Plastic Surgery Volume 1 Principles 4th

265
Autologous cartilage grafts and applications
Fig. 17.5  Technique 3: (A, B) The nasal defect is wide but short in its vertical dimension. This allows harvest from the anteroinferior helical rim. (C, D) Primary closure is
achieved by running the dog ear superiorly and inferiorly. (Reproduced from Singh DJ, Bartlett SP. Aesthetic management of the ear as a donor site. Plast Reconstr Surg.
2007;120:899–908.)
A
B
C
D
■ Technique 5: After harvesting composite graft wider than
1.5 cm from the base of helix, the closure of such a
wound becomes more intricate. A simple helical
advancement would result in significant distortion and
overprojection (Fig. 17.7). To overcome this, a V-shaped
wedge of skin is harvested, but a “half-star” pattern of
cartilage at the apex of the V is taken (Fig. 17.8B–E). In
addition, a postauricular incision with a scaphomastoid
suture is needed. With these procedures, this technique
can prevent cupping of the ear with closure (Fig. 17.8F).
■ When an extended skin graft component is needed with
the composite graft, the composite tissues can be taken
either preauricularly or postauricularly, as demonstrated
by the markings on Fig. 17.9.
The techniques described above provide simple methods to
close the wounds of auricular chondrocutaneous composite
grafts primarily with appreciable aesthetic outcome.12 Other
methods of rotational flap or skin graft have also been
described in the literature.13,14 Importantly, the functional and
aesthetic aspects should be well balanced for both the repair
site and donor site when a physician decides to harvest an
auricular composite chondrocutaneous graft of a particular
size and shape.
One concern for the clinical application of free auricular
composite grafts is their limited size due to the lack of immedi­
ate blood supply after transplantation. In contrast, microvas­
cularly transferred auricular composite grafts can overcome
this limit and are particularly useful when repairing large nasal
defects. Zhang et al. have recently reviewed their experiences
of surgical treatment of large nasal defects with vascularized
preauricular and helical rim flaps based on the superficial
temporal vessels in 63 clinical cases. The repaired deformities
include unilateral alar defect, alar and side wall defects, tip and
columellar defects, entire lower third of the nose missing, and
composite defects involving cheek and maxilla. The total flap
survival rate reached 97%. The results demonstrate that such
an approach, using vascularized preauricular and helical rim
flaps, is a reliable method for reconstructing nasal defects.15
Fig. 17.10 demonstrates the surgical technique for repairing an
alar lobule defect with vascularized helical rim composite
tissue and preauricular skin.15

266
CHAPTER 17 • Repair, grafting, and engineering of cartilage
Fig. 17.7  (A–C) With harvest of a small (1–1.2 cm) composite graft, closure can be achieved by helical rim advancement. However, overprojection of the ear may result,
as seen in this series of photographs. (Reproduced from Singh DJ, Bartlett SP. Aesthetic management of the ear as a donor site. Plast Reconstr Surg. 2007;120:899–908.)
B
A
C
C
D
A
B
Fig. 17.6  Technique 4: (A, B) Inset of
graft and aesthetic outcomes of nasal
reconstruction and (C) the auricular donor
site in comparison with (D) the
contralateral ear. (Reproduced from Singh
DJ, Bartlett SP. Aesthetic management of
the ear as a donor site. Plast Reconstr
Surg. 2007;120:899–908.)

267
Autologous cartilage grafts and applications
Fig. 17.8  Technique 5: (A,B) This nasal defect requires a larger composite graft with a width of 1.8 cm. The harvest site on the auricle is demonstrated by the purple lines.
(C–E) To overcome distortion and overprojection resulting from tissue harvest, closure for this size auricular defect involves a V-shaped wedge of skin and a half-star pattern
of cartilage excision at the apex of the V. The solid lines indicate skin incision and the dotted star pattern demonstrates cartilaginous incisions. In addition, a postauricular
incision with a scaphomastoid suture is performed. (F) This technique of wound closure prevents the cupping of the ear and relatively normal shape and projection of the
ear are maintained after the closure. (Modified from Singh DJ, Bartlett SP. Aesthetic management of the ear as a donor site. Plast Reconstr Surg. 2007;120:899–908.)
A
B
C
D
E
F
Fig. 17.9  Occasionally, an extended skin
graft is needed in addition to the composite
graft for reconstruction of nasal defects.
(A) The solid lines indicate skin incisions
and the dotted line indicates a cartilaginous
incision. (B) The solid lines represent the
harvest of a postauricular full-thickness skin
graft. (Reproduced from Singh DJ, Bartlett
SP. Aesthetic management of the ear as a
donor site. Plast Reconstr Surg.
2007;120:899–908.)
B
A

268
CHAPTER 17 • Repair, grafting, and engineering of cartilage
Fig. 17.10  A 17-year-old female with a deformity of the left alar lobule after laser treatment of a hemangioma. An ipsilateral reverse-flow flap consisting of both the helical
rim and the preauricular skin (2.2 × 1.8 cm) was used. The anastomosis was made to the proximal end of the superficial temporal vessels by means of 11-cm vascular
grafts from the descending branch of the lateral circumflex vessels. (A) Preoperative and (B) intraoperative views of the harvest of a reverse-flow flap based on the distal
superficial temporal vessels. (C) Postoperative view at 3-month follow-up. (Reproduced from Zhang YX, Yang J, Wang D, et al. Extended applications of vascularized
preauricular and helical rim flaps in reconstruction of nasal defects. Plast Reconstr Surg. 2008;121:1589–1597.)
B
C
A
Nasal cartilage graft
Although limited in its available amount, nasal cartilage has
been employed as a composite chondromucosal graft for
eyelid reconstruction. Septal cartilage is an important source
of nasal cartilage graft. As reported by Murrell, the septal
cartilage can be accessed via a hemitransfixion incision with
dissection around the caudal margin of the quadrangular
cartilage. After both sides of mucoperichondrium are raised,
the septal cartilage can be harvested.16 Importantly, as shown
in Fig. 17.11, an L-shaped septal strut should be preserved to
give nasal support and avoid nasal collapse.6,16 Nevertheless,
the amount of L-shaped strut cartilage necessary to provide
good support can vary greatly depending on the strength,
thickness, and dimensions of the nasal septum and other nasal
tissues (i.e., upper lateral cartilages, lower lateral cartilages,
nasal bones).16 As early as 1962, Millard published his work
of repairing eyelid defects with a chondromucosal graft har­
vested from nasal septum. Later, Mustardé also described the
technique in the book Repair and Reconstruction in the Orbital
Region.6
The other region available for harvesting nasal chondro­
mucosal graft is the upper lateral nasal cartilage, as reported
by Tessier in 1979.6 In clinical practice, repairing a large
defect of the upper eyelid is always a challenge. Scuderi and
colleagues have developed a surgical technique to address
such a concern. In their report, a pedicled nasal chondromu­
cosal flap was designed along the lateral nasal wall based
on the terminal branch of the dorsal nasal artery. The flap
includes the subcutaneous tissues down to the periosteum
and the cranial portion of the upper lateral cartilage. It can
be harvested either unilaterally or contralaterally and a skin
graft is applied for cutaneous coverage. In their reported 15
patients, the flap was viable in every patient and satisfactory
functional recovery was achieved, indicating that this one-
stage procedure can reconstruct a thin and mobile eyelid.17
Fig. 17.11  Harvest of nasal septal cartilage graft. It is important to keep the
L-shaped strut cartilage (shown in stripes) to provide nasal support. The area of
septum that attaches to the upper lateral cartilages should also be preserved.
Generally, nonsupportive quadrangular cartilage, which is pictured posterior to the
dashed vertical line, can be harvested freely. (Reproduced and modified from
Murrell GL. Dorsal augmentation with septal cartilage. Semin Plast Surg.
2008;22:124–135.)
Graft
Fig. 17.12 demonstrates its design and surgical procedure as
well as its clinical outcome.17 In addition to eyelid repair,
septal cartilage graft has been used for dorsal augmentation,16
tracheal repair,18 and extended septal graft for controlling the
projection shape of nose tip.19

269
Autologous cartilage grafts and applications
A
B
Ophthalmic
artery
Angular artery
Lateral
nasal artery
SC
LLC
ULC
NB
Dorsal nasal artery
Ophthalmic
artery
Angular artery
Lateral
nasal artery
SC
LLC
SC
NB
Pedicle (Dorsal
nasal artery)
Chondromucosal flap
Fig. 17.12  Schematic illustration of the design and procedure of a pedicled nasal chondromucosal flap for upper eyelid reconstruction. (A) Preoperative markings for
harvest of the upper lateral cartilage (ULC) in the nasal chondromucosal flap, which is based on the lateral terminal branch of the dorsal nasal artery. NB, nasal bone; LLC,
lower lateral cartilage; SC, septal cartilage. (B) After skin incision and undermining, the flap is raised after a portion of the upper lateral cartilage is harvested.
Continued

270
CHAPTER 17 • Repair, grafting, and engineering of cartilage
Fig. 17.13  Donor site for costal cartilage grafts in auricular framework and chin
graft construction. E, ear base block; H, helix; C1, chin graft, first layer; C2, chin
graft, second layer. (Reproduced and modified from Brent B. Repair and grafting of
cartilage and perichondrium. In: McCarthy JC (ed). Plastic Surgery. Philadelphia:
WB Saunders; 1990:559–582.)
C1
E
H
C2
C
D
E
F
(C) Clinical outcome of upper eyelid defect repaired with pedicled nasal chondromucosal flap. Carcinoma of the upper eyelid. (D) Harvest of the nasal
chondromucosal flap. (E,F) One-year postoperative result after reconstruction with the nasal chondromucosal flap. (Reprinted and modified from Scuderi N, Ribuffo D,
Chiummariello S. Total and subtotal upper eyelid reconstruction with the nasal chondromucosal flap: a 10-year experience. Plast Reconstr Surg. 2005;115:1259–1265.)
Fig. 17.12, cont’d
Rib cartilage graft
Costal cartilage may serve as the best donor site for cartilage
graft in terms of available tissue amount and mechani­
cal strength. The autologous rib cartilage can be virtually
contoured into any desired shape and it can retain form
and bulk after implantation if basic surgical principles
are followed.6 The costal cartilage graft is often used as a
cartilage framework for total ear reconstruction. Tanzer,20
Thomson et al.,21 and Brent22 have respectively described the
technique to harvest costal cartilage to construct auricular
framework. In the procedure, the synchondrosis of the sixth
and seventh cartilages as well as the eighth costal cartilage
is harvested, usually along with the perichondrium.20–23 In
contrast, Nagata’s method for total auricular reconstruction
requires the harvest of four costal cartilages in the first-stage
operation and one or two costal cartilages for the second-
stage operation.24–28 Fig. 17.13 is a schematic illustration of the
harvest of rib cartilage designed for ear reconstruction or chin
contouring.29
In clinical practice, harvest of costal cartilage can be associ­
ated with donor site morbidity. Uppal et al. have reported
their investigation on morbidity associated with the harvest
of costal cartilage in 42 patients who underwent ear recon­
struction. The commonest complaints included pain and
clicking of the chest wall, which usually peaked in the first
week postsurgery and gradually diminished over 3 months.
The other problems were scar and chest wall deformity.30
Among them, the most challenging problems are pneumotho­
rax during the surgical procedure and resulting abnormal
chest wall contour. This is particularly true for Nagata’s

271
Cartilage engineering
potential causes limiting its application include the technique
of harvest, the oxygenation of the recipient bed, and the pres­
ence of chondroprogenitors at the recipient site, which can
all have an influence on the composition of the new tissue.45
With proper microenvironments, perichondrium-mediated
cartilage regeneration remains possible, as revealed by experi­
mental studies. For example, Sari et al. reported neocartilage
formation in a folded-ear perichondium at 6 weeks postim­
plantation under the abdominal muscle in rabbit.45
One of the major clinical applications is for joint repair. As
early as 1975, Engkvist et al. reported a pilot experimental
study of regenerating articular cartilage of cavitas glenoida­
lis by transplanting ear perichondrium in a rabbit model.
Further, clinical trials were formed in 5 cases of arthritis
patients. After removal of the degenerated cartilage, rib
perichondrium was grafted and articular cartilage regenera­
tion took place.46 Following the initial trials, there were some
reports of the application of perichondrium graft for clinical
repair of human knee joint articular cartilage,47,48 but there
were few on repairing articular cartilage of human digits.49
Instead, transplantation of costal osteochondral grafts became
the major tissue graft for functional restoration of injured
digital articular cartilage.50,51
The other important clinical application might be nasal
reconstruction either as a perichondrium graft or as a peri­
chondrocutaneous graft. Recently, Boccieri and Marianetti
reported the application of perichondrium graft for nasal
surgery. Because of its thinness and malleability, perichon­
drium is particularly suitable for covering every part of the
cartilaginous graft, and is easy to fold into various layers if
greater thickness is required in filling certain areas and there­
fore is commonly used in secondary rhinoplasty.52,53 Fig. 17.17
demonstrates the surgical procedure of nasal tip reconstruc­
tion with auricular perichondrium.52
In addition, perichondrocutaneous graft is a common
graft type. Brent and Ott reported the graft and proposed
its application in reconstructive facial surgery in 1978.54 The
donor source for the graft should be restricted to concha in
order to prevent deformity.6 This graft has wide applications.
It has been reported to be used for facial reconstruction,55,56 for
reconstruction of nasal defect,57,58 for auricular defect repair,59
and repairing ectropion.60 Overall, although both perichon­
drium and perichondrocutaneous grafts have been widely
used for clinical repair of tissue defect, there remains no
solid clinical evidence to support the theory that transplanted
perichondrium is able to regenerate a significant amount of
cartilage tissue.
Cartilage engineering
Introduction and basic principle
Cartilage probably is one of the most commonly used tissue
grafts in plastic surgery and has been employed in auricular
reconstruction, rhinoplasty, and other surgical procedures, as
described above. However, autologous cartilage graft is
limited in its available amount and tissue harvest may cause
donor site morbidity.
Tissue engineering is a new biotechnology developed
from the late 1980s and early 1990s, which aims to repair
and regenerate human tissue via an engineering approach
method, which requires more and extra cartilage grafts for
auricular reconstruction.24–28
To address this concern, Kawanabe and Nagata developed
a new method for harvesting rib cartilage to avoid intra­
operative and postoperative complications and problems. In
this modified procedure, the costal cartilages were harvested
en bloc with the perichondrium left completely intact at the
donor site. In addition, after the fabrication of the auricular
cartilage frame, the remaining costal cartilage was cut into
small blocks that acted as spacers to fill the dead space formed
in the perichondrial pocket. The retained perichondrium not
only helps to avoid the injury of pleura, but may also promote
cartilage regeneration because of the presence of chondrogenic
stem cells.2,3 In an investigation of 270 cases of total auricular
reconstruction performed using Nagata’s method and the
new method of rib cartilage harvesting, the incidence of infec­
tion and pneumothorax was reduced to less than 1%. More
importantly, there were no postoperative chest wall deformi­
ties when costal cartilage was harvested with this new tech­
nique.31 Interestingly, a follow-up study revealed that returned
hyaline cartilages were mixed with fibrocartilage with visible
margins at 6 months postsurgery. Twelve months after the
first-stage operation, homogeneous hyaline cartilage was
observed histologically, and the regenerated cartilages were
similar to native costal cartilages in their hardness, which
enabled the second harvest.32 Fig. 17.14 gives a schematic
illustration of harvesting costal cartilage with the new method
of retaining perichondrium and returning the remaining
costal cartilage to the donor site.31
Modification of previously established methods of making
auricular framework is also considered as an alternative option
to avoid harvesting extra cartilage and thus help to prevent
chest wall deformity. As an example, Chin et al. recently
reported several modifications of auricular framework which
required less cartilage graft and meanwhile could also generate
individualized and harmonious ear frameworks to meet the
need for satisfactory three-dimensional (3D) outline of a recon­
structed auricle.33 The framework may consist of the helix, the
base frame, the Y-shaped antihelical complex, and the tragus
attached to an additional cartilaginous cube depending on
individualized need (Fig. 17.15). In addition, bone cement
is used as the support material during the ear elevation
stage, and therefore extracostal cartilage harvesting became
unnecessary.34 Fig. 17.16 demonstrates the clinical outcome of
reconstructed ear with modified frameworks based on the
individualized 3D auricle structures.33
In addition to auricular reconstruction, rib cartilage graft
also has other applications, including reconstruction of sig­
nificant saddle-nose deformity,35 treatment of maxillonasal
dysplasia (Binder’s syndrome),36 nipple reconstruction,37 for
septorhinoplasty,38 and in tracheal reconstruction.39
Autologous perichondrial graft
In 1959, Lester first reported the potential of neocartilage for­
mation by transplanted perichondrium.40 This phenomenon
has also been observed by others.41–44 Despite the enthusiasm
and optimism derived from the pioneering observations,
the clinical application of perichondrial grafts for cartilage
regeneration remains limited to certain conditions, such as
the reconstruction of degenerated knee joint cartilages. The

272
CHAPTER 17 • Repair, grafting, and engineering of cartilage
A
Fig. 17.14  (A) Schematic illustration of the new method of costal cartilage harvest. (Top, left) The muscular fascia of the rectus abdominis muscle and the external oblique
muscle are exposed. (Top, center) A longitudinal incision is made between the rectus abdominis muscle and the external oblique muscle with delicate dissecting scissors.
(Top, right) The perichondrium is completely exposed. (Center, left) The line of incision is marked in the center of the costal cartilage, and, taking special care not to injure
or damage the cartilage, the incision is made with a scalpel. (Center, center) The anterior surface of the perichondrium is first undermined with an elevator. (Center, right)
The undermining of the posterior surface of the perichondrium requires extreme caution because the risk of pneumothorax exists; therefore, the tip of the elevator is placed
on the cartilage or must face the cartilage to avoid accidental puncture of the thoracic wall. The approximately 1 cm of periosteum at the costochondral junction is
undermined for easier harvest of the costal cartilage. (Bottom, left) A Doyen rib raspatory is inserted underneath the undermined costal cartilage between the cartilage and
the perichondrium at the costochondral junction. The costal cartilage is excised with a scalpel and the costochondral junction is to be left behind at the donor site. (Bottom,
center) Harvesting of the costal cartilages is easier if the sixth and seventh costal cartilages are harvested from the side of the costochondral junction instead of the sternal
side. (Bottom, right) The appearance of the donor site immediately after the harvest of the costal cartilages. The perichondrium is left completely intact at the donor site.
Continued

273
Cartilage engineering
Fig. 17.15  The modified rib cartilage framework may contain these parts to
provide prominent structures. From left to right, the helix, the base frame, the
Y-shaped antihelical complex, and the tragus attached to an additional cartilaginous
cube. (Reproduced from Chin W, Zhang R, Zhang Q, et al. Modifications of
three-dimensional costal cartilage framework grafting in auricular reconstruction for
microtia. Plast Reconstr Surg. 2009;124:1940–1946. )
B
Intercostal nerve
(B) (Top, left) The perichondrium is sutured with 4-0 nylon at 5-mm intervals with the exception of one area for the return of the remaining costal
cartilage after fabrication of the three-dimensional costal cartilage framework, which is cut into 2–3-mm blocks. A small funnel is placed in the unsutured opening of the
perichondrium for returning the cut cartilage blocks. (Top, center) The appearance after the return of cut cartilage blocks. The cartilage blocks are visible at the opening of
the perichondrium. (Top, right) Intercostal nerve block is performed for reducing postoperative pain with 0.25% Marcaine under direct vision. A total of 5 mL is administered
per rib. (Bottom) The muscle and muscular fascia are sutured with 4-0 nylon and a Penrose drain is placed under the muscular layer. (Reproduced from Kawanabe Y, Nagata
S. A new method of costal cartilage harvest for total auricular reconstruction: part I. Avoidance and prevention of intraoperative and postoperative complications and
problems. Plast Reconstr Surg. 2006;117:2011–2018.)
Fig. 17.14, cont’d
As outlined by Stock and Vacanti,64 the basic concept of
tissue engineering includes a scaffold that provides an archi­
tecture on which seeded cells can organize and develop into
the desired organ or tissue prior to implantation. The scaffold
provides an initial biomechanical profile for the replacement
tissue until the cells produce an adequate ECM. During the
formation, deposition, and organization of the newly gener­
ated matrix, the scaffold is either degraded or metabolized,
eventually leading to a vital organ or tissue that restores,
maintains, or improves tissue function.
Generally, the tissue-engineering process involves three
major components: (1) seed cells: the component for matrix
production, deposition, and tissue formation; (2) scaffold: the
substance that provides a 3D construct for cells to reside,
proliferate, and produce matrix; (3) tissue formation environ­
ment: after being seeded on the scaffold, cells start to grow and
produce and deposit ECMs on the scaffold. In a proper envi­
ronment, with gradual degradation of the scaffold, cell prolif­
eration, matrix production, and proper tissue remodeling, an
engineered tissue gradually forms and becomes mature.
In the past 20 years, tremendous progress has been made
in basic and applied research of cartilage engineering, includ­
ing the search for different biomaterials as scaffolds, seeking
different cell sources and chondrogenic induction of stem cells
as well as advances in the techniques of cartilage tissue forma­
tion, such as in vitro reconstruction of cartilage tissue. Among
to produce autologous tissue. Actually, cartilage is closely
related to the development of tissue engineering because it
was used as a tissue target to prove the basic principle.61,62
Interestingly, the success in the engineering of human ear-
shaped cartilage in a nude mouse model vividly revealed that
this tissue-engineering technique has great potential in plastic
surgery for tissue repair and reconstruction.63

