BONE-GRAFT SUBSTITUTES: FACTS, FICTIONS & APPLICATIONS

AMERICAN ACADEMY OF ORTHOPAEDIC SURGEONS

70th Annual Meeting
February 5 - 9, 2003
New Orleans, Louisiana
COMMITTEE ON BIOLOGICAL IMPLANTS
Prepared by: Acknowledgements:
A. Seth Greenwald, D.Phil.(Oxon) Exactech, Inc.
GenSci OrthoBiologics, Inc.
Interpore Cross International
Medtronic Sofamor Danek
Osteotech, Inc.
Regeneration Technolgies, Inc.
Stryker Biotech
Synthes USA
Wright Medical Technology, Inc.
Zimmer, Inc.
Scott D. Boden, M.D.
Victor M. Goldberg, M.D.
Yusuf Khan, M.S.
Cato T. Laurencin, M.D., Ph.D.
Randy N. Rosier, M.D.
Presented with the permission of The Journal of Bone and Joint Surgery.
This material was first published, in slightly different form, in J Bone Joint Surg Am 83(Suppl. 2):98-103, 2001.
0
100
200
300
400
500
600
1998 2000 1999 2001 2002
Bone
substitutes/
Platelet helpers
Bone
dowels
Demineralized
bone matrix
products
Allograft bone
U.S. Sales
($millions)
BMP
0
5000
10000
15000
20000
25000
1994 1995 1996 1997 1998 1999 2000 2001
Donors
5,188
6,142
10,004
12,000
(est)
14,030
(est)
16,692
(est)
21,000
(est)
22,000
(est)
THE PHYSIOLOGY OF BONE GRAFTING
The biology of bone grafts and their substitutes is appreciated from an understanding of the bone formation
processes of Osteogenesis, Osteoinduction and Osteoconduction.
Graft Osteogenesis: The cellular elements within a donor graft, which survive transplantation and synthesize new
bone at the recipient site.
Graft Osteoinduction: New bone realized through the active recruitment of host mesenchymal stem cells from the
surrounding tissue, which differentiate into bone-forming osteoblasts. This process is facilitated by the presence of
growth factors within the graft, principally bone morphogenetic proteins (BMPs).
Graft Osteoconduction: The facilitation of blood-vessel incursion and new-bone formation into a defined passive
trellis structure.
All bone graft and bone-graft-substitute materials can be described through these processes.
BONE AUTOGRAFTS
Fresh autogenous cancellous and, to a lesser degree, cortical bone are benchmark graft materials that allograft and
bone substitutes attempt to match in in vivo performance. They incorporate all of the above properties, are harvested
at both primary and secondary surgical sites, and have no associated risk of viral transmission. Furthermore, they
offer structural support to implanted devices and, ultimately, become mechanically efficient structures as they are
incorporated into surrounding bone through creeping substitution. The availability of autografts is, however, limited
and harvest is often associated with donor-site morbidity.
A REALITY CHECK
It is estimated that more than 500,000 bone-grafting procedures are performed annually in the United States,
with approximately half of these procedures related to spine fusion. These numbers easily double on a global
basis and indicate a shortage in the availability of musculoskeletal donor tissue traditionally used in these
reconstructions. (Figure 1)
Figure 1: U.S. trends in musculoskeletal tissue donors
Source: United Network for Organ Sharing & MTF
Figure 2: U.S. sales of bone graft and bone substitutes
Source: Orthopedic Network News, industry estimates
This reality has stimulated a proliferation of corporate interest in supplying what is seen as a growing market in bone-
substitute materials. (Figure 2) These graft alternatives are subjected to varying degrees of regulatory scrutiny, and
thus their true safety and effectiveness in patients may not be known prior to their use by orthopaedic surgeons. It is
thus important to gain insight into this emerging class of bone-substitute alternatives.
No
No
No
++
+
+
+
No
Freeze-Dry
Demineralized
Allogeneic
Cancellous Chips
No No + +++ Frozen
Cortical
No + ++ No Freeze-Dry
No + ++ No Frozen
Cancellous
Allograft
++ ++ ++ +++ Cortical
+++ +++ +++ No Cancellous
Autograft
Osteogenesis
Osteo-
Induction
Osteo-
Conduction
Structural
Strength
Bone Graft
Figure 3: Comparative properties of bone grafts
Figure 4: (a) A 17-year old patient with osteosarcoma of the distal part of the femur with no extraosseous extension or metastatic
disease. Following chemotherapy, (b) limb salvage with wide resection was performed. Femoral reconstruction with the use of
an autogenous cortical fibular graft, iliac crest bone chips, morselized cancellous autograft and structural allograft combined with
internal fixation. (c) Graft incorporation and remodeling are seen at 3 years. (d) Limb restoration is noted at 10 years following
resection. (The intramedullary rod was removed at 5 years.)
(a) (b) (c) (d)
BONE ALLOGRAFTS
The advantages of bone allograft harvested from
cadaver sources include its ready availability in
various shapes and sizes, avoidance of the need to
sacrifice host structures and no donor-site morbid-
ity. Bone allografts are distributed through regional
tissue banks. Still, the grafts are not without contro-
versy, particularly regarding their association with
the transmission of infectious agents, a concern
virtually eliminated through tissue-processing and
sterilization. However, both freezing and irradia-
tion modify the processes of graft incorporation
and affect structural strength. A comparison of
the properties of allograft and autograft bone is
shown in Figure 3. Often, in complex surgical
reconstructions, these materials are used in tandem
with implants and fixation devices. (Figure 4)
BONE GRAFT SUBSTITUTES
The ideal bone-graft substitute is biocompatible, bioresorbable, osteoconductive, osteoinductive, structurally similar to
bone, easy to use and cost-effective. Within these parameters a growing number of bone alternatives are commercially
available for orthopaedic applications, including reconstruction of cavitary bone deficiency and augmentation in
situations of segmental bone loss and interbody spine fusion. They are variable in their composition and their claimed
mechanisms of action. Figure 5 shows a sampling of bone-graft substitute materials. Those containing growth factors
in their composition inclusive of rhBMP-2 and rhBMP-7 (OP-1) demonstrate osteoinduction in clinical application,
while the remainder are predominantly osteoconductive in their claims. All offer minimal structural integrity. A series
of case examples demonstrate their mechanisms of action through the healing process. (Figures 6, 7 and 8)
Figure 5: Summary of typical bone-graft substitutes that are commercially available
Figure 6: (a) A 60-year old female with a comminuted depressed fracture of the lateral tibial plateau. (b) Three weeks after ORIF
with filling of the resulting defect with OSTEOSET

