Case 1. A, Preoperative status. B, Parascapular osteocutaneous/teres major myofascial free flap. C, Inset with teres major muscle used to fill the cervical defect inferior to the neomandible and protect the vascular anastomoses and great vessels. D, E, and F, Postoperative result at 3 months.
Case 2. A, Preoperative status. B, Resection defect. C, Scapular/parascapular osteocutaneous/teres major myofascial free flap. D, Inset with teres major muscle used to enhance soft-tissue coverage of the mandibular reconstruction plate. E, Completed inset. F, Postoperative result at 5 weeks.
Case 3. A, Resection defect. B, Inset of parascapular osteocutaneous/teres major myofascial free flap with split-thickness skin graft. Teres major muscle was used to fill the facial defect and enhance the coverage of the reconstruction plate; the plate was covered with a split-thickness skin graft.
Tomlinson AR, Jameson MJ, Pagedar NA, Schoeff SS, Shearer AE, Boyd NH. Use of the Teres Major Muscle in Chimeric Subscapular System Free Flaps for Head and Neck Reconstruction. JAMA Otolaryngol Head Neck Surg. 2015;141(9):816-821. doi:10.1001/jamaoto.2015.1485
We present what we believe to be the first case series in which the teres major muscle is used as a free flap in head and neck reconstruction.
To describe our experience with the teres major muscle in free flap reconstruction of head and neck defects and to identify advantages of this approach.
Design, Setting, and Participants
A retrospective review was performed at 2 tertiary care centers between February 1, 2007, and June 30, 2012. Data analysis was conducted from July 31, 2014, through December 1, 2014.
Teres major muscle free flap for use in head and neck reconstruction.
Main Outcomes and Measures
Indications for use, complications, and outcomes including donor site morbidity.
The teres major free flap was used in 11 patients as a component of chimeric subscapular system free flaps for a variety of complex head and neck defects. The teres major muscle was used to fill soft-tissue defects of the neck, face, and nasal cavity; it provided substantial soft-tissue volume but was less bulky than the latissimus dorsi muscle. The teres major muscle was also used to provide protection for vascular anastomoses and/or great vessels and to enhance soft-tissue coverage of the mandibular reconstruction plate. In addition, the muscle was selected as a substrate for skin grafting where inadequate neck skin remained. Flap survival occurred in 10 of 11 flaps (91%). Two flaps (18%) demonstrated venous congestion that was managed successfully. Two patients (18%) developed minor recipient-site complications (submental fistula and infection with recurrent wound dehiscence and plate exposure). All donor sites healed well, with chronic, mild shoulder pain noted in 2 patients (18%) and no postoperative seromas observed in any patient.
Conclusions and Relevance
Addition of the teres major muscle to a subscapular system free flap is an option for reconstruction of a variety of complex head and neck defects, particularly when a moderate amount of soft tissue is required. In select cases, the teres major muscle may have advantages over the latissimus dorsi muscle.
The subscapular vascular system is the basis for a variety of free flaps in head and neck reconstruction. The system is notable for its flexibility since it provides a source of vascularized skin, muscle, bone, or any combination of these elements.1 When muscle is harvested with the flap, it is generally derived from the latissimus dorsi and/or serratus anterior muscles. These muscles are often large and bulky and may be unsuitable for repair of certain head and neck defects. The teres major muscle is readily accessible via surgical approaches to the subscapular system, has an axial blood supply, and is smaller than the latissimus dorsi and serratus anterior muscles. In addition, harvesting a fasciocutaneous or osteocutaneous scapular or parascapular flap requires division of the teres major muscle and its blood supply from the circumflex scapular vessels; thus, it is simple to harvest the teres major by simply maintaining the muscular branches from the circumflex scapular and making an additional muscular cut. Despite these potential advantages, there are few reports describing the teres major muscle in head and neck reconstruction.2,3 The present study describes what we believe to be the first case series in which the teres major muscle was harvested as part of a chimeric free flap of the subscapular system and used for reconstruction of various head and neck defects.
All free flap cases performed between February 1, 2007, and June 30, 2012, at the University of Iowa Hospitals and Clinics and the University of Virginia Health System were reviewed to identify patients who underwent reconstruction with use of a subscapular system free flap including the teres major muscle. The medical records of these patients were reviewed to obtain related clinical and outcome data. Data collection was approved by the institutional review boards of both institutions. Patients were not required to provide informed consent for data collection because the data were deidentified before extraction. Data analysis was conducted from July 31, 2014, through December 1, 2014.
Eleven patients were included in the series. A summary of patient and flap characteristics is presented in Table 1. The mean patient age was 60 years. Eight defects (73%) were due to malignant tumors, 1 defect (9%) was the result of trauma, and 2 defects (18%) were associated with osteoradionecrosis with chronic open wounds. All resections resulted in complex defects as summarized in Table 1. The teres major muscle was used principally to fill soft-tissue defects of the neck (8 [73%]), midface and nasal cavity (1 [9%]), and face and chin defects (2 [18%]). In 4 cases (36%) the teres major muscle was used to protect the vascular anastomoses and/or great vessels, and in 4 (36%) cases the muscle was used to enhance soft-tissue coverage of the mandibular reconstruction plate. The teres major muscle was used as a substrate for skin grafting in 5 cases (45%) in which inadequate neck skin remained. In 1 case (9%), it was used to support extremely thin cervical skin flaps. The arterial origin of the teres major pedicle was the circumflex scapular artery in 10 cases (91%) and the thoracodorsal artery in 1 case (9%).