274
CHAPTER 17 • Repair, grafting, and engineering of cartilage
the application potential in plastic surgery. The procedure
is described below to provide an example of translational
research of cartilage engineering.63
To generate a human ear-shaped cartilage, selection of
seed cell, scaffold materials, and an animal model should be
determined according to the principle. In this study, chon­
drocytes derived from calf cartilage, polyglycolic acid (PGA),
unwoven fibers, and a nude mouse model served as the basic
components to construct the cartilage.
To isolate the cells, cartilage fragments were first harvested
from the articular surface of glenohumeral and humeroulnar
joints and then subjected to collagenase digestion (3 mg/mL)
them, some have already revealed the great potential in plastic
surgery, for example, auricular reconstruction, rhinoplasty,
and facial contouring. The following gives a few examples
of generation techniques and the results of engineered carti­
lages that may potentially be used for cartilage repair and
reconstruction.
Engineering of auricular cartilage
An important achievement in cartilage engineering research
is the development of the technique that enables generation
of a human ear-shaped cartilage, because it vividly reveals
Fig. 17.16  Auricular reconstruction with modified frameworks. (A–C) A 7-year-old girl presented with congenital microtia. (A) The three-dimensional framework used: the
arrow indicates the Y-shaped antihelical complex attached on the base frame. (B) Preoperative oblique view of conchal-type microtia. (C) Close-up appearance of the
reconstructed auricle 8 months after grafting of the framework. (D–F) A 6-year-old boy presented with congenital microtia. (D) The completed costal framework with
Y-shaped antihelical complex and the tragus attached to an additional cartilaginous cube. (E) Preoperative lateral view of sausage-type microtia. (F) Postoperative oblique
view, 3 months after grafting of the framework. (Modified from Chin W, Zhang R, Zhang Q, et al. Modifications of three-dimensional costal cartilage framework grafting in
auricular reconstruction for microtia. Plast Reconstr Surg. 2009;124:1940–1946.)
C
B
A
F
E
D

275
Cartilage engineering
To prepare a cell–scaffold construct, the ear-shaped scaffold
was placed in a culture dish and seeded with 3 mL of chon­
drocyte suspension (1.5 × 108 cells) and placed in an incubator
for 4 hours to allow seeded cells to attach on the scaffold
and then culture medium (Hamm’s F-12 supplemented
with 10% fetal calf serum, 5 µg/mL ascorbic acid, 292 µg/mL
l-glutamine, 100 U/mL penicillin, and 100 µg/mL strepto­
mycin) was added, and the construct was incubated in vitro
at 37°C in 5% CO2 for 1 week. As shown, the cell–seeded
construct could maintain a good ear shape (Fig. 17.18A) and
scanning electron microscope revealed good cell attachment
on the scaffold and matrix production (Fig. 17.18B).
in culture medium at 37°C for 12–18 hours. The resulting
tissue digestion solution was filtered and centrifuged to
collect the chondrocytes and the cells were expanded in vitro.
To prepare the scaffold for the human ear shape, the ear of
a 3-year-old child was cast using alginate as the impression
material, and then a final cast of plaster was fashioned from
the alginate impression. Using the plaster cast as a mold, PGA
nonwoven mesh in about 100 µm thickness was coated with 1%
(w/v) solution of polylactic acid (PLA) in methylene chloride.
Following immersion, the fabric was removed and shaped
into the form of a human ear using the plaster prosthetic
mold.
Fig. 17.17  (A) Reconstruction of the nasal tip with lateral crural and shield grafts (Sheen-type) harvested from the cartilage of the auricular concha. (B) Graft of the
perichondrium stretched over and secured to cartilaginous grafts to make the contours smooth and disguise tip grafts. (C) Lateral view of the same graft. (D) Schematic
illustration of the perichondrium graft positioning and fixation. The presence of numerous grafts inevitably leads to sharp ridges and irregularities in the cartilaginous contour
that can be eliminated by means of the perichondrium graft. (Modified from Boccieri A, Marianetti TM. Perichondrium graft: harvesting and indications in nasal surgery.
J Craniofac Surg. 2010;21:40–44.)
A
B
C
D

276
CHAPTER 17 • Repair, grafting, and engineering of cartilage
resultant data were then input into a CAM system to print a
mold with 3D structure of a normal ear in half size. Then, PGA
unwoven fibers were inserted into the mold and coated with
0.3% PLA solution, thus generating a relatively solid ear-
shaped scaffold material. The resulting scaffold was laser-
scanned to generate a 3D image, which could be digitally
compared with the original ear 3D image to analyze the simi­
larity in 3D structure. As revealed in Fig. 17.22, the resulting
ear-shaped scaffold achieved a similarity level of above 97%
compared to the positive mold of original ear shape, indicat­
ing that the mold fabricated by CAD/CAM can fabricate a
scaffold accurately into an ear shape that is mirror-symmetrical
to the normal ear.
Afterwards, a total of 50 × 106 cells in 1 mL volume were
evenly seeded on to the ear-shaped scaffold and cultured in
vitro with medium change at regular time intervals. Interest­
ingly, a human ear cartilage could be generated in vitro after
12 weeks of culture with good elasticity (Fig. 17.23). The
engineered cartilage also revealed relatively mature histologi­
cal structure of cartilage with lacuna structure formation and
strong staining for Safranin-O and collagen II, as shown in
Fig. 17.24. More importantly, the human ear-shaped cartilage
formed in vitro could reach a morphological similarity of
82.6% to the positive ear mold, indicating that this technique
not only can generate cartilage tissue in vitro but is also able
After 1 week of in vitro culture, the cell–scaffold construct
was implanted into the subcutaneous tissue of a nude mouse
with the support of an external stent outside the skin to keep
the shape (Fig. 17.19). As shown in Fig. 17.20, an ear cartilage
structure is formed with the 3D shape that is almost identical
to that of a human ear after 12 weeks of in vivo implantation.
To verify the formation of engineered ear cartilage, tissues
were harvested and subcutaneous tissues were stripped and
the resulting gross view and histology confirmed a well-
formed ear-shaped cartilage (Fig. 17.21). Through this pioneer­
ing study, the result provides an example of how to translate
cartilage engineering research to plastic surgery application,
particularly for engineered auricular reconstruction.63
To explore the possibility of translating this technique to
clinical application, a technique of in vitro engineering of
human ear-shaped cartilage was developed with the assis­
tance of computer-aided design (CAD) and computer-aided
manufacture (CAM) technologies.65
To proceed, computed tomography was employed to scan
a patient’s normal ear to collect the geometric data; the infor­
mation was then processed by a CAD system to generate both
positive and negative image data of the normal ear. The
Fig. 17.18  Chondrocytes seeded on to the polymer ear mold in vitro. (A) Gross
appearance of the ear polymer mold seeded with chondrocytes (1.5 × 108).
(B) Scanning electron micrograph showing adherence of chondrocytes to
polyglycolic acid device before implantation. (Reproduced from Cao Y, Vacanti JP,
Paige KT, et al. Transplantation of chondrocytes utilizing a polymer–cell construct to
produce tissue-engineered cartilage in the shape of a human ear. Plast Reconstr
Surg. 1997;100:297–302.)
A
B
Fig. 17.19  An external stent was fixed on the outside skin of the polymer ear
implant to maintain polymer mold shape (A) and the view of an external stent (B).
(Reproduced from Cao Y, Vacanti JP, Paige KT, et al. Transplantation of chondrocytes
utilizing a polymer-cell construct to produce tissue-engineered cartilage in the shape
of a human ear. Plast Reconstr Surg. 1997;100:297–302.)
A
B

277
Cartilage engineering
Fig. 17.20  (A–E) Gross appearance of the constructs 12 weeks after subcutaneous implantation into nude mice. Note the three-dimensional shape that is almost identical
to that of a human ear. (Reproduced from Cao Y, Vacanti JP, Paige KT, et al. Transplantation of chondrocytes utilizing a polymer-cell construct to produce tissue-engineered
cartilage in the shape of a human ear. Plast Reconstr Surg. 1997;100:297–302.)
C
B
A
D
E
Fig. 17.21  (A) Gross view of engineered ear-shaped cartilage with (left) or without (right) external stent fixation. Note: fine contour of the new cartilaginous ear is
maintained in the group with stent application. (B) Histology of engineered ear-shaped cartilage (hematoxylin and eosin, ×320). (Modified from Cao Y, Vacanti JP, Paige KT,
et al. Transplantation of chondrocytes utilizing a polymer-cell construct to produce tissue-engineered cartilage in the shape of a human ear. Plast Reconstr Surg.
1997;100:297–302.)
A
B
to maintain a designed 3D tissue structure (see Fig. 17.23).65
Currently, continuous effort is being made in our center to
study the fate of the in vitro-engineered ear-shaped cartilage
after in vivo implantation and the progress in this area is
expected to pave the way for the eventual clinical application
of engineered human ear-shaped cartilage.
One of the unique properties of human ear cartilage is
its excellent elasticity and flexibility that allow for torsion
and bending without causing a fracture. Mimicking such
mechanical properties of the native tissue should also be an
important consideration for engineering ear-shaped carti­
lage and its application. In 2005, Xu et al. reported a new
method of generating flexible auricular cartilage.66 In their
study, auricular chondrocytes were isolated from swine ear
cartilage by enzyme digestion and fibrin polymer was used
as the scaffold. In addition, auricular perichondrium was

278
CHAPTER 17 • Repair, grafting, and engineering of cartilage
Fig. 17.22  Preparation and shape analysis of the ear-shaped scaffolds. (A) Three-dimensional (3D) image of the normal ear. (B) The mirror image of A. (C) The half-sized
resin-positive mold. (D) Laser scan image of C. (E) Color map of D. (F) Inner part of the resin-negative mold fabricated by 3D printing. (G) Outer part of the negative mold
cast from F with silicone rubber. (H) The ear-shaped polylactic acid/polyglycolic acid scaffold. (I) Laser scan image of H. (J) Color map of I compared to E. Color bar at
the right side represents similarity from highest (top, blue) to lowest (bottom, red). (Reproduced from Liu Y, Zhang L, Zhou G, et al. In vitro engineering of human ear-shaped
cartilage assisted with CAD/CAM technology. Biomaterials. 2010;31:2176–2183.)
A
B
C
D
E
F
G
H
I
J
Fig. 17.23  Shape evaluation of the ear-shaped constructs. The scaffold shows an accurate ear-like structure (A) with a high similarity level compared to the positive mold
(E). All the cell scaffold constructs largely maintain their original ear-like structures at 4 weeks (B), 8 weeks (C), and 12 weeks (D). Quantitative analysis shows over 84%
shape similarity in all the samples (F–H) compared to the positive mold. (Reproduced from Liu Y, Zhang L, Zhou G, et al. In vitro engineering of human ear-shaped cartilage
assisted with CAD/CAM technology. Biomaterials. 2010;31:2176–2183.)
A
B
C
D
E
F
G
H

279
Cartilage engineering
Fig. 17.24  Histological examinations of the in vitro ear-shaped constructs. At 4 weeks, the constructs form heterogeneous cartilage-like tissue along with undegraded
polyglycolic acid fibers (A, D, G). With prolonged culture time, the histological structure of the constructs gradually becomes compact, accompanied by increased numbers
of lacuna structures at 8 weeks (B, E, H). Homogeneous cartilage with abundant extracellular matrix and mature lacunae is observed at 12 weeks (C, F, I) with no visible
scaffold residuals in the constructs. The black arrows indicate the undegraded polyglycolic acid fibers. Scale bar = 100 µm. (Reproduced from Liu Y, Zhang L, Zhou G, et al.
In vitro engineering of human ear-shaped cartilage assisted with CAD/CAM technology. Biomaterials. 2010;31:2176–2183.)
C
B
A
F
E
D
I
H
G
harvested and lyophilized. During the engineering process,
equal volumes of both the chondrocyte fibrinogen suspension
and the thrombin solution were mixed together to give rise
to a chondrocyte–fibrin polymer suspension. The suspen­
sion was then placed into an ear-shaped well made from
bone wax to allow for polymerization. To produce a flexible
cartilage, the polymerized chondrocyte–fibrin construct was
sandwiched between two layers of lyophilized swine peri­
chondrium to form a trilayer, ear-shaped construct and then
implanted into athymic mice subcutaneously. As a control,
chondrocyte–fibrin construct without perichondrium was
implanted as well. At 12 weeks of implantation, engineered
cartilaginous tissue was formed in both the experimental and
control groups.
Importantly, engineered cartilage laminated with peri­
chondrium exhibited mechanical properties similar to those
of the native swine ear, and could tolerate severe torsion and
bending without fracture, although the 3D structure remains
to be improved. Histology also revealed good integration
between the engineered cartilage and lyophilized perichon­
drium (Fig. 17.25). In contrast, the engineered cartilage of
the control group became fractured after the test of torsion
and bending. The result of this study indicates that lamina­
tion with lyophilized perichondrium is a reliable method of
conferring elastic-like flexibility to an engineered ear carti­
lage. In the future, selection of a proper solid scaffold may
help to control the precise and detailed structure better in
order to generate flexible auricular cartilage with better 3D
structure.66
In addition to the experimental studies that employed
animal chondrocytes, human chondrocytes have also been
investigated for their ability to form engineered cartilage.
Kamil et al. reported a comparative study of normal chondro­
cytes and microtia chondrocytes from patients and the results
showed their similar abilities in cell proliferation and cartilage
tissue formation, indicating that cartilage derived from

280
CHAPTER 17 • Repair, grafting, and engineering of cartilage
Fig. 17.25  (A, B) Gross mechanical testing of the ear-shaped framework at 12 weeks after insertion demonstrates a continued high degree of flexibility. (C) After
mechanical testing, the framework easily recovers its initial shape. (D) Hematoxylin and eosin staining of flexible tissue-engineered cartilage (original magnification ×100;
bar = 100 µm), demonstrating the tight adherence at the interface between neocartilage and the lyophilized swine perichondrium laminate (arrows indicate interface).
(Modified from Xu JW, Johnson TS, Motarjem PM, et al. Tissue-engineered flexible ear-shaped cartilage. Plast Reconstr Surg. 2005;115:1633–1641.)
A
B
C
D
microtia patients may potentially serve as the cell source to
engineer human auricular cartilage.67
Interestingly, Yanaga et al. reported a clinical trial of auricu­
lar reconstruction with fabricated human cartilage. In this trial,
chondrocytes isolated from the cartilage remnant of microtia
patients were cultured in vitro using a multilayer culture
method, then the expanded cells along with their produced
gelatinous chondroid matrices were collected into a syringe
in the cell density of 0.5–1 × 107 cells per mL. Afterwards,
40–73 mL of cells/gelatinous matrices was injected into a sub­
cutaneous pocket on the fascia of a patient’s lower abdomen
using a 16-gauge indwelling needle in order to generate an
engineered cartilage block which could be sculptured into an
auricular framework (Fig. 17.26). The results showed that a
large piece of elastic cartilage was formed at the abdominal
environment at 6 months postimplantation, which was also
verified by histology examination. Importantly, this engi­
neered cartilage block provides sufficient cartilage to create
an ear framework. Furthermore, the framework could well
support the reconstructed auricle without obvious absorption
after being implanted for several years (see Fig. 17.26). This
preliminary study provides important supporting evidence
for the future clinical application of engineered auricular
reconstruction.68
Engineered cartilage for rhinoplasty and
facial contouring
In addition to auricular reconstruction, rhinoplasty is another
area where tissue-engineered cartilage may potentially be
applied clinically. In contrast to auricular reconstruction,
injectable cartilage is an important physical form for trans­
planting engineered cartilage.
In the early stage of tissue engineering research, injectable
cartilage was employed to explore the feasibility of cartilage
engineering using slowly polymerizing calcium alginate gels69
or Pluronic gel,70 amongst others. In recently published
studies, human chondrocytes have also been confirmed to be
able to form injectable cartilage using different injectable
scaffolds.71,72
Although most engineered injectable cartilage research
remains at the experimental stage using either animal cells or
an animal model, there are a few reports of the clinical applica­
tion of injectable cartilage.73,74 Yanaga et al. reported a clinical
trial of injecting cultured autologous chondrocytes for nasal
augmentation. First, a piece of cartilage was harvested from
auricular concha to extract chondrocytes and the cells were
expanded in vitro into a gel-type cell-containing mass. Then,
a subperiosteal skin pocket was created on the nasal dorsum

281
Cartilage engineering
A
B
C
D
E
F
Fig. 17.26  Engineering of human cartilage
block for auricular reconstruction.
(A) Injection of cultured chondrocytes
subcutaneously. (B) Formation of
engineered cartilage block with good
elasticity. (C) Histological verification of
engineered elastic cartilage. (D) An ear
framework, sculptured from the engineered
neocartilage block. (E) A 9-year-old boy
with microtia before operation. (F) 2 years
after auricular reconstruction with
engineered framework. (Modified from
Yanaga H, Imai K, Fujimoto T, et al.
Generating ears from cultured autologous
auricular chondrocytes by using two-stage
implantation in treatment of microtia. Plast
Reconstr Surg. 2009;124:817–825.)

282
CHAPTER 17 • Repair, grafting, and engineering of cartilage
Fig. 17.27  A 35-year-old woman received injected cartilage to correct saddle-nose deformity. (A–C) Preoperative views. The autologous cultured auricular chondrocytes
were grafted to the nasal dorsum. (D–F) Postoperative views after 34 months. The projection radix is higher than before grafting. The shape and contours of the nasal
dorsum have aesthetically improved. Good contours have been maintained, and there has been no absorption of the grafted chondrocytes. (Reproduced from Yanaga H,
Yanaga K, Imai K, et al. Clinical application of cultured autologous human auricular chondrocytes with autologous serum for craniofacial or nasal augmentation and repair.
Plast Reconstr Surg. 2006;117:2019–2030.)
C
B
A
F
E
D
of the patients and the chondrocytes were transplanted within
a gel mass and the wound was closed. The implantation site
was fixed with splint and tape for 3 weeks. Biopsy of injected
tissue confirmed the formation of engineered cartilage after 1
month of implantation.
The procedure was applied in 8 patients for primary
nasal augmentation or to correct a deformity resulting from
silicone implant exposure or deviation. The patients were
followed up for 6 months to 2 years and injected cartilage
could maintain its shape without resorption.73 Later the
same group further reported clinical trials in 32 patients for
both nasal augmentation and facial contouring, with excel­
lent clinical evaluation results and patient satisfaction in
most cases.74 Fig. 17.27 demonstrates the clinical outcome of
injected cartilage for nasal augmentation. Additionally, the
authors used injected autologous chondrocytes for chin aug­
mentation and to correct temporal and forehead depressed
deformity.73,74

283
Future directions
In addition to injectable cartilage, in vivo-engineered car­
tilage may potentially be used for nasal reconstruction. For
example, Farhadi et al. reported the engineering of human nasal
cartilage. In the study, chondrocytes were isolated from nasal
septum and seeded onto nonwoven meshes of esterified hyal­
uronan (Hyaff-11). The cell–scaffold construct was cultured in
vitro for 2 or 4 weeks and then implanted into nude mouse. The
results showed that the cartilage was well formed after 2 weeks
of in vivo implantation and the engineered cartilage exhibited
good mechanical property.75 Similar results were also reported
in another study.76 With enhanced mechanical strength, these
in vivo-engineered cartilages may provide the autologous car­
tilage grafts potentially for septum or nasal alar reconstruction
or nasal augmentation. In vitro engineering of cartilage with a
nasal tip has been reported many years ago, aiming for nasal
reconstruction with specifically shape-designed cartilage,77 but
its relatively weak mechanical strength might prohibit in vitro-
engineered cartilage from immediate transplantation in nasal
reconstruction.
Engineered cartilage for joint cartilage repair
and reconstruction
Engineered cartilage may also serve as a composite graft to
repair or reconstruct articular cartilage. There are two poten­
tial targets: temporomandibular joint (TMJ) and digital joints.
In 2001, Weng et al. reported a preliminary study of mandible
condylar reconstruction with tissue-engineered composites
of bone and cartilage. In the study, scaffold of PLA-coated
PGA was molded into the shape of human mandible condyle
and respectively seeded with osteoblasts and chondrocytes
followed by in vivo implantation in nude mouse. After 12
weeks, a condyle-shaped bone tissue was well formed with
articular cartilage on the surface and histology confirmed the
formation of trabecular bone and hyaline cartilage, suggest­
ing a potential approach for clinical TMJ reconstruction.78 In
addition, engineered cartilage was proposed for TMJ disc
reconstruction to potentially restore TMJ function.79
Cartilage presents as the major tissue component in inter­
phalangeal joints or metacarpophalangeal joints and plays an
important functional role. As early as 1999, Isogai et al.
reported a preliminary study on constructing small phalanges
and the whole joints using a tissue-engineering approach.80
The approach included the use of PLA or PGA as the scaffold
and fresh bovine periosteum was harvested to wrap around
the scaffold for bone engineering. Additionally, chondrocytes
and tenocytes were seeded on the scaffold to form articular
cartilage and associated tendon tissue. The results showed
that a preliminary composite tissue formed with the shape
and dimensions of human phalanges as well as the joints.80 A
study with much larger sample size was also performed to
engineer composite tissue of human phalanges and a small
joint, with successful results.81 Although these studies were
preliminary, the results revealed the promise of the future
application of engineered cartilage in joint reconstruction.
Future directions
Despite a few preliminary clinical trials of engineered carti­
lages in auricular and nasal reconstruction, there remains
much to do in both basic and applied research prior to true
translation of cartilage research to clinical application. To fulfill
the goal, the following areas may represent future directions.
Stem cell-based cartilage engineering
Although both auricular and nasal cartilages have been used
as the sources to extract chondrocytes, the available tissue
amounts are limited. Although more tissue volume can be
harvested from rib cartilage, the amount of cells that can be
harvested is limited due to low cellular density. Stem cells,
particularly adult stem cells, have become an alternative cell
source because of their multidifferentiation potential (includ­
ing chondrogenic) and strong proliferation capability.
The adult mesenchymal stem cells derived from bone
marrow and adipose tissue are the most feasible cell source
for cartilage engineering.82 Chondrogenic differentiation can
be achieved with growth factor induction83 or co-culture with
chondrocytes84 or with chondrogenic matrices.85 In recent
years, mimicking native histoarchitecture and the molecular
signaling of cartilage microenvironment in scaffold design
have been considered important to facilitate stem cell chon­
drogenic differentiation and cartilage formation.86 Therefore,
optimization of chondrogenic induction regime by proper
combination of paracrine factors, chondrogenic matrices, and
suitable topographical structure of scaffold is likely to enhance
chondrogenic differentiation of stem cells, and improve the
structure and function of stem cell-engineered cartilage.
In vitro engineering of cartilage with
enhanced mechanical strength
As described above, the application of plastic surgery requires
immediately available cartilage graft to generate an auricular
or a nasal frame for reconstructive surgical procedures. In
vivo engineering cartilage, first using the human body as a
bioreactor and then reshaping the in vivo-engineered cartilage
block into a desired frame, would require at least two-stage
operations.68 Furthermore, patients may undergo unnecessary
suffering if the first stage of in vivo engineering fails. Therefore,
in vitro cartilage engineering can not only avoid two-stage
operations, but also reduce the risk of failure because more
back-up engineered cartilages can be made simultaneously in
vitro without causing patient suffering.
Today, in vitro cartilage engineering has already been proven
feasible by experimental studies using either chondrocytes65
or mesenchymal stem cells.87 Nevertheless, the mechanical
strength of in vitro-engineered cartilage remains relatively
weak compared to that of in vivo-engineered cartilage.88 This
is likely caused by the differences between in vitro and in
vivo microenvironments, and the latter can promote the matu­
ration of engineered cartilage and thus lead to differential
expression of key matrix molecules such as collagen IX and
pyridinoline, and enhanced mechanical strength.88 Although
it remains difficult to define the environmental mechanism,
partially mimicking in vivo microenvironmental factors may
help to enhance the mechanical property of in vitro-engineered
cartilage. For example, when an articular cartilage was engi­
neered in a bioreactor with dynamic loading, it was found
that continuous dynamic loading could significantly increase
Young’s modulus of the engineered cartilage. In addition,
the loaded cartilage increased the production of cartilage
oligomeric matrix protein and collagens II and IX.89 In the