(Wright Medical Technology, Inc., Arlington, TN) pellets. (c) At 7 months

post-op, restoration of trabecular bone with complete dissolution of the graft material is noted.
(a) (b) (c)
Company
Commercially
available product
Composition
Commercially
available forms
Claimed mechanisms of
action
Burdens of proof FDA status
Exactech, Inc.
Opteform

DBM and cortical

cancellous chips in
gelatin carrier
Formable putty in
circular disks or
syringeable cylinders
Osteoconduction
Bioresorbable
Limited osteoinduction
Human studies
Case reports
Animal studies
Every lot tested in vivo
for osteoinduction
510(k) clearance required
Regulatory discretion
currently permits sale
Optefil
TM
DBM suspended in
gelatin carrier
Injectable bone paste
or powdered form
Osteoconduction
Bioresorbable
Limited osteoinduction
Human studies
Case reports
Animal studies
Every lot tested in vivo
for osteoinduction
510(k) clearance required
Regulatory discretion
currently permits sale
GenSci
OrthoBiologics
OrthoBlast
TM
Heat sensitive
copolymer with
cancellous bone
chips and DBM
Injectable paste or
putty
Osteoconduction
Bioresorbable
Limited osteoinduction
Case reports
Animal studies
Cell culture
510(k) clearance required
Regulatory discretion
currently permits sale
DynaGraft

Heat sensitive
copolymer with
DBM
Injectable gel, matrix
or putty
Osteoconduction
Bioresorbable
Limited osteoinduction
Human studies
Case reports
Animal studies
Cell culture
510(k) clearance required
Regulatory discretion
currently permits sale
Interpore Cross
International
ProOsteon

