Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

Contents lists available at ScienceDirect

Journal of Cranio-Maxillo-Facial Surgery journal homepage: www.jcmfs.com

Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap Kang-Mi Pang a, 1, Sung Weon Choi b, 1, Soo-Hwan Byun a, Jin-Yong Lee c, Hun-Jong Jung d, Koo-Young Lim e, Soung-Min Kim a, Jong-Ho Lee a, * a

Department of Oral and Maxillofacial Surgery, Graduate School of Dentistry, Seoul National University, Seoul, Republic of Korea Oral Oncology Clinic, Research Institute & Hospital, National Cancer Centre, Goyang, Republic of Korea Department of Oral and Maxillofacial Surgery, Korea University Guro Hospital, Seoul, Republic of Korea d Department of Occupation and Environment, Konkuk Postgraduate Medical School, Choong-Ju, Republic of Korea e Department of Oral and Maxillofacial Surgery, Hepsiba Clinic, Seoul, Republic of Korea b c

a r t i c l e i n f o

a b s t r a c t

Article history: Paper received 7 August 2014 Accepted 14 April 2015 Available online xxx

Background: Nonvascularised autogenous costochondral rib grafts are the gold standard for replacement of the mandibular ramus and condyle. However, condylar defects present a difficult condition to treat when soft tissue defects are involved. Thus, we used vascularised costochondral grafts (VCGs) with a cartilaginous cap based on the serratus anterior muscle flap to reconstruct these composite defects. The purpose of this study was to evaluate the advantages and effectiveness of VCGs based on long-term observation. Methods: We evaluated 15 patients who underwent mandibular condyle and ramus reconstruction using VCG after a mean follow-up of 75.9 months (range 46e156 months). Our 15 case of mandibular reconstruction with a serratus anterior/rib composite free flap due to congenital or acquired defects involved a total of 18 condyles (bilateral reconstruction in 3 cases and unilateral reconstruction in 12 cases). Results: Our success rate with the use of the serratus anterior/rib composite free flap was 100%, and there were no cases of resorption or malunion of the graft. The mean maximum mouth opening (MMO) at the last follow-up was 31.29 ± 7.56 mm (range 15e45 mm). Although two patients exhibited excessive growth of the graft, deficient growth of the graft was also found in our paediatric patients. Four patients who developed ankylosed TMJ during the follow-up period received additional gap arthroplasty. Conclusion: VCG based on a serratus anterior flap was an excellent treatment modality for patients with uni-or bilateral composite defects of the ramus and condyle, in which soft tissue and hard tissue, including both bone and cartilage, were necessary. © 2015 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights reserved.

Keywords: Vascularised costochondral graft Serratus anterior rib flap Ramus and condyle defect Condyle reconstruction

1. Introduction The temporomandibular joint (TMJ) has unique anatomical features, such as a meniscus and fibrocartilaginous tissue surrounding the joint, that are required for mandibular movement and stress distribution during mastication. Although nonvascularised

* Corresponding author. Department of Oral and Maxillofacial Surgery, Oral Cancer Centre, School of Dentistry, Seoul National University, 101 Daehak-ro, Jongno-gu, Seoul 110-749, Republic of Korea. Tel.: þ82 2 2072 2630; fax: þ82 2 766 4948. E-mail address: [email protected] (J.-H. Lee). 1 These authors contributed equally to this work.

costochondral grafts have been widely used for the reconstruction of condylar defects, microvascular free flaps are the preferred choice in cases involving a significant area of hard and soft tissue in a poor recipient bed resulting from multiple surgeries, infection, or irradiation. Therefore, vascularised tissues from the deep circumflex iliac artery (DCIA) or fibula osteocutaneous free flaps have been frequently used for reconstruction of the mandibular condyle and ramus. Several studies, however, have reported that these flaps are limited for reconstruction of composite defects for the following reasons: (1) they are too bulky for the TMJ area; (2) they cannot provide the fibrocartilaginous tissue that is a critical component of the TMJ; and (3) bilateral harvesting can lead to serious complications, particularly with DCIA flaps (Chen and Yan, 1983; Urken

http://dx.doi.org/10.1016/j.jcms.2015.04.014 1010-5182/© 2015 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

