct deformities and/or to provide initial structural support (Wang and Yeung 2017). In the spine, bone grafts are most often used to support biological healing with bony union of vertebral segments after a spinal fusion surgical procedure.

Bone grafts may come from a patient’s own bone (autograft), may come from a human cadaver or living donor via bone bank (allograft), or may be fabricated from a synthetic material such as ceramics or bioactive glass. Furthermore, combination materials including composites of allografts, growth factors, osteogenic cells, synthetic materials of ceramic and/or cements, bioactive glass, and peptide-based materials have been developed and are offered for clinical use in spine fusion. See Table 1 for a description of sources of grafting materials and their associated bone-forming properties.

Table 1 Description of specific graft materials and bone forming properties
Full size table

Design Requirements for Engineered Biomaterials (Table 2)

The selection of bone graft alternatives to be used for spinal fusion should be conducted carefully by considering the different healing environments, reviewing the preclinical and clinical data, and also considering the regulatory burden of proof for products not subjected to high levels of regulation (Boden 2002).

Table 2 Optimal characteristics of engineered biomaterials (O’Brien 2011)
Full size table

The development of products used for bone regeneration has followed the basic criteria of providing a biocompatible three-dimensional scaffold with controlled architecture capable of stimulating or supporting bone growth in the natural in vivo environment. The ability of the material to be amalgamated with cellular and signal (differentiation/growth factors)-based products is a key strategy in maximizing the efficacy and likely success of fusion. The primary characteristics and significance of bone graft substitutes is shown in Table 2.

Spinal Fusion

Spinal fusion is usually performed to provide stability to the spine when its biomechanics have been disturbed or altered. The surgical concept underlying spinal fusion is to reduce clinically important abnormal motion and add immediate and long-term stability, therefore decreasing or eliminating pain thought to be aggravated by the abnormal motion (Herkowitz et al. 2004; Adams 2013). Spinal fusion is performed in patients with degenerative diseases like spinal instability, vertebral fractures, degenerative disc disease, and scoliosis. After a surgical decompression procedure has been performed to relieve pressure on the nerve roots or spinal cord, a fusion procedure may be completed as well to address the instability and provide long-term bony stability and structural reinforcement. The two main types of spinal fusion procedures are posterolateral fusion (PLF) and interbody fusion (IBF) performed from among a large variety of surgical approaches and techniques (Makanji et al. 2018; Morris et al. 2018). Radiographic images of example patients after surgical fusion procedures in lumbar and cervical spine are provided (Figs. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, and 11).

Fig. 1
figure 1

Example of posterolateral fusion with consolidated bone mass (BB) approximately 1 year after spinal fusion procedure with instrumentation and autograft placed in the posterolateral bed

Full size image
Fig. 2
figure 2

Example of four-level posterolateral fusion with dense solid bone mass (BB) bilaterally at 1 year after posterior spinal fusion procedure with instrumentation and grafting with Fibergraft (Prosidyan), allografts, and bone marrow aspirate (BMA)

Full size image
Fig. 3
figure 3

A 59-year-old female patient was surgically treated for failed artificial disc in the lumbar spine. For treatment, a posterolateral fixation was performed with allograft cancellous chip bone (Medtronic) mixed with autologous local bone and pedicle screw fixation device (Medtronic). A radio-dense bone bridge was not observed between transverse process between L3 and S1 on the initial postoperative anterior-posterior radiographs (A). On radiographs taken at 3-year follow-up after removal of pedicle screws, there was a radio-dense bone bridge (BB black arrows) on anterior-posterior (AP) radiographs. On CT image (C), definitive radio-dense bone (BB) was observed with contact between transverse process, facet joint, and grafted bone

Full size image
Fig. 4
figure 4

A 60-year-old female patient was surgically treated with a diagnosis of spinal-stenosis L4-L5. Surgery: an indirect decompression of spinal nerve and fusion via oblique interbody fusion (OLIF) technique with PEEK cage containing DBM-based product (Medtronic). A hemilaminectomy via MIS surgery technique was performed using percutaneous screw fixation. On anterior-posterior (AP), lateral (L) radiographs taken at 1.5-year follow-up, a radio-dense bony line was observed between upper and lower vertebral body through the inserted cage. On sagittal CT view, a dense bridge (bony incorporation) was formed at fusion site. Wedge-shaped vertebral deformation of L1 and L2 compression fracture was observed. L2 fracture was treated with PMMA bone cement

Full size image
Fig. 5
figure 5

A 78-year-old male patient was surgically treated for TB spondylitis in L2. The infected vertebrae were removed through corpectomy process. Fusion procedures were performed using a distractible cage (DePuy Synthes) with allograft cancellous chip bone (Medtronic) mixed with autologous local bone and a percutaneous pedicle screw fixation device (Medtronic). Titanium distractible cage with keel on contact surface with vertebral endplate was used for load bearing architecture. On anterior-posterior (AP), lateral (L) radiographs taken at 2-year follow-up, a radio-dense bone bridge (black BB) was observed between upper and lower vertebral body. On CT image (C), definitive radio-dense bone (white BB) was observed connecting through the cage between the two vertebral endplates (white arrows indicate endplates)

Full size image
Fig. 6
figure 6

A 49-year-old male patient was surgically treated for herniated intervertebral disc C4–C5, C5–C6. A total discectomy was performed for decompression through an anterior surgical approach. A machined cortico-cancellous allograft (Medtronic) and cervical plate (Medtronic) were used in the fusion procedure. On anterior-posterior (AP), lateral (L) radiographs taken at 1-year follow-up, a radio-dense bone bridge (BB) was observed between upper and lower vertebral bodies though the inserted machined cortico-cancellous allograft

Full size image
Fig. 7
figure 7

A 34-year-old male patient was surgically treated for two-level herniated intervertebral disc C3–C4, C4–C5. A total discectomy was performed for decompression through an anterior approach. For the fusion procedure, iliac autogenous bone graft and a cervical plate (Medtronic) were used. On anterior-posterior (AP), lateral (L) radiographs taken at 1-year follow-up, a radio-dense bone bridge (BB) was observed between upper and lower vertebral body through the inserted iliac autogenous bone graft. On sagittal CT view, there is bone formed and complete incorporation of C3–C4–C5 (BB) at the fusion site

Full size image
Fig. 8
figure 8

A 64-year-old male patient was surgically treated for herniated intervertebral disc C6–C7. A total discectomy was performed for decompression through an anterior surgical approach. For fusion procedure, a machined cortico-cancellous allograft (Medtronic) and a cervical plate (Medtronic) were used. On anterior-posterior (AP), lateral (L) radiographs taken at 1.5-year follow-up, a radio-dense bone bridge (BB) was observed between upper and lower vertebral body through the inserted machined cortico-cancellous allograft. On sagittal CT view, a bone bridge (BB, complete bony incorporation) was formed at fusion site