500R Coral HA composite Granular or block

Osteoconduction
Bioresorbable
Human studies
Case reports
Animal studies
510(k) cleared
Medtronic
Sofamor Danek
InFuse
TM
rhBMP-2 protein
with absorbable
collagen sponge
Freeze-dried powder
and sponge in several
sizes
Bioresorbable sponge
Osteoinduction
Human studies
Animal studies
PMA approved for fusion
with spinal cage
MTF/Synthes DBX

DBM in a sodium
hyaluronate carrier
Injectable
paste, putty and
corticocancellous mix
Osteoconduction
Bioresorbable
Limited osteoinduction
Human studies
Case reports
Animal studies
510(k) clearance required
Regulatory discretion
currently permits sale
Osteotech Grafton

DBM combined with

Glycerol
Pellets, plugs,
formable putty and
injectable gel
Osteoconduction
Bioresorbable
Limited osteoinduction
Human studies
Case reports
Animal studies
510(k) clearance required
Regulatory discretion
currently permits sale
Regeneration
Technologies
OSTEOFIL

/
REGENAFIL

DBM combined with

non-toxic natural
gelatin carrier
Injectable paste,
injectable putty,
strips and blocks with
cortical cancellous
chips
Osteoconduction
Bioresorbable
Limited osteoinduction
Human studies
Case reports
Animal studies
Every lot tested in vivo
for osteoinduction
510(k) clearance required
Regulatory discretion
currently permits sale
Stryker Biotech OP-1 Implant
rhBMP-7 with type 1
bone collagen
Lyophilized powder
reconstituted to form
wet paste
Resorbable collagen scaffold
Osteoinduction
Human studies
Animal studies
HDE approval for long
bone nonunions
Synthes
Norian

SRS

Calcium phosphate Injectable paste

Osteoconduction
Bioresorbable
Compressive strength:
50 MPa
Human studies
Case reports
Animal studies
510(k) cleared
Calceon

6 Calcium sulfate Pellets

Osteoconduction
Bioresorbable
Animal studies 510(k) cleared
Wright Medical
Technology
OSTEOSET

Surgical grade
calcium sulfate
Various sized pellets
Osteoconduction
Bioresorbable
Human studies
Case reports
Animal studies
510(k) cleared
AlloMatrix
TM
DBM with surgical
grade calcium sulfate
powder
Injectable or
formable putty
Osteoconduction
Bioresorbable
Limited osteoinduction
Case reports
Animal studies
Cell culture
510(k) clearance required
Regulatory discretion
currently permits sale
Zimmer Collagraft
TM
Mixture of
hydroxyapatite,
tricalcium phosphate
and bovine collagen
Strip configurations
Osteoconduction
Bioresorbable
Limited osteoinduction
when mixed with bone
marrow
Human studies
Case reports
Animal studies
Cell culture
PMA approved
Figure 8: (a) A 23-year old male with an open, comminuted, grade II fracture of the left tibia. Prior treatments included autograft,
skin flap and multiple irrigation and debridement to treat infection. Amputation was scheduled after failure of these treatments.
(b) Six months following treatment with IM rod fixation and OP-1 Implant (Stryker Biotech, Hopkinton, MA). He was full weight
bearing and pain free 9 months post-operative. (c) Five years post-operative. (d) Ten years post-operative.
BURDEN OF PROOF
It is reasonable to assume that not all bone-substitute products will perform analogously. Thus, a quandary of choice
confronts the orthopaedic surgeon. As a first principle, it is important to appreciate that different healing environments
(e.g., a metaphyseal defect, a long-bone fracture, an interbody spine fusion, or a posterolateral spine fusion) have
different levels of difficulty in forming new bone. For example, a metaphyseal defect will permit the successful
use of many purely osteoconductive materials. In contrast, a posterolateral spine fusion will not succeed if purely
osteoconductive materials are used as a stand-alone substitute. Thus, validation of any bone-graft substitute in one
clinical site may not necessarily predict its performance in another location.
A second principle is to seek the highest burden of proof reported from preclinical studies to justify the use of an
osteoinductive graft material or the choice of one brand over another. Whether it is more difficult to make bone in
humans than it is in cell-culture or rodent models, with a progressive hierarchy of difficulty in more complex species,
has not been clearly determined. Only human trials can determine the efficacy of bone-graft substitutes in humans
as well as their site-specific effectiveness.
Figure 7: (a) AP and Lateral radiographs, 67-year old female with depressed fracture of the lateral tibial plateau. (b) AP and Lateral
radiographs 12 months after ORIF with filling the defect with Norian

