 |
|
The Scapular Osteofasciocutaneous Flap
A 12-Year Experience
Mark L. Urken, MD;
Andrew G. Bridger, MD;
Karen B. Zur, MD;
Eric M. Genden, MD
Arch Otolaryngol Head Neck Surg. 2001;127:862-869.
ABSTRACT
| |
Objective To elucidate the factors that play a role in the decision-making process
to use the scapular donor site, we have reviewed our 15-year experience with
57 clinical cases, to our knowledge the largest case series to date.
Design Retrospective, single-surgeon medical record review.
Patients and Methods Retrospective review of 57 consecutive cases (53 patients) involving
mandibular and maxillary reconstruction using bone-containing scapular free
flaps over a 15-year period. Composite flap composition as well as donor and
recipient site complications were recorded.
Results Forty-one reconstructions were performed for mandibular defects, 11
were performed for maxillary defects, and 5 for combined defects involving
the mandible and maxilla. Seven flaps were composed of 2 separate bone flaps
using the angular branch and the circumflex scapular artery. A total of 6
flaps were failures in 5 patients, giving an overall success rate of 89%.
Conclusions The subscapular system of flaps is a versatile donor site that offers
distinct advantages in the older patient population as well as in patients
with a preexisting gait disturbance. It is particularly advantageous in patients
requiring a large surface area of soft tissue to restore their defect.
INTRODUCTION
THE SCAPULAR osteofasciocutaneous flap is unique for its diversity of
available skin, bone, and muscle, as well as the mobility of the soft tissue
components relative to the bone. The complex vascular anatomy associated with
the axillary region has resulted in a variety of flaps, all of which are based
on a single vascular pedicle (Table 1).1 Baudet et al2 described
the first successful transfer of the lateral thoracic flap, one of the earliest
flaps harvested from the axillary region. Subsequently, dos Santos3 defined the subscapular vascular anatomy through a
series of cadaveric dissections and injections of dye, resulting in the clinical
application of free scapular osteocutaneous flap by Teot et al.4
It was not until Swartz et al5 published their
experience with 26 clinical cases that the scapular donor site became a popular
source of vascularized bone for head and neck reconstruction. Swartz et al
demonstrated the ability to reliably harvest vascularized bone, multiple cutaneous
paddles, and vascularized muscle, as a single composite free flap. Swartz
et al applied the scapular free flap to 5 maxillary reconstructions and 21
composite mandibular defects. The mobility of the soft tissue relative to
the bone facilitated the complex 3-dimensional reconstruction of a bilateral
maxillectomy defect, an orbital floor defect, and a hemipalate defect. For
the first time, complex composite defects could be restored in a single stage
using bone from the lateral scapular border. Similarly, Swartz et al demonstrated
the applicability of the scapula for the reconstruction of composite defects
of the mandible by using the latissimus dorsi muscle with a split-thickness
skin graft for an external cutaneous defect and the skin paddle for intraoral
reconstruction.
|
|
|
|
Table 1. Flaps Based on the Subscapular Artery and Vein
|
|
|
The unique ability to manipulate the vascularized bone graft relative
to the skin paddle was further enhanced in 1991 by Coleman and Sultan,6 who described an alternative blood supply to the tip
of the scapula. For the first time, Coleman and Sultan demonstrated that the
angular artery, a branch of either the thoracodorsal artery or the artery
to the serratus anterior muscle, enabled the harvest of 2 separate bone segments
based on 2 separate branches of the subscapular vascular system. Coleman and
Sultan found that the ability to harvest independently vascularized scapular
tip not only enhanced mandibular reconstruction but also proved highly useful
in the reconstruction of orbital and maxillary defects.
Shortly after the report of 26 cases of osseous reconstruction in 1986
by Swartz et al, there were several smaller series reported in the late 1980s
(Table 2). Aside from the experience
by Sullivan et al10 with 31 cases in 1990,
and Coleman and Sultan's report of 22 cases 1 year later, the scapular flap
waned in popularity, relegated as a secondary and even tertiary choice for
osseous reconstruction of the head and neck. The introduction of alternative
bone-containing free flap donor sites, such as the iliac crest and fibula,
is largely responsible for this decline. Extensive experience with the iliac
crest-internal oblique osteomusculocutaneous free flap by Urken and colleagues17-18 and application of the fibular osteocutaneous
free flap to head and neck reconstruction by Hidalgo and Rekow19-20
popularized 2 new sources of vascularized bone for head and neck reconstruction.