284
CHAPTER 17 • Repair, grafting, and engineering of cartilage
future, mechanical loading as well as other factors that are
able to facilitate collagen maturation may represent one of the
future directions for in vitro cartilage engineering.
Design and precise control of engineered
cartilage 3D structure
For engineered cartilage application in auricular and nasal
reconstruction, precise control of 3D structure is necessary to
achieve the goal. As previously described, a CAD/CAM
system is essential to collect the 3D information and to fabri­
cate the scaffold with a specific 3D structure that matches a
patient’s own auricular or nasal structure. In addition, the
fabrication of a suitable scaffold that is able to maintain the
precise 3D structure during cartilage formation may be an
important consideration. Furthermore, an inner core stent
with slow degradation rate may help to maintain the 3D
structure if the inner core starts to degrade after mature car­
tilage formation.
6. Brent B. Repair and grafting of cartilage and perichondrium. In:
McCarthy JC, ed. Plastic Surgery. Philadelphia: WB Saunders;
1990:559–582.
8. Han K, Kim J, Son D, et al. How to harvest the maximal amount
of conchal cartilage grafts. J Plast Reconstr Aesthet Surg.
2008;61:1465–1471.
12. Singh DJ, Bartlett SP. Aesthetic management of the ear as a donor
site. Plast Reconstr Surg. 2007;120:899–908. The ear is a popular donor
site for chondrocutaneous grafts in nasal reconstruction. The authors
review their experience with these procedures to offer insights in
maximizing aesthetic donor site outcomes.
17. Scuderi N, Ribuffo D, Chiummariello S. Total and subtotal upper
eyelid reconstruction with the nasal chondromucosal flap: a
10-year experience. Plast Reconstr Surg. 2005;115:1259–1265. A
technique for upper eyelid reconstruction with a nasal chondromucosal
flap is presented. Excellent flap viability and an average of 13-mm
levator excursion is reported.
23. Brent B. Auricular repair with autogenous rib cartilage grafts: two
decades of experience with 600 cases. Plast Reconstr Surg.
1992;90:355–374.
30. Uppal RS, Sabbagh W, Chana J, et al. Donor-site morbidity after
autologous costal cartilage harvest in ear reconstruction and
approaches to reducing donor-site contour deformity. Plast
Reconstr Surg. 2008;121:1949–1955.
31. Kawanabe Y, Nagata S. A new method of costal cartilage harvest
for total auricular reconstruction: part I. Avoidance and prevention
of intraoperative and postoperative complications and problems.
Plast Reconstr Surg. 2006;117:2011–2018.
63. Cao Y, Vacanti JP, Paige KT, et al. Transplantation of chondrocytes
utilizing a polymer-cell construct to produce tissue-engineered
cartilage in the shape of a human ear. Plast Reconstr Surg.
1997;100:297–302. This paper describes a strategy to tissue engineer
cartilage in the shape of a human ear on the dorsum of athymic mice.
Histological and morphometric outcomes are assessed.
65. Liu Y, Zhang L, Zhou G, et al. In vitro engineering of human
ear-shaped cartilage assisted with CAD/CAM technology.
Biomaterials. 2010;31:2176–2183. The authors report their efforts in
tissue engineering CAD/CAM-designed tissue engineered auricles on
PLA/PGA scaffolds in vitro.
68. Yanaga H, Imai K, Fujimoto T, et al. Generating ears from cultured
autologous auricular chondrocytes by using two-stage
implantation in treatment of microtia. Plast Reconstr Surg.
2009;124:817–825. A technique for in vivo chondrogenesis in a
heterotopic abdominal pocket is reported. Cartilaginous frameworks for
auricular reconstruction were successfully modeled and implanted from
this neocartilage in four patients, with promising reconstructive
outcomes.
Access the complete reference list online at

284.e1
References
References
1. Wheeless III CR, Nunley II JA, Urbaniak JR. Wheeless’ Textbook of
Orthopaedics. Available online from: <
.com>.
2. Gelse K, von der Mark K, Aigner T, et al. Articular cartilage repair
by gene therapy using growth factor-producing mesenchymal
cells. Arthritis Rheum. 2003;48:430–441.
3. Dounchis JS, Goomer RS, Harwood FL, et al. Chondrogenic
phenotype of perichondrium-derived chondroprogenitor cells is
influenced by transforming growth factor-beta 1. J Orthop Res.
1997;15:803–807.
4. Sushruta S. The Sushruta Samhita. Bhishagratna KK (translator).
Calcutta: Kaviraj Kunja Lal Bhishagratna; 1907.
5. Donald PJ. Cartilage grafting in facial reconstruction with
special consideration of irradiated grafts. Laryngoscope.
1986;96:786–807.
6. Brent B. Repair and grafting of cartilage and perichondrium. In:
McCarthy JC, ed. Plastic Surgery. Philadelphia: WB Saunders;
1990:559–582.
7. Ohlsén L. Cartilage regeneration from perichondrium.
Experimental studies and clinical applications. Plast Reconstr Surg.
1978;62:507–513.
8. Han K, Kim J, Son D, et al. How to harvest the maximal amount
of conchal cartilage grafts. J Plast Reconstr Aesthet Surg.
2008;61:1465–1471.
9. Brent B. The acquired auricular deformity. A systematic approach
to its analysis and reconstruction. Plast Reconstr Surg.
1977;59:475–485.
10. Ono I, Gunji H, Sato M, et al. A method of treatment of constricted
ears with a conchal cartilage graft to the posterior auricular plane.
Plast Reconstr Surg. 1993;92:621–627.
11. Firmin F, Sanger C, O’Toole G. Ear reconstruction following severe
complications of otoplasty. J Plast Reconstr Aesthet Surg.
2008;61(suppl 1):S13–S20.
12. Singh DJ, Bartlett SP. Aesthetic management of the ear as a donor
site. Plast Reconstr Surg. 2007;120:899–908. The ear is a popular donor
site for chondrocutaneous grafts in nasal reconstruction. The authors
review their experience with these procedures to offer insights in
maximizing aesthetic donor site outcomes.
13. Haug MD, Rieger UM, Witt P, et al. Managing the ear as a donor
site for composite graft in nasal reconstruction: update on
technical refinements and donor site morbidity in 110 cases. Ann
Plast Surg. 2009;63:171–175.
14. Schroeder WA Jr. Closure of conchal defects: The hinged
retroconchal island flap with overlying cutaneous transposition
flap. Dermatol Surg. 1995;21:560–562.
15. Zhang YX, Yang J, Wang D, et al. Extended applications of
vascularized preauricular and helical rim flaps in reconstruction of
nasal defects. Plast Reconstr Surg. 2008;121:1589–1597.
16. Murrell GL. Dorsal augmentation with septal cartilage. Semin Plast
Surg. 2008;22:124–135.
17. Scuderi N, Ribuffo D, Chiummariello S. Total and subtotal upper
eyelid reconstruction with the nasal chondromucosal flap: a
10-year experience. Plast Reconstr Surg. 2005;115:1259–1265. A
technique for upper eyelid reconstruction with a nasal chondromucosal
flap is presented. Excellent flap viability and an average of 13-mm
levator excursion is reported.
18. Bozkurt AK, Cansiz H. Tracheal reconstruction with autogenous
composite nasal septal graft. Ann Thorac Surg. 2002;74:2200–2201.
19. Byrd HS, Andochick S, Copit S, et al. Septal extension grafts: a
method of controlling tip projection shape. Plast Reconstr Surg.
1997;100:999–1010.
20. Tanzer RC. Total reconstruction of the auricle. The evolution of a
plan of treatment. Plast Reconstr Surg. 1971;47:523–533.
21. Thomson HG, Kim TY, Ein SH. Residual problems in chest donor
sites after microtia reconstruction: a long-term study. Plast Reconstr
Surg. 1995;95:961–968.
22. Brent B. The correction of microtia with autogenous cartilage
grafts: II. Atypical and complex deformities. Plast Reconstr Surg.
1980;66:13–21.
23. Brent B. Auricular repair with autogenous rib cartilage grafts: two
decades of experience with 600 cases. Plast Reconstr Surg.
1992;90:355–374.
24. Nagata S. A new method of total reconstruction of the auricle for
microtia. Plast Reconstr Surg. 1993;92:187–201.
25. Nagata S. Modification of the stages in total reconstruction of the
auricle: Part I. Grafting the three-dimensional costal cartilage
framework for lobule-type microtia. Plast Reconstr Surg.
1994;93:221–230.
26. Nagata S. Modification of the stages in total reconstruction of the
auricle: Part II. Grafting the three-dimensional costal cartilage
framework for concha-type microtia. Plast Reconstr Surg.
1994;93:231–242.
27. Nagata S. Modification of the stages in total reconstruction of the
auricle: Part III. Grafting the three-dimensional costal cartilage
framework for small concha-type microtia. Plast Reconstr Surg.
1994;93:243–253.
28. Nagata S. Modification of the stages in total reconstruction of the
auricle: Part IV. Ear elevation for the constructed auricle. Plast
Reconstr Surg. 1994;93:254–266.
29. Brent B, ed. The Artistry of Reconstructive Surgery. St Louis: Mosby;
1987.
30. Uppal RS, Sabbagh W, Chana J, et al. Donor-site morbidity after
autologous costal cartilage harvest in ear reconstruction and
approaches to reducing donor-site contour deformity. Plast
Reconstr Surg. 2008;121:1949–1955.
31. Kawanabe Y, Nagata S. A new method of costal cartilage harvest
for total auricular reconstruction: part I. Avoidance and prevention
of intraoperative and postoperative complications and problems.
Plast Reconstr Surg. 2006;117:2011–2018.
32. Kawanabe Y, Nagata S. A new method of costal cartilage harvest
for total auricular reconstruction: part II. Evaluation and analysis
of the regenerated costal cartilage. Plast Reconstr Surg.
2007;119:308–315.
33. Chin W, Zhang R, Zhang Q, et al. Modifications of three-
dimensional costal cartilage framework grafting in auricular
reconstruction for microtia. Plast Reconstr Surg.
2009;124:1940–1946.
34. Zhang Q, Zhang R, Xu F, et al. Auricular reconstruction for
microtia: personal 6-year experience based on 350 microtia ear
reconstructions in China. Plast Reconstr Surg. 2009;123:849–858.
35. Ahmed A, Imani P, Vuyk HD. Reconstruction of significant saddle
nose deformity using autogenous costal cartilage graft with
incorporated mirror image spreader grafts. Laryngoscope.
2010;120:491–494.
36. Bhatt YC, Vyas KA, Tandale MS, et al. Maxillonasal dysplasia
(Binder’s syndrome) and its treatment with costal cartilage graft: A
follow-up study. Indian J Plast Surg. 2008;41:151–159.
37. Cheng MH, Rodriguez ED, Smartt JM, et al. Nipple reconstruction
using the modified top hat flap with banked costal cartilage graft:
long-term follow-up in 58 patients. Ann Plast Surg.
2007;59:621–628.
38. Moshaver A, Gantous A. The use of autogenous costal cartilage
graft in septorhinoplasty. Otolaryngol Head Neck Surg.
2007;137:862–867.
39. Shinohara H, Yuzuriha S, Matsuo K, et al. Tracheal reconstruction
with a prefabricated deltopectoral flap combined with costal
cartilage graft and palatal mucosal graft. Ann Plast Surg.
2004;53:278–281.
40. Lester CW. Tissue replacement after subperichondrial resection of
costal cartilage: two case reports. Plast Reconstr Surg.
1959;23:49–54.
41. Skoog T, Ohlsén L, Sohn SA. Perichondrial potential for
cartilagenous regeneration. Scand J Plast Reconstr Surg.
1972;6:123–125.
42. Ohlsen L. Cartilage formation from free perichondrial
grafts: an experimental study in rabbits. Br J Plast Surg.
1976;29:262–267.

284.e2
CHAPTER 17 • Repair, grafting, and engineering of cartilage
43. Skoog T, Johansson SH. The formation of articular cartilage from
free perichondrial grafts. Plast Reconstr Surg. 1976;57:1–6.
44. Upton J, Sohn SA, Glowacki J. Neo-cartilage derived from
transplanted perichondrium: what is it? Plast Reconstr Surg.
1981;68:166–174.
45. Sari A, Tuncer S, Ayhan S, et al. What wrapped perichondrial and
periosteal grafts offer as regenerators of new tissue. J Craniofac
Surg. 2006;17:1137–1143.
46. Engkvist O, Johansson SH, Ohlsén L, et al. Reconstruction of
articular cartilage using autologous perichondrial grafts. A
preliminary report. Scand J Plast Reconstr Surg. 1975;9:203–206.
47. Bouwmeester SJ, Beckers JM, Kuijer R, et al. Long-term results of
rib perichondrial grafts for repair of cartilage defects in the human
knee. Int Orthop. 1997;21:313–317.
48. Bruns J, Steinhagen J. Treatment of deep hyalin cartilage defects
with autologous perichondrial grafts. Int J Sports Med.
2003;24:382–388.
49. Sully L, Jackson IT, Sommerlad BC. Perichondrial grafting in
rheumatoid metacarpophalangeal joints. Hand. 1980;12:137–148.
50. Sato K, Sasaki T, Nakamura T, et al. Clinical outcome and
histologic findings of costal osteochondral grafts for cartilage
defects in finger joints. J Hand Surg Am. 2008;33:511–515.
51. Fujiwara M, Kawakatsu M, Fukamizu H. Two-stage arthroplasty
with joint distraction and costal osteochondral grafting for
ankylosis of a metacarpophalangeal joint: nine years’ follow-up. J
Plast Reconstr Aesthet Surg. 2008;61:e1–e4.
52. Boccieri A, Marianetti TM. Perichondrium graft: harvesting and
indications in nasal surgery. J Craniofac Surg.
2010;21:40–44.
53. Boccieri A. The perichondrium graft in revision rhinoplasty. Plast
Reconstr Surg. 2008;122:216e–217e.
54. Brent B, Ott R. Perichondro-cutaneous graft. Plast Reconstr Surg.
1978;62:1–14.
55. Stucker FJ, Walsh WE, Dammert M, et al. The perichondrial
cutaneous graft: a facial reconstructive option for the ages (ages 1
week to 94 years). Laryngoscope. 2008;118:1753–1757.
56. Patterson AR, Brady G, Walker PD, et al. The perichondrial
cutaneous graft and “flip-flop” flap in facial reconstruction: a
series of 41 cases. Br J Oral Maxillofac Surg. 2008;46:114–118.
57. Kalbermatten DF, Haug M, Wettstein R, et al. New posterior
auricular perichondrial cutaneous graft for stable reconstruction of
nasal defects. Aesthetic Plast Surg. 2005;29:489–495.
58. Gloster HM Jr, Brodland DG. The use of perichondrial cutaneous
grafts to repair defects of the lower third of the nose. Br J Dermatol.
1997;136:43–46.
59. Stucker FJ, Sanders KW. A method to repair auricular defects
after perichondrial cutaneous grafting. Laryngoscope.
2002;112:1384–1386.
60. Stucker FJ Jr, Shaw GY. The perichondrial cutaneous graft. A
12-year clinical experience. Arch Otolaryngol Head Neck Surg.
1992;118:287–292.
61. Vacanti CA, Langer R, Schloo B, et al. Synthetic polymers seeded
with chondrocytes provide a template for new cartilage formation.
Plast Reconstr Surg. 1991;88:753–759.
62. Vacanti CA, Kim W, Upton J, et al. Tissue-engineered growth of
bone and cartilage. Transplant Proc. 1993;25:1019–1021.
63. Cao Y, Vacanti JP, Paige KT, et al. Transplantation of chondrocytes
utilizing a polymer-cell construct to produce tissue-engineered
cartilage in the shape of a human ear. Plast Reconstr Surg.
1997;100:297–302. This paper describes a strategy to tissue engineer
cartilage in the shape of a human ear on the dorsum of athymic mice.
Histological and morphometric outcomes are assessed.
64. Stock UA, Vacanti JP. Tissue engineering: current state and
prospects. Annu Rev Med. 2001;52:443–451.
65. Liu Y, Zhang L, Zhou G, et al. In vitro engineering of human
ear-shaped cartilage assisted with CAD/CAM technology.
Biomaterials. 2010;31:2176–2183. The authors report their efforts in
tissue engineering CAD/CAM-designed tissue engineered auricles on
PLA/PGA scaffolds in vitro.
66. Xu JW, Johnson TS, Motarjem PM, et al. Tissue-engineered flexible
ear-shaped cartilage. Plast Reconstr Surg.
2005;115:1633–1641.
67. Kamil SH, Vacanti MP, Vacanti CA, et al. Microtia chondrocytes as
a donor source for tissue-engineered cartilage. Laryngoscope.
2004;114:2187–2190.
68. Yanaga H, Imai K, Fujimoto T, et al. Generating ears from cultured
autologous auricular chondrocytes by using two-stage
implantation in treatment of microtia. Plast Reconstr Surg.
2009;124:817–825. A technique for in vivo chondrogenesis in a
heterotopic abdominal pocket is reported. Cartilaginous frameworks for
auricular reconstruction were successfully modeled and implanted from
this neocartilage in four patients, with promising reconstructive
outcomes.
69. Paige KT, Cima LG, Yaremchuk MJ, et al. Injectable cartilage. Plast
Reconstr Surg. 1995;96:1390–1398.
70. Saim AB, Cao Y, Weng Y, et al. Engineering autogenous cartilage
in the shape of a helix using an injectable hydrogel scaffold.
Laryngoscope. 2000;110:1694–1697.
71. Dobratz EJ, Kim SW, Voglewede A, et al. Injectable cartilage: using
alginate and human chondrocytes. Arch Facial Plast Surg.
2009;11:40–47.
72. Oliveira JT, Santos TC, Martins L, et al. Gellan gum injectable
hydrogels for cartilage tissue engineering applications: in vitro
studies and preliminary in vivo evaluation. Tissue Eng Part A.
2010;16:343–353.
73. Yanaga H, Koga M, Imai K, et al. Clinical application of
biotechnically cultured autologous chondrocytes as novel graft
material for nasal augmentation. Aesthetic Plast Surg.
2004;28:212–221.
74. Yanaga H, Yanaga K, Imai K, et al. Clinical application of cultured
autologous human auricular chondrocytes with autologous serum
for craniofacial or nasal augmentation and repair. Plast Reconstr
Surg. 2006;117:2019–2030.
75. Farhadi J, Fulco I, Miot S, et al. Precultivation of engineered
human nasal cartilage enhances the mechanical properties relevant
for use in facial reconstructive surgery. Ann Surg.
2006;244:978–985.
76. Haisch A, Duda GN, Schroeder D, et al. The morphology and
biomechanical characteristics of subcutaneously implanted
tissue-engineered human septal cartilage. Eur Arch
Otorhinolaryngol. 2005;262:993–997.
77. Kamil SH, Kojima K, Vacanti MP, et al. In vitro tissue engineering
to generate a human-sized auricle and nasal tip. Laryngoscope.
2003;113:90–94.
78. Weng Y, Cao Y, Silva CA, et al. Tissue-engineered composites of
bone and cartilage for mandible condylar reconstruction. J Oral
Maxillofac Surg. 2001;59:185–190.
79. Allen KD, Athanasiou KA. Tissue Engineering of the TMJ disc: a
review. Tissue Eng. 2006;12:1183–1196.
80. Isogai N, Landis W, Kim TH, et al. Formation of phalanges and
small joints by tissue-engineering. J Bone Joint Surg Am.
1999;81:306–316.
81. Landis WJ, Jacquet R, Hillyer J, et al. Design and assessment of a
tissue-engineered model of human phalanges and a small joint.
Orthod Craniofac Res. 2005;8:303–312.
82. Caplan AI. Adult mesenchymal stem cells for tissue engineering
versus regenerative medicine. J Cell Physiol. 2007;213:341–347.
83. Mueller MB, Fischer M, Zellner J, et al. Hypertrophy in
mesenchymal stem cell chondrogenesis: effect of TGF-beta
isoforms and chondrogenic conditioning. Cells Tissues Organs.
2010;192:158–166.
84. Lettry V, Hosoya K, Takagi S, et al. Coculture of equine
mesenchymal stem cells and mature equine articular chondrocytes
results in improved chondrogenic differentiation of the stem cells.
Jpn J Vet Res. 2010;58:5–15.
85. Bosnakovski D, Mizuno M, Kim G, et al. Chondrogenic
differentiation of bovine bone marrow mesenchymal stem cells
(MSCs) in different hydrogels: influence of collagen type II

284.e3
References
extracellular matrix on MSC chondrogenesis. Biotechnol Bioeng.
2006;93:1152–1163.
86. Spadaccio C, Rainer A, Trombetta M, et al. Poly-L-lactic acid/
hydroxyapatite electrospun nanocomposites induce chondrogenic
differentiation of human MSC. Ann Biomed Eng. 2009;37:1376–1389.
87. Liu K, Zhou GD, Liu W, et al. The dependence of in vivo stable
ectopic chondrogenesis by human mesenchymal stem cells on
chondrogenic differentiation in vitro. Biomaterials. 2008;29:
2183–2192.
88. Yan D, Zhou G, Zhou X, et al. The impact of low levels of collagen
IX and pyridinoline on the mechanical properties of in vitro
engineered cartilage. Biomaterials. 2009;30:814–821.
89. Ng KW, Mauck RL, Wang CC, et al. Duty cycle of deformational
loading influences the growth of engineered articular cartilage.
Cell Mol Bioeng. 2009;2:386–394.

