Parascapular Free Flaps for Head and Neck Reconstruction B. Chandrasekhar, MD, John A. Lorant, MO, Jose J. Terz, MD, Duarte, California
We report our experience with single-stage, primary reconstruction of the head and neck in 29 consecutive patients using parascapular free flaps. The commonest indications were for craniofacial defects ( 9 ) , oropharyngeal soft tissue defects ( 1 0 ) , and combined mandibular and soft tissue losses ( 4 ) . Ablative surgery was performed for squamous carcinoma ( 2 2 ) , melanoma ( 2 ) , and malignant fibrous histiocytoma ( 2 ) . Seven patients died of recurrent disease during a 3 ~ year follow-up. Seven patients are alive with recurrence. Flap complications included total loss ( 2 ) due to unsalvageable microvascular thrombosis, wound breakdown with oropharyngeal fistula ( 2 ) , mandibular osteomyelitis ( 1 ), trismus ( 2 ) , neck contracture ( 1 ), and donor site wound dehiscence ( 1 ). The overall success of this reconstruction was 93%. Primary wound healing was the general rule with lower morbidity than with other reconstructive techniques. T h e flap is thin, pliable, and conforms well to three-dimensional defects. The lateral border of the scapula can be incorporated on the same vascular pedicle for single-stage mandibular reconstruction. No muscle is sacrificed, and the posterior donor defeet is an added advantage. The parascapular flap is our first choice for reconstruction of m a j o r defects in the head and neck.
M
icrovascular flap reconstruction of the head and neck has distinct advantages over other techniques, and its place in the reconstructive surgeon's armamentarium is as yet being defined. Muscle, musculocutaneous, osteocutaneous, and fascial flaps are being increasingly utilized for primary, single-stage reconstruction [1-6]. The parascapular flap is a thin, fasciocutaneous flap with the vascular supply based on the circumflex scapular vessels [7]. We have used this flap for a variety of defects following cancer resection and report our experience in 29 patients over the last 3 89 years. Fromthe Departmentsof Plasticand ReconstructiveSurgeryand General and OncologicSurgery, City of Hope National Medical Center, Duarte, California. Requests for reprints should be addressed to B. Chandrasekhar, MD, Departmentof Plastic and ReconstructiveSurgery,City of Hope National Medical Center, 1500East DuarteRoad, Duarte, California 91010. Presentedat the 36th Annual Meetingof the Societyof Head and Neck Surgeons,Washington,DC, May 19-22, 1990. 450
PATIENTS AND M E T H O D S We studied 29 consecutive patients with large cancers in the head and neck who underwent resection and immediate reconstruction with parascapular free flaps. Twenty-one men and 8 women were included, with ages ranging from 29 to 91 years (mean: 50.1 years). Resection was used as the method of initial tumor control in 18 patients, with 9 of these receiving postoperative radiation. Eleven patients had recurrent disease following surgery alone (4), radiation alone (2), or a combination of these two modalities (5). These patients underwent salvage resection followed by immediate flap reconstruction. Every attempt was made to obtain tumor-free margins before the reconstruction was started. Fasciocutaneous flaps were used for oropharyngeal defects (I0), craniofacial defects (9), facial soft tissue deficits (1), carotid coverage (2), scalp defect (1), and for mediastinal tracheostomy reconstruction (1). Four flaps were harvested along with the lateral border of the scapula (7 to 10 cm) for simultaneous primary mandibular reconstruction. Satisfactory recipient vessels were dissected out before flap harvest. This usually entailed isolation of suitable branches of the external carotid artery and the external jugular vein if available. We used a standard approach to harvesting the parascapular flaps. This dissection was carried out in a semilateral position with the arm extended forward. Care was taken to obtain a long vascular pedicle with large-caliber vessels for reliable microvascular anastomoses. Flap transfer, microvascular anastomoses, flap inset, and donor site closure were the usual sequential steps in the reconstruction. Microvascular anastomoses were carried out using the operating