Figure 1. Absolute numbers of free flaps for oromandibular reconstruction used between May 2006 and May 2012. FFF indicates fibular free flap; LSBF, lateral scapular border flap; and STFF, scapular tip free flap.
Figure 2. Restoration of a long segment anterior mandible defect with a fibular free flap. A and B, Preoperative. C, Intraoperative defect. (The ruler is in inches [left scale] and centimeters [right scale].) D and E, Postoperative.
Figure 3. Use of the lateral scapular border flap in reconstruction of a complex soft-tissue defect combined with a significant bone defect. A, Defect. B, Flap. C, Bone inset. D, External skin paddle.
Figure 4. A typical case requiring short-segment bone with minimal mucosal defect reconstructed with scapular tip free flaps. A, Defect. B, Flap. (The ruler is in centimeters [top scale] and inches [bottom scale].) C, Postoperative.
Figure 5. Decision algorithm for osseous mandibular free flap reconstruction. This decision algorithm represents a guideline. The scapular tip free flap (STFF) is our preferred option for all patients with short-segment periangular bone defects. Patient age of 70 years is a guideline, and the overall comorbidities of each patient are considered. Other factors for consideration, such as pedicle length or revision surgery, are not represented within the algorithm. FFF indicates fibular free flap; LSBF, lateral scapular border flap.
Dowthwaite SA, Theurer J, Belzile M, Fung K, Franklin J, Nichols A, Yoo J. Comparison of Fibular and Scapular Osseous Free Flaps for Oromandibular ReconstructionA Patient-Centered Approach to Flap Selection. JAMA Otolaryngol Head Neck Surg. 2013;139(3):285-292. doi:10.1001/jamaoto.2013.1802
Author Affiliations: Department of Otolaryngology–Head and Neck Surgery, London Health Science Centre, Schulich School of Medicine & Dentistry, Western University, London, Ontario, Canada.
Importance Provides an approach to osseous free flap selection for reconstruction of segmental mandible defects that takes into consideration general medical status of the patient and reconstruction requirements; demonstrates the complementary qualities of fibular and subscapular system free flaps; and describes the different surgical indications for lateral border scapular and scapular tip free flaps.
Objectives To review our experience with osseous mandible reconstruction comparing the fibular and subscapular system free flaps, determine reconstruction-specific and general health variables that may differ between these patient groups, and present our approach to oromandibular reconstruction.
Design Retrospective study.
Setting Academic tertiary care medical center.
Participants A total of 110 patients (68 male, 42 female) undergoing single-stage oromandibular reconstructions with free-tissue transfers between May 1, 2006, and May 30, 2012.
Intervention Single-stage oromandibular reconstruction with free-tissue transfer.
Main Outcome Measures Differences in patient demographics, bone and soft-tissue aspects of the reconstruction, operative time, flap outcomes, and major postoperative complications between fibular, lateral scapular border, and scapular tip free flaps.
Results A total of 110 patients underwent 113 reconstructions, including 58 fibular free flaps (FFFs) (51.3%) and 55 subscapular system flaps (48.7%). Of the subscapular system free flaps, 27 flaps (49%) were scapular tip free flaps (STFFs) based on the angular artery branch of the thoracodorsal pedicle; the remaining 28 cases were lateral scapular border flaps (LSBFs). Patients undergoing reconstruction with FFFs were significantly younger than their subscapular system flap counterparts (56 vs 70 years, P < .001). Mean mandible defect lengths were similar for patients undergoing FFF and LSBF reconstruction (7.8 and 7.7 cm, respectively); STFFs were used to reconstruct significantly shorter defects (mean, 6.0 cm, P < .001). The FFFs were more commonly used for anterior mandible defects in which multiple osteotomies and limited soft tissue were required, while subscapular flaps were more commonly used for linear mandible defects with complex soft-tissue requirements. A single complete flap loss occurred in a patient who underwent reconstruction with an STFF; other complication rates were similar between groups.
Conclusions and Relevance The FFFs and subscapular flaps are complementary options for oromandibular reconstruction. The FFF is ideal for younger patients, extended defects, multiple osteotomies, and limited soft-tissue requirements. The subscapular system free flaps (LSBF and STFF) are excellent options for (1) elderly patients; (2) those with significant comorbidities, such as peripheral vascular disease; and (3) mandible defects associated with complex soft-tissue requirements. Furthermore, the STFF offers a reliable option to reconstruct short-segment defects, in particular, defects involving the angle of the mandible.