Full size image
Fig. 9
figure 9

A 49-year-old female patient was surgically treated for herniated intervertebral disc at C5–C6. A total discectomy was performed for decompression through the anterior approach. For the fusion procedure, a Zero-p system (DePuy Synthes) and DBM-based putty (DBX, DePuy Synthes) were used. On anterior-posterior (AP) and lateral (L) radiographs taken at 2-year follow-up, a radio-dense bone bridge (BB) was observed between upper and lower vertebral body

Full size image
Fig. 10
figure 10

A 59-year old patient was surgically treated for fracture dislocation injury at C3 vertebrae after fall from height. Treatment of fracture was performed by a decompression through corpectomy and fusion using auto-iliac crest strut bone graft with cervical metal plate fixation spanning C3–4–5). On anterior-posterior (AP), lateral (L) radiographs taken at 3-year follow-up, a radio-dense bone bridge (black BB) was observed between upper and lower vertebral bodies (C3–C5 a solid bone unit). Note no radio-opaque gap between graft material and endplate of adjacent vertebral bodies

Full size image
Fig. 11
figure 11

A 62-year-old male patient was surgically treated for OPLL from C4 to C6 cervical spine. Decompression of the spinal cord was performed through removal of vertebral body of C5 and C6. Fusion was performed using allograft strut fibula bone with a cervical metal plate (Medtronic). After initial postoperative radiography, a radio-opaque gap is seen between grafted material and endplate vertebral body. At 1-year follow-up, a complete incorporation between allograft and host bone is observed as a solid bone unit – no radio-opaque gap between grafted material and vertebral endplates. There is radio-density of grafted material indicating bony consolidation and incorporation with fusion from C4-C7

Full size image

In posterolateral fusion (PLF), the bone graft or bone graft substitute is surgically placed between the transverse processes, lateral to the side of the superior vertebral body and inferior vertebral body. During the healing process, the graft material is remodeled and incorporated into a solid bony “bridge” (BB) between the transverse processes and lamina. Once healed, the spine segment is stabilized, and motion between vertebral functional segments is eliminated or reduced (Fig. 1, example lumbar spine).

In interbody spinal fusion, compared to PLF, the bone grafts or bone graft substitutes are placed between the endplates of two adjacent vertebrae (e.g., Figs. 4 and 5, lumbar spine; Figs. 6, 7, 8, 9, 10, and 11, cervical spine). The bone graft’s and/or instrument with graft’s contact with the endplates of the adjacent vertebral bodies resists relatively high loading forces.

Due to these biomechanical differences in graft sites, interbody fusion bone grafts are commonly placed with cages that hold the graft in place and are designed to withstand the compressive forces of the vertebrae. When the bone graft or bone graft in a cage is placed between the endplates of the vertebral body, it creates a framework of mechanical support during the early time of graft incorporation. This mechanical fixation and support eventually aids in the biologic bony union connecting one vertebral body to the other. Similarly to posterolateral fusion, once the vertebrae are fused, the spine is stabilized, and movement between operated spine segments should disappear. A systematic recent review of 12 studies (565 IBF-treated patients) by Baker et al. (2017) concluded that interbody fusion was a good surgical option in spondylolisthesis patients with instability. Interbody fusion can be performed by several different surgical approaches and techniques such as anterior (anterior lumbar interbody fusion (ALIF)), posterior (posterior lumbar interbody fusion (PLIF)), transforaminal (transforaminal lumbar interbody fusion (TLIF)), and lateral (lateral lumbar interbody fusion (LLIF)).

After a fusion procedure, the bone healing process occurs in different phases: inflammation, soft callus formation, hard callus formation, and bone remodeling.

This includes hematoma formation, release of native growth factors/cytokines, and recruitment of inflammatory cells (e.g., macrophages and bone-forming cells); cell differentiation to bone-forming cells and mineralization of the extracellular matrix (ECM); bone resorption and remodeling; and formation of lamellar bone and hematopoietic marrow cavities (Rausch et al. 2017). These complex biologic processes of consolidation of grafted materials into new bone and remodeled into mature bone can be negatively affected by various systemic and local factors. Typical host or patient-based negative factors are advanced age (Ajiboye et al. 2017), concomitant use of tobacco or other drugs, poor nutritional status, and metabolic comorbidities (e.g., diabetes or osteoporosis) (Campbell et al. 2012; Ajiboye et al. 2017). Negative local factors are remaining structural instability, poor vascularity around surgical site, revision surgery, previous or current infection, and other local/surgical site considerations including surgical technical factors such as inadequate preparation of host bone, lack of fixation, inadequate bone graft volume and preparation, and improper use of graft materials (Yoo et al. 2015). Critical challenges for both interbody and posterolateral fusion are the excessive distances for the cells to migrate within and between host bone beds in order to attach to targeted neighboring anatomic bony structures; the limited durability of concentrations of growth factors, peptides, exogenous cells, biochemical, and other agents; and the biomechanical stability. These biologic challenges are particularly deterring in geriatric spine patients with severe osteoporosis.

To achieve successful bone fusion in the spine, surgeons vigilantly adhere to the requirements of bone regeneration and fracture healing mentioned above in deciding use of grafting materials. Bone formation requires three critical elements: osteoconduction, osteoinduction, and osteogenesis. Osteoconduction relies on a scaffold that supports cell ingrowth, facilitates vascularization, and provides a network for cells to attach. Osteoinduction relies on the provision of signals that act on the precursor cells and encourage cell migration, proliferation, and differentiation into bone-forming cells leading to rapid bone formation. Osteogenesis relies on the immediate provision of viable cells emanating from the host to the defect site differentiating into bone-forming cells. Autograft or autogenous bone possesses all three properties essential for bone formation and is therefore considered to be the gold standard graft material for inducing bone healing, consolidation, and fusion of the spine.

Current Materials for Spinal Fusion

The graft material used in spinal fusion procedures can be generally categorized into three main types of materials: autogenous bone graft (autograft) from the patient’s own body, allograft from human cadavers and/or living donors, and synthetic bone graft or substitutes (Table 1).

The use of autogenous bone graft has been a standard practice in spine surgery for over a century. The first reported use of autogenous bone graft for spine fusion was reported in 1911 when Fred Albee, MD, placed a tibia between spinal lamina in order to fuse and stabilize the spine (Albee 2007). Autograft has been considered the “gold standard” of bone grafting primarily because it contains all the elements required for successful fusion mentioned above: osteoconductive matrix, osteoinductive factors, and pluripotent bone-forming cells (Gupta and Maitra 2002; Whang and Wang 2003).