Table 2 summarizes flap complication at the recipient and donor sites. There was 1 complete flap loss, yielding a flap survival rate of 91% (10 of 11 flaps). Two flaps (18%) demonstrated venous congestion: one of these (9%) responded to treatment of the skin paddle with medical leeches without requiring return to the operating room and the other (9%) required thrombectomy and venous reanastomosis with a good outcome. At the recipient site, complications occurred in 2 patients (18%) and included a submental fistula and a postoperative infection with recurrent wound dehiscence and plate exposure. All donor sites healed well, with chronic, mild shoulder pain noted in 2 patients (18%); no other donor site morbidity was identified. No postoperative seromas were observed at the donor sites and no long-term functional limitations of the arm or shoulder were described.
A 69-year-old woman (patient 9) with T4aN2cM0 squamous cell carcinoma of the left mandibular alveolus (Figure 1A) underwent partial mandibulectomy with resection of the floor of the mouth and buccal soft tissue. The resulting cervical skin flap was thin and the patient had severe lower extremity vascular disease. A parascapular osteocutaneous and teres major myofascial free flap was harvested (Figure 1B) and the muscle was used to fill the cervical defect inferior to the neomandible and protect the vascular anastomoses and great vessels (Figure 1C). There were no postoperative donor- or recipient-site complications. At 3 months after reconstruction, intraoral (Figure 1D), cheek (Figure 1E), and neck (Figure 1F) contours were good and the patient had unrestricted arm mobility.
A 72-year-old man (patient 11) with a history of a T2N0M0 squamous cell carcinoma of the anterior floor of mouth underwent resection and postoperative radiotherapy and subsequently developed osteoradionecrosis with a chronic draining mental fistula that persisted through superficial debridement, antibiotic therapy, and hyperbaric oxygen therapy. Subsequent partial thickness mandibulectomy and myocutaneous pectoralis major flap ultimately failed with pathologic fracture (Figure 2A), requiring segmental mandibulectomy and removal of extensive mental and cervical soft tissue (Figure 2B). Reconstruction was performed with a scapular and parascapular osteocutaneous free flap and associated teres major myofascial free flap (Figure 2C-E); the teres major muscle was used to enhance soft-tissue coverage of the mandibular reconstruction plate. Postoperatively, there were 2 sites of minor wound breakdown at the flap/mental skin interface (without fistulae) that healed slowly with local wound care (Figure 2F at 5 weeks after reconstruction). There were no other postoperative donor or recipient-site complications.
A 69-year-old man (patient 7) with a T4aN2cM0 squamous cell carcinoma of the right buccal mucosa involving the full thickness of the cheek, oral commissure, mandibular body, and floor of the mouth was scheduled for resection and reconstruction with a parascapular osteocutaneous/latissimus dorsi myofascial free flap. A segmental mandibulectomy with through-and-through cheek resection was performed (Figure 3A). As the parascapular flap was raised, the teres major muscle was noted to be a better size match for the external cheek defect than the latissimus dorsi muscle, and thus was harvested in lieu of the latissimus dorsi. The chimeric flap was pedicled on the circumflex scapular vessels without disrupting the subscapular or thoracodorsal vessels. The flap was inset without difficulty (Figure 3B). The teres major muscle was used to fill the facial defect and enhance the coverage of the reconstruction plate; it was covered with a split-thickness skin graft. Postoperatively, there were no donor or recipient-site complications, and the result was acceptable from a cosmetic and functional standpoint.
The teres major is 1 of 6 scapulohumeral muscles whose role is to stabilize the glenohumeral joint and allow for circumduction of the humerus. The teres major muscle originates from the dorsal surface of the scapular tip and inserts on the anteromedial aspect of the proximal humerus. The teres major promotes inward rotation, adduction, retroversion, and extension of the arm, making it functionally equivalent to the latissimus dorsi.4 The nerve to the teres major arises directly from the posterior cord of the brachial plexus or as a branch of the thoracodorsal nerve and enters the muscle proximally on its deep surface. Harvesting the teres major muscle in conjunction with a scapular or parascapular flap does not require additional incisions along the flank. Functionally, harvest of the latissimus dorsi has been noted1 to restrict occupational, household, and sporting activities. It is reasonable to expect that removing the smaller teres major will reduce the risk of these adverse outcomes and we did not note these functional restrictions in our patients.