2

K.-M. Pang et al. / Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

et al., 1991). Therefore, vascular transfer of the costochondral graft based on the serratus anterior composite flap has been introduced as a method for composite tissue reconstruction of complex defects of the ramus and condyle (Guyot et al., 2004). A rib has previously been used as either a free or a pedicled composite flap for the reconstruction of defects around the mandible. Free vascularised rib transfers were harvested based anteriorly on the internal mammary vessels or posteriorly on the intercostal vessels. This type of approach has been described as difficult and time-consuming, and adaptation of the rib to the shape of the mandible using controlled fractures compromises the medullary blood supply (Ariyan, 1980). The ribs also receive a substantial part of their blood supply through the periosteum, which has been exploited in free flaps based anteriorly on the internal mammary vessels as well as the overlying muscles, including the latissimus dorsi and serratus anterior. (Harii et al., 1982; Maruyama et al., 1985; Richards et al., 1985; Takayanagi et al., 1988; Penfold et al., 1992; Netscher et al., 1998). As anatomic studies and clinical techniques have improved, however, many clinicians have selected the costochondral graft with the serratus anterior/rib composite flap as the best choice for reconstruction of mandibular defects due to several positive factors: a high graft survival rate, less resorption, and an adequate soft tissue volume for preserving facial contours through dead space filling (Kim and Blackwell, 2007). Although Richards et al. (1985) and Netscher et al. (1998) both used the serratus anterior/rib composite flap with the cartilaginous cap to reconstruct hemifacial microsomia and hemimandibular defects in head and neck cancer patients, respectively, long-term follow-up of their patients has not been reported. Thus, we analysed the long-term results in 15 patients since 1996, and the results of our study demonstrate the advantages and effectiveness of vascularised costochondral grafts (VCGs) based on the serratus anterior muscle flap for composite defects of the ramus and condyle. 2. Material and methods Among patients who underwent condylar reconstruction using vascularised costochondral serratus anterior composite flaps in the Department of Oral and Maxillofacial Surgery, Seoul National University Dental Hospital (SNUDH), from January 1996 to December 2009, 15 patients (male-to-female ratio 6:9) were eligible for this retrospective study, which was approved by the institutional review board of the SNUDH. The mean age of the patients was 27.3 years (range 4e48 years), and the mean follow-up period was 75.9 months (range 46e156 months). The patient characteristics

(Table 1), intraoperative findings, and follow-up results were collected from the patients' medical records of admission, surgery, and biopsy. Patients with a maximal mouth opening (MMO) of less than 15 mm were examined using computed tomography to differentiate between bony and fibrous ankylosis. Overall, we evaluated the viability and morphologic changes of the flap, TMJ function based on the measurement of MMO (maximal interincisal distance), and donor site morbidity. 2.1. Surgical procedure The vascularised costochondral serratus anterior flap was elevated in the supine position with the patient's arm abducted. The sixth or seventh rib was marked on the ipsilateral chest wall, and a curvilinear incision was made in the mid-axillary line (Fig. 1A). Skin and subcutaneous tissue were dissected until the anterior margin of the latissimus dorsi muscle was encountered. After further retraction, the anterior border of the latissimus and the vascular branches of the thoracodorsal artery, which supply the serratus anterior muscle, were traced. The serratus muscle belly supplying the sixth or seventh rib was then identified. The rib, together with the overlying serratus anterior muscle, was freed from the surrounding tissues until the costochondral junction was reached (Fig. 1B). The intercostal muscles attached to the rib were transected with an electric knife bipolarly until the pleural surface was encountered, indicated by the visible moving lung. Once both the inferior and superior borders of the rib were exposed, the inner surface of rib was dissected from the pleura using a curved mini-periosteal elevator by attaching a small part of the periosteum to the pleura to prevent its tearing, and finally, the anterior and posterior margins of the rib were dissected with a bone cutter. Rib elevation was started at the costochondral area with an elevator and a knife (Fig. 1C). The thoracic nerve above the selected rib was also transected, whereas the motor innervation of the serratus anterior muscle was left intact. When bilateral condyles were to be reconstructed, VCGs from the fifth rib, nourished by the lateral thoracic artery, and the seventh rib, nourished by the thoracodorsal system, were separately harvested at the same incision site. Alternatively, bilateral approaches were used to harvest two VCGs from both sides of the sixth or seventh rib. After harvesting the flap, a “bubble test” was performed to exclude pleural damage. When pleural tearing was noticed, the area was repaired after evacuating the pneumothorax, using a Nelaton catheter placed under suction. After harvesting the rib, while maintaining its cartilaginous portion, a 4-hole miniplate and

Table 1 Demographic data of patients receiving vascularised costochondral grafts based on serratus anterior muscle for mandibular condylar and ramal reconstruction. Patient