microscope and a 10• magnification. When the lateral border of the scapula was also harvested for simultaneous mandibular reconstruction, osteotomies and fixation of the contoured bone to the mandibular remnants were carried out before the microvascular anastomoses and flap reperfusion. Postoperative visual monitoring of the flaps was carfled out hourly for the first 3 days. Exploration of the microvascular anastomoses was carried out whenever a change in color or capillary refill was noted in the flaps. RESULTS A careful follow-up of all 29 patients was performed for at least 6 months or until their death from recurrent disease or other illnesses. Careful documentation of flap necrosis, suture line dehiscence, fistulas, exposure of vital structures, infection, and donor site morbidity allowed us to evaluate the reliability of this flap. Two patients died during the immediate postoperative period from internal carotid hemorrhage and their se-
THE AMERICAN JOURNAL OFSURGERY VOLUME160 OCTOBER 1990
PARASCAPULARFREEFLAPS
quelae. Seven patients died of recurrent disease during this study period. Seven patients are alive with recurrence locally (4) and at distant sites (3). Three other patients died of unrelated causes. Primary healing without any wound morbidity was achieved in 22 patients. Five patients returned to the operating room for re-exploration of the microvascular anastomoses with salvage achieved in three after revision of the microvascular anastomoses. One patient had infection around the microvascular anastomoses and his flap could not be salvaged. A second flap was lost due to delayed diagnosis of venous thrombosis. Other complications in the perioperative period associated with failure of flap reconstruction to achieve primary healing included wound separation and salivary fistula (2), wound infection (2), trismus (2), and mandibular osteomyelitis (1). Twenty-seven patients had primary closure of their donor sites. One patient had a dehiscence of this closure and underwent delayed skin grafting. Three others had primary skin grafting of large donor sites without complications. COMMENTS The vascular anatomy of the circumflex scapular vessels and their cutaneous distribution were first described by dos Santos [8]. Of the two possible cutaneous paddles, the scapular flap has a more transverse axis and is more difficult to close primarily than the parascapular flap. The parascapular flap is a thin fascioeutaneous flap and serves admirably to close a complex, three-dimensional defect in the oropharynx (Figure 1). Twisting and folding does not seem to affect the vascularity of the skin paddle because of the absence of bulk (Figure 2). It is generally hairless and therefore quite suitable for external facial coverage. The pedicle is usually 7 to 8 cm long, with vessels 2 to 3 mm in calibre. These are eminently suitable for microvascular anastomoses. Our dissatisfac-
~Subscapular a.
-~ I ;t'~'- Thorac:dorsal a.
~ C i r c u m f l e x
//~/~
$ Scapular a.
Osseous branches pular flap
Figure 1. Location, parascapular flap.
bloodsupply,and cutaneousaxis
of
the
tion with pectoralis and trapezius myocutaneous flaps led us to use the parascapular free flap for major head and neck reconstructions. Our series confirms the simplicity of flap harvest and the large size of skin paddles available for reconstruction (Figures 3 and 4). One flap loss was due to infection from an anatomically contiguous tracheal stoma, which constantly bathed the microvascular anastomoses. A more secure closure could have avoided this problem. A second flap loss resuited from delayed diagnosis of venous thrombosis of the vascular pedicle and an unsalvageable situation in the operating room. Better judgement and monitoring could have prevented these flap losses. In general, primary healing was the rule. Postoperative radiation therapy did not affect the reconstruction and was well tolerated. The biggest advantage was the lack of a bulky vascular pedi-
Figure 2. Glossopharyngeal tion showing the natural cor tongue and absence of bulk folding.
THE AMERICAN JOURNAL OFSURGERY VOLUME 160 OCTOBER 1990 451
CHANDRASEKHAR ET AL
Figure 3. A 25- X 16-cm parascapular flap reconstruction of the anterior and middle cranial fossae.
