ABSTRACT
Background: Back pain is a common chief complaint within the United States and is caused by a multitude of etiologies. There are many different treatment modalities for back pain, with a frequent option being spinal fusion procedures. The success of spinal fusion greatly depends on instrumentation, construct design, and bone grafts used in surgery. Bone allografts are important for both structural integrity and providing a scaffold for bone fusion to occur.
Method: Searches were performed using terms “allografts” and “bone” as well as product names in peer reviewed literature Pubmed, Google Scholar, FDA-510k approvals, and clinicaltrials.gov.
Results: This study is a review of allografts and focuses on currently available products and their success in both animal and clinical studies.
Conclusion: Bone grafts used in surgery are generally categorized into 3 main types: autogenous (from patient's own body), allograft (from cadaveric or living donor), and synthetic. This paper focuses on allografts and provides an overview on the different subtypes with an emphasis on recent product development and uses in spinal fusion surgery.
INTRODUCTION
Back pain is a common chief complaint within the United States and is caused by a multitude of etiologies. According to the National Center for Health Statistics, more than 650 000 spinal fusion surgeries are performed annually.1 The success of arthrodesis in spine surgery depends on multiple factors; however, an important component to success depends on the bone graft and graft substitutes used in surgery. 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/or to provide initial structural support.2 The graft material used in spinal fusion procedures can be categorized generally into 3 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.3,4 Autograft is considered the “gold standard”; however, the authors believe allograft and synthetics are currently replacing separate surgical site–harvested autografts as the standard because of patient donor site morbidity, advancement in the development of other products, limited quantity available, host limiting bone quality, and lack of training of young surgeons in the technique of autograft harvest.4 This paper focuses on allografts and recent advancements in product development and uses in spinal fusion surgery.
BONE GRAFT PROPERTIES
Osteogenesis, or bone formation, occurs via 2 mechanisms: endochondral ossification or intramembranous ossification. Intramembranous ossification involves direct conversion of mesenchymal tissue into bone, does not require cartilage as an intermediate, and does require bone morphogenic proteins and CBFA1 transcription factors.5 Endochondral ossification requires cartilage as an intermediate and can be divided into 5 stages. The stages of endochondral ossification are as follows:—mesenchymal cells differentiate to cartilage cells, formation of chondrocytes, proliferation of chondrocytes to form the model for the bone, formation of hypertrophic chondrocytes, and invasion of blood vessels.5 Both mechanisms of bone formation rely on complex intracellular signaling events, and each contributes to bone formation after spinal fusion surgeries.
Critical elements that are required for bone formation and are important in bone graft properties include osteoconduction, osteoinduction, osteogenesis, mechanical stability, and vascularization. Osteoconduction relies on a scaffold that supports cell ingrowth, facilitates vascularization, and provides a network for cells to attach.4 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.4 Osteogenesis relies on the immediate provision of viable cells emanating from the host to the defect site differentiating into bone-forming cells.4 Autologous bone is the only bone graft available that intrinsically contains all 3 properties and is therefore considered the “gold standard.” However, with advancements in allograft processing and development, recent products have theoretically been able to acquire all 3 properties of bone development.
ALLOGRAFT
Allograft bone is obtained from either living or deceased donors and then processed for sterility. Common preparation includes freezing or lyophilization (ie, freeze drying), which involves dehydration and vacuum packaging to store at room temperature.6 In general, allografts are primarily osteoconductive with minimal osteoinductive potential and are traditionally not osteogenic because the donor cells are eradicated during processing.7,8 Surgeons prefer allografts because they are readily accessible, available in various forms delivering handling properties, facilitate bone formation, and do not require donor site morbidity. However, traditionally available allografts consist of nonviable tissue and cannot stimulate bone formation without the addition of bone-stimulating factors and cells.9–11 These limitations lead to slower and less complete incorporation with native bone. Additionally, allografts have a potential risk of disease transmission even if the incidence is very low and the risk can be controlled during the procurement and sterilization process.7 Allogenic bone is traditionally available in many forms: cortico-cancellous, demineralized bone matrix (DBM), morselized and cancellous chips, and osteochondral and whole bone-segments.12
Recently, a new class of allograft has emerged called viable cellular allografts or cellular bone matrices (CBMs), which are designed to have all 3 properties of bone formation: osteoconduction, osteoinduction, and osteogenesis. CBMs are created using osteoconductive cadaveric bone with the retention or addition of allogeneic stem cells (ie, mesenchymal stem cells) to initiate an osteogenic process.13 The efficacy of mesenchymal stem cells has been shown to be as efficacious as rhBMP and allograft.12 Overley et al14 retrospectively examined 78 patients (98 fusion levels) and found no difference in radiographic fusion and rate of revision surgery in patients who underwent MS-TLIF with either rhBMP-2 or CBM as fusion adjuncts.