Autografts

Autograft for spinal fusion can be obtained via different surgical approaches and dissection methods. Firstly, resected lamina, spinous process, facet, and osteophytes during the surgical decompression process yield bone graft which is then morselized – “local bone graft” (LBG). Local bone is commonly limited in amount and quality as mixture consists of mostly cortical bone vs. cancellous bone (Tuchman et al. 2016). Secondly, a bone graft can be obtained from iliac crest using a separate surgical incision and various dissections (White and Hirsch 1971), which then can be used as strut or morselized bone. Iliac crest bone graft (ICBG) is relatively abundant, providing good-quality graft (mainly cancellous bone). However, iliac bone and local bone autografts have similar effectiveness in terms of fusion rates, pain scores, and functional outcomes in view of lumbar spine fusion (Tuchman et al. 2016).

Autograft bone is safe to use due to the low risk of disease transmission and offers the optimal chance of acceptance and effectiveness in the transplant site without immune reaction (Campana et al. 2014). However, the limitations with autogenous iliac bone graft such as relatively limited quantity, increased surgical time, and donor site morbidity are well recognized (Vaccaro et al. 2002). Due to these limitations, the use of autograft has declined.

The reduction in the use of autograft from the iliac crest in the recent practice has led to the increase in the use of local bone graft and has created new demands for the identification of cost-effective biologic materials that will “extend” the bone healing effects of local autograft (Ito et al. 2013). To achieve optimal outcomes, these materials should be biocompatible and biodegradable and have beneficial mechanical properties and microarchitecture that facilitates the biological healing process (Table 2).

Allografts

Allografts are primarily osteoconductive with minimal osteoinductive potential and traditionally not osteogenic because the donor cells are eradicated during processing (Campana et al. 2014; Duarte et al. 2017). Allografts have the advantage for a surgeon of easy procurement (off-the-shelf), availability (commercially available), and many varieties of structural and non-structural form. However, allografts consist of nonviable tissue and cannot stimulate bone formation without the addition of bone-stimulating factors and cells (Goldberg and Stevenson 1993; Garbuz et al. 1998; Stevenson 1999). These limitations lead to slower and less complete incorporation with native bone. Additionally, allografts have potential risk of disease transmission even if the incidence is very low and the risk can be controlled during procurement and sterilization process (Campana et al. 2014).

Allo-bone Graft: Cortico-cancellous Allograft (Table 3)

Allograft bone obtained from cadaver sources is added to the most widely used substitute or extender for autogenous bone graft. In the 1980s, femoral head from living donors (after total hip replacement surgery) was also introduced as another form of allograft and has demonstrated good clinical results in lumbar spine fusion (Urrutia and Molina 2013).

Table 3 Commercially available structural and/or nonstructural allograftsurala
Full size table

Allograft bone may be morselized to various sizes of particulate (i.e., chip bone) formed or machined to create structural spacers and then applied to site of desired bone formation. Cortical allograft is most often used as mechanical strut graft and is suited for interbody fusion, while cancellous allograft serves as a useful osteoconductive scaffold for bone formation.

The efficacy of allograft alone has been shown to have more clinical variability and lower fusion rates in challenging animal models and human studies of spinal fusion (Morris et al. 2018). These overall clinical results suggest that allograft be cautiously used in conjunction with either autograft or osteogenic material (e.g., bone marrow aspiration) to achieve good fusion rates and clinical outcomes (Morris et al. 2018). However, while the actual risk of transmission is negligible, issues of immunogenicity are present (Manyalich et al. 2009).

Allo-bone Graft: Demineralized Bone Matrix (DBM)-Based Product (Table 4)

The DBM technology is based on the observation by Urist MR (Urist 1965) that soluble signals contained within the organic phase of bone were capable of promoting bone formation . The processing of transforming ground cortical bone into DBM powder base involves the use of hydrochloric acid to progressively remove mineral while attempting to preserve the organic phase containing type 1 collagen, non-collagenous proteins, and inductive growth factors (Gupta et al. 2015). Even after processing, DBM possesses osteoconductivity and osteoinductivity, but as a putty-/paste-like substance, it lacks structural integrity (Gupta et al. 2015). Since DBM base powder is derived from human bone allograft, disease transmission related with implantation is low, yet possible, although still less than structural-type allografts (bacterial infection estimated at 0.7 for non-massive to 11.7% for massive bone) (Zamborsky et al. 2016; Kwong et al. 2005; Lord et al. 1988).

Table 4 Commercially available DBM-based products
Full size table

Due to lack of structural integrity and relatively low osteoinduction potential comparing to autograft, DBM mixed with a carrier (DBM-based product, DBMs) is frequently used as a bone graft extender/carrier in interbody fusion. Commonly, DBMs are mixed with morselized autografts and exogenous peptide/differentiation factors along with collagen matrix, bone marrow aspirate, and/or isolated native blood-derived growth factors to stimulate new bone growth. In previous clinical reports on spine surgery, DBMs with autograft, and DBMs with growth factors (bone marrow aspiration), DBMs mixed with peptides (rhBMP-2/ACS) may be substituted for ICBG (Kang et al. 2012; Morris et al. 2018). DBM-based products or DBM powder are rarely used as a stand-alone graft material (Kinney et al. 2010).

There are several limitations to overcome in the clinical use of DBM-based products. The clinical effectiveness of DBM-based products is known to be variable according to manufacturer, form of product, as well as different lot-based batches from the same product form and manufacturer (Bae et al. 2006, 2010). The possible features of DBM-based products that contribute to varied reliability are varying native BMPs, growth/differentiation factors (donor bone), and dosages (Bae et al. 2006, 2010); forms such as putty, gel, flexible sheets, or mixed with cortical chips; compositions of carriers, scaffolds, gels, and other fillers; particle sizes of final bone powder; quality of the donor bone; and manufacturers processing procedures and sterilization method of products (Peterson et al. 2004; Bae et al. 2006, 2010). Amid these limitations, DBM-based products provide a diverse range of DBM-based grafting options that have been commonly employed for specific applications. DBM-based products introduced to the market over the last two decades and currently used are presented in Table 4.