18 
Repair and grafting of bone
Iris A. Seitz, Chad M. Teven, and Russell R. Reid
SYNOPSIS
■ This chapter will concentrate on the principles and concepts of bone
repair, grafting, and reconstruction.
■ The embryology, physiology, microanatomy, and histochemistry of
bone will be reviewed.
■ Principles of mechanotransduction and cellular mechanisms of bone
turnover will also be discussed.
■ Pathophysiology of traumatic injury to bone (fractures, segmental loss,
defects) will be categorized and reviewed.
■ Bone-remodeling mechanisms (osteoconduction, osteoinduction,
osseointegration) will be described.
■ After a brief history of autogenous bone grafting, the clinical
application of bone transfer and transplantation will be captured by a
brief atlas of harvest of each subtype.
■ The reader will also be provided with a brief overview of bone
substitutes.
Microanatomy and histochemistry
Bone is a complex structure that serves many roles within
the body. It is crucial for maintaining mineral homeostasis,
provides protection to internal organs, and offers struc­
tural support for locomotion and stature. Osseous tissue
undergoes continuous remodeling, during which old bone
is degraded and replaced by new bone. Most organs in the
adult human undergo repair with scar tissue after injury.
Bone, however, regenerates, making it a unique organ
system.
The formation of bone occurs by either intramembranous
ossification or endochondral ossification. Direct condensa­
tions of mesenchymal stem cells (MSCs) initiate the process
of intramembranous ossification to create flat bones of
the craniofacial skeleton. During intramembranous ossifica­
tion, osteoblasts lining skeletal surfaces deposit new bone
upon previously laid bone in a process called apposition.
Alternatively, endochondral ossification is characterized by
the formation of an MSC-derived hyaline cartilage template
that is subsequently replaced by bone. Endochondral ossi­
fication is responsible for the development of tubular long
bones of the appendicular skeleton as well as the bones of the
vertebral column and pelvis.1–4
Bony tissue contains osteocytes located within open cavities
called lacunae. Osteocytes project radiating processes through
microscopic canaliculi that adjoin neighboring lacunae.
Osteocytic cytoplasmic processes are linked to one another
via gap junctions,5 through which osteocytes communicate.6
Cortical versus cancellous bone
The skeleton is comprised of cortical and cancellous bone (Fig.
18.1). Cortical bone is compact and forms the outer skeleton.
It accounts for 80% of osseous tissue. Cortical bone is dense
and strong to support the body and protect organs. A bilay­
ered cellular membrane (periosteum) covers cortical bone. The
outer layer is composed of a dense fibrous membrane contain­
ing flat fibroblast-like cells and serves as an attachment for
muscles and tendons. Large plump cells line the inner perios­
teal layer (cambium layer) and can differentiate into osteoblasts
upon stimulation. Bone also contains a thin layer of vascular
connective tissue (endosteum) that lines the medullary cavity.
The primary functional unit of cortical bone is the osteon
or haversian system (Fig. 18.2). An osteon contains concentric
layers of cortical tissue (lamellae) that surround a central
haversian canal that contains vascular and nervous tissue.
Nutrient vessels within haversian canals anastomose to blood
vessels in bone marrow and periosteum via the perpendicular
running Volkmann’s canals.
Cancellous or trabecular bone comprises approximately
20% of the skeleton. Relative to cortical bone, cancellous bone
is porous and is located deep within the medullary cavity,
typically at the ends of long bones. It consists of bony trabecu­
lae oriented along the lines of stress. Cancellous and cortical
bone are chemically equivalent; however, cancellous bone has
increased metabolic activity due to its greater surface area for

286
CHAPTER 18 • Repair and grafting of bone
transfer to the Golgi apparatus, where it is packaged and
secreted. Extracellularly, procollagen peptides are cleaved
into tropocollagen. The copper-dependent enzyme lysyl
oxidase mediates covalent cross-linking within and between
tropocollagen molecules to create mature collagen fibers. The
strength of bone is derived from interactions between collagen
fibers and inorganic minerals.
The cellular composition of bone
Three cell types are predominantly found in bone: osteoblasts,
osteocytes, and osteoclasts (Table 18.1).
Osteoblasts
Histology and function
Osteoblasts are derived from MSCs that have the capacity to
differentiate into several connective tissue cell types.15 Within
the appropriate osteogenic environment, MSCs differentiate
into osteoprogenitor cells, thereby committing to an osteo­
blastic lineage. Osteoprogenitor cells are ubiquitous in bone,
located within Volkmann’s and haversian canals, the cambium
layer of periosteum, endosteum, and within perivascular
tissue adjacent to bone.16
remodeling. Cancellous bone provides internal support for
bone marrow elements and cortical bone.
The chemical composition of bone
Inorganic phase
Bone is a calcified tissue composed of 60% inorganic matter,
30% organic matter, and 10% water. The inorganic component
of bone makes up approximately 40% of bony volume, while
the organic component and water comprise 35% and 25%,
respectively.7–10 The inorganic phase primarily consists of
the mineral crystal hydroxyapatite (Ca10(PO4)6(OH)2), which
measures 20–25 nm in length, 15 nm in width, and 2–5 nm in
thickness.7 It is the hydroxyl endmember of the apatite group;
however, impurities arise when the hydroxyl group is replaced
by potassium, magnesium, or sodium; or when carbonate
replaces the phosphate group.7,10–14 By storing them as a
mixture within impure hydroxyapatite, bone provides a res­
ervoir of various minerals within the body.
Organic phase
The organic component of bone (demineralized organic bone
matrix or osteoid) is deposited by osteoblasts during bone
formation. Osteoid is primarily composed of type I collagen,
but also includes many noncollagenous proteins.9 Collagen is
a trilaminar protein consisting of two α1 chains and one α2
chain that form a distinct right-handed helical structure.
Within the rough endoplasmic reticulum of osteoblasts, pro­
collagen undergoes hydroxylation and glycosylation before
Fig. 18.1  Diagrammatic representation of axial compact cortical bone
microarchitecture.
Compact, cortical bone
Spongy, cancellous bone
Haversian canal
Volkmann’s canal
Trabeculae
Fig. 18.2  (A) Cross-section of compact bone as seen by microradiography. Note
that the osteons are oriented in the longitudinal axis of long bone. The dark masses
represent recent decalcified osteons. The light masses represent older calcified
osteons. (B) A single osteon of the haversian system.
Uncalcified
osteon
Calcified
osteon
Bone marrow
Osteoblast
Canaliculi
Haversian canal
with its vessels
Lacunae
A
B

287
Microanatomy and histochemistry
signaling of osteogenesis. Although the exact mechanisms
of osteogenesis are complex and only partially elucidated,
advances have been made regarding the initiation and
molecular control of this process. Wnt/β-catenin signaling
has been shown to control the differentiation of both osteo­
blasts and osteoclasts, which may be important in postnatal
bone acquisition.41 Wnt/β-catenin signaling is also important
during skeletogenesis in the fetus, and is considered to be
partially responsible for both osteoblast and chondrocyte
differentiation.42,43 Notch signaling is a highly conserved
signaling system involving cell–cell communication. In addi­
tion to cell fate division and homeostatic maintenance, the
notch pathway is believed to be important in osteogenesis
because of notch1–BMP-2 interactions that promote osteo­
genic differentiation.44 Additionally, Engin et al.45 showed
with osteoblast-specific gain of Notch function studies that
Notch not only stimulates terminal osteoblast differentiation,
but that it also plays a role in the proliferation of immature
osteoblasts. In addition, loss of notch signaling is associated
with the expression of an osteoporotic phenotype.45
Hedgehog signaling plays a role in many embryonic pro­
cesses and also is involved with the maintenance of stem cells
in adults. There are three mammalian Hedgehog orthologs:
Desert, Sonic, and Indian. Sonic Hedgehog and Indian Hedge­
hog both have important roles in osteogenesis.33,34,46-49 Specifi­
cally, Indian Hedgehog is critical for endochondral bone
formation,50 whereas Sonic Hedgehog appears to be important
for skeletal patterning.51
Transcriptional regulation
Many transcription factors play a significant role in osteogen­
esis. Runt-related transcription factor 2/core-binding factor
alpha 1 (RUNX2/CBFα-1) and osterix have important effects
on osteoblast differentiation.52–57 RUNX2 is the mammalian
homolog of the Drosophila transcription factor Runt and is
Osteoblasts are plump, basophilic cells located on the
external surfaces of bone at areas of active bone formation
during development and fracture repair. They have a vast
endoplasmic reticulum for robust collagen production and
contain many mitochondria and a large Golgi apparatus to
package and secrete procollagen. Osteoblasts are responsible
for the production of bony matrix and facilitate the mineral­
ization of osteoid.16–18 Some osteoblasts will eventually become
surrounded with bony matrix and terminally differentiate
into osteocytes.
During osteoprogenitor cell differentiation, cells halt pro­
liferation and start expressing proteins characteristic of an
osteoblastic lineage. Preosteoblasts secrete proteins (e.g., alka­
line phosphatase (ALP)) that denote early osteogenesis,4,19,20
and also begin to produce bony matrix proteins. As the dif­
ferentiation process ensues, cells display a basophilic appear­
ance, stain strongly for ALP, and lose the ability to proliferate.
Mature osteoblasts deposit significant osteoid. They also play
a role in many other processes as indicated by their secretion
of numerous cytokines and noncollagenous proteins that have
pleiotropic effects, as well as factors important for myelo­
poiesis.16,21 Differentiated osteoblasts affect bone metabolism
and mineral homeostasis by interacting via surface receptors
with parathyroid hormone and 1,25-dihydroxyvitamin D.
Most osteoblasts ultimately undergo apoptosis or become
bone-lining cells. However, 20% differentiate into osteocytes.22
Regulation of osteoblast differentiation
Major signaling pathways
Under the control of multiple signaling pathways, pre­
osteoblasts differentiate into mature osteoblasts. The Wnt
pathway,23–26
the
TGF-β/BMP
superfamily,19,27–29
notch
signaling,30–32 hedgehog proteins,33–36 and fibroblast growth
factors37–40 (FGFs) have all been implicated in the molecular
Table 18.1  Bone cells
Cell type
Morphology
Location
Function
Source
Precursor
cell
Differentiation
product
Osteoblast
Rounded,
basophilic cells
Stain strongly for
alkaline
phosphatase
External surfaces
of bone in areas
of active
remodeling and
bone formation
Bone surfaces in
fractures
Produce bone matrix
Periosteum
Endosteum
Bone marrow
? Others
Preosteoblast
? Bone-lining
cells
Osteocyte
Osteocyte
Stellate cells with
thin cytoplasmic
processes
Embedded in
lacunae
Exact function
unknown
Potential functions:
Mechanosensory
Mineral homeostasis
Bone resorption
Osteoblasts
that
become
embedded
in osteoid
Osteoblasts
Terminally
differentiated
cell
Osteoclast
Large,
multinuclear
cells with ruffled
border
Stain positive for
tartrate-resistant
acid
phosphatase
Endosteal and
periosteal
surfaces of bone
in areas of active
remodeling
On fractured bone
surfaces
Bone resorption
Bone marrow
Spleen
? Lung,
peritoneum,
peripheral
blood
Hematopoietic
stem cell
None

288
CHAPTER 18 • Repair and grafting of bone
communication between mechanical stimuli and effector cells
(osteoblasts and osteoclasts). Recent evidence has shown that
osteocyte processes also attach to ECM proteins and may
mediate mechanotransduction by the amplification of shear
stress experienced at the ECM.80 In addition, mechanical
loading alters osteocyte gene expression. Osteocytes rapidly
produce increased levels of c-fos, IGF-1, prostaglandin (PG),
and nitrous oxide (NO) when mechanically stimulated.81–83
These and many other molecules (see below) have numerous
effects on bone turnover. Molecular studies have implicated
the activation of the Wnt/β-catenin pathway in response to
loading.84,85 The osteocytic Wnt/β-catenin pathway appears
to be triggered by crosstalk with various signaling pathways
when the osteocyte senses a load strain, thereby decreas­
ing negative regulators of the Wnt/β-catenin pathway,82 an
important regulator of bone mass. Finally, Tatsumi et al.86 have
shown that targeted ablation of osteocytes induces osteoporo­
sis with defective mechanotransduction.
One potential function of the osteocyte, termed periosteo­
cytic osteolysis, has remained elusive. Osteocytes may fulfill
an osteoclast-like activity by resorbing varying amounts of
calcified bone matrix surrounding their lacunae. Evidence for
this comes from studies that have shown a larger than expected
size of lacunae in conditions characterized by increased bone
resorption.87–89 Additionally, the administration of parathyroid
and thyroid hormone has been shown to induce microradio­
graphic enlargement of lacunae, thought to be the result of
the hormones’ actions on osteocytes.90,91 Recently, Wysolmer­
ski92 suggested that osteocytes have a role in coordinating
bone and mineral metabolism during reproductive cycles.
Specifically, osteocytes reversibly remodel perilacunar and
pericanalicular bone during lactation by removing mineral­
ized tissue in order to liberate calcium (and potentially
phosphorous) for milk production. This tissue is then replaced
during the recovery period after weaning.
Osteoclasts
Histology and function
Osteoclasts are multinucleated cells primarily responsible
for bone resorption, a process necessary for bone growth,
tooth eruption, fracture repair, and calcium homeostasis.
Osteoclasts typically have between three and 25 nuclei per
cell and are approximately 40 µm in diameter. Histologically,
osteoclast cytoplasm has a characteristic homogeneous,
“foamy” appearance due to a high concentration of vesicles
and vacuoles. Osteoclasts are derived from hematopoietic
progenitor cells and are related to the monocyte/macrophage
lineage. Osteoclasts are found on the endosteal and periosteal
surfaces of bone in areas of active remodeling.
Bone resorption is a complex process by which crystal­
line hydroxyapatite is dissolved and organic bone matrix is
degraded. In vitro bone resorption models using primary osteo­
clast isolates provided the basis for the resorption cycle.93,94 The
cycle is a complex series of actions which includes osteoclast
migration to the resorption site and attachment to bone, mem­
brane polarization, dissolution of hydroxyapatite followed by
degradation of organic bone matrix, removal of degradation
products, and osteoclast inactivation or apoptosis.95 Clini­
cally, bone resorption is an extremely important process, as
evidenced by pathologic conditions characterized by dysfunc­
tional resorption. For example, increased osteoclastic function
believed to be evolutionarily conserved in humans to serve a
critical role throughout the many steps of osteoblastogenesis,
including osteoblast induction, proliferation, and maturation.
Homozygous deletions of the Runx2 gene are uniformly lethal
in mice due to a complete lack of mineralized bone matrix.53
Runx2 haploinsufficiency causes cleidocranial dysplasia, an
autosomal-dominant disease in humans characterized by
hypoplastic clavicles, dental deformities, shortened stature,
brachycephaly, hypertelorism, and other skeletal defects.58
While the exact mechanisms that control the regulation of
Runx2 have not been fully decoded, studies have shown that
various histone acetyltransferases are important Runx2 cofac­
tors.59,60 Additionally, microRNAs appear to regulate Runx2
protein expression.61 MicroRNAs play critical roles in stem
cell function,62,63 and thus may have important clinical impli­
cations for diseases due to dysfunctional MSC/Runx2 interac­
tions. Many other regulators also play an important role in
Runx2 function64–70; however, their exact roles are currently
under study.
The zinc finger-containing protein osterix (Osx) is also a
key transcription factor to bone formation.55,71 Osx-null mice
form neither cortical nor cancellous bone. Unlike Runx2-null
mice, however, Osx-null mice still produce normal cartilage.72
As a result of the findings that Osx expression is absent in
Runx2 knockout mice, whereas Runx2 expression is normal
in Osx-null mice, the Osx gene is thought to be downstream
to Runx2.55 Nishio et al.73 confirmed this using overexpression
experiments to show that Runx2 transactivates the Osx gene
promoter significantly, indicating that there is a Runx2-binding
element within the promoter region.
The
signaling
pathways
and
transcription
factors
responsible for osteoblastogenesis are exquisitely detailed
elsewhere.74,75
Osteocytes
Histology
Osteocytes are terminally differentiated cells of the osteoblast
lineage and comprise 90–95% of all bone cells. Despite being
the most abundant cell in bone, osteocytes are the least well
characterized. Osteocytes are located within lacunae. Multiple
cytoplasmic processes radiate from osteocyte somas and
travel through canaliculi between lacunae. Gap junction con­
nections between adjacent cytoplasmic processes enable
osteocytes to communicate with one another. Osteocytes com­
municate with osteoblasts by way of this canalicular network
as well. Processes also extend to the haversian canals so that
osteocytes can rid themselves of waste products and receive
nutrients necessary for survival. Relative to osteoblasts, osteo­
cytes have qualitatively similar organelles, but the organelles
are smaller in both size and number.76
Osteocyte function
Bone is a dynamic substance that maintains its viability by
constantly adapting to the environment in which it lives.
Victim to mechanical loading and various traumas, bone must
have a way to recognize such stressors. It is believed that
osteocytes are mechanosensory cells that translate (via mecha­
notransduction) physical stress into chemical and/or electrical
signals that stimulate bone remodeling.77–79 This hypothesis
has been derived from various lines of evidence. Anatomi­
cally, osteocytes form a syncytium with surrounding cells
via their cellular processes. These connections may provide

289
Microanatomy and histochemistry
phenotype122, while OPG knockout mice display an osteopo­
rotic phenotype.123,124
The RANK molecule lacks enzymatic activity, and thus
must recruit adaptor proteins when stimulated by RANKL to
promote differentiation.125–127 TNF receptor-associated factor
(TRAF) family members appear to be important adaptor
proteins, with TRAF6 serving the most critical role.128–130
TRAF6 knockout mice develop severe osteopetrosis as a result
of either the dysfunction or total absence of osteoclasts.131,132
Moreover, TRAF6 activates NF-κB, another essential modula­
tor of osteoclast differentiation.133,134 NF-κB enters the nucleus
of osteoclasts stimulated with RANKL, and regulates tran­
scription of target genes critical for osteoclastogenesis.135
Extracellular matrix
The extracellular matrix (ECM) is the scaffolding from which
bone derives much of its strength and architecture. Most
molecules found within the ECM are produced by osteoblasts.
The ECM contains nearly 30 types of collagen, as well as
noncollagenous phospho- and glycoproteins. Type I collagen
is the predominant form found within the ECM, making up
approximately 90% of bone matrix. The significance of type I
collagen is highlighted by patients with osteogenesis imper­
fecta (OI), a disease characterized by bone fragility and
repeated fracture. OI often occurs in patients with pro-α1 or
pro-α2-chain gene mutations, or from a mutation in one of the
enzymes responsible for posttranslational hydroxylation of
collagen.136 In addition, bone mineralization takes place within
the ECM. Bone mineralization is a well-orchestrated process
during which hydroxyapatite crystals are laid down within
the organic matrix of bone and calcify proteins that are
secreted by osteoblasts.
In addition to type I collagen, many noncollagenous pro­
teins are found within the ECM (Table 18.2). Osteopontin is a
ubiquitous phosphoprotein that exists in multiple forms. It
has been localized to most cells within bone, including osteo­
blasts, osteocytes, osteoclasts, and bone precursor cells.137–139
Osteopontin plays a role in the nucleation of hydroxyapatite
crystals during matrix mineralization and also in osteoclast
adhesion during bone remodeling.140–145
Osteonectin is a glycoprotein that binds to calcium and
type I collagen and also initiates and regulates bone mineral­
ization.146,147 BSP is another important noncollagenous protein
that, in contrast to osteopontin and osteonectin (both of which
are expressed by many tissues throughout the body), is mainly
expressed by cells within the skeleton.148 BSP is an acidic
phosphoprotein that binds to collagen and also promotes
hydroxyapatite nucleation.149 BSP-null mice express a pheno­
type characterized by small and undermineralized bones.150
Osteocalcin (bone Gla protein) is a vitamin K-dependent
γ-carboxyglutamic acid-containing protein, similar to factors
II, VII, IX, and X of the coagulation cascade. The most abun­
dant noncollagenous protein in bone, osteocalcin is secreted
by osteoblasts and odontoblasts and is an important marker
of increased bone turnover.151 Osteocalcin plays a role in
bone turnover,152,153 osteoclast differentiation,154 and energy
metabolism.155,156
Osteoblasts also produce proteoglycans, the most abundant
of which is biglycan (BGN). BGN, a leucine-rich noncollage­
nous protein, is believed to play a role in the mineralization of
bone given that bgn-deficient mice express an osteoporotic
phenotype.157 However, BGN seems to play an important role
leads to excess bone resorption, thereby causing osteoporosis.
In contrast, osteoclast underactivity decreases bone resorption,
which causes osteopetrosis.
Bone resorption is key for both bone remodeling at sites
of injury and also as a mechanism of calcium homeostasis.
Osteoclasts are stimulated to resorb bone by growth factors
and cytokines secreted by local osteoblasts.96 Once osteo­
clasts localize to the area in need of resorption they form
a tight seal that separates the area of bone to be resorbed
from the extracellular environment. Evidence has pointed to
integrins (in particular integrin αvβ3) and cadherins as playing
a prominent role in the attachment of this sealing zone.97–99
Myosin and actin-binding proteins are also important for
such attachments.100,101 After sealing zone attachments are
completed, osteocytes polarize such that the area adjacent
to the resorbing bone surface becomes ruffled. The ruffled
border is a specialized membrane domain formed by the
fusion of intracellular acidic vesicles and functions as the
osteocyte’s resorbing organelle.102,103 Finger-like projections
from the ruffled border increase the effective surface area
of resorption. Before proteolytic cleavage of the organic
matrix occurs, osteoclasts dissolve hydroxyapatite crystals by
targeted secretion of hydrochloric acid through the ruffled
border toward the resorption pit (Howship’s lacuna).95 After
the dissolution of inorganic matter, many proteolytic enzymes,
including matrix metalloproteinase-9 and cathepsin K, are
secreted into the resorption lacuna to degrade the organic
components of bones.104,105 Both phases of bone are subse­
quently removed from the resorption lacuna via transcytosis
through the ruffled border and sent to the secretory domain of
the osteoclast, which then expels these degradation products
into the extracellular space outside the osteoclast.106–108 The
osteocytic marker tartrate-resistant acid protease (TRAP) has
been localized to transcytotic vesicles of resorbing osteo­
clasts, and is believed to generate reactive oxygen species
that degrade matrix degradation products within these
vesicles.109
Osteoclast differentiation
The mechanisms surrounding osteoclast differentiation are
complex and have been the subject of intense study. Depend­
ing on local stimuli, hematopoietic stem cells may differentiate
into osteoclasts, macrophages, or dendritic cells. Macrophage
colony-stimulating factor (M-CSF) and receptor activator for
nuclear factor κB (RANK) and RANK ligand (RANKL) are
early mediators that signal osteoclastogenesis.110–115 In murine
studies, M-CSF is functionally lacking in op/op mice, which
express an osteopetrotic phenotype.116,117 In addition, in order
for osteoclastogenesis to commence, RANKL (found on the
cellular membrane of osteoblasts) must interact with RANK,
a receptor found on osteoclast precursors.118–120 RANK–RANKL
interaction must occur in order for osteoclast precursor cells
to begin expressing phenotypic markers (e.g., TRAP) that
characterize osteoclasts. Similar to op/op mice with function­
ally absent M-CSF, RANK and RANKL knockout mice also
express an osteopetrotic phenotype.113,115,121
Osteoprotegerin (OPG), a TNF-related soluble protein
secreted by osteoblasts, is also an important regulator of
osteoclastogenesis. OPG has effects on bone density and
bone mass118,122 by acting as a decoy receptor for RANK,
thereby blocking RANK–RANKL interaction. Transgenic
mice that overexpress OPG suffer from an osteopetrotic

