Journal Logo

Supplement Article

Bone Graft Substitution and Augmentation

Nauth, Aaron MD, MSc*; Lane, Joseph MD; Watson, J. Tracy MD; Giannoudis, Peter MD§

Author Information
Journal of Orthopaedic Trauma: December 2015 - Volume 29 - Issue - p S34-S38
doi: 10.1097/BOT.0000000000000464



There are a variety of bone grafts and bone graft substitutes available when treating patients with nonunion or bone defects. Appropriate selection of a graft requires recognition of the specific bone healing needs of the patient and an understanding of the particular bone healing properties of the bone graft or bone graft substitute. This review provides a framework for selecting an appropriate bone graft or substitute, in addition to a detailed review of a number of promising alternatives to autogenous iliac crest bone graft (AICBG), including reamer–irrigator–aspirator (RIA) bone graft, bone marrow concentrate, bone morphogenetic proteins (BMPs), and calcium phosphate cements.


Selecting the appropriate bone graft or bone graft substitute for your patient requires careful recognition of the bone healing needs of that patient's specific clinical problem and a thorough understanding of the different properties that the available bone grafts and substitutes possess. For example, the desired properties of a bone graft or substitute for the treatment of a contained metaphyseal defect in the proximal tibia that arises after elevation of a depressed tibial plateau fracture (Fig. 1) are distinctly different from those required to treat an infected segmental bone defect in the femur (Fig. 2). Bone grafts and bone graft substitutes are generally evaluated by their ability to address the critical components of fracture healing, principally: osteogenesis, osteoconduction, osteoinduction, and vascularity. Osteogenesis refers to the delivery of stem and precursor cells, which can differentiate into bone. Osteoinduction refers to the stimulation of bone formation by powerful proteins (such as BMPs). Osteoconduction refers to the provision of a scaffold to support the growth of bone and vasculature. Additionally, the provision of mechanical stability, harvest-site morbidity, clinical results, and cost are important issues that need to be considered when selecting a bone graft or a substitute.

Preoperative (A) and 6 months postoperative (B) radiographs of a 63-year-old female patient with a lateral split-depression tibial plateau fracture treated with lateral plating and calcium phosphate grafting of the subsequent bone defect (white arrow). In this setting, which includes a subarticular contained metaphyseal defect, the desired graft properties are osteoconduction and compressive strength, both of which are provided by calcium phosphate cement.
A, Initial radiograph of an infected femoral bone defect in a 42-year-old male that occurred after resection of spontaneous osteomyelitis, with placement of antibiotic beads and external fixation. B, Postoperative radiograph after further debridement, cephalomedullary nailing, and insertion of an antibiotic cement spacer using the induced membrane technique. C, Three-month postoperative radiographs after insertion of RIA bone graft combined with platelet-derived growth factor (PDGF) showing early healing of the bone defect (note: this represents an off-label indication for PDGF). This biologically challenging bone defect requires osteoconduction, osteoinduction, osteogenesis, and vascularity, which are provided by the combination of RIA bone graft and PDGF.

(see Table, Supplemental Digital Cotent 1, provides an overview of the clinically available bone grafts and bone graft substitutes along with a comparison of the properties that each possess. AICBG remains the gold standard, as it possesses all the critical components of fracture healing, and furthermore provides mechanical stability, has a low cost, and has extensive support in the literature.1 However, harvest-site morbidity has been an issue with AICBG. RIA is a novel system for harvesting intramedullary reamings as bone graft from the canal of the femur or tibia. There is emerging basic science evidence, which suggests that RIA possesses equivalent osteoconductive and vascular properties to AICBG, with potentially superior osteoinductive and osteogenic properties.2,32,3 In addition, recent clinical evidence indicates that RIA harvest can produce larger volumes of graft with less harvest-site morbidity and pain.4 Selecting a bone graft or bone graft substitute is based on the specific clinical needs of the patient and the desired graft properties to address those needs (Figs. 1, 2). Several important bone graft substitutes are discussed in detail below.