The teres major blood supply generally arises from a branch of the lateral circumflex scapular artery and is drained by 2 venae comitantes that run parallel to the artery.3 In a few cases, the teres major blood supply arises from the thoracodorsal artery; therefore, in most people, if the flap can be pedicled on the circumflex scapular vessels without harvesting the subscapular vessels, the thoracodorsal artery can be preserved during teres major harvesting, thereby sparing the primary blood supply to the latissimus dorsi muscle.1,3,4 Our series includes a case (patient 8) in which the thoracodorsal and circumflex scapular arteries arose distinctly from the axillary artery. In this situation, using the latissimus dorsi as originally planned would have required an additional arterial anastomosis, but a chimeric flap was formed instead using the teres major muscle on a single arterial supply (circumflex scapular artery), simplifying the reconstructive process. The subscapular vessels, which must be harvested to include the latissimus dorsi muscle in a chimeric scapular osteocutaneous flap, can be excessively large for anastomosis with cervical vessels; this situation was noted in one case in our series (patient 11). Because the teres major was used rather than the latissimus dorsi, the subscapular artery was ligated and the proximal portion of the thoracodorsal artery was harvested in continuity with the circumflex scapular artery. The teres major was used in a retrograde fashion to provide a better size match and greater pedicle length within the neck.
The teres major muscle has been routinely used in pedicle flap reconstruction of the shoulder, breast, and chest wall.4- 9 Despite its reliable anatomy and excellent exposure in standard approaches to scapular flaps, there have been few reports2,3 of its use in head and neck reconstruction. In the present cohort, use of the teres major muscle was driven by pragmatism. In all cases, the need for muscle was anticipated and the preoperative plan was to harvest muscle as part of a chimeric subscapular flap. A common plan was to use the scapular skin paddle for internal mucosal lining and the muscle component to fill the associated external skin defect and provide a scaffold for skin grafting. In these cases, the teres major muscle was found intraoperatively to be a more appropriate size than the latissimus dorsi muscle for reconstruction of the head and neck defect, without the additional time and morbidity associated with harvesting the latissimus dorsi muscle. We noted no postoperative seromas at the donor site when harvesting the teres major muscle; however, the rates of seroma formation for latissimus dorsi muscle flaps were reported to be between 1% and 80% in a recent systematic review on those flaps.10 Thus, the teres major muscle may be considered a convenient alternative to the latissimus dorsi muscle in situations in which the latissimus dorsi is considered for reconstruction. In practice, choosing between these 2 muscles is largely based on intraoperative assessment of the defect size and the relative size of the 2 muscles. In some cases (2 cases in the present series), use of both muscles is advantageous.
The teres major muscle has a few potential drawbacks. It is unclear whether there are cutaneous perforators feeding the skin overlying the teres major muscle. Thus, unlike the latissimus dorsi, the teres major is not suitable for harvest with its own skin paddle. The teres major flap has a shorter pedicle length than the latissimus dorsi flap, which may limit its mobility within a defect to a greater extent. Prior studies3 have established the mean length of the teres major pedicle after arising from the circumflex scapular artery to be approximately 3 cm. In contrast, pedicle lengths of 8 to 16 cm have been reported for the latissimus dorsi,11 which allows for greater range within the reconstructive site, particularly if the entire latissimus muscle dorsi is harvested. However, harvesting the entire muscle often necessitates incorporating the secondary and tertiary angiosomes of the latissimus dorsi muscle, which may have ramifications for wound healing. Research is necessary to reliably characterize the vascular anatomy of the teres major blood supply to establish its mean pedicle length in many patients.
The present study illustrates the feasibility of incorporating the teres major muscle into chimeric subscapular system free flaps for reconstruction of complex head and neck defects. The teres major muscle is well suited for defects involving both oral mucosa and external skin when a moderate amount of soft tissue is required. Harvesting the teres major is a straightforward extension of the scapular or parascapular free flap harvest and may reduce morbidity relative to latissimus dorsi harvest. Thus, addition of the teres major muscle to a subscapular system free flap is a practical option for reconstruction of a variety of complex head and neck defects.
Corresponding Author: Nathan H. Boyd, MD, Division of Otolaryngology–Head and Neck Surgery, Department of Surgery, Mail Stop Code 10 5610, University of New Mexico Health Science Center, One University of New Mexico, Albuquerque, NM 87131 (firstname.lastname@example.org).
Submitted for Publication: December 12, 2014; final revision received April 6, 2015; accepted June 7, 2015.
Published Online: August 27, 2015. doi:10.1001/jamaoto.2015.1485.
Author Contributions: Drs Tomlinson and Boyd 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: Tomlinson, Jameson, Pagedar, Boyd.
Acquisition, analysis, or interpretation of data: Tomlinson, Jameson, Schoeff, Shearer, Boyd.
Drafting of the manuscript: Tomlinson, Shearer, Boyd.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Tomlinson.
Administrative, technical, or material support: Jameson.
Study supervision: Tomlinson, Jameson, Pagedar, Boyd.
Conflict of Interest Disclosures: None reported.
Additional Contributions: We thank the patients for granting permission to publish this information. We acknowledge Gerry F. Funk, MD, Department of Otolaryngology–Head and Neck Surgery, University of Iowa Hospitals and Clinics, for his supportive mentoring and his creativity and enthusiasm regarding the use of the teres major muscle as a free flap. We also acknowledge Lucy H. Karnell, PhD, for her help obtaining institutional review board approval at the University of Iowa Hospitals and Clinics. There was no financial compensation.