SRS

(Synthes USA, Paoli, PA). No loss of reduction of the

plateau surface is noted, fracture completely healed.
(a) (b)
(a) (b) (c) (d)
BURDEN OF PROOF (Contd.)
A third principle requiring burden of proof specifically pertains to products that are not subjected to high levels of
regulatory scrutiny, such as demineralized bone matrix (DBM) or platelet gels containing autologous growth factors.
Such products are considered to involve minimal manipulation of cells or tissue and are thus regulated as tissue
rather than as devices, unless they are configured with an additive and then require 510(k) clearance. As a result,
there is no standardized level of proof of safety and effectiveness required before these products are marketed and
are used in patients. While these products may satisfy the technical definition of minimal manipulation, there is a
risk that they will not produce the expected results in humans when there has been little or no testing in relevant
animal models.
FUTURE
Recent FDA approvals include the use of rhBMP-2 for assisted spinal fusion and rhBMP-7 (OP-1) as an autograft
substitute for tibial non-unions. The FDA Orthopaedic Device Advisory Panel has also recommended extending the
indication for rhBMP-2, in conjunction with a collagen sponge, for the treatment of long bone fractures. These clinical
applications demonstrate impressive osteoinductive capacity and pave the way for broader clinical applications.
Their methods of administration include direct placement in the surgical site, but results have been more promising
when the growth factors have been administered in combination with substrates to facilitate timed-release delivery
and/or provide a material scaffold for bone formation. FDA regulatory imperatives will continue to determine their
availability. Their cost/benefit ratio will ultimately influence clinical use.
Further advances in tissue-engineering, the integration of the biological, physical and engineering sciences, will
create new carrier constructs that regenerate and restore tissue to its functional state. These constructs are likely to
encompass additional families of growth factors, evolving biological scaffolds and incorporation of mesenchymal
stem cells. Ultimately, the development of ex vivo bioreactors capable of bone manufacture with the appropriate
biomechanical cues will provide tissue-engineered constructs for direct use in the skeletal system.
TAKE HOME MESSAGE
The increasing number of bone-grafting procedures performed annually in the U.S. has created a shortage of
cadaver allograft material and a need to increase musculoskeletal tissue donation.
This has stimulated corporate interest in developing and supplying a rapidly expanding number of bone substitutes,
the makeup of which includes natural, synthetic, human and animal-derived materials.
Fresh autogenous cancellous and, to a lesser degree, cortical bone are the benchmark graft materials that, ideally
both allograft and bone substitutes should match in in vivo performance. Their shortcomings include limited
availability and donor-site morbidity.
The advantages of allograft bone include availability in various sizes and shapes as well as avoidance of host-
structure sacrifice and donor-site morbidity. Tissue-processing, however, modifies graft incorporation as well as
structural strength. Transmission of infection, particularly the human immunodeficiency virus (HIV) has been
virtually eliminated as a concern.
The ideal bone-graft substitute is biocompatible, bioresorbable, osteoconductive, osteoinductive, structurally similar
to bone, easy to use and cost-effective. Currently marketed products are variable in their composition and their
claimed mechanisms of action. It is reasonable that not all bone-substitute products will perform the same.
Recent FDA approvals for specific uses of recombinant human growth factors (rhBMP-2 and rhBMP-7 (OP-1)) are
based on demonstrated osteoinductive capacity in human trials. Other applications will likely emerge.
A quandary of choice confronts the orthopaedic surgeon. Caveat emptor! Selection should be based on reasoned
burdens of proof. These include examination of the product claims and whether they are supported by preclinical
and human studies in site-specific locations where they are to be utilized in surgery.