Bone marrow aspirate contains bone-forming cells, physiologically relevant amounts of growth factors, established osteogenic and osteoinductive activities, and proven clinical efficacy. Bone marrow cells can be used as fresh aspirate, concentrated by filtration, or culture expanded as autologous stem cells or allogenic cells. Mesenchymal stem cells (MSCs) are present in bone marrow aspirate and have the potential to form site-specific tissue. In the appropriate environment, these progenitor cells can form bone, cartilage, tendon, and ligament. Concentrated MSCs are 3-fold more effective than bone marrow aspirate in their osteogenic activity. Both bone marrow aspirate and autologous stem cells are effective in stimulating osteogenesis and osteoinduction in the setting of fracture healing. These cell populations are often combined with an osteoconductive scaffold or osteoinductive proteins, particularly in the setting of a bone defect.

Hernigou et al5 concentrated bone marrow aspirate from the iliac crest of tibial nonunion patients and reported a 7-fold increase in the concentration of progenitor cells. In a series of 60 patients with established tibial nonunions, 53 went on to successful union when given a percutaneous injection of concentrated bone marrow aspirate. All the successful cases had more than 1,500 progenitors per cubic centimeter. The 7 failed cases had less than 1,500 progenitors per cubic centimeter. This series clearly demonstrated that the number of progenitor cells available in the graft influenced healing rates.

Desai and Lane et al reported a further extension of this technique in a series of 49 consecutive patients, including all gap sizes, all long bones (humerus, femur, tibia), and a combination of bone marrow aspirate concentrate (BMAC) with either BMP or demineralized bone matrix (DBM).6 DBM contains smaller amounts of BMP; however, it also contains vascular endothelial growth factor—a noted angiogenic growth factor. The average age of the patients was 54 years (67% females), with 20 of the 49 patients being exposed to at least 1 year of bisphosphonate therapy. The gap size averaged 5 mm but ranged from 1 to 20 mm. Sixty percent of the fractures were treated within 6 months of injury. Twenty-two percent were open fractures, and 26% were comminuted.

Overall, 79.6% of the fractures treated with BMAC in this series healed. BMP and DBM had healing rates of 71% and 86%, respectively, and were significantly different (P = 0.033). Gaps of <5 and >5 mm had healing rates of 81% and 78%, respectively, with no significant difference detected. Univariate analysis showed that union rate was not affected by age, fracture location, fracture gap size, or bisphosphonate use. There was a decreased likelihood of healing when BMP was used in combination with BMAC versus DBM with BMAC (P = 0.033). Similarly, late intervention (>6 months) resulted in poorer healing (P = 0.04). Increased age, a larger gap size, and bisphosphonate use all negatively affected time to healing, but not to overall union.

This series demonstrated that bone marrow concentrate was effective at treating long-bone nonunions (including those with gaps >5 mm) when combined with BMP or DBM. DBM was statistically superior to BMP in this series. In addition, BMP use was associated with increased swelling/seroma formation and was markedly more expensive. Consequently, DBM seems to be the preferred adjuvant to bone marrow concentrate.

There have been reports in the literature on autologous stem cells (either culture-expanded or isolated by cell-sorting techniques) for bone healing and regeneration; however, this area of clinical investigation remains in its infancy.7,87,8

Issues related to bone marrow concentration or stem cell transplantation that are yet to be resolved relate to patient donor limitations and isolation techniques. At this time, there are no readily available laboratory tests that identify good versus poor donors, and the clinician must therefore rely on the history and character of the donor. Colony-forming units are also insufficient for characterizing bone marrow concentrate. The appropriate mix of cells in the concentrate has yet to be determined. Furthermore, the ideal technique for cell concentration has not been identified. All the currently approved concentration techniques do provide concentration of MSCs, but their efficiency varies greatly. To overcome deficiencies in donors, current research is now being directed toward allogenic stem cells from ideal donors and has expanded far beyond the capacity of concentration alone. Nevertheless, bone marrow concentrate in its currently available form seems to offer a significant potential to enhance healing.


Some of the most important members of the transforming growth factor β superfamily are BMPs. BMPs are not only known to play a crucial role in organogenesis during embryonic development but also regulate important physiological processes of bone and cartilage tissue by having a direct effect on migration, proliferation, and differentiation of osteoprogenitor cells.9 After the important work of Gourjon et al,10 Urist et al,11 and other intense experimental and clinical research activities, the US Food and Drug Administration approved the application of BMP-2 and BMP-7 for the treatment of open tibial fractures and tibial nonunions, respectively.