Unlike the scapular donor site, the vascular anatomy of the iliac crest and
fibula are consistent so that harvesting from these sites is straightforward.21-22 In addition, the iliac and the fibular
donor sites allow for a 2-team approach, thereby decreasing operative time.
As the fibula and iliac became increasingly popular in the late 1980s and
early 1990s, the shortcomings associated with the scapular donor site led
to a decrease in its popularity. This is reflected in the substantial decline,
over the past decade, in the number of reported series using the osteofasciocutaneous
scapular flap for maxillary and mandibular reconstruction (Table 2).
|
|
|
|
Table 2. Experience With the Osteofasciocutaneous Scapular Free Flap
|
|
|
We have continued to use the scapular donor site as an important tool
for reconstruction of the upper and lower jaws. In an effort to elucidate
the factors that play a role in the decision-making process to use this donor
site, we have reviewed our 15-year experience with 57 clinical cases, to our
knowledge the largest case series to date. We have retrospectively examined
patient-related factors and defect-related factors to help define the current
indications for the use of this flap in contemporary reconstruction of the
mandible and maxilla.
PATIENTS AND METHODS
A retrospective review was performed of 57 consecutive cases (53 patients)
of mandibular and maxillary reconstruction using bone-containing scapular
free flaps over a 15-year period. All of the cases reviewed were performed
at Mount Sinai Medical Center, Department of Otolaryngology–Head and
Neck Surgery, New York, NY, a tertiary referral center for otolaryngology.
Any patient who had undergone a reconstruction of the mandible, maxilla, or
both using a bone-containing scapular free flap was included in this study.
Medical records were reviewed for age, sex, pathologic diagnosis, preexisting
comorbidities, donor site and recipient site complications, and factors related
to donor site choice. Defect characteristics were reviewed, including the
extent of the soft tissue defect, the extent of the bony defect, and the location
of these defects.
REPORT OF CASES
CASE 1
A 74-year-old man had had a history of peripheral vascular disease and
biopsy-proven squamous cell carcinoma of the right mandibular alveolus. The
patient underwent a composite resection (Figure 1) of the right mandibular body and floor of mouth and bilateral
neck dissections. Because of the patient's history of significant peripheral
vascular disease, he was not a candidate for a fibula free flap. In an effort
to expedite postoperative ambulation, he was also not considered as a candidate
for an iliac crest free flap. Consequently, a scapular free flap was chosen
for the reconstruction. Because of the patient's body habitus and excessive
subcutaneous adipose tissue, a skin paddle was undesirable for the intraoral
reconstruction. Instead, the lateral border of the scapula was used to reconstruct
the mandibular defect and the intraoral reconstruction was achieved with the
dorsal thoracic fascia, based on the circumflex scapular artery and vein (Figure 2). Five weeks postoperatively, the
dorsal thoracic fascia has begun to mucosalize intraorally (Figure 3), and the patient has achieved an acceptable cosmetic result
(Figure 4). This case demonstrates
the option of incorporating the vascularized dorsal thoracic fascia for intraoral
reconstruction when the body habitus and skin paddle thickness are undesirable
for intraoral lining.
|
|
|
|
Figure 1. Patient 1. Composite resection
specimen.
|
|
|
|
|
|
|
Figure 2. Patient 1. Scapular osteofascial
composite flap composed of the lateral scapular border and the dorsal thoracic
fascia.
|
|
|
|
|
|
|
Figure 3. Patient 1. Intraoral view 5 weeks
postoperatively demonstrating the dorsal thoracic fascia serving to reline
the oral cavity.
|
|
|
|
|
|
|
Figure 4. Postoperative view of patient
1.
|
|
|
CASE 2
A 52-year-old man had had a history of an extensive basal cell carcinoma
of the left temporal region, involving the temporal skin and the ascending
ramus of the mandible (Figure 5).