Table 1 provides a description of the different types of allografts and their corresponding characteristics. For a more thorough description of all classes of bone grafts, please refer to the chapter by Yang et al4 in the Handbook of Spine Technology or the review article by Gruskin et al.3 A comprehensive review of bone graft characteristics can be found in the chapter by Bae et al23 in AAOS Comprehensive Orthopaedic Review 2.
Cortico-Cancellous Allograft
Cortico-cancellous allografts are the most commonly used allograft today. They are strictly osteoconductive without any osteoinductive or osteogenic properties. These grafts can be prepared as whole pieces, such as rings of femoral head/neck used traditionally for interbody fusion, or prepared as chips to aid in void-filling scenarios or posterolateral fusion.24 Cortical allograft is most commonly used as a mechanical strut graft, whereas cancellous allograft functions as a osteoconductive scaffold for bone formation.4 In a study by Park et al,25 46 patients underwent ACDF with either a cortico-cancellous allograft or iliac crest autograft, and there was no significant difference in fusion status between the 2 groups. Another study by Suchomel et al26 evaluated fibular allografts versus autologous iliac crest grafts in 80 patients undergoing ACDF procedures and found in single-level procedures that there was no difference in fusion rates and graft collapse between autograft and allograft. Table 2 provides a list of commercially available cortico-cancellous allografts used for spinal fusions and specifics on each product.
Demineralized Bone Matrix
DBM is derived from human allografts and prepared by acid extraction of innate minerals to create an osteoconductive organic matrix with differing quantities of proteins that aid in osteoinduction.5 DBM is a composite of collagens (mostly type I), noncollagenous proteins and growth factors, residual calcium phosphate mineral (1%–6%), and some cellular debris.3 The use of DBM was developed in 1965 by Urist,29 who observed that soluble signals contained within the organic phase of bone were capable of promoting bone formation. After processing, DBM lacks structural integrity but retains osteoconductive and osteoinductive properties.3 DBM base is available, and when mixed with other substances these DBM-based products come in many forms, including powders, granules, gels, putties, and strips.30 Importantly, the concentration of native BMP in DBM products differs significantly by manufacturer, donor lot, and batch, making it difficult to study the efficacy of DBM in clinical trials.30–33 In an athymic rat model, Bae et al34 observed significant lot-to-lot variability of a single DBM-based product, commercially available “off-the-shelf” with regard to BMP concentrations and associated in vivo bone formation for fusion rates.35 Therefore, it is important to note the efficacy of DBM's osteoconduction and osteoinduction properties in clinical studies is limited by mainly narrative study designs with limited levels of evidence, small sample size, and lack of appropriate controls.26
In current clinical practice, because DBM-based products lack structural integrity, they are exclusively/mostly used in spinal applications as bone graft extenders, typically mixed with surgical site local bone or morselized harvested bone from the iliac crest (ICBG) autografts, and/or exogenous peptide/differentiation factors (rhBMP-2) to promote bone growth.4 Kang et al36 completed a 2-year prospective randomized clinical trial comparing outcomes of Grafton DBM with local bone to those of ICBG in a single-level instrumented posterior lumbar fusion. In the study, 46 patients (30 Grafton, 16 ICBG) were evaluated, and primary outcome was solid posterolateral lumbar fusion. Results indicated no significant difference in overall fusion rates between the 2 study groups (86% for Grafton, 92% for ICBG). Lower blood loss was recorded in the patients who received an implant of DBM-base matrix (Grafton DBM-Matrix), but with equal or slightly greater improvement in Oswestry Disability Index scores for the DBM-base matrix patients.