Exogenous Inductive Differentiation Growth Factors and Other Peptides (Table 5)

Bone Morphogenetic Protein

Bone morphogenetic proteins (BMPs) are soluble members of the transforming growth factor-β superfamily that are involved in the differentiation, maturation, and proliferation of mesenchymal stem cells (MSCs) into osteogenic cells (Miyazono et al. 2005). To describe the acting mechanism, BMPs act via serine-threonine kinase receptors found on the surface of target cells and often transduce their signal via the SMAD pathway, leading to nuclear translocation and subsequent expression of target genes involved in osteogenesis (Hoffmann and Gross 2001; Sykaras and Opperman 2003). The reaction mechanism of BMP is mainly osteoinduction and reactively much less osteogenic potential (Campana et al. 2014). The graft material includes rhBMP-2 (exogenous protein) along with absorbable collage sponge (ACS, rhBMP-2 carrier). The carrier (ACS) has BMP binding competence in order to decrease diffusion away from the desired site for bone formation and increase controlled continual delivery of protein at the site. Although numerous carriers such as metals, collagen, ceramic such as tricalcium phosphate (TCP) and HCO, bioactive glass (BG), and polymers have been described (Agrawal and Sinha 2017), the most commercially available scaffold is an absorbable type 1 collagen sponge (ACS) bovine derived (Kannan et al. 2015).

Table 5 Commercially available bone inductive peptides, proteins-based products, recombinant versions
Full size table

For several decades, over 20 BMPs have been identified and described. Among them, BMP-1, BMP-2 (BMP-2A), BMP-3 (osteogenin, less osteoinductive) BMP-4 (BMP-2B), BMP-5, BMP-6, BMP-7, and osteoinductive factor (OIF) have been shown to induce bone formation (Wozney 1989, 2002). However, only two commercial forms of recombinant human BMPs currently are available for clinical use (Kannan et al. 2015). Recombinant human forms of BMP-2 (Infuse®; Medtronic) and BMP-7 (OP-1; Stryker) have been developed and approved both in the USA and Europe for commercial purposes by employing mammalian cells transfected with the corresponding human BMP sequence (Campana et al. 2014). Extensive research (over 30 years) has been conducted in support of the US-FDA approval process of rhBMP-2 and rhBMP-7 ; translational problems include scaling up, as super physiologic concentrations of rhBMP-2 are needed to meet MEDs in widely applicable orthopaedic indications in humans (Vallejo et al. 2002).

Internationally, in Korea, several products were developed employing various production processes for BMP-2 and different carriers. Its approved by Korea Food and Drug Administration (KFDA); to date its not approved by US-FDA. For spine fusion, the product carrier is granular HA and is based on Escherichia coli-derived rhBMP-2 (E.BMP-2, CGBio, Korea; E.BMP-2/HA, Novosis®, Korea) designed to improve the protein yield over the production process of using mammalian origin cell lines, such as Chinese hamster ovary (CHO) cells that incur low yield and high cost. There are several animal and clinical studies demonstrating the effectiveness and safety of Novosis® (Lee et al. 2012; Kong et al. 2014; Kim et al. 2015). According to the study of Cho et al. (2017), a fusion rate of 100% for E.BMP-2/HA (Novosis®) was comparable with that of 94.1% for AIBG demonstrating clinical efficacy and safety in PLF. E.BMP-2 production of rhBMP-2 -based products and clinically used or in investigation are a rhBMP-2/Beta-TCP putty type (NCT01764906, Novosis® Korea), another Beta-TCP product containing rhBMP-2 (ExcelOS-inject, ExcelOS 14-01, NCT02714829, BioAlpha Inc., Korea), and a collagen gel +DBM containing rhBMP-2 (50 ug/cc) (rhBMP-2 produced from CHO cells, RafugenTM BMP-2, Cellumed Co Ltd., Seoul, Korea) employed as graft for interbody spinal fusion (pivotal RCT completed, 2017; submitted KFDA 2018, approved for dental application KFDA 2013).

US Regulatory approval by the FDA was initially granted for rhBMP-2/ACS (Infuse, Medtronic) in single-level anterior lumbar interbody fusion procedures in 2002 (Burkus et al. 2002). rhBMP-2 was then approved for tibia nonunion as an alternative to autograft in 2004 and for oral maxillofacial reconstructions in 2007 (Rengachary 2002). During last decade, rhBMP-2 has been commonly used off-label in posterolateral lumbar fusion surgery (Morris et al. 2018).

RhBMP-7 , an osteogenic growth factor related to BMP-2, was first approved by the FDA in 2001 for use as an alternative to autograft for long bone fracture repair. In 2004, approval was expanded to cover PLLF (Morris et al. 2018). RhBMP-7 or OP-1 was approved for limited use under humanitarian device exemption (HDE) (no longer marketed in the USA (https://www.transparencymarketresearch.com/bone-morphogenetic-protein-market.html). In the last decade, several types of rhBMPs were developed and commoditized to medical market. RhBMP-based products were introduced to the market over the last two decades. Currently used products are presented in Table 5.

The osteogenic/osteoinductive potential of rhBMPs was strongly investigated in both preclinical and clinical studies, with a reported performance that is comparable to autogenous cancellous bone, with fusion rates between 80% and 99% (Campana et al. 2014). There are approximately 80 clinical studies on rhBMP-2 testing various surgical indications. According to a Level I comparison study of ICBG vs. rhBMP-2 with collagen sponge and ceramic granule by Dawson et al. (2009), at 24 months the rhBMP-2-/CS-/CM-treated patients had significantly higher solid fusion rates than those in the iliac crest autograft group (95% vs. 70%). Additionally, patients in the rhBMP-2/CS/CM group reported significantly greater improvement in clinical outcomes than did those in the iliac crest autograft group. According to the studies of Vaccaro et al. (2004, 2005), the use of rhBMP-7 (as OP-1 putty from) in conjunction with bovine collagen and carboxymethylcellulose (carrier) showed similar or slightly superior clinical result in spine fusion (posterolateral non-instrumented fusion) compared with autograft from the iliac crest.

However, limitations for general use of BMPs and complete substitution for autograft remain. First, rhBMPs have marked species-specific concentration requirements for osteogenesis, and thus results from preclinical studies are not considered as valuable background information for human application. Second, the dose-dependent efficacy in humans of rhBMPs has been observed in previous studies, and various clinical trials are aimed toward elucidating the optimal dosage of rhBMP-2/ACS (Govender et al. 2002). However, the optimal dosage/concentration for various off-label applications has rarely been reported or suggested in spine surgery. Third, during clinical trials, several major and minor adverse effects like ectopic bone formation in the neural canal, dysphagia when used in cervical fusion applications, prevertebral swelling, seroma/hematoma formation, radiculitis, osteolysis, heterotopic ossification, retrograde ejaculation, increased rates of new malignancy, and implant subsidence due to end-plate osteolysis are reported (Shields et al. 2006; James et al. 2016). Because of these limitations, numerous ongoing areas of investigation target alterations in dosage for optimal minimal dosage, scaffold to maintain concentration, and the implementation of supplemental proteins or growth factors to regulate the nonspecific action of rhBMP-2 (Agrawal and Sinha 2017; Burke and Dhall 2017; Poorman et al. 2017). Outside the USA, alternative-type protein products are in development (BoneAlbumin™, plasma protein) used to enhance bone allograft (Gmbh, OrthoSera, Austria).