At present, BMPs are considered the most potent inductive molecules promoting bone repair. Several clinical studies have shown that both BMP-2 and BMP-7 can promote the growth of large volumes of bone where an impaired fracture healing response has been established.12–1412–1412–14 Unfortunately, despite abundant evidence for the effectiveness of BMPs, these molecules have not dominated strategies for bone regeneration in the clinical setting. Moreover, in 2013, Olympus Biotech withdrew BMP-7 from the market. This event created a great deal of uncertainty about the future of BMPs, and questions were raised as to whether they still have a role to play in the clinical management of fracture healing and bone regeneration.

One therefore has to consider the following questions: What went wrong with BMPs? Why did these molecules fail to position themselves as the key players for promoting bone repair? Do they still have a role to play?

There are several issues that may have contributed to the poor reputation of BMPs. First, the industry rushed to position these molecules as the most powerful osteoinductive agents for fracture healing. They were presented as a “magic bullet” to restore bone continuity and the functional capacity of the affected extremity with limited knowledge of their functional mechanisms. Second, the degree to which they can build bone by themselves is unclear, which led to unrealistic expectations. Third, many important issues lacked clarification, such as the ideal dosing of BMPs, timing of administration, and indications for combination with other graft materials. Fourth, many clinicians did not appreciate that several prerequisites must be present in order for BMPs to be successful, including the availability of osteoprogenitor cells at the fracture/nonunion site, containment of the active molecule where needed, and an adequate blood supply to the site of desired bone regeneration. In addition, clinicians failed to recognize that activation of the osteoclastic pathway can be induced by BMPs instead of the osteoblastic pathway required for osteogenesis, which can lead to bone resorption and failure of treatment.15

Finally, when it was discovered that BMPs are not truly superior to the gold standard of all grafting materials, AICBG, it was claimed that BMPs are equivalent in efficacy to AICBG without the morbidity associated with bone graft harvest. However, even this claim was not scientifically valid, as BMPs possess only one property for fracture repair (osteoinduction). In fact, they need a carrier for their delivery and the naturally existing anti-inhibitory molecules of fracture repair can neutralize their action.16 In contrast, AICBG contains 3 of the critical components of fracture healing (osteoinduction, osteoconduction, and osteogenesis). Moreover, AICBG is naturally available from the patient, whereas BMPs are made from recombinant DNA technology. Unlike BMPs, AICBG does not require determination of the ideal carrier, the optimal length of the graft at the fracture/nonunion site, or how containment can be optimized. Finally, with AICBG, the clinician does not have to worry about allergic reactions or even formation of serum antibodies, as has been reported with BMPs.

With regard to the molecules' failure to become the key players for bone regeneration, several additional factors beyond those listed above should be considered, including limited evidence from level I clinical trials, the high cost associated with their usage, and concerns related to the development of carcinogenesis.17 The latter issue is still under debate and will likely be the subject of ongoing discussion in the years to come.

Thus, do BMPs still have a role to play? The answer is yes. They are important constituents of the conceptual framework for a successful bone repair response, known as the “diamond concept.”18 Whether they will continue to be the key players is still to be seen. The development of BMPs that can act without the interference of inhibitory molecules, the introduction of active synthetic peptides possessing similar powerful inductive properties, and the investigation of other molecules that have the capacity to exert comparable biologic responses will determine whether BMPs are here to stay in the long term.19,2019,20


Osteoconductive matrices implanted adjacent to bone are the basis for bone void fillers. The substrate must mimic the cancellous bony architecture and has very specific surface kinetics to facilitate the migration, attachment, and proliferation of MSCs, which then differentiate into osteoprogenitor cells. The broad category of calcium ceramics includes a number of suitable substrate materials, such as calcium sulfate, calcium phosphate, synthetic tricalcium phosphate, beta tricalcium phosphate, and coralline hydroxyapatite. These materials are indicated for subarticular contained metaphyseal bone defects, where compressive strength and osteoconduction are the desired properties.