290
CHAPTER 18 • Repair and grafting of bone
other tissues within the body. As described previously,
bone is subject to constant remodeling, a process medi­
ated by osteoclasts and osteoblasts. Physiologically, bone
homeostasis is maintained by the counterbalancing of bone
resorption with formation. When one of these processes
becomes dysfunctional, patients may manifest a distinct
pathology. Excessive bone resorption resulting in an exces­
sive loss of bone is the fundamental pathogenesis underlying
osteoporosis.161 Conversely, decreased bone resorption due
to decreased osteoclast activity leads to an osteopetrotic
phenotype.162 Decreased bone resorption due to increased
osteoblast activity, however, will produce an osteosclerotic
phenotype.
The structure of bone is a function of its material composi­
tion as well as the genetic blueprint of an individual. In
addition, various loads to which bone is subjected define its
outside the skeleton as well, as bgn-knockout mice manifest
spontaneous aortic rupture and dissection,158 dental and mus­
cular abnormalities, thinning of skin, and osteoarthritis.159
A variety of enzymes produced by bone cells also function
within the ECM. The most prominent is ALP. Like osteocalcin,
ALP is often used as a marker of bone turnover.160 While the
exact function of ALP is unknown, it is believed to be impor­
tant for bone mineralization, during which it regulates apatite
formation.
Principles of bone homeostasis
and turnover
Unique to bone is its ability to undergo regeneration rather
than repair with scar formation, as is typified by most
Table 18.2  Extracellular matrix molecules in bone
Molecule
Protein structure
Cell source
Present in
nonosseous
tissues
Function
Phenotype of
knockout
animals or
deficiency in
humans
Collagen type I
Trilaminar protein
consisting of 2 α1
chains and 1 α2
chain
Many
Mature and immature
osteoblasts in
bone
Yes
Primary scaffold of bone
90% of matrix
Osteogenesis
imperfecta in
humans
Alkaline
phosphatase
Metalloenzyme
Mature and immature
osteoblasts
Different isoforms
expressed by
cardiac and
hepatic cells
Yes
Enzyme
Regulator of extracellular
matrix mineralization
Regulator of cellular
migration, adhesion,
and differentiation
Hypophosphatemia
and defects in
skeletal
mineralization in
knockout mice
Osteopontin
Phosphorylated
glycoprotein
Osteoblasts/
osteoclasts
Tumor cells (e.g.,
breast)
Yes
Osteoclast activation
and extracellular
matrix mineralization
Anchor osteoclasts to
mineralized matrix
Alterations in
extracellular
matrix
remodeling in
knockout mice
Osteonectin
Glycoprotein
Osteoblasts
Endothelial cells,
megakaryocytes
Yes
Binds calcium and
collagen type I
Regulator of
mineralization
Unknown
Bone
sialoprotein
Phosphorylated
glycoprotein
Almost exclusively
produced by
skeletal cells
(osteoblasts,
osteocytes,
hypertrophic
chondrocytes)
No
Exact function unknown
Regulator of cellular
adhesion
Unknown
Osteocalcin
Vitamin K-dependent
γ-carboxyglutamic
acid-containing
protein
Osteoblasts
Odontoblasts
Hypertrophic
chondrocytes
No
Regulator of bone
turnover
Regulator of osteoclast
migration
Binds hydroxyapatite
Osteopetrosis in
knockout mice
Biglycan
Proteoglycan
Mature and immature
osteoblasts
Nonosseous cells
Yes
Exact function unknown
Regulates function
apatite formation
Osteoporosis and
small, thin, short
limbs in
knockout mice

291
Principles of bone homeostasis and turnover
structure. This process of functional adaptation mediates
repair of damaged bone and also facilitates prevention of
damage before it may occur. Bone exhibiting suppressed
remodeling and resorption suffers increased microdamage
and fracture accumulation.163,164 It has been demonstrated that
sites of microcracks are subsequently associated with new
bone remodeling more often than expected by chance,165–167
which has led to the belief that bone remodeling and repair
are often targeted to specific locations.168
Wolff’s law and mechanotransduction
Wolff (1892) derived a mathematical formula to describe how
changes in the form and function of bone could lead to a
change in its internal architecture and external structure.169
Wolff believed that during the functional adaptation of bone
to new loads that occur during one’s life (e.g., as a result of
trauma), the trabeculae within bone would reorient to align
with the new principal stress trajectories of these environ­
mental loads. Wolff’s contributions were key to the early
understanding of bone adaptation. More recently, authors
have recognized that bones are subject to dynamic applied
loads.170
Considerable research has examined the initial events that
stimulate bone adaptation. It is believed that the stimulus for
bone remodeling is strain (the physical deformation of bone
tissue).171 Osteocytes, the mediators of mechanotransduc­
tion,172 respond molecularly to mechanical loads. They undergo
direct deformation by bending and also are deformed by the
electrical potential induced by the flow of surrounding inter­
stitial fluid.173,174 In addition, osteocytes are subjected to shear
stress by the physical flow of interstitial fluid around their
cellular membranes.175–179
When strained, osteocytes convert the mechanical
stimulus into chemical or electrical activity in a process
thought to require various signaling molecules including NO
and PGs.172 Interestingly, NO synthase, a calcium/calmodulin-
dependent enzyme, is activated under conditions of increased
intracellular calcium; intracellular calcium concentrations
appear to increase in response to fluid flow.180,181 Klein-Nulend
and colleagues182 demonstrated that a pulsating fluid flow
model led to the release of NO from osteocytes. Taken together,
these studies indicate that NO may be important in osteocyte
mechanotransduction.183 Additional studies have shown that
pulsating fluid flow treatment also stimulates osteocytes
to synthesize increased levels of PGE2 and PGI2,184,185 as well
as increased levels of cyclooxygenase-2 (COX-2) mRNA.186
COX-2 production is also stimulated by mechanical
loading.187,188 Thus, COX-2 induction and the production
of PGs are thought to be important mediators in the conver­
sion of mechanical shear forces into signals that trigger bone
turnover.
Bone regeneration: the role of the stem cell
In the last decade, the field of regenerative medicine has
exploded with the discovery of new techniques to isolate cells
capable of differentiating into many different tissues. In par­
ticular, the MSC (Fig. 18.3) has considerable therapeutic
potential for fracture repair and other bone pathologies. For
concordance, Dominici and colleagues189 have proposed that
MSCs meet the following criteria: (1) isolated MSCs must
Fig. 18.3  Schematic representation of mesenchymal stem cell (MSC)
differentiation pathways. MSCs are pluripotent stem cells that are capable of
differentiating into several lineages with distinct growth and differentiation cues.
These lineages include osteogenic, chondrogenic, adipogenic, and myogenic
lineages. Lineage-specific differentiation is a well-coordinated process that is
regulated by master regulators, such as MyoD for myogenesis, peroxisome
proliferator-accepted receptor (PPAR)-γ for adipogenesis, SOX9 for chondrogenesis,
and RUNX2 and Osterix for osteogenesis. Osteogenic differentiation can be staged
by measuring alkaline phosphatase (early marker) and osteopontin and osteonectin
(late markers). (From Tang N, Song W-X, Luo J, et al. Osteosarcoma development
and stem cell differentiation. Clin Orthop Relat Res. 2008;466:2114–2130.)
Pluripotent
mesenchymal
stem cell
Committed
osteoprogenitor
Proteoblast
Early
osteoblast
Mature
osteoblast
MyoD
Myocytes
Adipocytes
Chondrocytes
Osteocyte
PPARγ
SOX9
Growth and
differentiation cues
Proliferation
Differentiation
Alkaline phosphate
Osterix
Osteocalcin
RUNX2
Osteopontin
adhere to plastic in culture; (2) MSCs should express the
CD105, CD73, and CD90 surface antigens and not express the
CD34, CD45, CD14 or CD11B, CD79A or CD19 and human
leukocyte antigen-DR (HLA-DR) surface antigens; and (3)
MSCs must have the capacity to differentiate into osteoblasts,
adipocytes, and chondroblasts. MSCs and MSC-like cells are
derived from bone marrow, but can also be expanded from
skeletal muscle,190 adipose tissue,191 dental pulp,192 the circula­
tory system,193 synovium,194 amniotic fluid,195 urine,196 the
umbilical cord,197 and fetal tissues.198,199
The capability of MSCs to differentiate into and assist
with the restoration and repair of bone has been extensively
studied.200–203 MSCs are also used clinically in patients with
various bone disorders.204 In a clinical trial204 that examined
MSC administration to children with OI, bone mineral
density, growth velocity, and ambulation all improved after
MSC administration. However, the concentration of MSCs
detected in bone, skin, and other tissues was less than
1%. These data and evidence from additional studies205,206
have led authors to postulate that MSCs may actually
secrete soluble factors that alter tissue microenvironment,
which may be an important mechanism by which MSCs
repair tissue.207,208

292
CHAPTER 18 • Repair and grafting of bone
signal peptide, a prodomain, and a mature peptide.214 BMPs
are thought to be secreted from cells in their active form.
BMPs are natural ligands for type I and type II serine/
threonine kinase transmembrane cellular surface receptors.
There are three type I and three type II receptors that are
capable of binding BMPs.212 Upon ligand (BMP) binding, a
heterotetrameric complex is formed, after which type II recep­
tor kinase domains phosphorylate Gly-Ser domains in the
type I receptor kinases. This process activates type I receptors
to recruit and phosphorylate intracytoplasmic SMAD proteins
(SMADs 1, 5, and 8). Following SMAD phosphorylation,
SMAD 1, 5, or 8 will bind to SMAD 4, and this SMAD4-
phosphorylated SMAD 1, 5, or 8 complex will relocate to the
cell nucleus. Transcription factor activation occurs, leading to
the transcription of bone morphogenetic target genes.215,216
Bone morphogenetic protein function
BMP exerts pleiotropic effects throughout the lifespan of
an organism. In the early stages of human development,
BMP plays a key role in skeletogenesis.211 Also during this
time period, BMP signals epidermal induction,217 directs the
development of neural crest cells,218 and induces a sympa­
thetic adrenergic phenotype.219 Bmp-2-null mice typically die
between 7 and 10 days of gestation due to cardiac defects,
even before the formation of bone begins.
Molecular mechanisms of bone regeneration
Bone regeneration consists of a complex interplay of
molecular processes that promote MSC migration, pro­
liferation, and differentiation. Recently, there has been a
great deal of research aimed at the identification of these
molecular processes, and advances in our understanding of
the molecular mechanisms of bone regeneration have been
made. Many have been able to identify important signal­
ing molecules as well as transcriptional regulators of bone
regeneration (Table 18.3).
Bone morphogenetic protein
In the 1960s, Urist209 discovered the function of BMP from his
work with demineralized bone matrix (DBM). To date,
approximately 20 BMP isoforms have been characterized.
BMPs are pleiotropic members of the TGF-β superfamily.
BMPs, while important for brain and bone formation in
utero,210–212 have generated much of their recognition from
their osteoinductive properties. BMPs have also been impli­
cated in human disease.213
BMPs are synthesized as large precursor molecules that
consist of characteristic subsections. A series of seven cysteine
residues is located at the carboxyl termini of BMPs, and is
important for proper protein folding. BMPs also contain a
Table 18.3  Growth factors and fracture repair
Molecule
Ectopic
bone
formation
Segmental
defect
healing
Effect on
fractures
Combination
with other
growth factors
Effective in
lower
vertebrates
Effective
in
primates
Potential
clinical use
BMPs
Yes
Endochondral
++++
Membranous
++++
Increase callus
volume
Increase
mechanical
strength
? Increase rate
of healing
Synergistic with
TGF-β
Yes
Yes
Spinal fusion
Alveolar bone
deficiency
Segmental
defects
Augment bone
graft healing
Dental implants
TGF-β
No
Endochondral
++
Membranous
++
Increases callus
size
Increases
mechanical
strength
Synergistic with
BMP
Increases
FGF-2
expression
Increases VEGF
expression
Yes
No
(limited
study)
Role alone
unclear
FGFs
No
Endochondral
+/−
Membranous
+/−
Increase callus
and bone
volume
Increase
mechanical
strength
Increase VEGF
expression
? Synergistic
with TGF-β
Yes
Yes
(limited
study)
Potential for
augmenting
angiogenesis
in
compromised
wounds
PDGF
No
Endochondral
−
Membranous
−
Increases callus
volume
Increases
mechanical
strength
Increases VEGF
expression
Synergistic with
TGF-β for
chemotaxis
Yes
No
Unknown
++++, very active; +/−, minimally active; −, no activity. BMPs, bone morphogenetic proteins; TGF-β, transforming growth factor beta; FGFs, fibroblast growth factors;
PDGF, platelet-derived growth factor; VEGF, vascular endothelial growth factor.

293
Principles of bone homeostasis and turnover
BMP is the only signaling molecule capable of singly induc­
ing de novo bone formation. It has been hypothesized that
BMP-2, -6, and -9, and to a lesser extent, -4, and -7, various
BMP isoforms, have the greatest osteogenic capacity.220 Luu
and colleagues19 infected HEK293, C2C12, and C3H10T1/2
cells with adenoviral vectors containing various BMP isoforms
(AdBMP). The authors measured the expression of early
(ALP) and late (osteocalcin) osteogenic markers in vitro. They
also assessed in vivo induction of heterotopic ossification by
implanting AdBMPs into nude mice and evaluating radio­
graphic and histologic results at 3 and 5 weeks. BMP-2, -4, -6,
and -7 were shown to induce increased ALP elevation com­
pared to other BMPs. In addition, BMP-2, -4-, -6, -7, and -9
stimulated increased osteocalcin expression compared to
others. The authors also found that BMP-2, -6, -7, and -9
induced the greatest degree of osteogenesis in vivo. The
authors concluded that BMP-2, -6, and -9 have the greatest
osteogenic potential, and that BMP-4, and -7 also have a sig­
nificant degree of osteogenic potential.
BMP-2 promotes osteoblast differentiation by stimulating
Runx2 in a process mediated by Smad proteins.221 In addition,
BMP-2 affects osteoblast differentiation via activation of the
β-catenin signaling pathway.222 Interestingly, endogenous
activation of β-catenin induces ALP mRNA expression;
however, BMP-2 must be present for β-catenin signaling to
induce osteocalcin expression as well.223 Thus, β-catenin-
dependent differentiation processes likely require BMP-2 to
promote the later stages of osteoblast differentiation. Recently,
it has been demonstrated that BMP-2-mediated osteoblast
differentiation is also dependent upon Akt2 selective signal­
ing.224 IGFs are important for osteoblast differentiation and
normal bone growth,225–228 and Akt2 is a key molecule within
the insulin signaling pathway. Mukherjee et al.224 showed that,
although BMP-2 signaling is normal, osteoblast differentia­
tion does not occur in Akt2-knockout mice. Delivery of Runx2
to Akt2-deficient mice restores osteoblast differentiation; thus,
Akt likely serves a specific role in osteoblast differentiation
through its regulation of Runx2 gene expression.
Furthermore, BMP-2 appears to improve healing of large
osseous defects. Many studies have evaluated recombinant
human BMP-2 (rhBMP-2) in in vivo animal models. In nonhu­
man primate and other mammalian models, rhBMP-2 has
been shown to augment bone growth successfully and
also close critical-sized osseous defects.229–231 The efficacy of
rhBMP-2 in humans has also been assessed.232,233 Slosar et al.234
demonstrated that anterior lumbar allografts filled with
rhBMP-2 experienced higher fusion rates than lumbar
allografts without rhBMP-2 in patients who underwent
lumbar fusion. The authors demonstrated that at 6 months
and at 2 years, allografts with rhBMP-2 shortened the time­
frame for expected improvement, and also improved patients’
Oswestry Disability Index scores. Thus, it has been concluded
allografts with rhBMP-2 can offer reliable and clinically sig­
nificant benefits.
Adenoviral delivery of BMP (AdBMP) is also effective in
stimulating osteogenic differentiation of mesenchymal pro­
genitor cells. Our group isolated and subsequently reversibly
immortalized primary murine calvarial cells that retained a
progenitor cell phenotype.235 Immortalized calvarial cells
(iCALs) treated with AdBMP-2 demonstrated an increase in
early (e.g., alkaline phosphatase activity, osteocalcin mRNA
transcription) and late (e.g., matrix mineralization) marker
expression of osteogenic differentiation compared to iCALs
treated with AdGFP (control). Additionally, in vivo stem cell
implantation assays revealed increased ectopic bone produc­
tion from iCALs exposed to BMP-2 compared to control. In a
similar study that evaluated the capacity of BMP-9 to induce
osteogenic differentiation of calvarial progenitor cells, Teven
and colleagues236 demonstrated that AdBMP-9-treated iCALs
display enhanced osteogenic differentiation compared to
AdGFP-treated cells (control) in both the in vitro and in vivo
settings. The translational utility of these experiments relates
to identifying effective agents for bony (e.g., calvarial) tissue
regeneration. However, the clinical use of adenovirally-
delivered BMPs has been limited and therefore further
investigation is warranted.
Presently, rhBMP-2 (INFUSE, Medtronic, Inc.) is one of two
rhBMPs approved by the Food and Drug Administration
(FDA) for human application. The other is rhBMP-7 (rhOP-1).
As described, rhBMP-2 is often used in spinal fusion, as well
as orthopedic trauma and dental procedures. rhBMP-7 has
shown clinical benefits in spinal fusion and tibia repair.237,238
Interestingly, although BMP-2 may be more osteoinductive
than BMP-7 based on in vitro analyses,29 there has been rela­
tively little in vivo data comparing the two. Barr and col­
leagues239 have found that rhBMP-7 promotes similar bone
quality but greater bone volume induction than rhBMP-2 in
murine muscle pouch assays analyzed by microscopic com­
puted tomography and histology. However, other studies
have found different results.240 Conflicting results may be due
to differences in experimental design.
Transforming growth factor-β
TGF-β has protean effects on cellular processes within the
human body. TGF-β has been implicated in cell cycle regula­
tion, angiogenesis, wound healing, and skeletogenesis. TGF-β
also appears to have significant roles in human disease.241
There are three distinct human TGF-β isoforms (TGF-β1,
-β2, and -β3). Though they exhibit differences, each isoform
shares similar functions with one another, including func­
tions pertaining to the regulation of osteogenesis and osseous
repair.
TGF-β, like BMP, contains a cluster of conserved cysteine
residues.242 Each isoform is produced as a precursor molecule
consisting of TGF-β and a propeptide region. In contrast to
BMPs, however, TGF-β is secreted as an inactive molecule
and stored in its latent form within the ECM. TGF-β becomes
active after the noncovalent disulfide bond linking TGF-β and
its propeptide region is broken.
There are four TGF-β receptors identified: types I, II,
and III, and endoglin. TGF-β-initiated signaling, like BMP-
mediated signaling, uses SMAD protein intermediates. Active
TGF-β binds to either type III or type II receptors, both of
which promote the phosphorylation of type I receptors.
Phosphorylation of SMAD2 or SMAD3 follows, which in turn
leads to the binding of SMAD4 to the phosphorylated SMAD2
or SMAD3 proteins. This complex translocates to the cellular
nucleus, where transcription factors activate specific target
gene transcription, thereby mediating the effects of TGF-β at
the cellular level.
Evidence that TGF-β may play a role in osteogenesis
has come from experiments demonstrating that TGF-β is
produced by osteoblasts243 and stimulates the production of