With developments and refinements in free-tissue transfers, several options are now available for reconstructing segmental oromandibular defects. This has enabled ideal flap selection to optimize the functional and aesthetic outcomes for patients. In most cases, the particulars of the defect determine the type of flap chosen. Along with other advantages and disadvantages posed by the various options, donor site morbidity is an important consideration. Furthermore, an aging population poses additional challenges in flap selection and postoperative management.
Contemporary choices for osseous free flap reconstruction include the fibular free flap (FFF), subscapular system free flaps (including lateral scapular border flaps [LSBFs] and scapular tip free flaps [STFFs]), iliac crest free flap (ICFF), and radial forearm osseous free flap (ROFF). Over the past 2 decades, the FFF has become the most popular because of its anatomic reliability, ease of elevation, ability to use a 2-team approach, and acceptable donor site morbidity.1,2 Although the soft-tissue component may have limitations, most defects are readily reconstructed with the reliable fasciocutaneous paddle.3- 5 Most important, the FFF provides quality and length of vascularized bone that is unrivaled by other bone flaps. Although some surgeons prefer the ICFF or the ROFF, these options have become less popular for various reasons, including donor-site morbidity and limited bone stock (ROFF).6- 8
The subscapular system provides an excellent potential donor site for oromandibular reconstruction. The bone stock is sufficient for most segmental mandibular defects, and the bone may be harvested as a lateral border based on the circumflex scapular pedicle or as the scapular tip based on the angular branch of the thoracodorsal pedicle. Each of these subscapular options has unique advantages and disadvantages, but overall, the soft tissue abundance and versatility are unmatched by other composite flaps. An additional benefit is the reduced amount of vascular disease within the subscapular system, even in older patients.9,10 An important advantage in eliminating lower extremity surgery may be the allowance of early ambulation in the postoperative setting, which may be particularly important for elderly patients and those with significant comorbidities.11 One specific advantage of the STFF is pedicle lengths of up to 17 cm, which may be ideal in selected situations.12- 14
The objective of this study was to review 6 years of experience with osseous reconstruction of segmental mandibular defects using both FFF and subscapular system free flaps (LSBF and STFF), analyze the changing trends over time, and discuss our current philosophy of flap selection.
A retrospective analysis was performed on all patients undergoing mandible reconstruction with either FFF or subscapular system free-tissue transfer (LSBF and STFF) at the Victoria Hospital, London Health Sciences Centre, between May 1, 2006, and May 30, 2012.
Data relating to patient demographics, tumor staging, flap selection, flap-related complications, and general medical complications were collated. Specific information regarding defect length, number of osteotomies, use of vein grafts, soft-tissue paddle design and dimensions, revascularization, and operative times was also recorded. Group differences with respect to age, defect length, duration of operative procedure, and skin paddle size were examined using 1-way analysis of variance, and differences associated with sex and numbers of osteotomies were analyzed with nonparametric statistics.
A total of 113 flaps were constructed in 110 patients (68 male, 42 female) who underwent single-stage oromandibular reconstruction with free-tissue transfers. Three patients required 2 separate flaps at different times during the review period. The flaps consisted of 58 fibular (51.3%) and 55 subscapular system (48.7%) flaps. Of the subscapular system free flaps, 27 (49%) were STFFs based on the angular artery branch of the thoracodorsal pedicle, and the remaining 28 were harvested from the lateral scapular border (LSBF). The absolute and relative numbers of each of these free flaps during the study period are illustrated in Figure 1. Patient and free flap characteristics are summarized in Table 1.
The mean age across the entire cohort was 62 years (range, 16-90 years). Patients receiving either type of subscapular flap were significantly older than FFF patients (70 vs 56 years, P < .001). There was no significant age difference between the 2 subscapular system flap groups (Table 1). Mean mandible defect lengths were similar for patients undergoing FFF and LSBF reconstruction (7.8 and 7.7 cm, respectively); STFFs were used to reconstruct significantly shorter defects (mean, 6.0 cm, P < .001, Table 1).