The patient underwent a left lateral temporal bone resection, resection of
the left ascending ramus and condyle of the mandible, and extensive resection
of the temporal skin and deep soft tissues of the temporal region and the
facial nerve. An 8 x 15-cm parascapular skin flap was raised with a
vascularized bone segment for the restoration of the mandibular defect (Figure 6). The parascapular flap was deepithelialized
and used to restore form lost as a result of the infratemporal fossa resection.
A 20 x 8-cm myocutaneous latissimus dorsi flap was raised and used to
resurface an extensive cutaneous scalp defect. A segment of the thoracodorsal
nerve was used as a facial nerve graft. Osseointegrated implants were placed
in the remaining temporal bone for the placement of an auricular prosthesis.
|
|
|
|
Figure 5. Patient 2. Preoperative view demonstrating
extensive invasion of a basal cell carcinoma.
|
|
|
|
|
|
|
Figure 6. Patient 2. Scapular flap design
with 20 x 8-cm latissimus dorsi myocutaneous flap based on the thoracodorsal
artery, and 8 x 15-cm parascapular skin paddle based on the circumflex
scapular artery.
|
|
|
This case demonstrates the versatility of the subscapular system of
flaps. Only the scapular donor site allows for the harvest of a vascularized
bone segment and 2 separate large skin flaps. An extensive latissimus skin
paddle can be reliably harvested while the deepithelialized scapular flap
is used to augment the underlying soft tissue defect. Postoperatively, the
patient regained facial symmetry and contour as well as complete coverage
of the scalp defect (Figure 7).
|
|
|
|
Figure 7. Patient 2. Postoperative view
demonstrating contour (A) and coverage of the temporal defect (B).
|
|
|
CASE 3
A 72-year-old active female tennis player had a squamous cell carcinoma
of the right alveolar ridge involving the body and retromolar trigone of the
mandible. Because of this patient's active lifestyle and desire to continue
playing tennis, the scapular donor site was chosen. The patient is right-handed
so a left-sided scapular flap was used for the reconstruction of a right-sided
composite defect of the mandible and floor of mouth (Figure 8). The patient had an excellent aesthetic and functional
result (Figure 9), and by 10 weeks
postoperatively the patient was able to participate in tennis without a significant
residual donor site deficit (Figure 10).
|
|
|
|
Figure 8. Patient 3. Intraoperative view
demonstrating reconstruction of composite defect with scapular osteofasciocutaneous
free flap.
|
|
|
|
|
|
|
Figure 9. Patient 3. Postoperative view
demonstrating facial symmetry.
|
|
|
|
|
|
|
Figure 10. Patient 3. Ten weeks postoperatively,
the patient is able to resume playing tennis.
|
|
|
This case demonstrates that the scapular donor site offers several unique
advantages over the iliac crest and fibular donor sites. Because bone is not
harvested from the lower extremity, patients are able to ambulate earlier
in the postoperative course, decreasing the chance of a deep vein thrombosis
and pulmonary embolus.
CASE 4
A 67-year-old man who had a history significant for childhood poliomyelitis
currently suffers from postpoliomyelitis syndrome. He relies on 2 canes and
a body brace for ambulation. This patient was initially seen with mandibular
osteoradionecrosis and an exposed reconstruction plate (Figure 11), after being previously treated for squamous cell carcinoma
of the floor of mouth and mandible with a surgical resection and external
beam irradiation. Because of this patient's profound gait disturbance, he
was not considered a candidate for an iliac crest or fibular free flap reconstruction.
A scapular free flap was raised consisting of a vascularized bone segment
and a myocutaneous latissimus dorsi flap (Figure 12). The vascualrized bone segment was used to reconstruct
the mandibular defect while the latissimus dorsi muscle was placed into the
neck and over the mandibular reconstruction. The skin paddle was deepithelialized
and used to augment the soft tissue deficit.
|
|
|
|
Figure 11. Patient 4. Osteoradionecrosis
with exposure of reconstruction plate.
|
|
|
|
|
|
|
Figure 12. Patient 4. A, Parascapular flap
and latissimus dorsi free flap outlined on the patient's back prior to harvest.