Sex

Age

Aetiology

Joint involved

Pre-op status

Interpositioning

Additional treatment

Post-op MMOa

Follow-up (mo)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

F M M F F M M F F M F M F F F

45 26 19 47 48 18 40 11 32 47 6 10 13 4 43

Trauma Malignancy Malignancy Trauma Malignancy Trauma Trauma Benign tumour Trauma Benign tumour Malignancy Malignancy Trauma Hemifacial microsomia Benign tumour

Bilateral Left Right Left Left Bilateral Bilateral Right Left Left Right Left Right Right Right

Ankylosis, bird face

No No No No Meniscus preserved No No Temporalis fascia No Silastic sheet No No Silastic sheet Meniscus preserved Silastic sheet

Coronoidectomy Parotidectomy Parotidectomy No Parotidectomy No No Coronoidectomy Coronoidectomy No No No Orthognathic surgery No Parotidectomy

Ankylosis 44 Ankylosis 30 38 Ankylosis 31 Ankylosis 29 28 27 33 42 21 21

156 84 144 46 87 71 64 55 54 77 69 70 60 54 48

a

Condyle resorption Ankylosis, bird face Ankylosis, bird face Ankylosis, bird face Ankylosis

Ankylosis

MMO shows the mean width of the maximal interincisal mouth opening in millimetres.

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

K.-M. Pang et al. / Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

3

Fig. 1. Harvesting of free serratus anterior and rib myo-osseous flap. (A) In the supine position with the arm abducted, a sixth rib incision line along the mid-axillary line was drawn. (B) The rib, together with the overlying serratus muscle, was freed from the surrounding tissues. (C) Harvested free serratus anterior and rib myo-osseous flap with attached thoracodorsal pedicle. (D) Rapid prototyping model for patients with hemifacial microsomia. (E) For contouring of the mandible, a surgical template was prepared. (F) The harvested bony fragment was designed and contoured with a 4-hole miniplate and screws.

screws were used to design the harvest (Fig. 1E, F). The TMJ and the mandibular defect were prepared by endaural and submandibular approaches. The two incisions were tunnelled underneath the facial nerve, and the proper recipient vessels were prepared. Thereafter, the flap was contoured to fit the recipient's need by osteotomy while maintaining its cartilaginous cap at 2e3 mm in length with perichondrium and transferring it to the recipient site. The cartilaginous portion of the rib was placed into the mandibular fossa, fixed with 3e0 nonabsorbable suture (PDS®, Johnson & Johnson, UK), and screwed and/or plate fixed to the host mandible to maintain the correct position of the graft (Fig. 1E, F). Maxillomandibular fixation was also used to stabilise the graft in its correct position. The thoracodorsal vessels were usually anastomosed to the superior thyroid artery and to branches of the internal jugular vein. 2.2. Postoperative mouth opening exercise Maxillomandibular fixation (MMF) was placed postoperatively for approximately 21 days. After release of the MMF, gradual and gentle mouth movement was started. A soft diet and passive or active jaw-opening exercises were recommended for 1 month, depending on the patient's tolerance. Forcible mouth opening and a hard diet were prohibited for at least 6 weeks to prevent incomplete bony union. Advanced exercises, such as protrusive and excursive movements, were then allowed. Progression of MMO was measured using a ruler at each follow-up visit (Guzel et al., 2007). 3. Results Our 15 cases of mandibular reconstruction with the serratus anterior/rib composite free flap, due to congenital or acquired

defects, involved a total of 18 condyles (bilateral reconstruction in 3 cases and unilateral reconstruction in 12 cases). The clinical characteristics of these patients are summarised in Table 1. One case involved hemifacial microsomia, whereas the remaining 14 cases were categorised by acquired defects, such as infection, trauma, and malignant or benign tumours: three patients had condylar hypoplasia associated with early childhood trauma; two patients had condylar defects resulting from previous surgeries; and one patient had condylar necrosis due to osteomyelitis after a condylar fracture. Among the five patients with malignant tumours, three cases were osteosarcomas, and the others were metastatic follicular carcinoma and spindle sarcoma. The benign tumour cases included two recurrent ameloblastomas and one odontogenic keratocyst. There were three unilateral and three bilateral TMJ ankyloses with severe facial retrusion of the affected site. For the unilateral reconstructions, the contralateral sixth or seventh rib was harvested with the serratus anterior muscle. For bilateral reconstruction, one patient underwent a procedure on the contralateral fifth rib pedicled by lateral thoracic vessels and on the seventh rib by the thoracodorsal vessels on the same side. The other two bilateral reconstruction cases received the left sixth or seventh rib to right condyle and right sixth or seventh rib to left condyle procedures (Fig. 2). The sixth rib was used in 11 joints, the seventh rib was used in 6 joints, and the fifth rib was used in 1 joint. Usually, one or two slips of serratus anterior muscles were harvested together with rib bone and cartilage caps, depending on the size of the soft tissue defect. The latissimus dorsi muscle was combined with the same thoracodorsal vessels in four patients; however, to cover the facial soft tissue after extensive malignant tumour removal (Fig. 3A). Overall, our success rate with serratus anterior/rib composite free flap rate was 100%.