294
CHAPTER 18 • Repair and grafting of bone
craniosynostosis.38 Significant evidence exists demonstrating
that FGF–FGFR signaling is essential to the developing axial
skeleton. Additionally, the signaling cascade is necessary
for cranial bone intramembranous ossification as well as
for the maintenance of cranial suture homeostasis. A review
of the roles of FGF signaling in physiologic development
and in the development of skeletal diseases can be found
elsewhere.256
There are four known FGFRs (FGFR-1, -2, -3, -4). Fol­
lowing FGF ligand binding, receptor dimerization occurs,
which promotes the transphosphorylation of each receptor
monomer by an intrinsic tyrosine kinase domain. Phos­
phorylated tyrosine residues on FGFRs become docking
sites for adaptor proteins necessary for downstream signal­
ing.257 Human gene studies have identified associations
between skeletal disorders and FGFR mutations, including
Pfeiffer syndrome (FGFR1), Crouzon syndrome (FGFR2), and
achondroplasia (FGFR3).256
FGF has been implicated in a variety of osteogenic mecha­
nisms. Coffin et al.258 found that increased concentrations of
FGF-2 are found at epiphyseal growth plates of endochondral
bone, specifically within the proliferation and hypertrophic
zones. FGF-2 also accelerates fracture repair and closure of
critical-sized defects.259 Interestingly, there seems to be a
differential effect of FGF-2 treatment based on delivery. Con­
tinuous FGF-2 treatment stimulates osteoblast proliferation;
however, it decreases levels of differentiation markers (e.g.,
ALP) and augments osteoclast formation, thus resulting in
net bone resorption. In contrast, intermittent FGF-2 treatment
enhances bone formation.
The effects of FGF-2 in combination with other growth
factors have also been reported. Maegawa and colleagues260
examined MSCs deposited in osteogenic media supplemented
with BMP-2, FGF-2, or both. MSCs supplemented by both
FGF-2 and BMP-2 concurrently expressed the highest levels
of ALP, osteocalcin mRNA, and bone matrix formation. In
addition, Sabbieti et al.261 investigated whether FGF-2 modu­
lates the anabolic response of bone to PTH. Primary calvarial
osteoblasts were isolated from Fgf-2+/+ (wild type) and Fgf-
2−/− mice. By immunocytochemistry, Fgf-2-null osteoblasts
expressed decreased Runx2 protein synthesis as well as sig­
nificantly less Runx2 expression within perinuclear and
nuclear spaces than wild-type variants when exposed to PTH.
Given the importance of Runx2 on osteogenesis, it was con­
cluded that Fgf-2 expression is required for PTH to promote
an anabolic response by bone.
Platelet-derived growth factor
Platelet-derived growth factor (PDGF) plays an important
role in a number of biological processes, including embryo­
logical development, inflammatory reactions, angiogenesis,
organogenesis, and wound healing. Four unique isoforms of
PDGF have been characterized (PDGF-A, -B, -C, -D), which
are inactive in their monomeric forms, and thus must dimer­
ize to become active. Five dimeric PDGF compounds have
been identified (PDGF-AA, -BB, -CC, -DD, and -AB).262 All
PDGF molecules share a highly conserved PDGF/vascular
endothelial growth factor (VEGF) homology domain, a motif
also found in VEGF.263 PDGFs interact with two receptor
tyrosine kinases (PDGF receptor-α (PDGFR-α) and PDGFR-
β), which also must dimerize to activate.264 Receptor
collagen and osteopontin.244,245 In vivo studies have corrobo­
rated the osteoinductive effect of TGF-β.246 TGF-β1, in particu­
lar, has been the subject of a considerable body of literature
with respect to its osteogenic potential, in part because it is
the dominant TGF-β isoform expressed by bone cells.247 In
vitro analyses have demonstrated that TGF-β1, by recruiting
and stimulating the proliferation of osteoblast progenitors,
increases bone formation.248 Interestingly, TGF-β1 seems to
have a differential effect on osteoblast differentiation. Early
phases of differentiation appear to be promoted by TGF-β1,
while later phases are blocked.249
In addition, TGF-β1 modulates the processes of osteoclas­
togenesis and bone resorption. At high TGF-β1 concentra­
tions (0.1–10 ng/mL), osteoclastogenesis appears to be
downregulated.248 TGF-β1 binds to its receptor on osteoblasts
and augments the expression of OPG while also decreasing
RANKL expression. A lower RANKL : OPG ratio results in
maximal RANKL occupation by OPG, thus decreasing matu­
ration of osteoclasts. TGF-β1 appears to serve a bifunctional
role, however, as low concentrations are associated with
high levels of both RANKL and RANK, and subsequent
osteoclastogenesis.
TGF-β1 may also be important for the coupling of bone
resorption and formation, as is found at bone resorption sites.
Tang and colleagues250 injected green fluorescent protein
(GFP)-labeled osteogenic bone marrow stromal cells (BMSCs)
into the femur cavity of Tgfb1+/+ and Tgfb1−/− immunodeficient
mice. Using anti-GFP antibodies, injected BMSCs were
detected at both 1 and 4 weeks. After 1 week, BMSCs
were localized to the bone surfaces in Tgfb1+/+ mice; BMSCs
were widely dispersed in the bone marrow (i.e., not at bone
surfaces) of Tgfb1−/− mice. After 4 weeks, BMSCs were embed­
ded in trabecular bone in Tgfb1+/+mice but not in Tgfb1−/− mice.
The authors concluded that TGF-β1 plays an essential role in
the coupling of bone resorption to bone formation by their
induction of osteogenic BMSCs to sites of resorption.
In addition, TGF-β expression is elevated during frac­
ture repair251 and has been considered for exogenous use
to augment bone growth and repair. Studies support that
TGF-β isoforms significantly and rapidly induce the forma­
tion of endochondral bone at extracranial sites.252 Studies
on the effects of TGF-β on calvarial tissue are inconclusive.
Many authors have found little osteoinductive potential
associated with TGF-β.253 More recently, ectopic TGF-β was
shown to augment calvarial defect healing in rabbits and
promote normal suture reformation.254 Interestingly, combi­
nation therapy of BMP-2 and TGF-β did not demonstrate
increased calvarial defect repair compared to either therapy
individually.
Fibroblast growth factor
The FGF family is comprised of structurally related cytokines
that mediate several physiologic processes including cellular
proliferation, migration, and differentiation; mitogenesis;
angiogenesis; embryonic development; and wound healing.
There are at least 18 known members of the FGF family,
all possessing a characteristically high affinity for heparin.
Most isoforms share a similar internal core region that has
important FGF receptor (FGFR)-binding properties.255 Muta­
tions in FGFs or FGFRs are involved in the development of
various skeletal dysplasias, including achondroplasia and

295
Healing of fractures
dimerization results in autophosphorylation at a specific
tyrosine residue on a PDGFR, a necessary step for signal
transduction to proceed. Because of unique temporospatial
expression patterns, it is believed that PDGFR-α and PDGFR-
β mediate distinct events during development.265
PDGF is a potent mitogen and chemotactic agent for cells
and tissues of mesenchymal origin and is crucial in bone
homeostasis and repair. At sites of bone fracture, platelets
congregate and release PDGF. PDGF attracts many other cell
types important for tissue repair and the formation of granu­
lation tissue. PDGF has also been shown specifically to attract
osteogenic cells derived from calvaria, periosteum of long
bones, trabecular bone, and BMSCs.266 Additionally, angio­
genesis is crucial for bony healing (to be discussed later) and
PDGF, in addition to its direct mitogenic effect on osteoblasts,
indirectly enhances bone regeneration by stimulating angio­
genic cytokines.267
PDGF also stimulates bony healing via interactions with
other growth factors. Levi et al.268 showed that IGF-1 exposure
promoted osteogenic differentiation of human adipose-
derived stromal cells but PDGF-A alone did not. In combina­
tion, IGF-1 and PDGF-A enhanced osteogenesis to a greater
degree than IGF-1 alone, possibly due to enhanced IGF-1
transcription in the presence of PDGF-A.
The effects of PDGF on osseous repair in the clinical setting
have recently been investigated. One randomized controlled
trial assessed the effectiveness and safety of administering
recombinant human PDGF-BB (rhPDGF-BB) with beta-
tricalcium phosphate (β-TCP) for periodontal osseous
defects.269 Compared to β-TCP, β-TCP in combination with
rhPDGF-BB significantly enhanced linear bone growth (LBG)
and percent defect fill at 6 months. Subjects receiving
rhPDGF-BB also demonstrated a significant gain of clinical
attachment level (CAL) versus those who did not. These early
data were promising and therefore the authors extended the
study to test long-term stability and efficacy of PDGF-BB.
After 36 months, an analysis of 83 subjects with localized
severe periodontal osseous defects demonstrated that, based
on composite outcomes for CAL gain and LBG, PDGF-BB in
a synthetic scaffold matrix promotes long-term stable clinical
and radiographic improvements.270 Another randomized pilot
study offered rhPDGF-BB within a β-TCP matrix to patients
in need of ankle/hindfoot fusion.271 Compared to subjects
receiving autologous bone grafting, the PDGF group demon­
strated greater osseous bridging across the fusion site.
Healing of fractures
Bone repair is the physiologic process by which the body
facilitates the healing of fractures. In many ways, fracture
repair recapitulates the events that occur during skeletogen­
esis. Successful fracture healing is dependent upon many
factors. The mechanism of disruption, the fracture pattern,
and the method and timing of fixation all affect the outcome
of bone repair. For example, fractures that create more than
two fragments of the same bone (comminuted fractures)
require swift surgical intervention to heal successfully. In
addition, the biologic milieu surrounding a fractured bone
may play a role in the fate of the fracture.
Originally distinguished based on their histology, two
types of bone repair have been described (Fig. 18.4). Primary
Fig. 18.4  Schematic representation of primary and secondary (callus) bone repair.
(A) Primary bone repair: nondisplaced bone fragments heal without cartilaginous
intermediate. (B) Secondary (callus) bone repair: displaced (or unstable) bone
segments heal with cartilaginous intermediate. (Modified from Mehrara BJ,
McCarthy JG. Repair and grafting of bone. In: Mathes SJ, ed. Plastic Surgery. Vol.
2. Philadelphia: WB Saunders; 2005:639–718.)
Hematoma
Woven bone
Woven bone
Remodeled bone
Remodeled bone
Hematoma
Hypertrophic
chondrocytes
Periosteum
Nutrient artery
re-established
Nondisplaced
fracture segments
Displaced
fracture segments
A
B
bone repair refers to direct cortical healing of two fracture
ends without a cartilaginous intermediate. In contrast, during
secondary (callus) bone repair, a cartilaginous intermediate is
formed by mechanisms that occur within the periosteum, soft
tissues, and bone marrow. Most fracture repair is by second­
ary bone healing.
Primary bone repair
Primary bone repair involves the direct deposition of woven
bone by osteoblasts to re-establish mechanical continuity
between fracture fragments. Similar to intramembranous
ossification, the periosteum provides osteoprogenitor cells
and undifferentiated MSCs during primary bone repair. In
order for primary repair to occur, fracture fragments are
reduced to anatomic position, usually by rigid internal
fixation. The interfragmentary strain should also be at a
minimum.272 Bone healing occurs under axial compression
and fails under tension (Fig. 18.5).
During primary bone repair, new bone is synthesized in
parallel to the long axis of the bone by osteoprogenitor cells.
This process is initiated by the formation of cutting cones at

296
CHAPTER 18 • Repair and grafting of bone
the junction of live and necrotic skeletal segments. A cutting
cone is a system of cells, mainly composed of active osteo­
clasts. It burrows its way through cortical bone across the
fractured bone segments, thereby allowing the ingrowth of
blood vessels and undifferentiated MSCs. MSCs then differ­
entiate into bone-producing osteoblasts. New haversian
systems are created and provide pathways for blood vessel
penetration. Initially, bone is laid down in a woven-matrix
configuration. Within a few weeks, the woven bone becomes
more lamellar-like in its orientation. As the proportion of
lamellar bone increases over the course of months, complete
fracture repair occurs. The benefits of rigid plate osteosynthe­
sis include increased stability, decreased incidence of non- and
Fig. 18.5  The compression mode of miniplate fixation. (A) A four-hole plate. Note
that the inner holes are eccentrically shaped so that, as the screws are tightened,
the head falls into the wider portion and compresses the bone margins (B).
(C) Completed view of the skeletal fixation.
A
B
C
malunion, and decreased incidence of infection.273 Orthopedic
and cardiac surgeons commonly perform rigid internal fixa­
tion.274 Reconstructive surgeons also use rigid fixation tech­
niques to repair or reconstruct the craniofacial skeleton.275,276
Secondary (callus) bone repair
Untreated fractures or those treated with external or intramed­
ullary fixation or by sling or cast immobilization (i.e., without
rigid fixation) heal by secondary repair. During secondary
or callus bone repair, colonies of MSCs form cartilage at the
fracture prior to bone formation. There are three overlapping
phases of secondary repair: inflammatory, reparative, and
remodeling (Fig. 18.6). However, at least five distinct stages
have been described.277
After a fracture, the disruption of osseous, soft, and vascu­
lar tissues initiates the inflammatory phase. Lasting roughly
seven days, the inflammatory phase peaks at approximately
48 hours post-injury. Pain and swelling during this time
promote immobilization at the fracture site.278 A hematoma
forms, further immobilizing the area via its fibrin network.
The hematoma is also a source of signaling molecules that
initiate repair.279 Platelets release PDGF and TGF-β, which
promote MSC proliferation and osteogenic differentiation.
Recruited neutrophils, lymphocytes, monocytes, and macro­
phages remove necrotic debris and stimulate angiogenesis.
Approximately 3–4 days post-fracture – before the inflam­
matory phase subsides – the reparative phase commences. This
phase lasts weeks to months and is characterized by the devel­
opment of a fracture callus that will stabilize and unite fracture
segments before being replaced by bone. The callus is chiefly
composed of cartilage, but also contains woven bone, osteoid,
fibrous connective tissue, and blood vessels. MSCs from the
surrounding periosteum and marrow migrate to the injured
site and mature into many cell types, including osteoblasts,
chondrocytes, and fibroblasts, which form the callus.
The fracture callus undergoes both intramembranous and
endochondral ossification.277 Endochondral ossification occurs
during the second week of repair, when abundant cartilage
overlying the fracture site (formed by chondrogenesis of the
callus) undergoes calcification. Blood vessels grow into this
calcified cartilaginous callus, bringing with them osteoblast
progenitor cells. As the calcified cartilage is resorbed by
chondroclasts, osteoblasts begin to lay woven bone. Intra­
membranous ossification, on the other hand, occurs adjacent
to the fracture site.
The remodeling phase is characterized by the replacement
of woven bone with lamellar bone and may remain active for
several years. Osteoclasts resorb poorly located trabeculae
and new bone is formed along the lines of stress. In response
to varying mechanical loads, bone will gradually modify at
the fracture region until optimum regeneration is achieved.
During remodeling, pain subsides and the normal activities
can be resumed.
Variables influencing bone repair
Blood supply
Blood flow to bone represents approximately 10–20% of total
cardiac output. The pattern of vascularization to the majority
of the appendicular skeleton is similar in organization to the

297
Healing of fractures
Fig. 18.6  Schematic diagram of callus fracture repair in endochondral bone.
(A) Early stages of fracture repair characterized by hematoma formation,
inflammatory cell migration and differentiation (polymorphonuclear neutrophils),
and scattered primitive mesenchymal cells. (B) As healing progresses, a callus is
formed between the bone edges, and hypertrophic chondrocytes can be seen. In
addition, newly formed blood vessels sprout from the nutrient artery and local blood
vessels. (C) Primitive woven bone has bridged the fractured bone segment.
(D) Woven bone is remodeled into lamellar bone.
Fracture hematoma
Hypertrophic
chondrocytes
Periosteum
Newly formed blood vessel
Nutrient artery
disrupted
Primitive
mesenchymal cells
Callus
Necrotic bone edge
Primitive woven bone
Inflammation
Repair
Repair
Remodeled bone
A
B
C
D
Fig. 18.7  The three sources of blood supply to the long bone. The nutrient artery
provides the principal source of blood supply to the marrow cavity and the inner
cortex. The diaphyseal periosteal vessels supply the outer cortex of the diaphysis.
The metaphyseal epiphyseal periosteal vessels penetrate the cortex of the bone in
the adult and anastomose with the nutrient artery, providing adequate supply to the
marrow cavity and inner cortex in cases of disruption of the nutrient artery.
Articular cartilage
End-arterial
terminals
Medullary
sinusoids
Central venous
channel
Principal nutrient
artery and vein
Periosteal capillaries
in continuity with
cortical capillaries
Metaphyseal arteries
and terminals of the
medullary arterial
system
Large emissary
vein
Transverse
epiphyseal
venous channel
Venous sinusoids
and metaphyseal
veins
vascular pattern of a typical long bone.280 A tubular long bone
in the adult human receives blood by three distinct arterial
systems that interconnect extensively with one another (Fig.
18.7). The primary blood supply to the inner two-thirds of
bony cortex and the medullary cavity comes from the diaphy­
seal nutrient artery. As the nutrient artery traverses the cortex,
it passes through to the medullary cavity, where it divides into
ascending and descending medullary branches. Metaphyseal
and epiphyseal arteries represent the second significant source
of blood. These vessels arise from arteries supplying adjacent
joints and deliver blood to cancellous bone found at the end
of long bones. The third main source of blood comes from
periosteal arteries, which carry supply to the outer one-third
of cortex.

298
CHAPTER 18 • Repair and grafting of bone
As blood travels through bone from endosteum to perios­
teum, it flows through sinusoids and arterioles within the
haversian system. Exchange vessels (i.e., vessels connecting
the arterial and venous systems) within osteons lie parallel to
the axis of the bone. Collecting sinuses and veins with similar
names to their arterial counterparts receive deoxygenated
blood from exchange vessels. A high intramedullary pressure
Table 18.4  Growth factors and angiogenesis
Growth
factor
Structure
Angiogenesis
Function
Role in fracture
healing
Regulation of
expression
Receptor
BMP
Single large
propeptide
molecule
Cleavage
enables
dimerization
Member of the
TGF-β
superfamily
Conserved
series of
seven
cysteine
residues at
carboxyl
terminal
At least 15
known
isoforms
Indirect
regulator of
angiogenesis
Mitogen and
chemoattractant
for osteoblasts
and
mesenchymal
cells
Apoptosis
Patterning
Expressed by
mesenchymal cells
early in fracture
repair
Also expressed by
osteoblasts and
osteoclasts
Exogenous BMPs can
heal critical-sized
defects
Ectopic bone formation
? Accelerate fracture
repair
? Actions synergistic
with TGF-β
Patterning genes
and transcription
factors (e.g.,
Cbfa1)
Serine-
threonine
kinase
receptors
TGF-β
Propeptide
Cleavage
necessary
for
activation
At least three
different
isoforms
Indirect
regulator of
angiogenesis
Chemoattractant
and mitogen for
osteoblast
precursors and
osteoblasts
Inhibits osteoblast
differentiation
Regulates
expression of
other growth
factors
important in
angiogenesis
and
osteogenesis
Increases matrix
synthesis
Produced for
osteoblasts,
mesenchymal cells,
osteoclasts
Increases callus
formation and
volume
Subperiosteal injection
promotes both
chondrogenesis and
membranous
ossification
Exogenous TGF-β may
heal some critical-
sized defects
Patterning genes
and transcription
factors
Mechanical strain
Serine-
threonine
kinase
receptors
FGF-2
Heparin-
binding
glycoprotein
At least nine
different
isoforms
Direct regulator
of
angiogenesis
Angiogenesis
Regulator of bone
development
and skeletal
patterning
Can regulate
osteoblast
differentiation
and proliferation
(actual effect is
dose-
dependent)
Increases
angiogenesis in
fracture site
Increases expression
during fracture repair
Exogenous FGF-2 can
increase mechanical
strength of fractures
? Accelerates fracture
repair or promotes
healing of critical-
sized defects
Inflammatory cells
and acute
inflammation
Patterning genes
and transcription
factors
Tyrosine
kinase
receptors
is likely responsible for the centrifugal flow of blood through
bone.281
Angiogenesis is critical for bone maturation and repair.282
Although the mechanisms that regulate angiogenesis during
fracture repair have not been fully elucidated, many important
angiogenic cytokines have been described (Table 18.4). One
molecule thought to play a pivotal role in angiogenesis within

299
Healing of fractures
developing and healing bone is VEGF. Other factors described
previously (BMP, TGF-β, FGF-2, and PDGF) have also been
implicated in angiogenic processes necessary for osteogenesis.
VEGF and FGF-2 are the only two directly acting regulators
of angiogenesis; other growth factors serve indirect roles. In
addition, the hypoxia-inducible factor-1α (HIF-1α) pathway
is crucial to the coupling of angiogenesis with osteogenesis.283
Discoveries associated with this pathway have led authors
to hypothesize that skeletal angiogenesis and osteogenic–
angiogenic interactions are rate-limiting processes in bone
growth.
A potent regulator of both vasculogenesis and angiogenesis,
VEGF mediates many important functions during the
embryologic stage of development and throughout the life­
time of organisms. Deletion of a single VEGF allele is associ­
ated with lethality in utero.284 VEGF is structurally related to
PDGF and is expressed by many cell types, often at high
concentrations within highly vascularized tissues. VEGF
mediates its effects on cells through interactions with VEGF
receptors (VEGFR-1, VEGFR-2, VEGFR-3).
The hypothesis that VEGF plays an essential role in skeleto­
genesis mainly comes from studies examining endochondral
bone formation. At the epiphyseal growth plates of long
bones, hypertrophic chondrocytes express high levels of
VEGF.285 Injection of an anti-VEGF monoclonal antibody
Table 18.4  Growth factors and angiogenesis—cont’d
Growth
factor
Structure
Angiogenesis
Function
Role in fracture
healing
Regulation of
expression
Receptor
VEGF
Dimeric
glycoprotein
Direct regulator
of
angiogenesis
Increases vascular
permeability
Increases vascular
sprouting
Increases
endothelial cell
migration,
proliferation,
adhesion
Increased expression
during fracture repair
Expressed by
osteoblasts,
osteoclasts, and
mesenchymal cells
Increases blood
vessels in growth
during fracture repair
Microenvironment
(e.g., pH, hypoxia,
lactic acid
concentration)
Growth factors (e.g.,
TGF-β, FGF, BMP,
IGF), inflammatory
cytokines (e.g.,
PGE2)
Tyrosine
kinase
receptors
PDGF
Dimeric,
disulfide-
bonded
polypeptide
chain
α and β
subunits
determine
isoform (AA,
AB)
Indirect
regulator of
angiogenesis
Promotes
chemotaxis of
osteoblasts and
inflammatory
mediators
Decreases cellular
differentiation
Increases
collagen
synthesis
Increases
collagen
degradation
and turnover
May promote
osteoblast
proliferation
Released by
degranulating
platelets and acts to
increase osteoblast
chemotaxis, and
possibly proliferation
Expressed by
osteoblasts,
macrophages, and
mature/immature
chondrocytes during
fracture repair
Promotes synthesis of
angiogenic
molecules (e.g.,
VEGF)
Promotes matrix
deposition and
turnover
Inflammation and
injury
Mechanical strain
TGF-β (decreased
expression)
Tyrosine
kinase
receptors
BMP, bone morphogenetic protein; FGF-2, fibroblast growth factor; IGF, insulin-like growth factor; PDGF, platelet-derived growth factor; PGE2, prostaglandin E2; TGF-β,
transforming growth factor-beta; VEGF, vascular endothelial growth factor.
results in inhibition of vascular invasion of the nonhuman
primate growth plate, thereby hampering longitudinal bone
growth.286 Moreover, mice treated with a VEGF inhibitor
manifest significantly decreased bone capillary invasion and
trabecular bone formation.287 Cessation of anti-VEGF treat­
ment is associated with capillary invasion and restoration of
bone growth. A comprehensive review produced by Yang
et al.288 discusses VEGF in the context of bone angiogenesis
and development, emphasizing its role on intramembranous
and endochondral ossification.
The therapeutic utilization of VEGF and other angiogenic
cytokines is a subject of great promise in the treatment of
patients with both complicated and uncomplicated fractures.
Luo and colleagues289 examined the effects of VEGF and
BMP-2 on bone defect repair around dental implants. The
study found that bone marrow stromal cells within chitosan/
collagen scaffolds were responsive to AdBMP-2 administra­
tion but VEGF alone failed to stimulate bone formation.
Administration of AdBMP-2 and VEGF protein together
resulted in significantly increased bone formation and bone-
to-implant contact compared to the AdBMP-2 group, suggest­
ing a synergistic effect of combination BMP-2 gene and VEGF
protein therapy on bone healing. In addition, Liu and Olsen290
found that mice with a congenital deficiency of VEGF in
osteoblastic precursor cells demonstrate reduced bone mass