The angular branch was included to enhance the vascular supply to the scapular
tip. B, Lateral scapular border and parascapular skin paddle based on the
circumflex scapular artery and latissimus dorsi muscle based on the thoracodorsal
artery.
|
|
|
This patient is presently 3 years postoperative, ambulating with 2 canes
and a body brace without difficulty (Figure
13). This case demonstrates that harvesting from the scapular donor
site did not destabilize the shoulder nor did it impede this patient's ability
to ambulate effectively. However, patients with a preexisting gait disturbance
may not compensate after harvest of a composite free flap from the lower extremity
or pelvic girdle.
|
|
|
|
Figure 13. Patient 4. Postoperative view
demonstrating successful mandibular reconstruction and closure of the external
skin defect.
|
|
|
RESULTS
From March 1, 1987, to April 30, 2000, 53 patients underwent 57 bone-containing
scapular free flap reconstructions. Patients ranged in age from 16 to 89 years
(mean age, 57 years). There were 24 female and 29 male patients with pathologic
diagnoses including squamous cell carcinoma (25 patients), osteoradiorecrosis
(11 patients), sarcoma (11 patients), salivary gland carcinoma (2 patients),
ameloblastoma (1 patient), fibroma (1 patient), hemangioma (1 patient), and
posttraumatic deformity (1 patient). Forty-one reconstructions were performed
for mandibular defects, 11 were performed for maxillary defects, and 5 were
performed for combined defects of the mandible and maxilla.
All 57 scapular free flaps consisted of the lateral border of the scapula,
and 7 flaps were composed of 2 separate bone flaps based on the circumflex
scapular arterial perforators and the angular artery. In 4 of these cases,
the scapular tip was used to reconstruct the palate, maxilla, or inferior
orbital floor. Latissimus dorsi muscle was harvested as part of a composite
flap in 26 cases. Three of these muscle flaps were reinnervated (thoracodorsal
nerve to facial nerve) as part of a facial reanimation procedure. Nine scapular
free flaps were harvested with 2 separate skin paddles (scapular and parascapular),
and 4 flaps were harvested as a vascularized bone graft without a skin paddle.
Osseointegrated dental implants were placed primarily in 6 patients and secondarily
in 4 patients.
There were 3 recipient site wound infections, which were treated successfully
with intravenous antibiotic agents and local wound care. There were 2 donor
site wound infections; one required intraoperative debridement and the second
was treated successfully with local wound care. Two of 57 free flaps required
postoperative debridement of the skin paddle; however the bone remained vascularized,
and there were 6 flap failures in 5 patients (overall flap success of 89%).
Three of the 5 patients who sustained flap failure had multiple medical problems
including peripheral vascular disease, documented coronary artery disease,
and chronic obstructive pulmonary disease.
COMMENT
The application of microvascular free tissue transfer to the reconstruction
of oromandibular defects has revolutionized the management of oral cancer.
The early phase of this era saw an explosion in the number and variety of
donor sites from which to harvest vascularized bone containing free flaps.
With the introduction of each new donor site, the enthusiasm curve for each
of these composite flaps started off at a feverish pitch. As experience with
each of these composite flaps grew, a more reasoned view of the advantages
and disadvantages became apparent and allowed surgeons to place these donor
sites in their appropriate position relative to available reconstructive options.
Multiple factors are at play in determining the utility of a particular composite
flap donor site (Table 3).
|
|
|
|
Table 3. Factors Affecting the Usefulness of a Composite Free Flap
Donor Site
|
|
|
Excluded from Table 3 is
the surgeon's facility and comfort with the dissection of that donor site,
as that is something that can readily be learned and mastered. As experience
with a particular composite flap grows, these factors become more evident
and the enthusiasm for a particular donor site can usually be reliably measured
by the number and frequency of articles published in the literature. Since
its introduction into the head and neck surgeon's reconstructive options,
the subscapular system has been recognized for the tremendous range of soft
tissue flaps that can be transferred and their 3-dimensional mobility relative
to the bone. It is safe to say that this donor site offers the most versatility
of any donor site. However, the inconvenience of patient positioning during
harvest has led most surgeons to use other donor sites far more frequently,
with a growing sense of abandonment of the subscapular system. It is for this
reason that we chose to review our series of oromandibular reconstructions
using this flap, which, to our knowledge, represents the largest series to
date published in the literature (Table
2).