Peptide-Based Materials

Although naturally derived extracellular matrix (ECM) has demonstrated some degree of success in selected studies, it is challenging to modify, characterize, and control the presentation of natural ECM biomaterials (Shekaran and Garcia 2011). The limitations of ECM molecules have spurred the use of ECM-derived peptides or recombinant fragments that incorporate the minimal functional sequence of their parent protein to convey bioactivity to implant materials.

Cerapedics

P-15 is a synthetic 15-amino acid peptide derived from the (766)GTPGPQGIAGQRGVV(780) sequence found in the α1(I) chain of type I collagen. Several preclinical studies have demonstrated that P-15 enhances cell adhesion, osteoblastic gene expression, and mineralization when implanted on anorganic bone matrix (ABM) in vitro and accelerates early bone formation in porcine and rat cranial defects (Shekaran and Garcia 2011). In a head-to-head comparison of DGEA peptide and P-15-coated hydroxyapatite discs implanted into rat tibiae, both peptides improved new bone formation, but P-15 failed to enhance bone implant contact. A recent study in the larger bovine model, ABM/P-15 (ABM an allograft in this application), failed at 4.5 months after uninstrumented posterior lumbar spine surgery; 68% fusion in allograft implanted sheep vs. 0% fusion as determined by bridging between transverse processes was found in ABM/P-15 implanted sheep (Axelsen 2019).

For human applications of a xenograft carrier (a sinterized cancellous bovine bone matrix), the implemented carrier has been employed. P-15 peptide-coated ABM has been used in human periodontal osseous defects resulting in better clinical outcomes than open flap debridement alone and has also been used in a pilot clinical study for long-bone defects (Shekaran and Garcia 2011). In a prospective, randomized, single-blinded trial of single-level ACDF using P-15/CBM in an allograft spacer versus local autograft in an allograft spacer, 89.0% vs. 85.8% fusion rates were reported, respectively, at 1-year follow-up with equivalent clinical outcomes and complications (Hsu et al. 2017).

B2A

B2A is a bioactive synthetic multi-domain peptide that augments osteogenic differentiation via increasing endogenous cellular BMP-2 by pre-osteoblast receptor modulation at spine fusion site (Lin et al. 2012). The empirical formula of B2A is C241H418N66O65S2 containing 42 amino acids and 3 lysine analogue residues of 6-aminohexanoic (Glazebrook and Young 2016). This peptide has osteoinductive potential; it is used with a scaffold. The osteoconductive scaffold is a ceramic granule from which B2A elutes in vivo. Two commercialized products PREFIX® (Ferring Pharmaceuticals, Saint-Prex, Switzerland) and AMPLEX® (Ferring Pharmaceuticals, Saint-Prex, Switzerland) are based on this converged technology. After grafting of B2A with ceramic granules, complete absorption of B2A occurs within approximately 6–8 weeks (Glazebrook and Young 2016).

There is great interest in the benefits of conjugation technology for modulating release kinetics in grafting materials. However, there are limited preclinical and clinical studies on the safety and effectiveness of B2A/ceramics. B2A/ceramic granule was tested in two animal studies (rabbit and sheep). B2A/ceramic significantly improved the fusion rate in PLLF and PLIF over simple autograft bone graft (Smucker et al. 2008; Cunningham et al. 2009). In a clinical study, higher fusion rates were observed in B2A-coated ceramic granule (formulated as PREFIX®)-grafted patients than in ICBG-grafted patients after an interbody fusion procedure (Sardar et al. 2015). Studies are limited; Clinicaltrials.gov indicates two registered multicenter studies with “unknown” status. Validating the safety and efficacy of this bone graft material necessitates high-quality clinical studies and/or multicenter studies enrolling large number of patients. To date, there is only one published pilot study (Sardar et al. 2015, Canada) with a small/insufficient sample size.

Synthetic Materials and Drafts (Table 6)

Synthetic graft materials are typically employed during fusion surgery as bone graft extenders and sometimes substitutes. Traditionally, these materials provide an osteoconductive scaffold with ideally no reactive inflammatory immunogenic response from host tissues. The known advantageous properties of synthetic materials like ceramics include osteoconductive, biodegradable, no risk of infection, no donor site morbidity, unlimited supply, relatively easy sterilization, easiness of molding sized and shape, and lack of immunogenicity and toxicity (Gupta et al. 2015; Kannan et al. 2015). More recently, the emerging novel synthetics involve new technological advances in material science and/or incorporate a menagerie of cross-product materials in order to address the molecular biologic demands for bone induction, consolidation or healing, and fusion mass incorporation. Design innovation may lead to a true potent autograft substitute.

Table 6 Commercially available synthetic bone void fillers, extenders, and substitutes products
Full size table

For making an ideal bone graft extender or graft substitute, several characteristics should be considered. The development of products used for bone regeneration has followed the basic criteria of providing a biocompatible three-dimensional scaffold with controlled architecture capable of stimulating or supporting bone growth in the natural in vivo environment (O’Brien 2011). The ability of the material to be used in conjunction with other cellular and signal-based therapies (peptides, growth factors) is a key strategy in maximizing the efficacy and likely success of fusion. The primary characteristics of bone graft substitutes are shown in Table 6.

Calcium Phosphate Materials

Calcium phosphates are a common base for synthetic graft materials. This is primarily because 70– 90% of inorganic material in the body is a type of calcium phosphate. Calcium phosphate materials have been cleared in the USA for use as “bone void fillers” (FDA MQV, MBP) that can be used for spine fusion and orthopedic applications. The common types of calcium phosphate materials are beta tricalcium phosphate Ca3(PO4)2 (TCP) and hydroxyapatite Ca10(PO4)6 (OH)2 (HA).