Operative times (ie, procedure stop time minus procedure start time) did not differ significantly between the fibular and scapular groups, with a mean procedure length of 9 hours 19 minutes for FFFs compared with a combined mean operative time of 9 hours 47 minutes for subscapular system flaps (unpaired, 2-tailed t test, P = .08). When examined across the 3 groups, a small significant difference was detected: LSBF operative times (mean, 10 hours 19 minutes) tended to be longer than those of FFF and STFF (mean, 9 hours 19 minutes and 9 hours 18 minutes, respectively). There was no significant difference between the FFF and STFF groups, and none of the pairwise comparisons reached statistical significance (Table 1).
The FFF was used more often for anterior mandible defects, when multiple osteotomies were required (two-thirds of patients required osteotomies), and for limited soft-tissue requirements (Figure 2). Subscapular flaps were more commonly used for linear mandible defects with complex soft-tissue requirements (Figure 3). The higher rate of osteotomies observed with LSBF compared with STFF reflects the difference in mean length between these 2 flap options, with an osteotomy required in only 1 case in the STFF group. Two LSBF cases required vein grafts. One of these was in a patient who had received chemoradiation and was undergoing oromandibular reconstruction, and the other was in a complex maxillomandibular reconstruction requiring a double flap.
The need for soft-tissue paddles differed across the 3 types of flaps. In approximately half the STFF cases (14 of 27), no skin paddle was incorporated. The mucosal defect was either closed primarily or reconstructed with muscle that was allowed to remucosalize. In comparison, 11 of 58 FFF cases did not incorporate a skin paddle. Only 1 LSBF case was harvested as a bone-only flap. With respect to the size of the skin paddle, FFF cases incorporated significantly smaller skin paddles compared with those of the LSBF, with mean paddle areas of 48.6 cm2 and 107.0 cm2, respectively (P < .001) (Table 1).
A single complete flap loss occurred in a patient who received an STFF. Flap complications, including early vascular compromise, head and neck or donor-site hematomas, seromas requiring intervention, nonunion/malunion, and hardware exposure or removal are outlined within Table 2. Perioperative complications are also detailed for both cohorts. There were no perioperative mortalities in either group.
With several options available, many factors must be considered in selecting the optimal reconstruction for patients undergoing segmental oromandibular resection. When a free flap reconstruction is being planned, issues at the forefront of flap selection include bone and soft-tissue requirements. In addition, however, changing demographics with an aging population will present different challenges to the health care team. Factors such as donor-site morbidity, the general medical condition of the patient, and the need for early ambulation may be considered in the decision-making process.
For cases of oromandibular reconstruction, the FFF has gained and retained wide popularity because of the consistent anatomy, reliable soft-tissue paddle, and convenience of a 2-team approach. The most significant advantage of the FFF is the quality, length, and versatility of the bone. Several factors may preclude the use of the fibula as a first-line option, such as abnormal leg anatomy or evidence of vascular insufficiency. Furthermore, for patients with significant comorbidities, the desire for early ambulation may mitigate the use of flaps obtained from the legs. The present review demonstrates this principle in our institution, where similar defects were reconstructed differently based on general patient factors, with the mean age of the fibular patient almost 15 years younger than that of the subscapular cohort (Table 1). A retrospective analysis such as this is not able to determine whether flap selection made on the basis of general patient well-being improved the overall outcome. However, our current data demonstrate similar low rates of flap-related complications and general medical perioperative complications across both cohorts.
The location and length of the mandible defect significantly affected selection of the reconstructive method. The benefits of the FFF for extended defects and reconstructions requiring closing osteotomies have been well described.15- 18 However, when anatomic or patient factors preclude the use of the fibula, the LSBF is a reliable alternative. Furthermore, when concurrent extensive soft tissue is needed, the subscapular system is superior to the FFF.
For short-segment linear defects up to 8 cm, defects that involve the mandibular angle, and revision surgery, the STFF has been shown to provide reliable bone stock with excellent pedicle length19 (Figure 4A and B). The harvesting osteotomies in these cases can be tailored such that no excess bone is taken. The shoulder function following STFF has been demonstrated to be excellent.20
For defects in excess of 8 cm or when osteotomies are required, our preferred flaps are the FFF and LSBF, both representing reliable options for reconstructions up to 12 cm. In this regard, our series of LSBF was similar to the series by Louis et al,21 in which the maximal length measured 13 cm. The mean defect length of approximately 8 cm was similar in our series for both the FFF and LSBF. These differed significantly from defects reconstructed with STFF, in which the longest STFF used was 8 cm (Table 1). Other specific clinical indications in which the STFF is beneficial have been published.19
For defects longer than 12 cm, the fibula remains our first-line option. Descriptions of extended defects reconstructed by incorporating the lateral border and the scapular tip have been published.22,23 In our present series, we were fortunate that the fibula was an option in all patients with defects larger than 12 cm.