Figure 4. Same patient, early postoperative result, showing primary healing.
cle in the neck as seen in all myocutaneous flaps reconstructions. This resulted in superior aesthetics and avoided later revisional procedures. In this study, we believed that the frequency of complications was reduced because of better wound healing and introduction of richly vascularized tissue into the surgical defects. Osteocutaneous flaps were harvested for simultaneous primary mandibular reconstruction for bony defects up to 10 cm. The lateral border of the scapula with its osseous branches was dissected free along with the parascapular skin paddle for this purpose (Figure 1). We confirm the reliability of the blood supply of the osteotomised segments as reported by others [4]. Resection of full-thickness facial defects where large portions of the cutaneous coverage and buccal mucosa 452
were removed created a reconstructive challenge. We chose to harvest a combination of scapular and parascapular skin paddles based on the same vascular pedicle in such situations. Use of one of the paddles for lining and the other for cutaneous coverage allowed a more secure closure without bulk. The posterior donor defect was a particular advantage in women and could often be hidden by a bra strap. Donor site morbidity has been relatively low and well accepted by our patients. The major disadvantage is the patient positioning during flap harvest and donor site closure. Dissection of the vascular pedicle tends to be tedious because of numerous muscular and bony branches that need to be controlled. Our success with this flap is superior to our use of
THE AMERICAN JOURNAL OF SURGERY VOLUME160 OCTOBER1990
PARASCAPULARFREE FLAPS
myocutaneous flaps used for similar indications. The tethering effect of the vascular pedicle limits the reach of myocutaneous flaps and often their skin paddles have to be designed over areas of questionable vascularity to allow for rotation and inset without tension. Further, the bulky muscle pedicle in the neck can interfere with the tracheal stoma. Thin, fasciocutaneous free flaps contour well to tongue remnants and do not restrict tongue mobility, thus having a very positive effect on speech and swallowing (Figure 2). Microvascular experience and expertise are essential to the success of this type of reconstruction and are generally freely available today. More importantly, judgment is needed to properly evaluate recipient vessels and resort to other techniques of closure when they are deemed to be inadequate. Our biggest problem initially has been the use of poor-quality recipient vessels, resulting in microvascular thrombosis and flap failure. Protection of the microvascular anastomoses from oral or tracheal contamination is also vital in preventing thrombosis at the microvascular anastomoses.
REFERENCES 1. Pennington DG, Stern HS, Lee KK. Free flap reconstruction of large defects of the scalp and calvarium. Plast Reconstr Surg 1989; 83: 655-61. 2. Rosen IB, Manktelow RT, Zuker RM, Boyd B. Application of free osteocutaneous flaps in management of post-radiation recurrent oral cancer. Am J Surg 1985; 150: 474-9. 3. Schlenker JD, Robson MC, Parsons RW. Method and results of reconstruction with free flaps following resection of squamous cell carcinoma of the head and neck. Ann Plast Surg 1981; 6: 362-73. 4. Swartz WM, Banis JC, Newton ED, Ramasastry SS, Jones NF, Acland R. The osteocutaneous scapular flap for mandibular and maxillary reconstruction. Plast Reconstr Surg 1986; 77: 530-45. 5. Soutar DS, Scheker LR, Tanner NS, McGregor IA. The radial forearm flap: a versatile method of intraoral reconstruction. Br J Plast Surg 1983; 36: 1-8. 6. Zuker RM, Manktelow RT, Palmer JA, Rosen lB. Head and neck reconstruction following resection of carcinoma using microvascular free flaps. Surgery 1980; 88: 461-6. 7. Nassif TM, Vidal L, Bovet J, Baudet JL. The parascapular flap: a new cutaneous microsurgical free flap. Plast Reconstr Surg 1982; 69: 591-600. 8. dos Santos LF. The vascular anatomy and dissection of the free scapular flap. Plast Reconstr Surg 1984; 73: 599-604.