Cammisa et al37 completed a multicenter, prospective, side-to-side (right versus left) comparison of a DBM-based gel (Grafton DBM gel) combined with iliac crest autograft (2:1 ratio) placed on one side of the fusion construct versus iliac crest autograft alone on the other side in 120 patients who underwent posterolateral spinal fusion (PLF) procedures. At 24 months nearly equivalent fusion rates between the sides implanted with a composite of DBM-based gel (Grafton DBM gel) + one-third iliac autograft were 52% fused (42 of 81 sides) versus contralateral sides at 54% fused (44 of 81 sides) after being implanted with autograft alone. Specifically, radiographically fused rates of 40.7% bilateral (33 of 81 consistently both right and left sides fused) or 24.7% unilateral (only autograft side fused in 14% [11 of 81] versus DBM-based gel + autograft composite fused in 11% [9 of 81]). Although 34.6% (28 of 81) were not radiographically fused, the pseudarthrosis revision surgery rate was <1% (1 of 81). Interestingly, the fusion rate in this study is substantially lower than the accepted solid fusion rate of PLF surgery (90%),38–40 and the authors ascribe this discrepancy to a difficult patient population, strict radiologic criteria for fusion, and only evaluating bone graft lateral to the instrumentation on anterioposterior film.
These authors conclude that DBM-based allograft products may be used to augment the amount of autograft bone graft needed for successful lumbar fusion. Cammissa et al37 report that one third the quantity of autograft may be used with this DBM-based gel graft extender to achieve consolidated bony fusion, and Kang et al36 used 15 to 20 cm3 of autograft with DBM compared with 25 to 30 cm3 ICBG for successful fusion. The studies by Kang et al and Cammissa et al provide level 1 evidence that Grafton can be used as a bone graft extender for lumbar spinal fusion.37
Interestingly, Grafton is the only bone graft extender to have level 1 evidence and shows different efficacy in the lumbar spine versus the cervical spine. As the above studies showed Grafton to be effective in lumbar spinal fusions, a study by An et al showed level 1 evidence that Grafton DBM is not useful for cervical spinal fusion.41,42 In a randomized control trial, An et al42 compared 77 ACDF patients with either Grafton DBM + tricortical bone versus tricortical bone alone. Nonunion developed in 46% of patients in the Grafton group compared with 26% of patients in the standalone tricortical bone group.
Table 3 provides a list of commercially available DBM-based products used for spinal fusion and specifics on each product.
Viable Cellular Allografts (Cellular Bone Matrices)
The advancement in the field of stem cell procurement has generated the development of allogenic bone grafts containing live mesenchymal stem cells (MSCs), also known as cellular bone matrices.63 Mesenchymal stem cells were identified in 1966 by Fridenstein et al in bone marrow and have been shown to differentiate into chondroblasts and osteoblasts.63,64 These commercially available bone allografts are composed of osteoconductive partially demineralized cadaveric bone as matrix carriers with components of cryopreserved allogeneic cells (MSCs) that promote osteogenesis and osteoinduction.6,65 MSCs can be isolated from bone marrow, placenta, umbilical cord blood, connective tissue, skin, synovial fluid, fat, and teeth.63,66 MSCs are capable of evading the immune system because they uniquely do not express human leukocyte antigen class II molecules, which are essential for activation of the cellular immune response.63,67–69
In the United States, the process to manufacture these materials involves the American Association of Tissue Banks (AATB) approval processes for cadaveric human bone recovering (contract with independent US Food and Drug Administration [FDA]–registered tissue recovery groups), processing, storing, and preserving cellular components of the bone, or addition of cells, and removal of noncellular proteins. Marketed under FDA-HCT wherein the regulation of product directive is safety, safety is exercised by restricted donor screening. Unlike other DBM-based allografts approved via 510(k) or premarket approval pathways, CBMs are not required to be terminally sterilized, relying on the donor screening and aseptic processing to ensure safety. The exact procedures vary by manufacturer. The HCT/P classification does not require lot-to-lot cell composition or validation of growth factor production. (Per FDA guidance documents on HCT/P products, to “rely on the metabolic activity of living cells for their primary function” would render a product as a biologic drug [section 360], which would require a biologic license application and clinical trials.)