Calcium sulfate has a nonporous crystalline structure and a structure-independent rate of incorporation that is very consistent in terms of dissolution/incorporation. The water-based hemihydrate form, better known as plaster of paris, is used as a self-setting biomaterial, as it has a rapid set time and can be easily injected because of its low initial viscosity and flow kinematics. However, this material has a rapid degradation rate that can be problematic when used to support articular subchondral surfaces.21 Transition to full weight bearing occurs at 3–4 months postsurgery, and many cases of late articular collapse have been reported because of the rapid incorporation and loss of mechanical compressive strength after implantation.22 This combination of rapid degradation rate, speedy loss of compressive strength, and lack of bioactivity has limited its application for bone defect management.

The phosphate ceramics can also be manufactured as injectable cements.23,2423,24 The advantage of cements over blocks, granules, or powders is the ability to custom-fill defects and produce increased compressive strength. The porosity of these materials is the primary factor in determining their ability to foster ingrowth and osseointegration. These highly porous interconnected materials have abundant sites available for cellular interactions with host osteogenic cells and to promote osseointegration. If the material is designed with minimal porosity, the rate of osseointegration will be prolonged because of paucity of cellular interactions, but the corresponding compressive strength will also be very high. Thus, their ability to provide structural support is dependent on the degree of porosity inherent in each material and can be highly manipulated. These materials have the advantage of incorporating at a slower rate than calcium sulfate materials.

Investigators continue to manipulate calcium phosphate substrates in an attempt to find the optimal resorption/incorporation rates with optimal strength ratios maintained. Many injectable products that are now commercially available consist of a composite of calcium phosphate and sulfate constituents. These hybrid materials demonstrate the best injection properties, intermediate incorporation times, and satisfactory strength properties, and thus, they should be the material of choice at this time.

Clinical application of injectable calcium phosphate cements can be problematic because of the Venturi effect, in which initial injection into the void goes well; however, as the pressure inside the bone cavity equalizes, the material backs up and out toward the cannula making further injection difficult. Dual cannula techniques can help minimize this problem. One cannula is placed into the void with suction applied to produce a negative pressure environment while the injection cannula simultaneously implants the material. No pressure backup occurs, and the material fills the void completely. Alternatively, new screw designs are available, which allow injection directly through the implanted cannulated screws. These are also designed to provide a low-pressure environment and allow consistent filling of the void. The use of injectable cements enables support of the reduced joint surface without open bone grafting and decreases exposure for grafting and subchondral defect augmentation. Despite a lack of good quality randomized control trials, there is arguably sufficient evidence supporting the use of injectable bone graft substitutes.25


Appropriate selection of a bone graft or bone graft substitute requires a comprehensive understanding of the specific biologic needs of the patient's clinical problem. This understanding, coupled with knowledge of the specific fracture healing properties of the different bone grafts or bone graft substitutes, can lead to effective selection and treatment success. AICBG remains the gold standard of treatment, although there are an emerging number of alternatives (including RIA bone graft, bone marrow concentrate, BMPs, and calcium phosphate cements) that can be extremely effective when used in the appropriate clinical context.