Our need for osteotomies with the different flaps is a reflection of our preference for flap use for the varying defect sites. In most cases, anterior defects were reconstructed with FFF. In contrast, short-segment linear defects were reconstructed with the subscapular system. During the past 5 years, there has been a gradual shift from the LSBF to the STFF for the short-segment defect. Especially for defects that involve the mandibular angle, the shape of the STFF bone harvest may mitigate the need for osteotomies. This is reflected in our present series, with only 1 of 27 STFFs requiring an osteotomy.
The likelihood of dental rehabilitation with enosseous implantation was not considered in our flap selection algorithm. It is well known that the FFF can accept implants. Although there is less experience with scapular bone, the LSBF has been shown8,24- 26 to compare favorably with other composite free flaps with regard to dental implantation in long-term follow-up studies. Various techniques have been described to improve the vertical height of both FFF and scapular bone. The fibula height can be enhanced by “double barreling,” and vertical distraction osteogenesis is also possible with both scapula and fibula.27 In addition, patients undergoing reconstruction with STFF can undergo dental implantation, as demonstrated by Clark et al.20 In our series, very few patients received dental implantation, but the reasons were unrelated to the type of flap.
The significant advantage of the subscapular system over the FFF is in the versatility of the soft-tissue paddles that can be harvested as part of a single free flap. This allows refinements in soft-tissue repair and may also negate the need for multiple free flaps in selected cases.28,29 Elegant and complex descriptions of multiple soft-tissue paddles associated with FFF have been reported.30 Nevertheless, the availability of considerable tissue bulk, broad surface area, and excellent mobility of the soft tissue relative to the bone makes the scapular chimeric flap the first-line choice for single-flap reconstruction of complex 3-dimensional oromandibular defects.
This present series demonstrated significantly larger mean skin paddle dimensions of the LSBF compared with the FFF skin flap (107.0 vs 48.6 cm2, respectively). Although it was apparent that significantly larger volumes of soft tissue were available with the subscapular system of flaps, the volume was not quantified. In addition to increased volume, the thoracodorsal division of the subscapular system allows the harvest of reliable soft-tissue components that can be positioned largely independent of the osseous component. With the use of the thoracodorsal pedicle in either LSBF or STFF, soft-tissue paddles usually can be designed as a local perforator-based myofascial or myocutaneous flap incorporating the latissimus dorsi or serratus anterior. In these instances, there is tremendous flexibility in the volume of soft tissue that can be harvested using these donor sites, with the donor site able to be primarily closed. The other option for soft-tissue design with the STFF is the incorporation of local myofascial components that insert onto the scapular tip—either the serratus anterior or the teres major muscles. In most cases in which STFF was used, the intraoral lining was reconstructed only with muscle, which was allowed to remucosalize.
The general medical status of the patient as well as careful assessment for vascular disease at potential donor sites is vital in selecting the appropriate flap for oromandibular reconstruction. Despite its popularity, the FFF may not always provide the best option for oromandibular reconstruction because of higher rates of peripheral vascular disease compared with the subscapular system. The presence of vascular disease within the residual posterior and anterior tibial pedicles may be a risk for inadequate blood supply to the leg and foot postoperatively and needs to be assessed either clinically or radiologically before surgery. Combined rates of anatomic anomalies and peripheral vascular disease precluding the safe harvest of the fibula flap have been reported as between 21% and 25%.31,32
In contrast, the subscapular system provides reliable composite free flap options with low rates of vascular disease, which makes it an excellent choice for elderly patients and those with evidence of peripheral vascular disease. Our current series demonstrates a significant difference in the age of patients undergoing reconstruction with FFF (mean, 56 years) compared with the subscapular system (LSBF, 68 years; STFF, 71 years).
An important benefit of the subscapular system over FFF is the ability to safely ambulate earlier following surgery, especially for elderly patients. This may reduce the risk of perioperative complications, including deep venous thrombosis and pneumonia. Our current series observed no significant difference in these complications between the different cohorts, even in the background of age differences.