It is our belief that there are certain clinical situations where this
donor site emerges as the flap of choice to restore oromandibular defects.
Some of these situations are patient related, while others are dictated by
the particular defect. The age range for patients afflicted with oral cancer
often requires that the head and neck surgeon render treatment to patients
in the seventh, eighth, and occasionally in the ninth decades of life. Patients
in this age group are more difficult to mobilize after surgery for all of
the reasons associated with the aging process, in addition to the comorbidities,
which often coexist with the disease in the oral cavity. It is our belief
that one of the primary goals in patients who undergo surgery in the latter
years of life should be to mobilize them quickly in the postoperative period
to help minimize further complications. Harvest of a composite flap from the
subscapular system permits a far more rapid overall rehabilitation of patients
than the harvest of a fibular or iliac flap. The same philosophy holds true
for the patient of any age who is seen with an existing gait disturbance in
whom the harvest of a composite flap from either the fibula or the ilium may
compromise a fragile balance that could prove extremely difficult to rehabilitate
to achieve the same level of preoperative ambulation.
Severe peripheral vascular disease identified on preoperative history
or on vascular imaging, or the presence of vascular anomalies of the lower
extremities, serves as a contraindication to fibular flap harvest and points
toward the subscapular system as a safer donor site. Although vascular anomalies
of the subscapular system are found in particular on the venous side,21 it is rare for the vessels of the subscapular system
to be adversely affected by atherosclerosis.
There are certain features of a particular oromandibular defect that
we believe lend themselves to the use of a subscapular composite flap. The
range of soft tissue types and the maneuverability relative to the bone make
it particularly conducive to the more complex defects of the oromandibular
region. The potential surface area of the soft tissue flaps, which can be
transferred, is far greater than any other available donor site. In heavyset
patients with more subcutaneous fat throughout their body, cutaneous flaps
are often too thick for use in the oral cavity. The subscapular system provides
an opportunity to harvest either the latissimus dorsi muscle or the thoracolumbar
fascia, which are both well vascularized and significantly thinner soft tissue
flaps in virtually all patients.2, 23
The angular branch arising from the thoracodorsal artery or the branch to
the serratus anterior muscle offers a unique opportunity to harvest a second
independent vascularized segment of bone from the scapular tip. First described
by Deraemaecker et al24 in 1988 and later applied
clinically by Coleman and Sultan,6 the ability
to harvest 2 segments of bone that are vascularized by independent branches
of the parent artery and vein is a truly unique feature of the subscapular
donor site. While there are rare instances that one would need 2 independent
segments of vascularized bone in oromandibular reconstruction, there are distinct
advantages of a second blood supply to the scapular tip. The creation of osteotomies
along the lateral scapular border is often necessary to appropriately contour
the scapular border to the shape of the missing segment of the mandible. The
vascular supply to the distal segments of that bone is derived from the muscle
and periosteal circulation, which must be left attached to the bone while
creating these osteotomies. In the original article by Swartz et al5 concerning application of the subscapular composite
flap to oromandibular reconstruction, they reported several postoperative
bone scans in which the distal segments of bone, following the creation of
contouring osteotomies, did not appear to be well vascularized. The incorporation
of the angular branch in situations where multiple osteotomies of the lateral
scapular border are required would help to ensure the vascular supply to the
entire neomandible. The length and caliber of the vascular pedicle of the
subscapular flap are usually suitable for most head and neck reconstructions.