300
CHAPTER 18 • Repair and grafting of bone
differentiate to produce bone when exposed to osteoinductive
factors. Significant quantities of these progenitor cell types are
found in adipose tissue191 as well as the neonatal umbilical
cord and infant palatal periosteum.301 The ability to harvest
and manipulate these cells via osteoinduction has important
therapeutic implications for both children and adults suffer­
ing critical-sized bony defects.
There are numerous strategies to facilitate the osteoinduc­
tion of effector cells in order to produce bone. Examples
include the use of bone grafts, vascularized bone flaps, bioac­
tive scaffolds, osteogenic growth factors, and biophysical
stimulation.302 A detailed discussion of bony regeneration and
osteoinduction can be found elsewhere.303,304
Osteoconduction
Osteoconduction is the ability of a material to serve as a scaf­
fold onto which bone can attach and grow. During primary
bone repair, the opposing fracture fragments mediate osteo­
conduction. In secondary bone repair, callus ECM acts as a
scaffold onto which new bone can attach and develop. In the
field of regenerative medicine, many implanted materials
serve an osteoconductive role. An osteoconductive substrate
must be porous to allow for the ingrowth of bone and fibro­
vascular tissue during osteoinduction.305 Many osteoconduc­
tive materials are available to enhance bony healing in the
clinical setting. Reconstructive surgeons can choose from
allografts (processed human bone), autogenous bone grafts,
purified collagen, calcium phosphate (CaP) substitutes, syn­
thetic polymers, and more. In some cases, osteoconductive
materials may also have osteoinductive properties. Native
bone, for example, stimulates both osteoconduction and
osteoinduction. Additionally, purely osteoconductive materi­
als (e.g, CaP substitutes) are enhanced by being combined
with osteoinductive factors (e.g., BMP) before placement into
a defect.306
Osseointegration
The phenomenon of osseointegration describes a stable
anchorage between bone and implant that results in a struc­
tural and functional connection between the two. Similar to
primary fracture healing, during osseointegration native bone
and the implant unite without a cartilaginous intermediate. A
key difference between primary fracture repair and osseoin­
tegration is unification of native tissue and a foreign body
during osseointegration. Therefore, the structure and compo­
sition of the implant are integral to the process.
In order for osseointegration to manifest, certain prerequi­
sites must be met.307 The implant must be bioinert or favorably
bioactive. Bioinert materials do not cause adverse reactions in
native tissue. Titanium is an example of a bioinert material
often used in dental, craniofacial, and orthopedic procedures.
A material that is favorably bioactive promotes a positive
tissue reaction by facilitating osseous tissue production or
chemical bond formation between implant and native tissue.
Favorable bioactive agents are often used to coat implants with
an otherwise unfavorable mechanical quality. The degree and
strength of bone–implant contact depend on the surface prop­
erties of an implant. The extent of bone–implant interface has
been shown to be positively correlated to implant surface
roughness.308 A rougher implant may be better able to tolerate
and increased bone marrow fat consistent with an osteoporosis-
type phenotype. Therefore, VEGF has become a potential
target for therapies to reduce bone loss.
Fracture fixation
Critical to the success of fracture repair in long bones is the
extent of movement at the fracture site after fixation.291
Delayed union and nonunion are both associated with excess
motion; however, some movement, known as micromotion,
can enhance bone healing. The mechanism by which absolute
immobilization hampers fracture repair may be related to
increased resorption in a stress-protected environment. In
addition, the degree of mechanical strain endured at the site
of a fixed fracture has differential effects on chondrogenesis
and osteogenesis.292
Age
Characteristic changes within bone occur during the aging
process. Pediatric fractures tend to heal at an accelerated rate
compared to adult fractures.293 This may in part be due to the
effect of age on angiogenesis.294 To understand this, Lu and
colleagues295 evaluated tibial fractures created in 4-week-old,
6-month-old, and 18-month-old (juvenile, middle-aged, and
elderly, respectively) rats. At 7 days post-fracture, there was
a higher surface density of blood vessels in juvenile fracture
calluses compared to middle-aged and elderly rats. In addi­
tion, juvenile rats expressed increased HIF-1α and VEGF at
3 days post-fracture versus older rats. These data support
the notion that angiogenic potential during fracture healing
changes over time. Age may also influence fracture repair via
effects on periosteal structure, cell population, and chondro­
genic potential; stem cell function; and biologic signaling.296,297
Bone remodeling
Osteoinduction
Osteoinduction is a process by which undifferentiated plu­
ripotent cells are stimulated to become cells within the
osteoblast lineage.298 It occurs naturally during skeletogenesis
and fracture repair.
Although osteoinduction is an intrinsic biologic process, it
can also be manipulated exogenously. For example, purified
and recombinant human isoforms of BMP are commonly used
in experimental settings to induce bony formation. Various
biomaterials have also been used for the same purpose. Bio­
glass and other osteoinductive materials have shown promise
when used in craniomaxillofacial applications.299 After bioac­
tive glass is mixed with saline or blood, an apatite surface
layer forms at the biomaterial–bone interface, which incorpo­
rates collagen and proteins from surrounding native bone. In
addition to producing a chemical bone at the interface, the
apatite layer stimulates production of osteogenic cytokines
from surrounding osteoprogenitor cells, resulting in new
bone formation.300
A major concern in the treatment of significant osseous
defects is the amount of native tissue available for repair.
Large defects may not be amenable to autologous reconstruc­
tion without undue donor site morbidity. Therefore, an area of
intense study is the identification of undifferentiated cells that

301
Bone remodeling
Histology
Successful distraction osteogenesis results from membranous
ossification in the absence of a cartilaginous intermediate.323–327
In the early phases of regenerate formation, distraction angio­
genesis is a more accurate term to describe the cellular
response to ischemia and traction. This phase is characterized
by the appearance of multipotent precursor cells and type I
collagen matrix expression in the distraction zone. Recently,
Cetrulo and colleagues were able to demonstrate the honing
of MSCs to the distraction zone in rat mandibles after
intra-arterial transfer,328 an indication that paracrine factors
expressed in ischemia (e.g., HIF-1α) are chemoattractants and
eventual agents of osteoinduction. Specifically, endothelial
precursor cells are enriched at midactivation through 1–2
weeks into the consolidation period, when distracted rats
were compared to controls.328 CD31, CD34, and Flk-1, markers
of neovascularization, stem cells, and endothelial precursors,
respectively, have all been demonstrated by immunohisto­
chemistry. Further in the activation period, dense collagen
fibrils are formed and organized in the direction of distraction
as a template for mineralization.325,329–331
It has been recognized that bone precursor cells respond to
mechanical cues in order to differentiate. Distraction appears
to be no exception to this tenet. In a rat model, Rhee and
colleagues demonstrated increased intranuclear expression
of signal transduction factors ERK1/ERK2 early in the dis­
traction phase.332 This expression paralleled upregulation of
BMP-2/BMP-4, two relatively potent osteoinduction agents.
Mineralization by active mature osteoblasts starts 10–14 days
after the onset of distraction, and begins at the osteotomy
fronts, proceeding centrally towards the relatively avascu­
lar fibrous interzone (Fig. 18.8).313,325,329,333 This leads to the
formation of the primary mineralization front. Thin-walled
vascular channels and bony spicules are slowly converted
to normal lamellar bone and marrow. Interestingly, during
the latter phase of activation, the infantile regenerate is still
malleable and can remodel according to external forces. One
can therefore take advantage of the gradual process and
judiciously manipulate the distracted bone clinically in order
to obtain a more precise occlusion, for example. This clinical
principle, which is based on the gradual histomorphogenesis
of the distraction zone, is termed molding the regenerate.334,335
Several studies have confirmed that molding the distraction
bone, clinically by use of an orthopedic appliance, does not
lead to disturbance of the regenerate or fibrous nonunion.334,335
Variables affecting osteogenesis
To understand the variables that have a negative or positive
impact on osteogenesis in the context of distraction, one must
understand the factors that promote successful regenerate
formation and bony union. These tenets are aptly described by
Ilizarov336: (1) device stability; (2) a latency period; (3) a gradual
distraction (activation) period; and (4) a sufficient consolida­
tion period. First, osteocyte viability reflects osteoblastic activ­
ity and thus the selection of a strong area of bone for osteotomy
and distractor placement is critical. A stable device placed in
stably healing bone results in the formation of strong cortical
bone within the regenerate and minimizes the chance of fibrous
nonunion. It follows that any thermal or mechanical injury to
the bone can damage the blood supply to the bone, resulting in
shear stress. In addition, Butz and colleagues309 assessed the
intrinsic biomechanical properties (hardness and elastic
modulus) of bone tissue integrated to titanium implants with
varying degrees of surface roughness. Bone deposited around
acid-etched titanium implants displayed enhanced biome­
chanical properties compared to bone deposited around
implants with smoother, machined surface topography. Simi­
larly, compared with machined surfaces, implant surfaces
subjected to anodization exhibit increased bone-to-implant
contact.310
Distraction osteogenesis
Distraction osteogenesis, the formation of new bone from the
gradual separation of osteotomized fronts, was pioneered by
the work of Codvilla at the turn of the 20th century and
expanded by Ilizarov in orthopedic surgery.311–313 Through
a set of rigorous experimentations, Ilizarov313 was the first
to establish the hypothesis that gradual controlled separa­
tion of osteotomized bone results in a “tension stress effect”
leading to angiogenesis and new bone formation. These initial
observations and classic experiments were based upon the
axial skeleton. It was not until the early 1970s that the concept
of craniofacial distraction was introduced, first in a canine
model,314 in which the authors demonstrated correction of an
iatrogenic crossbite by an external distraction device (1 mm/
day for 2 weeks). The application of distraction to the human
craniofacial skeleton was finally introduced by McCarthy
et al. in the early 1990s.315–317 Since its inception in the human
mandible, the indications of distraction osteogenesis have
expanded from congenital cases to cases of craniofacial skel­
etal deficiency with tumor-, trauma-, or iatrogenic-related
etiologies.
Distraction osteogenesis requires three main stages for
successful augmentation of bone formation: latency, activa­
tion, and consolidation. The first stage, latency, occurs
immediately after osteotomy and refers to the period during
which bone healing is initiated at the bony gap, periosteal
integrity is restored, and callus formation begins. Latency
typically ranges between 1 and 7 days, and is typically gauged
according to the patient’s age; the younger the patient, the
shorter the latency period. During the second stage, activation
(distraction), osteogenesis is induced with the generation of
immature bone. Distraction occurs by the turning of an axial
screw at a predetermined rate for a certain number of days
until target bone augmentation is achieved. The third and
final stage, consolidation, refers to the phase during which
immature bone remodels into mature bone. In practical terms,
consolidation is the period from the end of distraction to the
removal of the device. The parameters for these stages are
established by the craniofacial surgeon in response to specific
patient profiles and the region of the skeleton undergoing
distraction.
As understanding of and experience with distraction
osteogenesis advance, associated adverse events occur
less frequently.318 However, this number may be under-
reported.319,320 Complications that arise may be due to surgeon
technique, patient factors (see below), hardware failure, or
unknown factors. Many authors have attempted to classify
complications for specific body areas (e.g., mandibular dis­
traction)318,320–322; however, no consensus standard classifica­
tion scheme exists.

302
CHAPTER 18 • Repair and grafting of bone
failed to underscore any difference in protocols in terms of
osteogenesis.340,341 Clinically, Hollier and colleagues demon­
strate a low complication rate with the elimination of the
latency period and implementation of a rapid distraction rate
(2 mm/day) in their pediatric population.342 It is important to
note that the nonunion rate in this series was 4.5% and was
attributed to the only distracted bone graft in the study. It is
generally accepted that reduced latency periods and faster
distraction rates formulate an acceptable protocol in children,
due to improved blood supply and osteogenic potential. A
study by Slack et al.343 showed that children (ages 4–12) under­
going mandibular distraction demonstrate no difference in
clinical outcomes up to one year postoperatively whether
their distraction protocol contains a 48-hour latency period or
no latency period. Flexibility in the timing and rate of distrac­
tion needs to be inherent in any successful protocol, with
respect to patient profile.
Patient factors
Age
It has been well recognized that increasing age contributes
negatively to the overall outcome of distraction osteogenesis,
presumably secondary to limited ability of older subjects to
recruit osteoprogenitor cells to the site of injury. In support of
this concept, studies have demonstrated that the rate of bone
formation and bone mineral deposition in pediatric distrac­
tion patients (385 mm/day) is far superior to adults (213 mm/
day).340 This age-related difference in osteogenic potential,
however, is not a contraindication to distraction osteogenesis,
as older individuals can undergo surgery successfully. In this
context, latency and consolidation periods may be extended
to allow for protracted regenerate formation in the older
patient.
Blood supply
Several studies have demonstrated that, on a molecular level,
the establishment of neovascularization is critical to the for­
mation and eventual mineralization of the regenerate. Accord­
ingly, factors that impair vascularization of the distracted
field may result in failed bone formation. Osteocyte survival
depends on the proximity of less than 0.1 mm to nutrient
vessels.344 The factors critical in osteogenesis of the axial skel­
eton (e.g., periosteal preservation) appear to be less important
for osteogenesis in the craniofacial skeleton.312,323,333,337,339,345,346
There are, however, factors that unify both regions.
Radiation/chemotherapy
Radiotherapy, by compromising vascularity, cellularity, and
local oxygen supply, has been known to impair osteogenesis.
Limited clinical347–349 and animal studies350–353 have been per­
formed in this regard; these studies have had mixed results.
In the experimental realm, animals exposed to 36 Gy of radia­
tion prior to the distraction process show a definitive decrease
in bone mineralization microdensitometric analysis.354 These
authors therefore propose finding an optimal radiation dosing
protocol, which would make distraction osteogenesis a viable
alternative in the head and neck cancer patient. Regardless of
the optimal conditions for radiation delivery and in consider­
ation of the reduced vascularity within the irradiated field,
ischemic fibrogenesis. Fibrogenic damage leads to irregular
and disorganized collagen formation from arterial insuffi­
ciency or cystic degeneration from venous outflow obstruction.
Previous concerns of periosteal damage resulting in poor
osteogenesis have recently diminished as increasing experi­
ence among craniofacial surgeons has demonstrated that, even
with complete osteotomy, the periosteum is sufficiently osteo­
genic for strong bone production.
Inappropriate timing of the distraction stages can result in
osteogenic failure. In terms of the latency period, this trans­
lates into either too short a period (immediate distraction) or
a prolonged period (delayed distraction). Important for pri­
mordial callus and ultimately regenerate volume, latency
periods range between 0 and 7 days, depending upon patient
variables, such as age, osteotomy site, blood supply, and other
factors that may affect regenerate formation (see below). The
majority of experimental evidence points to an optimal latency
period of 5–7 days. Theoretically, immediate distraction may
prevent adequate hematoma and primordial callus genera­
tion, culminating in a regenerate of suboptimal volume/
mechanical strength, and, in extreme circumstances, fibrous
nonunion. It follows that disruption of initial fracture hema­
tomas, which simulates an abbreviated latency period, results
in a 26% decrease in callus cross-sectional area, and a signifi­
cant impairment in the mechanical strength of the newly
formed bone.337 On the other hand, deferment of the distrac­
tion period >14 days results in premature consolidation of the
distraction site and device failure.323,329,338,339
Overall, latency periods have been a source of debate
among craniofacial surgeons and their significance held in
question ever since the inception of the technique. To this end,
studies comparing latency periods of 7, 14, and 21 days have
Fig. 18.8  Schematic drawing of bone formation during distraction osteogenesis.
After approximately 10 days of distraction and during the distraction period, the
regenerated tissue can be divided into three zones: (1) an avascular fibrous
interzone (FZ); (2) two mineralization zones (MZ) or fronts; (3) and two regenerate
zones (RZ). Regenerate zones are adjacent to residual host bone segments (RHBS).
(From Samchukov ML, Cope JB, Cherkashin AM. Craniofacial Distraction
Osteogenesis. St. Louis: Mosby; 2001.)
RHBS
RZ
RZ
MZ
MZ
FZ
RHBS

303
Clinical application of bone transfers
intracortical osteolysis.358 Successful grafting requires a well-
vascularized bed and adequate graft fixation.359 Mechanical
stability of the healing defect allows for revascularization of
the graft.360 Mechanical stress on a rigidly fixed graft prevents
graft resorption.361–363 An intact periosteum also improves
graft survival.361
Cancellous versus cortical grafts
Cancellous and cortical bone grafts have different applica­
tions in skeletal reconstruction. Compared to cortical bone,
cancellous bone has little immediate strength or structural
support but has higher osteogenic, osteoinductive, and osteo­
conductive properties. It is more quickly incorporated and
revascularized than cortical bone, usually within two weeks.
In the transplanted cancellous bone graft, viable osteoblasts
and endosteal cells line the graft surface. Cancellous bone
grafts serve as an osteoconductive substrate to support creep­
ing substitution.364–367 They are indicated to bridge gaps less
than 6 cm in nonstress-bearing areas. Typical sources of can­
cellous bone grafts are iliac crest, cranial diploe, upper tibial
epiphysis, and distal radius.
Cortical bone, alternatively, has limited osteogenic
potential.368 Creeping substitution is the main mechanism
of cortical bone graft incorporation, and the graft largely
relies on plasmatic imbibition from the recipient bed to
support the outer layer of osteocytes.365 The incorporation
and revascularization of cortical grafts usually take one to
two months.369–371 However, cortical grafts provide immediate
structural support and are able to bridge defects up to 12 cm.
Cortical grafts should be tailored to the defect. Generally,
rigid fixation is required to prevent fracture of the healing
graft.372 Cortical grafts are often used for onlay augmentation
of contour deficits. Typical graft sites include fibula, rib, and
iliac crest.
Clinical considerations
To optimize the chance of survival, the bone graft should be
handled as follows373:
1. Graft exposure time to air at room temperature should
be minimized (ideally to less than one hour to maintain
cell viability).374 However, longer periods of ischemia
can be tolerated by bone cells if the medullary nutrient
blood supply is later reconstituted, as in vascularized
bone.375
2. To enhance bone viability by four to six hours, a graft
should be covered in a blood-soaked sponge with a
moist saline gauze on top. If longer periods of graft
exposure are required, cell viability can be maintained if
the graft is stored in 10% human serum albumin and a
90% balanced salt solution at 3°C in Collins–Terasaki
solution.375
3. Grafts should be kept cool, as temperatures over 42°C
may cause cell death.
4. Antibiotic washes are not only bactericidal but are also
cellucidal and therefore should be used with caution.
5. Dead space around the graft should be avoided.
6. The cancellous portion of the graft should be placed in
contact with the cancellous bone in the bed.
7. Ideally, the graft should be placed in a previously
prepared vascularized bed to increase graft survival.
such patients may need an extended latency period, slower
distraction rate, and longer consolidation period to achieve
similar outcomes in osteogenesis to the nonirradiated
patient.
Many patients with malignant bone tumors will receive
chemotherapy as part of their treatment plan. Chemothera­
peutic agents are often cytotoxic not just to tumor cells but
also to normal cells. Timing appears to be important with
respect to the detrimental effects of chemotherapy on bone
formed by distraction osteogenesis. Monsell and colleagues355
found in distracted rabbits that preoperative chemotherapy
reduced regenerate bone mineral density, bone mineral
content, and volumetric bone mineral density but did not alter
regenerate structural integrity (i.e., the callus formed is of
good quality). In contrast, perioperative chemotherapy does
not alter bone mineralization but negatively impacts mechani­
cal properties including energy at yield and yield strain.355,356
The latter may have clinical implications and further studies
in humans are required.
Clinical application of bone transfers
Indication for bone transfers
A variety of clinical approaches are currently used to repair
skeletal defects in plastic surgery. In order to provide an
optimal outcome, detailed analyses of the bony defect and
mechanical requirements of the reconstructed part are needed.
Reconstruction of bony deficiencies can be performed using
autografts (obtained from the patient), allografts (obtained
from another individual), xenografts (obtained from another
species), bone substitutes, or implants. The ideal is to use
autologous bone grafting when possible. The type of bone
graft is chosen based on the clinical scenario; medical condi­
tion of the patient; previous radiation, infection, or severe
scarring of the recipient site; donor site availability; and
patient and surgeon preference.
Common indications for the use of bone grafts are critical-
sized bony defects from trauma, tumor resection, and con­
genital disease. Other indications include fracture nonunion,
arthrodeses, limb-lengthening procedures, and spinal fusion.
The structure and biomechanical properties of the graft deter­
mine the clinical application.
Bone graft healing and graft survival
Bone graft incorporation and fracture healing go through
similar stages. Graft incorporation consists of graft resorption
and replacement by vascular ingrowth and new bone (i.e.,
creeping substitution).357 Bone graft incorporation is subject to
several factors: type of graft (e.g., autogenous/allogeneic,
vascularized/nonvascularized), quality of transplanted bone
and recipient site, mechanical properties of the graft, and
systemic and local disease. The molecular and mechanical
environment of the graft site, the contact between the graft
and recipient bed, and the osteoinductive, osteoconductive,
and osseointegrative capability of the graft influence bone
graft healing.
Bone formation during remodeling is stimulated by com­
pression, which leads to periosteal and endosteal apposition;
traction, in contrast, promotes periosteal resorption and