This present study did not specifically measure factors relating to ambulation, such as time to begin ambulation, time to full weight bearing, or time to unassisted ambulation. However, the physiotherapy protocol and the patients' advancement of activities are dramatically different between patients undergoing FFF and scapular flaps. At our center, patients who undergo scapular flaps begin activity as tolerated, with no restrictions on weight bearing unless limited by preexisting conditions. In contrast, the affected leg of all FFF patients is splinted for 7 days, and the patients ambulate only with a gait aid, such as a standard roller walker and cast sandal. Most patients are able to bear weight approximately 25% to 50% in the first week if ambulating, and most patients are discharged with an appropriate gait aid that enables normal gait patterns, such as a standard walker, roller walker, or cane.
Length of surgery may be an important consideration, although free flap operations have been done with acceptable success rates in elderly patients.33- 35 In this data set, there were no significant differences in operative time between fibular and scapular flaps (9 hours 19 minutes vs 9 hours 47 minutes). There was, however, a small significant difference in the operative times of LSBF operations (10 hours 19 minutes), which lasted approximately 1 hour longer than FFF and STFF operations. Although analysis is beyond the scope of this review, the longer operations that incorporated the LSBF may be a consequence of more complex defects, especially with respect to soft-tissue reconstruction. Whether this small difference in longer operations was clinically important was not specifically addressed by this present review.
During the past 6 years, some interesting trends were noted. The rate of FFF has remained consistent, suggesting consistent selection criteria for FFF. In addition, the ratio of FFF to subscapular system flaps has remained consistent. However, there has been an important shift within the subscapular system of flaps. As more experience has been gained with the STFF, we have come to understand the distinct advantages of this flap as well as the clinical situations in which it is best applied. These include the elderly patient, short-segment linear defect, defects involving the mandibular angle, and revision surgery. Whereas in the past the LSBF may have been used, the shift has been away from LSBF in favor of the STFF. The STFF harvest has tended to be easier and quicker than the LSBF because of less muscular dissection and easier surgical approach. The STFF seems to cause less shoulder dysfunction than the LSBF, likely because of less disruption of the rotator cuff muscles. Although a direct comparison of shoulder function between LSBF and STFF was not done in this study, others have shown acceptable outcomes with near-normal range of motion following STFF.9,20 Shoulder functional assessment continues to be an area of active investigation at our center. At present, the LSBF is being used less frequently, most commonly for longer bone requirements or extensive soft-tissue defects where the FFF may be less desirable.
We have reported a single institution's evolving management paradigm for patients requiring oromandibular reconstruction, and our decision guidelines are presented in Figure 5. If the reconstruction is deemed to require a free-tissue transfer, our preference is to use a bone-containing free flap, even for the elderly patient. The FFF is our flap of choice in most young patients and in cases of extended bone defects of 12 cm or greater, anterior defects that require multiple osteotomies, limited soft-tissue requirements, and normal vascular anatomy. The subscapular system of flaps is our preference for elderly patients and patients with significant peripheral vascular disease, abnormal leg anatomy, or extensive soft-tissue requirements. In particular, the STFF is used for short-segment linear defects, mandibular angle defects, and when a long pedicle length is required.
Correspondence: John Yoo, MD, Department of Otolaryngology–Head and Neck Surgery, London Health Science Centre, Schulich School of Medicine and Dentistry, Western University, 800 Commissioners Rd E, Ste B3-433A, London, ON N6A 5W9, Canada (firstname.lastname@example.org).
Submitted for Publication: July 11, 2012; final revision received October 16, 2012; accepted November 13, 2012.
Author Contributions: Drs Dowthwaite, Belzile, and Yoo had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Dowthwaite, Fung, Franklin, and Yoo. Acquisition of data: Dowthwaite, Theurer, Fung, Franklin, and Yoo. Analysis and interpretation of data: Dowthwaite, Theurer, Belzile, Fung, Franklin, Nichols, and Yoo. Drafting of the manuscript: Dowthwaite, Theurer, and Yoo. Critical revision of the manuscript for important intellectual content: Dowthwaite, Theurer, Belzile, Fung, Franklin, Nichols, and Yoo. Statistical analysis: Dowthwaite, Theurer, and Yoo. Obtained funding: Yoo. Administrative, technical, and material support: Dowthwaite, Theurer, Fung, and Nichols. Study supervision: Fung, Franklin, and Yoo.
Conflict of Interest Disclosures: None reported.
Previous Presentation: This study was presented at the American Head and Neck Society Eighth International Conference on Head and Neck Cancer; July 23, 2012; Toronto, Ontario, Canada.