THE AMERICAN JOURNAL OF SURGERY VOLUME 160 OCTOBER 1990 453
We report our experience with single-stage, primary reconstruction of the head and neck in 29 consecutive patients using parascapular free flaps. The commonest indications were for craniofacial defects ( 9 ) , oropharyngeal soft tissue defects ( 1 0 ) , and combined mandibular and soft tissue losses ( 4 ) . Ablative surgery was performed for squamous carcinoma ( 2 2 ) , melanoma ( 2 ) , and malignant fibrous histiocytoma ( 2 ) . Seven patients died of recurrent disease during a 3 ~ year follow-up. Seven patients are alive with recurrence. Flap complications included total loss ( 2 ) due to unsalvageable microvascular thrombosis, wound breakdown with oropharyngeal fistula ( 2 ) , mandibular osteomyelitis ( 1 ), trismus ( 2 ) , neck contracture ( 1 ), and donor site wound dehiscence ( 1 ). The overall success of this reconstruction was 93%. Primary wound healing was the general rule with lower morbidity than with other reconstructive techniques. T h e flap is thin, pliable, and conforms well to three-dimensional defects. The lateral border of the scapula can be incorporated on the same vascular pedicle for single-stage mandibular reconstruction. No muscle is sacrificed, and the posterior donor defeet is an added advantage. The parascapular flap is our first choice for reconstruction of m a j o r defects in the head and neck.
M
icrovascular flap reconstruction of the head and neck has distinct advantages over other techniques, and its place in the reconstructive surgeon's armamentarium is as yet being defined. Muscle, musculocutaneous, osteocutaneous, and fascial flaps are being increasingly utilized for primary, single-stage reconstruction [1-6]. The parascapular flap is a thin, fasciocutaneous flap with the vascular supply based on the circumflex scapular vessels [7]. We have used this flap for a variety of defects following cancer resection and report our experience in 29 patients over the last 3 89 years. Fromthe Departmentsof Plasticand ReconstructiveSurgeryand General and OncologicSurgery, City of Hope National Medical Center, Duarte, California. Requests for reprints should be addressed to B. Chandrasekhar, MD, Departmentof Plastic and ReconstructiveSurgery,City of Hope National Medical Center, 1500East DuarteRoad, Duarte, California 91010. Presentedat the 36th Annual Meetingof the Societyof Head and Neck Surgeons,Washington,DC, May 19-22, 1990. 450
PATIENTS AND M E T H O D S We studied 29 consecutive patients with large cancers in the head and neck who underwent resection and immediate reconstruction with parascapular free flaps. Twenty-one men and 8 women were included, with ages ranging from 29 to 91 years (mean: 50.1 years). Resection was used as the method of initial tumor control in 18 patients, with 9 of these receiving postoperative radiation. Eleven patients had recurrent disease following surgery alone (4), radiation alone (2), or a combination of these two modalities (5). These patients underwent salvage resection followed by immediate flap reconstruction. Every attempt was made to obtain tumor-free margins before the reconstruction was started. Fasciocutaneous flaps were used for oropharyngeal defects (I0), craniofacial defects (9), facial soft tissue deficits (1), carotid coverage (2), scalp defect (1), and for mediastinal tracheostomy reconstruction (1). Four flaps were harvested along with the lateral border of the scapula (7 to 10 cm) for simultaneous primary mandibular reconstruction. Satisfactory recipient vessels were dissected out before flap harvest. This usually entailed isolation of suitable branches of the external carotid artery and the external jugular vein if available. We used a standard approach to harvesting the parascapular flaps. This dissection was carried out in a semilateral position with the arm extended forward. Care was taken to obtain a long vascular pedicle with large-caliber vessels for reliable microvascular anastomoses. Flap transfer, microvascular anastomoses, flap inset, and donor site closure were the usual sequential steps in the reconstruction. Microvascular anastomoses were carried out using the operating microscope and a 10• magnification. When