1. Sen MK, Miclau T. Autologous iliac crest bone graft: should it still be the gold standard for treating nonunions? Injury. 2007;38(suppl 1):S75–S80.
2. Henrich D, Seebach C, Sterlepper E, et al.. RIA reamings and hip aspirate: a comparative evaluation of osteoprogenitor and endothelial progenitor cells. Injury. 2010;41(suppl 2):S62–S68.
3. Sagi HC, Young ML, Gerstenfeld L, et al.. Qualitative and quantitative differences between bone graft obtained from the medullary canal (with a Reamer/Irrigator/Aspirator) and the iliac crest of the same patient. J Bone Joint Surg Am. 2012;94:2128–2135.
4. Dawson J, Kiner D, Gardner W II, et al.. The reamer-irrigator-aspirator as a device for harvesting bone graft compared with iliac crest bone graft: union rates and complications. J Orthop Trauma. 2014;28:584–590.
5. Hernigou P, Poignard A, Beaujean F, et al.. Percutaneous autologous bone-marrow grafting for nonunions. Influence of the number and concentration of progenitor cells. J Bone Joint Surg Am. 2005;87:1430–1437.
6. Desai P, Hasan SM, Zambrana L, et al.. Bone mesenchymal stem cells with growth factors successfully treat nonunions and delayed unions. HSS J. 2015;11:104–111.
7. Kim SJ, Shin YW, Yang KH, et al.. A multi-center, randomized, clinical study to compare the effect and safety of autologous cultured osteoblast(Ossron) injection to treat fractures. BMC Musculoskelet Disord. 2009;10:20.
8. Liebergall M, Schroeder J, Mosheiff R, et al.. Stem cell-based therapy for prevention of delayed fracture union: a randomized and prospective preliminary study. Mol Ther. 2013;21:1631–1638.
9. Dimitriou R, Giannoudis PV. Discovery and development of BMPs. Injury. 2005;36(suppl 3):S28–S33.
10. Senn on the healing of aseptic bone cavities by implantation of antiseptic decalcified bone. Ann Surg. 1889;10:352–368.
11. Urist MR. Bone: formation by autoinduction. Science. 1965;150:893–899.
12. Kanakaris NK, Calori GM, Verdonk R, et al.. Application of BMP-7 to tibial non-unions: a 3-year multicenter experience. Injury. 2008;39(suppl 2):S83–S90.
13. Friedlaender GE, Perry CR, Cole JD, et al.. Osteogenic protein-1 (bone morphogenetic protein-7) in the treatment of tibial nonunions. J Bone Joint Surg Am. 2001;83-A(suppl 1):S151–S158.
14. Govender S, Csimma C, Genant HK, et al.. Recombinant human bone morphogenetic protein-2 for treatment of open tibial fractures: a prospective, controlled, randomized study of four hundred and fifty patients. J Bone Joint Surg Am. 2002;84-A:2123–2134.
15. Giannoudis PV, Kanakaris NK, Einhorn TA. Interaction of bone morphogenetic proteins with cells of the osteoclast lineage: review of the existing evidence. Osteoporos Int. 2007;18:1565–1581.
16. Giannoudis PV. Fracture healing and bone regeneration: autologous bone grafting or BMPs? Injury. 2009;40:1243–1244.
17. Pountos I, Panteli M, Georgouli T, et al.. Neoplasia following use of BMPs: is there an increased risk? Expert Opin Drug Saf. 2014;13:1525–1534.
18. Giannoudis PV, Einhorn TA, Marsh D. Fracture healing: the diamond concept. Injury. 2007;38(suppl 4):S3–S6.
19. Pountos I, Panteli M, Panagiotopoulos E, et al.. Can we enhance fracture vascularity: what is the evidence? Injury. 2014;45(suppl 2):S49–S57.
20. Lamprou M, Kaspiris A, Panagiotopoulos E, et al.. The role of pleiotrophin in bone repair. Injury. 2014;45:1816–1823.
21. Peters CL, Hines JL, Bachus KN, et al.. Biological effects of calcium sulfate as a bone graft substitute in ovine metaphyseal defects. J Biomed Mater Res A. 2006;76:456–462.
22. Yu B, Han K, Ma H, et al.. Treatment of tibial plateau fractures with high strength injectable calcium sulphate. Int Orthop. 2009;33:1127–1133.
23. Holmes RE, Bucholz RW, Mooney V. Porous hydroxyapatite as a bone-graft substitute in metaphyseal defects. A histometric study. J Bone Joint Surg Am. 1986;68:904–911.
24. Russell TA, Leighton RK. Comparison of autogenous bone graft and endothermic calcium phosphate cement for defect augmentation in tibial plateau fractures. A multicenter, prospective, randomized study. J Bone Joint Surg Am. 2008;90:2057–2061.
25. Bajammal SS, Zlowodzki M, Lelwica A, et al.. The use of calcium phosphate bone cement in fracture treatment. A meta-analysis of randomized trials. J Bone Joint Surg Am. 2008;90:1186–1196.

bone graft substitute; stem cells; bone marrow aspiration; calcium phosphate; bone morphogenetic protein

Supplemental Digital Content

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.