CBMs are commercially provided as frozen products and must be stored at −80°C, and they require thawing prior to surgical implantation. Neither the reproducibility of cell recovery, after thaw, nor the viability of the cells following implantation has been established for commercially available products or production lots of them. The average number of cells across products is claimed to range from 66 000 to 3 million. Attempting to preserve the viable cells, these products are not terminally sterilized, like 510(k) DBMs or premarket approval products, but rely on aseptic processing to ensure safety. Table 4 provides a list of commercially available CBMs used for spinal fusion and specifics on each product.
The efficacy of viable cellular allografts in spinal fusion is difficult to determine. Given the properties of mesenchymal stem cells, their ability to promote osteogenesis, and their ability to evade the immune system, it is reasonable to think they would be advantageous for bone fusion. Several in vivo studies have demonstrated theoretical benefits of using CBMs. Cui et al90 compared cloned osteoprogenitor cells to mixed marrow cells and found that cloned cells produced a greater amount of mature osseous tissue at an earlier time point during spine fusion in an athymic rat model. Gupta et al91 used an ovine posterolateral lumbar fusion model and found similar fusion rates with osteoprogenitor-enriched graft compared with autograft.
To date, there have been very few non–industry-sponsored clinical trials. McAnany et al92 evaluated 57 patients who underwent a 1- or 2-level ACDF using interbody allograft with Osteocel (NuVasive, San Diego, California). The patients were matched to a control group of 57 patients where only interbody allograft was used. At the 1-year follow-up, 87% in the Osteocel cohort had solid fusion compared with 94.7% in the control group.81
There are many factors that can influence the efficacy of CBMs and therefore result in limitations to these products. Hernigou et al93 showed that bone marrow aspirates containing fewer than 1500 MSC/cc were ineffective for the treatment of tibial nonunion, suggesting that this is the minimal MSC concentration for bony healing.63 Preparation of MSCs is not standardized, and variation in donor age, donor site, and viability of stem cells after thawing the allograft can all influence the effectiveness of CBM.
Although human clinical data are lacking, the athymic rat model allows for direct testing of CBM bone graft products. In a well-established rat model, fusion is assessed by manual palpation of a bony mass 6 to 8 weeks after implantation of DBM-based or CBM-based graft placed between the transverse processes during a posterolateral fusion procedure.33,34,76 Using this model, Bhamb et al33 reported at 8 weeks a 0 of 16 fusion rate in rats implanted with CBM Osteocel Plus Pro (NuVasive) compared with 88% to 100% fusion rate after noncellular human DBM–based products were implanted (Acell Evo3, DBX Mix, DBX Strip, Grafton Crunch, Grafton Flex, and Grafton Matrix), and 13% (2 of 16) were manually fused when implanted with syngeneic bone graft. Lin et al76 at 6 weeks detected manual palpation fusion in 73% (11 of 15) of Cellentra (Zimmer Biomet, Warsaw, Indiana), 53% (8 of 15) of Trinity Elite (OrthoFix, Lewisville, Texas), 13% (2 of 15) of Vivi-Gen (Dupey Synthes, Raynham, Massachusettes) and 0 of 15 for each of Osteocel Plus Pro, Bio4 (BIO, Stryker, Kalamazoo, Michigan), and Map3 (RTI Surgical, Marquette, Michigan) implanted rats; 33% (5 of 15) were manually fused for syngeneic bone graft–implanted rats. Johnstone et al80 recently evaluated manual palpation results after posterolateral fusion performed with several commercially available human CBM grafts; the highest manual palpation fusion rates were 71% (10 of 14) for Trinity Evolution and 77% (10 of 13) for Trinity Elite compared with 7% (1 of 14) for Osteocel Plus Pro and 40% (6 of 15) for syngeneic bone–implanted rats.
The findings of these studies highlight the variability among CBM commercial products and potentially among production lots.32 Yet, interestingly, there are common observations across these studies: Syngeneic bone graft, a proxy for ICBG in this model, containing some live cells yielded low fused rates (13%, 33%, and 40%) across the studies, whereas the Osteocel Plus preparations consistently yielded almost no sites fused (0%, 0%, and 7%). Trinity Elite (53% and 77%) and Trinity Evolution (71%) formulations, along with Cellentra (73%), seem to yield somewhat comparable results. Preparations, sterilization, formulation, manufacturing processes, and the donor bone itself contribute to differences among the products. Unlike for clinical use, for purposes of testing in rats, the products are not mixed with autograft bone. Clinically these allografts are not used in isolation; bone from the surgical dissection is morselized and mixed with allograft bone graft extenders of DBM-based or CBM products. Autograft potentially compensates for allograft products' debility.