the lateral border of the scapula was also harvested for simultaneous mandibular reconstruction, osteotomies and fixation of the contoured bone to the mandibular remnants were carried out before the microvascular anastomoses and flap reperfusion. Postoperative visual monitoring of the flaps was carfled out hourly for the first 3 days. Exploration of the microvascular anastomoses was carried out whenever a change in color or capillary refill was noted in the flaps. RESULTS A careful follow-up of all 29 patients was performed for at least 6 months or until their death from recurrent disease or other illnesses. Careful documentation of flap necrosis, suture line dehiscence, fistulas, exposure of vital structures, infection, and donor site morbidity allowed us to evaluate the reliability of this flap. Two patients died during the immediate postoperative period from internal carotid hemorrhage and their se-
THE AMERICAN JOURNAL OFSURGERY VOLUME160 OCTOBER 1990
PARASCAPULARFREEFLAPS
quelae. Seven patients died of recurrent disease during this study period. Seven patients are alive with recurrence locally (4) and at distant sites (3). Three other patients died of unrelated causes. Primary healing without any wound morbidity was achieved in 22 patients. Five patients returned to the operating room for re-exploration of the microvascular anastomoses with salvage achieved in three after revision of the microvascular anastomoses. One patient had infection around the microvascular anastomoses and his flap could not be salvaged. A second flap was lost due to delayed diagnosis of venous thrombosis. Other complications in the perioperative period associated with failure of flap reconstruction to achieve primary healing included wound separation and salivary fistula (2), wound infection (2), trismus (2), and mandibular osteomyelitis (1). Twenty-seven patients had primary closure of their donor sites. One patient had a dehiscence of this closure and underwent delayed skin grafting. Three others had primary skin grafting of large donor sites without complications. COMMENTS The vascular anatomy of the circumflex scapular vessels and their cutaneous distribution were first described by dos Santos [8]. Of the two possible cutaneous paddles, the scapular flap has a more transverse axis and is more difficult to close primarily than the parascapular flap. The parascapular flap is a thin fascioeutaneous flap and serves admirably to close a complex, three-dimensional defect in the oropharynx (Figure 1). Twisting and folding does not seem to affect the vascularity of the skin paddle because of the absence of bulk (Figure 2). It is generally hairless and therefore quite suitable for external facial coverage. The pedicle is usually 7 to 8 cm long, with vessels 2 to 3 mm in calibre. These are eminently suitable for microvascular anastomoses. Our dissatisfac-
~Subscapular a.
-~ I ;t'~'- Thorac:dorsal a.
~ C i r c u m f l e x
//~/~
$ Scapular a.
Osseous branches pular flap
Figure 1. Location, parascapular flap.
bloodsupply,and cutaneousaxis
of
the
tion with pectoralis and trapezius myocutaneous flaps led us to use the parascapular free flap for major head and neck reconstructions. Our series confirms the simplicity of flap harvest and the large size of skin paddles available for reconstruction (Figures 3 and 4). One flap loss was due to infection from an anatomically contiguous tracheal stoma, which constantly bathed the microvascular anastomoses. A more secure closure could have avoided this problem. A second flap loss resuited from delayed diagnosis of venous thrombosis of the vascular pedicle and an unsalvageable situation in the operating room. Better judgement and monitoring could have prevented these flap losses. In general, primary healing was the rule. Postoperative radiation therapy did not affect the reconstruction and was well tolerated. The biggest advantage was the lack of a bulky vascular pedi-
Figure 2. Glossopharyngeal tion showing the natural cor tongue and absence of bulk folding.
THE AMERICAN JOURNAL OFSURGERY VOLUME 160 OCTOBER 1990 451
CHANDRASEKHAR ET AL
Figure 3. A 25- X 16-cm parascapular flap reconstruction of the anterior and middle cranial fossae.