TCP was one of the earliest synthesized forms of calcium phosphate materials that was used as an osteoconductive bone void filler. TCP in the form of granules or blocks is available as a three-dimensional structure with interconnected pores from 1 to 1,000 microns. However, TCP and all calcium phosphate materials are brittle, as they do not possess the tensile properties of bone. Therefore, TCPs and calcium phosphates have been used in areas of relatively low tensile stress or non-load-bearing applications. Thus, the calcium phosphate-based materials are not recommended alone for use in load-bearing applications (Park et al. 2013). It is important to recognize that most osteoconductive products have been approved for use only in posterolateral spine fusion applications and not in interbody fusion applications. Since TCP has only osteoconductive effects, these TCP-type products may be used in conjunction with biologic osteoinductive or osteogenic supplements of autograft, BMPs, growth factors, mesenchymal stem cell (MSC) derivatives, etc. (see combined products, Table 7) (Gupta et al. 2015; Duarte et al. 2017).

Table 7 Commercially available combination grafting products, naturally occurring peptides, growth differentiating factors, cellularized grafts, cellular bone matrices (CBMs)
Full size table

The most widely recognized TCP product is Vitoss® Bone Graft Substitute (Stryker, Allendale, NJ). This material was first commercialized in 2004, and its application in different formats has established it as the preferred TCP material. Another TCP-based material that has been reported is the Augment® Bone Graft from Wright medical. Augment® Bone Graft combines recombinant human platelet-derived growth factor B homodimer (rhPDGF-BB) with a bio-resorbable synthetic bone matrix (β-TCP). This product has been developed for use in bone repair. It is reported that the use of this product eliminates the need for using autograft, proposed as a “substitute.” However, Augment® Bone Graft is only indicated for use as an alternative to autograft in the ankle or hindfoot (Augment® bone graft – FDA. https://www.accessdata.fda.gov/cdrh_docs/pdf10/P100006d.pdf). There are several TCP-based products combined with different carriers to provide improved handling characteristics (see combined products, Table 7).

HA is another calcium phosphate material of significance, since x-ray diffraction and chemical studies have demonstrated that the primary mineral phase in bone is HA. HA is a biomaterial for medical devices and is available in the form of nanocrystalline powders, porous granules, and dense blocks. It can be manufactured from natural coral, bovine cortical bone, or synthesized by chemical reactions. HA is stronger (less brit tle) than TCP providing high compression strength but is still somewhat brittle. Due to its brittle quality, HA use is limited in load-bearing applications (Zdeblick et al. 1994; Park et al. 2013). Unlike autograft, allograft, and TCP, the absorption rate of HA is very slow (with incomplete absorption/resorption), and HA remains at the site of implantation for years (Zadegan et al. 2017a). In most circumstances, this prolonged resorption may not be advantageous. Grafting materials are ideally completely resorbed and replaced by new bone eventually. If the material does not resorb, it can act as an obstacle or inhibit new bone formation. Historically, coralline HA has been used effectively as a bone graft extender in patients as an adjunct to autologous bone for PLLF (Morris et al. 2018). The critical amount of graft volume per area of functional level (spine) has not been reported. Yoo et al. suggest that an amount of at least 12 mL of bone graft is needed to achieve a satisfactory bone fusion in minimal invasive TLIF surgery regardless of mixture ratio of HA with autograft bone (Yoo et al. 2015). There are several HA-based products combined with different carriers to provide improved handling characteristics (Tables 6 and 7).

According to study of Nickoli MS et al., ceramic-based bone grafts (TCP) with an osteoinductive stimulus represent a promising bone graft extender in lumbar spine fusion (Nickoli and Hsu 2014). In a meta-analysis review of 1,332 patients in 30 studies, from 1980 to 2013, ceramics used in combination with local autograft resulted in significantly higher fusion rates compared with all other adjuncts and bone marrow aspirate and platelet concentrates (Nickoli and Hsu 2014). Previous clinical studies on HA-based bone graft such as HA when used alone, or in combination with BAG (bioactive glass), BMA (bone marrow aspirate), or rhBMP-2 have been shown to improve function to the and reduce preoperative pain same extent as ICBG, yet have been associated with suboptimal radiographic fusion rates in lumbar spine (Singh et al. 2006; Acharya et al. 2008; Ploumis et al. 2010).

Silicate-Substituted Calcium Phosphate

Silicate-substituted calcium phosphate (Si-CaP) constitutes a newer generation of ceramics produced by adding silicate which has been found to play role in bone metabolism to previously developed calcium phosphate ceramics (Gao et al. 2001). This combination provides superior biocompatibility and osteoconductivity . In addition combining Si-CaP with a graft provides negative surface charge that results in enhanced osteoblast activity and neovascularization of the bone which lead to more ideal spine fusion as a substitute of ICBG (Campion et al. 2011; Alimi et al. 2017).

Silicated hydroxyapatite has been prepared by the addition of a small amount of silicon (0.4% to 0.8% by wt.) into the structure of HA. The role of silicate-based materials in improving tissue implant interactions has been reported (Zhou et al. 2017). Silica-substituted HA, such as Actifuse™ from Baxter, is available in the form of granules, pastes, and blocks. The performance of these products has been investigated in preclinical models and clinical study. According to study of Jenis and Banco (2010), a silica-substituted hydroxyapatite (Actifuse™) with BMA has been shown to be effective as a graft substitute as ICBG with significant pain improvement in PLLF. According to study of Licina P et al. (Licina et al. 2015), silicate-substituted calcium phosphate (Actifuse™) and rhBMP-2 with ceramic granule were comparable in view of achieving PLLF.

Clinical data are limited for various types of lumbar surgery and the numbers of enrolled patients in trials. For confirming the efficacy and safety of Si-CaP and/or silicated hydroxyapatite as a bone-grafting substitute, further investigations using greater numbers of subjects will be necessary. And the radio-opaque nature of Si-CaP allows for intra- and post-operative localization, but this radio-dense characteristic immediately after surgery resembling bone and the long residence time exceeding a year has decreased the accurate assessment of the process of bone formation.

Bioactive Glass (Table 8)

Bioactive glass (BAG) is a class of glass-based graft substitute or extender products having a compositional range that allows the formation of nanocrystalline hydroxyapatite (ncHA) as a surface layer when exposed to an aqueous phosphate-containing solution, such as simulated body fluid. The ncHA layer that forms within an aqueous phosphate-containing solution plays a significant role in forming a strong bond with natural bone.