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

4

K.-M. Pang et al. / Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

Fig. 2. Patient with a previous condylectomy history for relieving a mouth opening limitation. The maximal mouth opening was 15 mm before the operation. (A) Preoperative panoramic view. (B) Immediate postoperative radiograph. Both condyles were reconstructed. (C) Postoperative panoramic radiographs at 8 years 6 months show the increased radiopacity of the grafted bone. (D) Clinical photographs 8 years 6 months postoperatively. (E) This patient has maintained a maximum interincisal mouth opening of 30 mm.

3.1. Intraoperative findings There were no cases in which we encountered vascular anatomic variations that prevented flap harvesting. An intact meniscus was found only in two cases, and the others showed no meniscus, due to previous pathology or surgery or its removal during tumour ablation and VCGs. The temporalis fascia in one case or a silastic sheet in three cases was placed to cover the bare bone on the mandibular fossa. The silastic sheets were removed later, usually after 3 months. To increase mouth opening in three TMJ ankylosis patients, unilateral or bilateral coronoidectomy was also performed, and, as a result, their MMOs exceeded 40 mm. Parotidectomy was performed in four patients to remove tumours and to preserve the facial nerve. In three patients, incidental tearing of the pleura was managed by chest tube insertion without sequellae of pneumo- or haemothorax.

3.2. Long-term outcomes of TMJ function and graft growth Because all of the grafts healed uneventfully, flap debridement was not required. There was one case of postoperative facial nerve palsy caused by traction and preparation of recipient site, in which the facial nerve function was fully restored after 3 months. One case showed shoulder winging after harvesting two VCGS from the fifth and seventh ribs on the same side. There was also one case of hypertrophic scarring at the axillary incision site. The mean MMO at the last follow-up was 31.29 ± 7.56 mm (range 15e45 mm), and four patients showed limited mouth opening of less than 15 mm

(Table 2). Although most of the patients were able to open their mouths properly, protrusive and excursive movements remained limited (Fig. 4).

3.3. Clinical characteristics of patients with limited mouth opening Among the four patients with limited mouth opening of less than 15 mm at the last follow-up after VCG, two were diagnosed with bony ankylosis and the other two with fibrous ankylosis. The interval between VCG and another surgery to correct the limited mouth opening ranged from 10 to 94 months. Two patients underwent bilateral reconstructive surgery, and the other two patients underwent unilateral surgery. Three of these patients were less than 20 years of age. To improve the limited mouth opening, gap arthroplasty was performed in all four patients using an interpositioning material, such as the temporal fascia or a silastic sheet. Patient 1 underwent bilateral gap arthroplasty with the temporal fascia and a silastic sheet at 94 months after VCG. Patient 3 underwent both gap arthroplasty with a silastic sheet and left coronoidectomy due to ankylosis on the right side as a sequela of the VCG for osteosarcoma. The 17-month postoperative MMO was 33 mm. In the case of patient 6, bilateral gap arthroplasty with the temporal fascia and coronoidectomy did not improve his mouth opening; therefore, additional gap arthroplasty with a silastic sheet was performed, and after 30 months, the MMO was maintained at greater than 20 mm. Similar to patient 6, patient 8 also underwent gap arthroplasty with the temporal fascia twice due to the development of ankylosis after

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

K.-M. Pang et al. / Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

5

Fig. 3. Six-year-old girl who had a spindle cell sarcoma in the right condyle area. After mass resection, the defects were reconstructed using vascularised costochondral serratus anterior combined with latissimus dorsi flap. (A) Harvested vascularised costochondral serratus anterior combined with latissimus dorsi flap. (B) Intraoperative photograph showing the defects after mass resection. (C) Rib bone was contoured according to the prefabricated template. (D) Grafting of the harvested combined flap. The facial nerve was also reconstructed using the harvested sural nerve. (E) Immediate postoperative radiograph. (F) One year and 6 months after graft shows a tendency towards condylar elongation. (G, H) Panoramic radiographs at postoperative 4 years 6 months and 6 years 4 months. Overgrowth of the graft is definite, and condylar elongation is 20% more than the unoperated hemimandible, with the operated side open bite and gradual facial asymmetry.

the first surgery. Immediately after the second operation, a bite block was placed in the patient's mouth for 3 days (perioperative MMO 40 mm). Subsequently, 2 h of active mouth exercise and 2 h of wearing the block during the day and throughout the night during sleep were suggested for 2 weeks. The bite block was then used only at night, and active exercise was encouraged for another 3 months. The patient's MMO at 35 months postoperatively was 22 mm (Table 3).