In conclusion, there is no definitive clinical evidence that viable cellular allografts promote increased fusion compared with regular allograft DBM-based products. The claimed advantage of osteoinductive and osteogenic properties remains theoretical. The CBM allografts are available at an increased cost compared with other allografts, and more research is needed to justify each product's use instead of standard allografts.
CONCLUSION
Use of bone grafting techniques performed in surgical procedures for spinal fusion have been reported since the beginning of the 20th century.94 Novel instrumentation and surgical techniques are designed and inspired by advances in grafting technologies. The history of allografts dates back many decades and has greatly evolved since Urist's first observation on how bone demineralization impacts the incorporation at the graft-host interface. The advancements in allograft products have purposely been designed to facilitate surgical procedures for fracture healing and spinal fusion. Allografts vary in size, shape, consistency, strength, viable cellular components, and many other properties. For the past decades, with a clinical history of use of allografts and DBM bases, diverse materials and composites have been continually combined with various materials and allograft forms to improve material properties, and further developed as novel grafting options.4 Once bone allografts in various forms are approved or cleared by the responsible agency of the intended market country, they are rapidly adopted into specific surgical application. However, graft-contributing complications may still occur, and sometimes systematically with the use of a particular form or product. The target allograft then moves from bedside to “bench” for reevaluation. In the application process for approval, grafts' safety is evaluated, yet the osteoinductive and conductive effectiveness burden may not have been investigated, which likely has unintended consequences.
As discussed in this paper, there is a great unmet need for improvement in allografts. The ideal bone graft provides a biocompatible scaffold that promotes osteoconduction, osteoinduction, and osteogenesis. Allografts predominantly contain osteoconductive and some osteoinductive potential. However, with the advancement of viable cellular allografts, there is the theoretical addition of osteogenesis. Allografts with blood-derived augmentation (either with BMAC or PRP) were not directly discussed in this paper because this subject is the primary focus of another article in submission. With time and technologic advantages in stem cells and differentiation factors, more products will be developed to enrich the bone graft–to–fusion process and increase the rapidity and success rate of bone healing after spinal fusion procedures.
Footnotes
Disclosures and COI: The authors received no funding for this study and declare no conflicts of interest.
- This manuscript is generously published free of charge by ISASS, the International Society for the Advancement of Spine Surgery. Copyright © 2021 ISASS
References
- 1 .↵
- 2 .↵
- 3 .↵
- 4 .↵
- 5 .↵
- 6 .↵
- 7 .↵
- 8 .↵
- 9 .↵
- 10 .
- 11 .↵
- 12 .↵
- 13 .↵
- 14 .↵
- 15 .
- 16 .
- 17 .
- 18 .
- 19 .
- 20 .
- 21 .
- 22 .
- 23 .↵
- 24 .↵
- 25 .↵
- 26 .↵
- 27 .
- 28 .
- 29 .↵
- 30 .↵
- 31 .
- 32 .↵
- 33 .↵
- 34 .↵
- 35 .↵
- 36 .↵
- 37 .↵
- 38 .↵
- 39 .
- 40 .↵
- 41 .↵
- 42 .↵
- 43 .
- 44 .
- 45 .
- 46 .
- 47 .
- 48 .
- 49 .
- 50 .
- 51 .
- 52 .
- 53 .
- 54 .
- 55 .
- 56 .
- 57 .
- 58 .
- 59 .
- 60 .
- 61 .
- 62 .
- 63 .↵
- 64 .↵
- 65 .↵
- 66 .↵
- 67 .↵
- 68 .
- 69 .↵
- 70 .
- 71 .
- 72 .
- 73 .
- 74 .
- 75 .
- 76 .↵
- 77 .
- 78 .
- 79 .
- 80 .↵
- 81 .↵
- 82 .
- 83 .
- 84 .
- 85 .
- 86 .
- 87 .
- 88 .
- 89 .
- 90 .↵
- 91 .↵
- 92 .↵
- 93 .↵
- 94 .↵