Table 8 Composition and properties of bioactive glasses and glass-ceramics used clinically for ontological, musculoskeletal, and dental grafting applications (Baino et al. 2018; Hench and Jones 2015)
Full size table

BAG has an established history of bone bonding that occurs as a result of a rapid sequence of reactions on its surface when implanted into living tissues (Hench and Jones 2015). There are two mechanisms of bioactivity for bioactive glass products. Bone bonding is attributed to the (1) formation of an HA layer, which interacts with collagen fibrils of damaged bone to form a bond (Hench and Jones 2015), while the action of the (2) dissolution products from the bioactive glass is reported to simulate osteogenesis (Hench and Polak 2002). When hydrated, a layer of silica gel forms on the surface of the bioactive glass. The adhesion of amorphous calcium , phosphate, and carbonate ions to the silica surface leads to an eventual crystallization of a bone-like HA as early as 24 hours. Bone-forming cells migrate and colonize the surface of the bioactive glass and promote the production of a new bone-like matrix (Beckham et al. 1971). Gao et al. (2001) observed increased expressed detectable mRNA levels of BMP-2 from Saos-2 osteoblastic cells when cultured on two types of BAG (BAG containing 6% Na2O, 12% K2O, 20% CaO, 4% P2O5, 5% MgO and 53% SiO2 and biocompatible glass (BCG) containing 6% Na2O, 12% K2O, 15% CaO, 4% P2O5, 5% MgO and 58% SiO2 (wt.%)) than on control inert glass (Gao et al. 2001).

The mechanism for the formation of the ncHA layer is now quite well understood and well characterized, but the biological interactions at the ncHA–host bone interface are still under intense investigation in view of potential employment with stem cells (Tsigkou et al. 2014).

In addition, the high pH and the subsequent osmotic effect caused by dissolution of the bioactive glass have been suggested as an antibacterial material quality (Stoor et al. 1998; Allan et al. 2001). Recently, Sanchez-Salcedo et al. (2017) introduce the design and synthesis of a new nano-structured zwitterionic mesoporous bioactive glasses (MBGs) with incorporation with amino acid for antibio-fouling capability that inhibits bacterial adhesion (formation of biofilm) wherefrom they report successful results in vitro.

BAG has been used for a variety of clinical applications since it was first created in 1969 (Hench and Jones 2015). There are many types of BAG (Table 6) and glass-based products used (Hench and Jones 2015) in periodontal repair and orthopaedic applications (Table 8).

The originally developed composition was bioactive glass 45S5 (Food and Drug Administration (FDA) approved in 1993 (Jones 2015). 45S5 bioactive glass consists of 45 wt.% SiO2, 24.5 wt.% CaO, 24.5 wt.% Na2O, and 6.0 wt.% P2O5 which demonstrated effective biological properties. NovaBone® , a product based on this 45S5 technology, has been approved as a bone graft substitute in 1999 (Jones 2013; Hench and Jones 2015). The NovaBone® material is considered an early generation of bioactive glass. This is due to the lack of inherent porosity of the NovaBone® granules or granules in which porosity has been manufactured by the fusion of smaller granules. NovaBone® was compared to autograft in posterior spinal fusion procedures for treatment of adolescent idiopathic scoliosis in 88 patients (Ilharreborde et al. 2008). NovaBone® showed improved clinical results in terms of reduced infection, donor site complication, and fewer mechanical failures in a 4-year follow-up. However, its clinical use for spine fusion applications has not been reported widely.

A commercially available bioactive glass product is BonAlive® (BonAlive Biomaterials, Turku, Finland), which was programmed in Finland based on S53P4 bioactive glass. BonAlive® received European approval for orthopedic use as a bone graft substitute in 2006 (Jones 2015). The S53P4 bioactive glass contains 53 wt.% SiO2, 23 wt.% Na2O, 20 wt.% CaO, and 4 wt.% P2O5. According to Frantzen et al. (2011) a prospective long-term study (11 years) of Frantzen et al., the fusion rate of all fusion sites for BAG-S53P4 with autograft as a bone substitute was 88% at the L4/L5 level and 88% at the L5/S1 level compared to 100% for autograft in degenerative spondylolisthesis patients. Similar results were seen after surgical treatment of a spondylitis patient (Lindfors et al. 2010). BonAlive® was also compared to autograft in the same patients in PLF procedures for treatment of spine burst fractures. At the 10-year follow-up, 5 out of 10 implants had full fusion compared to all 10 autografts (Rantakokko et al. 2012).

Fibergraft® BG Morsels (Prosidyan Inc., USA) is a 100% BAG material (no additives) specifically FDA cleared for orthopedic and spine grafting applications. Traditional bioactive glass does not allow for ease of handling and has slow resorption due to low porosity. Fibergraft® BG Morsels is the first osteostimulative (or bioactive) material engineered to take advantage of the unique properties of bioactive glass. The morsels are engineered with overlapping and interlocking bioactive glass fibers with pores dispersed throughout. The material structure and ultra-porous, nano-, micro-, and macro-porosity provides direct connectivity for cell in-growth and material resorption, enabling new bone formation.

A 95% radiographic success rate was reported in a retrospective study of Fibergraft® BG Morsels use when mixed with local autograft and bone marrow aspirate in 63 patients at 1 year after 1-, 2-, and 3-level posterolateral fusions (Barcohana et al. 2017). Additionally, a high rate of 88.5% (46/52 levels with complete fusion) together with a 5.8% (3/52, levels partial fusion) in anterior cervical fusion was demonstrated after use of Fibergraft® BG Morsels mixed with BMA, bone dust, and or local bone in 27 patients (51 levels of fusion) at approximately 6 months after anterior cervical discectomy and fusion study (Fortier et al. 2017).

Fibergraft® BG Morsels (Prosidyan Inc., USA) is also provided in a putty form as Fibergraft® BG Putty and in a Matrix form as Fibergraft® BG Matrix. All Fibergraft® products are specifically FDA cleared for orthopedic and spine grafting applications. The BG Putty can be used for Minimally Invasive Surgery (MIS) applications, while the BG Matrix can be combined with bone marrow aspirate and used as a compression-resistant strip that can be molded to the shape of the defect.

Clinical and in vivo studies on commercially available bioactive glass particulates show that BAG can perform better than other bio-ceramic particles and have performed similarly to autograft in multiple in vivo studies (Walsh et al. 2017; Bedi 2017).

Unmet Challenges for Engineered Bioactive Glass Matrices

The major scientific and technical challenges exist with previously developed bioactive glass. Glass based materials lack osteogenesis, are difficult in clinical handling, not load bearing due to brittleness, and have slow resorption due to low porosity (Hench and Jones 2015; Jones 2015). To overcome these limitations and use BAG as effective substitute for autograft, several experiments were attempted to combat these limitations.

First, to enhance osteogenesis, tissue regeneration through gene activation by controlled release of inorganic ions from BAG is required. However, the role of the dissolution products from implanted BAG on bone marrow-derived mesenchymal stem cells (MSC) is not yet controllable. In some studies dissolution products induced osteogenic differentiation into osteoblast-like cells, and in others, it did not (Reilly et al. 2007; Karpov et al. 2008; Brauer et al. 2010). To control this problem, the fundamental mechanisms involved in ionic stimulation in the stem cell nucleus and the exact mechanism of “how the bioactive glass particles/dissolution products” should be explained (Hench and Jones 2015).