304
CHAPTER 18 • Repair and grafting of bone
Preservation of the insertion of the abdominal and gluteal
muscles on the iliac crest is key. Dependent upon the amount
of graft taken, the outer, inner, or both cortical plates of the
iliac wing need to be preserved. The amount of cancellous
bone that can be harvested from the ilium is quite large –
down to the cotyloid ridge above the glenoid fossa. A large
subcrest corticocancellous graft up to 11 cm, a 6 × 10 cm cortico­
cancellous inner plate, and an approximately 5 × 8 cm cortico­
cancellous outer plate can also be harvested. Harvest of the
anterior iliac crest (see Fig. 18.9) is convenient due to supine
patient position. However, the posterior iliac crest provides a
larger amount of graft and avoids aberrant entry into the
abdomen. To prevent avulsion fractures of the anterior supe­
rior iliac spine (ASIS), the anterior approach should stop 3 cm
posterior to the ASIS. The posterior approach should stop
4 cm anterior to the posterior superior iliac spine to prevent
injury to the sacroiliac joint.368
The anterior approach can be performed in supine position
or with a roll underneath the patient’s hip to raise it. The
following landmarks are drawn: ASIS and the tubercle of the
iliac crest. The incision is placed over the iliac crest and is
Techniques of harvest: autologous
bone grafts
Ilium
The ilium is the most commonly used site for corticocancel­
lous bone grafts (Figs. 18.9, 18.10). The clinical indications are
wide due to the versatility of the ilium, the large amount of
cortical and cancellous bone available, and the possibility to
transfer it as a vascularized bone flap. The iliac crest is easily
accessible and has limited morbidity if proper technique is
employed. Complications include chronic donor site pain,
paresthesias, and gait abnormalities.376
The harvest technique should preserve the shape of the
iliac crest by careful and proper splitting. Correct handling of
the iliac wing permits its regeneration and allows for further
or repeat harvest. Additionally, wiring the split iliac crest back
together can decrease postoperative pain. Care must be taken
to avoid stretching the lateral femoral cutaneous nerve by
unnecessary superior dissection, to avoid tearing the gluteal
muscles, and to avoid removing the iliac crest itself.
Fig. 18.9  A technique for removal of cancellous bone from the ilium. (A) Incision of the periosteum after exposure of the crest. (B) Separation of the cortex with an
osteotome. (C) Cross-section showing lines of separation of the inner and outer cortex from the cancellous bone. (D) Exposure of cancellous bone. (E) Cross-section
showing exposure. (F) Reunion of the inner and outer cortex by stainless steel wire suture. ((A–F) From Cutting CB, McCarthy JG, Knize DM. Repair and grafting of bone. In:
McCarthy JG (ed). Plastic Surgery. Vol 1. Elsevier; 1990. (F) After Tessier P, Kawamoto H, Matthews D, et al. Taking bone grafts from the anterior and posterior ilium – tools
and techniques: a 6800-case experience in maxillofacial and craniofacial surgery. Plast Reconstr Surg. 2005;116(Suppl):25S–37S; see also Wolfe SA, Kawamoto HK. Taking
the iliac-bone graft. J Bone Joint Surg Am. 1978;60:411.)
A
B
C
D
E
F
Anterior superior
iliac spine

305
Clinical application of bone transfers
6–10 cm in length. By pulling the skin medially, the incision
actually resides 3–5 cm below and lateral to the iliac crest. No
subcutaneous dissection is necessary; the abdominal soft
tissue is retracted medially and the crest, tuberosity, and
anterior spine are palpated. A cartilaginous cap is frequently
encountered overlying the bone and this must be split to
access the cortical plate. The next step is to split the iliac crest;
the split is extended just behind the anterior spine. The medial
leaf of the split crest is created and moved medially as a single
piece with the attachments of the abdominal muscles. The
same maneuver is performed for the lateral leaf with attention
to the convexity of the lateral crest. The grafts are extracted
according to defect size and shape.377 Excessive retraction can
potentially cause injury to the lateral femoral cutaneous nerve,
which travels retroperitoneal on the deep surface of the iliac
muscle, down to the inguinal ligament near the ASIS. This can
result in postoperative paresthesias.
Prior to closure, hemostasis is achieved using electrocau­
tery or bone wax (e.g., bleeding from the central artery of the
iliac bone).
Gelfoam can be placed in the periosteal pocket, and the
iliac leafs are wired back together. Muscles, subcutaneous
tissue, and skin are closed in a layered fashion over a suction
drain. A pain pump catheter delivering local anesthetics
can be inserted in the wound for better postoperative pain
control.378 In similar fashion, the posterior iliac crest can be
harvested from the supine or prone position. When harvesting
this region of the crest in the prone position, one must take
care to avoid injury to the cluneal nerves.378 Also, in order to
prevent injury to the sacroiliac joint, the most posterior mark
with the osteotome is made 4 cm anterior to the posterior iliac
spine.368
Fig. 18.10  Instruments for ilium harvest.
(A) Obwegeser periosteal elevator, 6 mm.
(B) Obwegeser periosteal elevator, 12 mm.
(C) Wheatlander retractor. (D, E) Farabeuf retractors.
(F) Digman bone-holding forceps. (G–M) Osteotomes,
4–12 mm, straight and curved. (N) Mallet.
(O, P) Senn retractors. (Q, R) Ragnel retractors.
(S) Cottle crusher.
Complications are rare if proper technique is used, as
shown in a summary of 5600 patients with a resultant com­
plication rate of 0.5%.377
Tibia and fibula
The tibia was mainly used as a source of corticocancellous
bone graft during World War I and II. Its use has decreased
in popularity due to donor site morbidity and pathologic
fractures. The tibia remains a favorable donor site in various
centers in Scandinavia and the UK when only small amounts
of cancellous bone are needed (e.g., alveolar bone graft­
ing).379,380 A detailed description of technical aspects of tibial
harvest has been given (Fig. 18.11).381
The use of fibula grafts has decreased with the availability
of allograft. Currently, the fibula is mainly used as a vascular­
ized transfer for mandible reconstruction, as well as midface
and extremity reconstruction.
Greater trochanter and olecranon
The greater trochanter and the olecranon are potential sites of
cancellous bone harvest, specifically when small amounts are
needed. The harvest technique is similar to a bone biopsy. A
small incision is made overlying the area of interest, dissection
is carried down to the bone, and the periosteum is removed
from the desired area. An osteotome can be used to transect
the cortex or a small drill bit can be used to create perforations
in an elliptical pattern, which then can be connected with an
osteotome. Through this opening, cancellous bone can then be
harvested with a curette. After completion of the harvest, the
cortical bone piece can be reinserted, or if need be, can be used
as part of the graft. Hemostasis is achieved using gelfoam and

306
CHAPTER 18 • Repair and grafting of bone
Fig. 18.12  Ribs 5–7 are the most commonly used donor ribs. (From Tessier P,
Kawamoto H, Matthews D, et al. Taking long rib grafts for facial reconstruction
– tools and techniques: III. A 2900-case experience in maxillofacial and craniofacial
surgery. Plast Reconstr Surg. 2005;116:38S.)
5
6
7
Fig. 18.11  Tibial bone grafts. (A) Incision for removal of the graft. (B) Removal of
the graft with an osteotome.
A
B
thrombin. A layered closure is performed and a compression
dressing applied.376
Rib
Rib can be used as graft or as a vascularized transfer. Rib
grafts are flexible, which allows for easy bending and wire
fixation. Also, rib integrates well to recipient bone. However,
ribs are fragile and do not allow for stable screw fixation.
Furthermore, the amount of cancellous bone is minimal and
a second harvest is implausible due to poorly regenerated rib.
When harvesting rib, short skin incisions should be used
and one rib should be skipped to preserve contour and stability
of the thoracic wall (Figs. 18.12, 18.13). If possible, the surgeon
should avoid subcutaneous dissection and transection of the
rectus muscle. Also, one must be cognizant of bleeding from
the intercostal vessels as well as air leaks. The harvest can be
performed with the patient in the supine or lateral position.
The skin incision should be located over the rib that is to be
harvested or over the skipped rib if two ribs are required. Dis­
section is carried down to the rib, and the periosteum is incised
longitudinally on the lateral aspect of the rib. Subperiosteal
elevation should then be performed up to the costochondral
junction. The length of the rib to be harvested is determined
based on the amount needed: a long segment from one rib is
preferred over a short segment from two ribs. If cartilage is
needed, it can be included in the harvest.
After harvest, the wound should be irrigated and examined
for an air leak and then closed in a layered fashion. This takes
place by suturing the edges of the periosteum and performing
a layered closure of the muscles, subcutaneous tissue, and
skin. A suction drain and local pain pump can be added if the
surgeon desires.
The associated donor site morbidities and risks include
pneumothorax, which can easily be prevented if careful and
proper technique is used. In addition, the chest wall might
become deformed if consecutive ribs are harvested. Post­
operative breathing difficulties secondary to pain can affect
postoperative recovery time. Overall the complication rate
is low but varies within the literature. Complications may
include pain, intraoperative bleeding and air leaks, secondary
bleeding, and rarely, chest wall deformities or scoliosis.382
Calvarium
Calvarial grafts are ectomesenchymal, with a rich diploic
vascular system that allows for rapid revascularization of the
haversian systems. This facilitates osteocyte survival, leading
to less resorption and increased structural maintenance.383
Calvarial grafts are ideal candidates for calvarial, midfacial,
nasal, and orbital reconstruction.384 They are obtained as split-
or full-thickness grafts. In young children, however, a split
graft may not be feasible because of a thin calvarium. Harvest
can occur via in situ splitting of the outer table or full-thickness
harvest and on-the-table splitting to increase the surface area
of graft available.385 It is important to use correct instruments

307
Clinical application of bone transfers
Fig. 18.13  Instruments for rib harvest. (A) Bone-
cutting forceps. (B) Stille–Listor angled bone-cutting
forceps. (C) Alexander–Farabeuf periosteotome.
(D, E) Doyen elevators. (F) Frykholm bone rongeur.
(G) Rib shears. (H) Tessier bone bender.
Fig. 18.14  Instruments for calvarial harvest.
(A) Obwegeser periosteal elevator, 6 mm.
(B) Obwegeser periosteal elevator, 12 mm.
(C, D) Farabeuf retractors. (E) Tessier bone bender.
(F) Digman bone-holding forceps. (G–K) Dautry–
Munro osteotomes, straight and curved. (L) Straight
osteotome. (M) Stille osteotome, 15 mm. (N) Mallet.
that are functional (Fig. 18.14). The surgeon must be aware of
the location of the sagittal and coronal sutures, as the corti­
cotomy should not occur less than 1 cm from the coronal
suture or less than 1.5 cm from the sagittal suture. It is addi­
tionally important to locate the perforating vessel posteriorly
near the sagittal suture.
The technique of calvarial bone graft harvest includes a
coronal incision that is made through scalp and galea. The
pericranium is then incised and elevated. Bone wax and/or
gelfoam can be used to halt bone bleeding. The landmarks
include the coronal suture, the sagittal suture (with its poste­
rior perforating vessels), the anterior temporal crest, and the

308
CHAPTER 18 • Repair and grafting of bone
intraosseous lateral vein. When taking a split in situ outer-
table graft, the design can be outlined and traced with an
oscillating saw followed by burring until diploe is seen. The
alternating straight and curved osteotomes are used to split
the calvarium. After the splitting of the outer table, further
strips of diploe can be harvested until the inner table is
encountered specifically at the thicker, posterior border near
the lambdoid suture. If during this process the dura becomes
exposed, it can be covered with bone chips. The pericranium
can be closed over Surgicel® and the scalp should be closed
in layers over suction drains.
There are several factors that make an intracranial approach
for calvarial bone graft harvest a better option. Specifically, an
intracranial approach should be undertaken in children with
a thin calvarium, when a large graft is needed, in a patient
with an irregular skull shape, or if, for graft design purposes,
the inner table is more feasible. Availability of a neurosurgeon
is recommended for this approach. Once the full-thickness
piece is obtained, it can be split on the back table. One should
not use a saw but rather an osteotome to reduce the waste
of bony substance and to reduce heat damage to the graft.
It is important to make sure that no cerebrospinal fluid
leak is present and that hemostasis is achieved. One split
segment is returned to the donor site; the other is used for
reconstruction.
Overall, complications are rare (0.25%). Intraoperative inci­
dents such as bone bleeding, dural bleeding, and dural lacera­
tion are usually controlled during surgery. Postoperative
complications include minimal pain and occasional paresthe­
sia, subscalp hematoma and calvarial irregularities, infection,
and rare brain injury with transient neurological sequelae.386
Fig. 18.15  The deep circumflex artery vascularized bone flap. Vascularized bone based on the deep circumflex iliac artery and vein (DCIA/DCIV) is harvested and can be
taken with an overlying skin paddle.
Deep circumflex
iliac artery and vein
External
oblique muscle
Ascending
branches
Internal oblique
muscle
Transverse
abdominis muscle
DCIA
DCIV
Vascularized bone flaps
Vascularized bone flaps avert the reparative phase of nonvas­
cularized grafts and do not depend on recipient bed vascular­
ity. Specifically in circumstances of prior radiation, extensive
trauma, or chronic scarring, vascularized bone flaps are
superior to nonvascularized bone grafts.387–390 The survival of
osteocytes in vascularized bone flaps leads to a decrease in
the remodeling process during revascularization as well as
the maintenance of bone mass and strength.391–394 Other
advantages include increased blood flow to compromised
recipient sites; utilization of composite tissue flaps, including
skin, muscle, and nerves; and the growth potential when
including a growth plate in the flap.395–400 An important indica­
tion for the use of vascularized bone flaps is segmental defects
of greater than 6–8 cm after traumatic or oncologic bone loss;
composite tissue loss; bony nonunion; avascular necrosis;
osteomyelitis; or biologic failure.401–404
Vascularized iliac transfer
The ilium, also a great source of graft material, has a robust
periosteal blood supply in addition to the nutrient artery.
Many studies have evaluated the best blood supply to the
ilium.405,406 The superficial circumflex iliac vessels are often
utilized in conjunction with a groin fasciocutaneous flap, but
the deep circumflex iliac vessels offer a better blood supply to
the bone (Fig. 18.15). Furthermore, a piece of vascularized iliac
crest can be harvested in conjunction with the anterolateral
thigh flap based off the lateral femoral circumflex system,
with the ascending branch giving a small vascular pedicle to

309
Clinical application of bone transfers
Fig. 18.16  The fibula free flap. Vascularized fibula is based on the peroneal artery and vein pedicle and can be raised with or without an overlying skin paddle.
Popliteal artery
and vein
Peroneal artery
and vein
Deep peroneal nerve,
anterior tibial artery and vein
Tibia
Extensor digitorum
and extensor hallucis
longus muscle
Peroneus brevis
and longus muscle
Flexor hallucis
longus muscle
Flexor digitorum
longus muscle
Soleus muscle
Gastrocnemius
muscle
Posterior tibial artery
and vein, tibial nerve
Posterior tibialis
muscle
Peroneal artery
and vein
Fibula with
skin paddle
Posterior tibial
artery and vein
Anterior tibial
artery and vein
Fibular
flap
the anterior superior iliac crest.407,408 Details of vascularized
bone transfers can be found elsewhere in this series.
Vascularized fibula
The free fibula transfer is utilized mainly in mandibular and
occasionally in midface reconstruction as well as for long-
bone defects or nonunions. The fibula provides a straight
piece of cortical bone up to 30 cm. The principal blood supply
is via the nutrient artery, a branch of the peroneal artery with
a diameter of 1.5–3 mm (Fig. 18.16). The fibula is also supplied
by periosteal branches from the peroneal artery proximal to
the nutrient vessel and musculoperiosteal branches distally.
Dissection is performed through an anterior or posterior
approach and bone can be transferred with adjacent muscles
(e.g., soleus, peroneal, flexor hallucis longus). Fasciocutaneous
perforators supply a skin paddle up to 10 × 20 cm.409 Though
the surgical steps are described elsewhere, it is important to
leave 6 cm of fibula distally to maintain a stable ankle joint.
Also, rigid fixation of the transferred bone at the recipient
site will improve healing. Complications include instability
of the ankle joint, persistent pain, neurological compromise,
and wound breakdown. A skin graft may be required if a skin
paddle was included. Various series have shown significant
complication rates, approaching 20% to 50%.410,411
A unique advantage of free vascularized fibula transfer is
the ability to include the physis within the flap based on the
anterior tibial artery and vein.412 This is particularly useful in
children with a malignant bone tumor located in an area that
requires extirpation of all or part of a long bone including the
native growth plate.
Vascularized scapula
The vascularized scapular bone flap can be combined with a
variety of possible composite flaps that include varying
amounts of muscle and skin. The scapula can be transferred
as vascularized bone based on the nutrient artery, which
enters inferior to the lateral attachments of the acromion to
the scapula plate. Chimeric free flaps containing vascularized
scapula, as well as rib and/or iliac crest, are particularly
useful for “hostile” defects, composite defects in the setting
of cigarette smoking, radiation, or scarring.413,414
The scapula bone flap can be harvested off the lateral
scapular border based on the circumflex scapular circulation
or off the medial scapular ridge also based on the circumflex
scapular circulation as well as on periosteal blood supply.415–417
The scapular bone can also be based on the angular branch of
the thoracodorsal system, which supplies the scapular wing.418
The mean pedicle length of the circumflex scapular system is

310
CHAPTER 18 • Repair and grafting of bone
Fig. 18.17  (A) The outer table of the parietal bone is
harvested with a curved osteotome after a trough is created
around the periphery of the bone flap. The blood supply of
the bone flap is maintained by preserving the periosteal
blood vessels in continuity with the superficial temporal
artery system within the galea superficial
musculoaponeurotic system and the deep temporal fascia.
(B) Sutures passed between the overlying galea,
periosteum, and calvaria are helpful in keeping the
periosteum attached to the bone and preserving the blood
supply.
A
B
7.5 cm. It can, however, be increased up to 13–15 cm by clip­
ping the circumflex scapula perforators, which allows the
bone to be supplied by the angular vessels, thus avoiding the
need for vein grafts.419
During flap dissection the teres major and minor muscles
need to be divided, which can cause shoulder dysfunction.
Another potential complication is scapular winging from divi­
sion of serratus anterior. Vascularized scapular bone might
offer an advantage over fibula or iliac crest in older patients in
whom ambulation is essential for postoperative recovery.420
Vascularized rib
The rib provides a curved, malleable piece of bone up to
30 cm long that receives its vascular supply through nutrient
and periosteal vessels. The nutrient vessel arises from the
posterior intercostal vessels and the periosteal blood supply
from various nearby sources (intercostal, internal mammary,
lateral thoracic, and thoracoacromial). The overlying muscle,
skin, and fat of each region can be transferred with the flap.421
A composite flap can be particularly helpful for reconstruction
of complex three-dimensional (3D) cranial defects.422,423 A
dissection that includes the nutrient vessel within the flap
requires thoracotomy, adding difficulty and associated risks.
Therefore, most vascularized rib transfers are based only on
the periosteal blood supply.
Vascularized calvarium
The vascularized calvarial flap is useful for unfavorable
recipient sites (irradiated or scarred) as well as in midface
reconstruction.424 An understanding of the anatomy of the
temporoparietal region is important to utilize this area for
reconstructions with vascularized bone. Calvarial bone
is covered with pericranium, overlying temporal muscle,
temporal fascia (deep and superficial layers), subgaleal fascia,
superficial temporal (temporoparietal) fascia, subcutaneous
tissue, and skin. Different flap types based off the superfi­
cial and deep temporal system can be useful to reconstruct
midface and other facial defects.424
A temporoparietal fascial flap can be harvested with pari­
etal bone vascularized by periosteal vessels from perforators
of the superficial temporal system (Fig. 18.17). A vascularized
calvarial flap can also be harvested with a fused segment of
the deep and superficial temporal fascial flap 2 cm above the
orbital rim.385 This flap is based on the middle temporal artery.
A third option is a temporalis myo-osseous flap, which trans­
fers the temporalis muscle and distal periosteum with the
underlying bone as one unit. This is based off the deep tem­
poral vessels.425 The parietal bone can be harvested as a partial
(outer table) or full-thickness flap while avoiding the sagittal
sinus (Fig. 18.18).
The surgical steps include a preauricular incision extend­
ing into the hairline, the elevation of flaps and dissection
down to temporoparietal fascia, and a decision of flap level
elevation and flap harvest with the appropriate pedicle, as
described above. The pedicle is dissected into the pericranium
of the desired calvarial bone. The bone flap design is based
on the defect and reconstructive needs of the patient. Multiple
strips can be harvested for complex three-dimensional defects.
Stable fixation of the cranial bone flap at the recipient site is
essential for reliable early bone integration.
Principles of bone transfer
Principles of bone grafting must be understood to optimize
outcomes. Knowledge of theses principles arises from three
components: individual experience, experience of others
(literature review), and knowledge acquired from basic
science research.426 There are general principles and specific