Figure 4. Same patient, early postoperative result, showing primary healing.
cle in the neck as seen in all myocutaneous flaps reconstructions. This resulted in superior aesthetics and avoided later revisional procedures. In this study, we believed that the frequency of complications was reduced because of better wound healing and introduction of richly vascularized tissue into the surgical defects. Osteocutaneous flaps were harvested for simultaneous primary mandibular reconstruction for bony defects up to 10 cm. The lateral border of the scapula with its osseous branches was dissected free along with the parascapular skin paddle for this purpose (Figure 1). We confirm the reliability of the blood supply of the osteotomised segments as reported by others [4]. Resection of full-thickness facial defects where large portions of the cutaneous coverage and buccal mucosa 452
were removed created a reconstructive challenge. We chose to harvest a combination of scapular and parascapular skin paddles based on the same vascular pedicle in such situations. Use of one of the paddles for lining and the other for cutaneous coverage allowed a more secure closure without bulk. The posterior donor defect was a particular advantage in women and could often be hidden by a bra strap. Donor site morbidity has been relatively low and well accepted by our patients. The major disadvantage is the patient positioning during flap harvest and donor site closure. Dissection of the vascular pedicle tends to be tedious because of numerous muscular and bony branches that need to be controlled. Our success with this flap is superior to our use of
THE AMERICAN JOURNAL OF SURGERY VOLUME160 OCTOBER1990
PARASCAPULARFREE FLAPS
myocutaneous flaps used for similar indications. The tethering effect of the vascular pedicle limits the reach of myocutaneous flaps and often their skin paddles have to be designed over areas of questionable vascularity to allow for rotation and inset without tension. Further, the bulky muscle pedicle in the neck can interfere with the tracheal stoma. Thin, fasciocutaneous free flaps contour well to tongue remnants and do not restrict tongue mobility, thus having a very positive effect on speech and swallowing (Figure 2). Microvascular experience and expertise are essential to the success of this type of reconstruction and are generally freely available today. More importantly, judgment is needed to properly evaluate recipient vessels and resort to other techniques of closure when they are deemed to be inadequate. Our biggest problem initially has been the use of poor-quality recipient vessels, resulting in microvascular thrombosis and flap failure. Protection of the microvascular anastomoses from oral or tracheal contamination is also vital in preventing thrombosis at the microvascular anastomoses.
REFERENCES 1. Pennington DG, Stern HS, Lee KK. Free flap reconstruction of large defects of the scalp and calvarium. Plast Reconstr Surg 1989; 83: 655-61. 2. Rosen IB, Manktelow RT, Zuker RM, Boyd B. Application of free osteocutaneous flaps in management of post-radiation recurrent oral cancer. Am J Surg 1985; 150: 474-9. 3. Schlenker JD, Robson MC, Parsons RW. Method and results of reconstruction with free flaps following resection of squamous cell carcinoma of the head and neck. Ann Plast Surg 1981; 6: 362-73. 4. Swartz WM, Banis JC, Newton ED, Ramasastry SS, Jones NF, Acland R. The osteocutaneous scapular flap for mandibular and maxillary reconstruction. Plast Reconstr Surg 1986; 77: 530-45. 5. Soutar DS, Scheker LR, Tanner NS, McGregor IA. The radial forearm flap: a versatile method of intraoral reconstruction. Br J Plast Surg 1983; 36: 1-8. 6. Zuker RM, Manktelow RT, Palmer JA, Rosen lB. Head and neck reconstruction following resection of carcinoma using microvascular free flaps. Surgery 1980; 88: 461-6. 7. Nassif TM, Vidal L, Bovet J, Baudet JL. The parascapular flap: a new cutaneous microsurgical free flap. Plast Reconstr Surg 1982; 69: 591-600. 8. dos Santos LF. The vascular anatomy and dissection of the free scapular flap. Plast Reconstr Surg 1984; 73: 599-604.
THE AMERICAN JOURNAL OF SURGERY VOLUME 160 OCTOBER 1990 453