Second, particles and putties containing a variety of BAG particulates are in widespread clinical use, but large interconnected macroporous scaffolds for regeneration of large bone defects were not developed. To overcome and address this, the bottom-up sol–gel process, where gelation of nanoparticles in a sol (polycondensation) forms a glass network by avoiding sintering of crystalized Bioglass 45S5, was initially developed (Li et al. 1991). After, a room temperature gelation process was employed, allowing pores interconnection with a compression strength equivalent to porous bone (Jones et al. 2006). Melt-derived glass scaffolds were introduced to make macroporous scaffolds (Wu et al. 2011). According to a review by Hench and Jones (2015), none of described techniques are being further developed for use by medical device companies even though sol–gel and melt-derived scaffolds still exist.

Third, tissue-engineered constructs for replacement of large bone defects have been investigated for many years but are still not available as routine clinical products. To achieve this, a stable vasculature is necessary during initial grafting. Tsigkou et al. (2010) demonstrated that it is possible in mice models (Tsigkou et al. 2010). More research is needed to test the possible enhancement of angiogenesis optimal activity duration in humans (Azevedo et al. 2015).

Fourth, load-bearing devices that can be used in orthopedics over the long term, which also regenerate living bone, are still not available clinically. Therefore, the 3D printing technology was adapted to bioactive glass scaffolds to generate interconnected pores similar in diameter to the porous foam scaffolds but with higher compressive strengths (Fu et al. 2011; Kolan et al. 2011). However, BAG scaffolds are still brittle and therefore not suitable for all grafting applications, such as sites that are under cyclic loads.

Mixed Use Graft Materials with Antibacterial Effects (Table 7)

Infection Prevention and Treatment of Previous Surgical Site Infection

For improvement of bone graft materials including substitutes, dual-functional graft materials have been designed. Among several possible additional options, prevention or treatment of surgical site infection with/without bone destruction is needed for clinical application (Turner et al. 2005; Anderson et al. 2014). Risk factors associated with surgical conditions (relatively wide soft tissue dissection, muscular damage, long operation time, and limited control of bleeding during operation) and patient characteristics and health status (old age, comorbidities like diabetes mellitus, renal failure and vasculopathy, and smoking, etc.) in spine fusion operations.

For prevention or control of the post-operative infection, systemic and localized bactericide are necessary. However systemic delivery of antibiotics to infected site or vulnerable to infection is limited by abnormal blood supply in operated site, drug toxicity to organs, antimicrobial-resistant form of bacteria, etc. (Shiels et al. 2017). Due to mentioned causes, newly designed graft materials have been developed for local bactericidal carrier, which may increase the safety and satisfaction after treatment (Lentino 2003; Radcliff et al. 2015).

A variety of materials including calcium-based substitutes, synthetic polymers, DBM, and protein-based materials have been proposed as alternative delivery vehicles with bone fusion function (McLaren 2004; Nelson 2004). Because the most common pathogen responsible for spinal infections after surgery is the gram-positive bacteria Staphylococcus aureus , the antibiotic candidates for biomaterials for infection-targeted delivery (or prevention) may be limited to vancomycin, aminoglycoside series like tobramycin, gentamicin, amikacin, and quinolone series like ciprofloxacin (Turner et al. 2005; Logoluso et al. 2016; Shiels et al. 2017; Boles et al. 2018; Wells et al. 2018).

Several animal studies have shown that calcium sulfate pellets are substantially resorbed and replaced with new bone formation by 6 weeks and a similar rate of pellet resorption has been reported clinically (Turner et al. 2001; McKee et al. 2002). According to study by Shiels SM.et al., vancomycin continued to be released from the DBM over the course of 6 days while maintaining sufficient eluate concentrations to maintain a zone of inhibition similar or larger than a vancomycin control in spine fusion in rabbit (Shiels et al. 2017).

There are several obstacles to overcome in order to use this newly designed bone graft material in clinical spine fusion. First, the ideal shape, desired materials of bone graft, and release concentrations are not established. McLaren et al. questioned the effect of laboratory sampling methods on characterizing the elution of tobramycin from calcium sulfate and the reliability of in vitro elution data in predicting the in vivo release of antibiotics (McLaren et al. 2002). Second, local site effects by eluted antibiotics are of concern. Since neither the optimal level of antibiotic nor the duration of its release has been established, the effect of high local levels of antibiotics on the ability of grafted material to enhance bone healing is largely unknown. In a rabbit study, the use of vancomycin-loaded DBM showed a decrease in the fusion rate compared to DBM when used in a sterile wound (Shiels et al. 2017). Furthermore, an in vitro study suggests that vancomycin has toxic effects on hMSCs, a cell population particularly important for bone formation (Chu et al. 2017). Finally, clinical studies on the use of antibiotic-impregnated graft materials for spine fusion in humans are few. Pilot studies focused on the use of antibiotic-impregnated graft material in total joint arthroplasty and osteomyelitis (Logoluso et al. 2016) (Table 7).

Conclusion

A wide variety of bone graft materials are used in spinal surgery applications. Increasingly, over the past decade, diverse materials and composites are being developed as grafting options for use in spinal surgery. Consideration of the ideal properties of a grafting material and the material’s mechanism of action, structural and handling characteristics, FDA classification and related approval or registration, and available clinical and preclinical data will optimize appropriate grafting choice for a certain surgical application for spinal fusion. Moreover, bone grafts do not fuse immediately; instead, they provide a foundation or scaffold for the patient’s body to grow new bone in anatomical sites wherein bone did not previously exist such as in a spinal fusion site.

The development of products used for bone regeneration has followed the basic criteria of providing a biocompatible three-dimensional scaffold with controlled architecture capable of stimulating or supporting bone growth in the natural in vivo environment. The ability of the material to be used in conjunction with other cellular and signal (growth factors)-based therapies is a key strategy in maximizing the efficacy and likely success of fusion. However, while many bone graft substitutes perform well as bone graft extenders, only autogenous bone grafts are osteogenic and BMPs are osteoinductive .

Variations in anatomical location, surgical application (meticulous surgical preparation including adequate decortication), instrumentation type, and the patient’s risk factors (metabolic and nutritional status, vitamin D, diabetes, smoking, drug and alcohol abuse) are critically important factors to consider in choosing an ideal grafting agent or bone graft to achieve a successful biologic bone union.