However, there are clinical situations in which the recipient vessels in the
neck are deficient, and in these, transfer of the scapular tip, based on the
angular branch, offers a unique advantage for reconstruction. The significant
additional pedicle length afforded by transferring vascularized bone supplied
by the angular artery helped to avoid the use of vein grafts in one such patient
in this series. The transfer of an independent segment of vascularized bone
based on the angular artery also provides the maximum separation and freedom
of maneuverability of the scapular and parascapular soft tissue flaps relative
to the bone. There are select clinical situations in which enhanced flexibility
may be advantageous, in comparison with the usual more limited relationship
of the scapular and parascapular flaps to the lateral scapular border supplied
by the circumflex scapular artery and vein. In the more complex palatomaxillary
reconstructions, we have used 2 independent segments of bone, based on the
circumflex scapular and angular branches, to achieve a more accurate restoration
of this 3-dimensional anatomy. In these reconstructions, we have used the
scapular tip as a horizontal shelf of the palate and the lateral scapular
border to restore the vertical dimension of the maxillary buttress.25 In such situations, the complete freedom of maneuverability
of these 2 independent bone segments is helpful to achieve a 3-dimensional
anatomical restoration of the palatomaxillary defect.
In 1994, we conducted a study to evaluate the bone stock that is available
for placement of dental implants in the most common donor sites used to harvest
vascularized bone-containing free flaps.26
In that series we evaluated 28 cadaveric specimens and determined the mean
index of implantability, that is, an index derived from height, width, and
cross-sectional area associated with each donor site. The results of that
study revealed that the scapular donor site ranked second behind the iliac
crest for reliability of the bone stock that can be harvested and its suitability
for placement of dental implants. In the current series, there were few patients
undergoing oromandibular reconstruction with the scapular composite flap in
whom dental rehabilitation using dental implants was achieved. The decision
to use osteointegrated implants is a multifactorial one that involves not
only the availability of sufficient bone stock, but also the motivation of
the patients and the anticipated functional benefit to the patient with respect
to the ability to achieve functional mastication following what is frequently
multimodality cancer therapy. In addition, financial means to afford prosthetic
restoration is also a necessary factor that must be included in the decision-making
process. Many of the patients in this series were older or had more advanced
disease, which made them less suitable candidates for dental rehabilitation.
In analyzing the current series, it is believed that these factors were more
important in explaining the relative scarcity of patients with restored dental
implants compared with our larger series using the iliac and fibular donor
sites as well.
In this series, the free flap failure rate was 11%. Most failures occurred
early in our experience with the subscapular system of free flaps. Since gaining
experience with this donor site, our success rate has vastly improved.
CONCLUSIONS
We have reported the largest series of scapular composite free flaps
used in head and neck reconstruction. With the senior author's (M.L.U.) experience
in performing more than 300 composite free flaps for oromandibular reconstruction,
the current perspectives on the role of the subscapular system in contemporary
head and neck surgery are provided. Despite the inconvenience of patient positioning
during surgery, there are very defined indications in which the subscapular
flap remains the reconstructive method of choice to restore the mandibular
infrastructure in combination with the oral lining and for the overlying skin.
AUTHOR INFORMATION
Accepted for publication April 6, 2001.
Presented at the annual meeting of the American Head and Neck Society,
Fifth International Conference on Head and Neck Cancer, San Francisco, Calif,
August 1, 2000.
Corresponding author and reprints: Mark L. Urken, MD, Box 1189, Mount
Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029 (e-mail: mark.urken{at}mssm.edu).
From the Department of Otolaryngology–Head and Neck Surgery,
Mount Sinai Medical Center, New York, NY.
REFERENCES
1. Urken ML, Sullivan MJ, Biller HF. Free flaps, composite flaps. In: Atlas of Regional and Free Flaps for Head and
Neck Reconstruction. New York, NY: Raven Press Ltd; 1995:213.
2. Baudet J, Guimberteau JC, Nascimento E. Successful clinical transfer of two free thoraco-dorsal axillary flaps. Plast Reconstr Surg. 1976;58:680-688.
ISI
| PUBMED
3. dos Santos LF. The vascular anatomy and dissection of the free scapular flap. Plast Reconstr Surg. 1984;73:599-604.
ISI
| PUBMED
4. Teot L, Bosse JP, Moufarrege R, Papillon J, Beauregard G. The scapular crest pedicled bone graft. Int J Microsurg. 1981;3:257-262.
5. Swartz WM, Banis JC, Newton ED, Ramasastry SS, Jones NF, Acland R. The osteocutaneous scapular flap for mandibular and maxillary reconstruction. |