3.4. Change in graft volume and growth Among our five growing patients who received VCG at the ages of 4e13 years, two patients showed overgrowth of the vascularised graft. Using panoramic radiography, we found that the elongation was 20% more than on the nonoperated side in a 6-year-old girl over a follow-up of 5 years 3 months. The CT of the other patient showed 14.5-mm elongation in the VCG site compared to 2.6 mm at

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

6

K.-M. Pang et al. / Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

Table 2 Summary of postoperative results. Vascularised costochondral graft Mean MMOa Flap survival rateb Mouth-opening limitationc Transient facial nerve palsy Chest tube insertion due to pleural tearing Shoulder winging Hypertrophic scar in the axilla

31.29 ± 7.56d 100% n¼4 n¼1 n¼3 n¼1 n¼1

a MMO shows the mean width of the maximal interincisal mouth opening in millimetres. b Survival rate examined, limited to the graft used for mandibular condyle reconstruction. c Number of patients with a maximal mouth opening of less than 15 mm. d Data are the means of maximal interincisal mouth opening, excluding the data from patients with limited mouth opening.

the unoperated site over the follow-up period of 3 years 6 months (Fig. 3). A 4-year-old girl with hemifacial microsomia showed nearly balanced growth without any overgrowth of the VCG. In this patient, the volume of the graft increased at the horizontal and vertical ramus areas, maintaining good occlusion (Fig. 5). The other two patients showed a lack of growth of VCGs compared to their healthy sides. Of the two patients, one showed ankylosis during the follow-up period, and the other patient underwent postoperative irradiation of the graft site.

4. Discussion Most composite segmental mandibular defects are reconstructed today using fibular osteocutaneous free flaps or iliac crest free flaps (Wang et al., 2011; Hayden et al., 2012). However, costochondral grafts can be preferably selected for composite defects of the mandibular ramus and condyle because these flaps have superior structural and functional similarities to the native condyle

Fig. 4. Patient with right TMJ ankylosis due to trauma. The right condyle was reconstructed with vascularised costochondral serratus anterior composite flap and simultaneous orthognathic surgery. (A) Pre-operative panoramic radiograph shows hypoplasia of the right condyle and mandible deviation. (B) Immediate postoperative radiograph. (C) Radiograph at 2 years after surgery. Resorption and remodelling of grafted bone was observed. (D) Clinical photographs at postoperative 2 years. (E and F) Lateral excursion of the condyle. Lateral excursion to the ipsilateral side showed the limited range of motion.

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

22 mm/35 37 mm

21 mm/30

33 mm/17 41 mm

45 mm

28.5 mm/7 40 mm

Yes

MMO, mean maximal interincisal mouth opening; F, female; M, male.

No 3 mm 10 3 mm #8 (11/F)

Right

No No 14 mm 10 25 mm #6 (18/F)

Bilateral

No No 13 mm 32 18 mm #3 (19/M)

Right

24 mm #1 (45/F)

Bilateral

94

5 mm

No

No

After 7 Y 10 M Both gap arthroplasty, temporal fascia interpositioning and insertion of silastic sheet After 2 Y 8 M Gap arthroplasty with insertion of silastic sheet, left coronoidectomy After 10 M Both coronoidectomy After 3 Y 3 M Both gap arthroplasty, temporal fascia interpositioning After 4 Y 10 M Both gap arthroplasty with insertion of silastic sheet After 10 M Gap arthroplasty with insertion of silastic sheet After 2 Y 11 M Right condylectomy with insertion of silastic sheet, left condyle exploration

Post-operative MMO/follow-up period from last operation (mo) Meniscus remaining MMO immediately before first additional operation Elapsed time until additional operation (mo) Affected site Preoperative MMO Patient (age [y]/sex)

Table 3 Clinical characteristics and management of patients with postoperative ankylosis during follow-up.

Temporal fascia interpositioning

Additional operation

Perioperative MMO

K.-M. Pang et al. / Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

7

compared to the fibula or ilium. The nonvascularised costochondral graft was established as the “gold standard” in condyle reconstruction (Norman, 1978, 1982), and vascularised grafts could only be justified when there was a deficiency of the soft tissue of either congenital or acquired aetiology. The vascularised serratus anterior rib flap offers many advantages over free vascularised rib transfers based anteriorly on the internal mammary vessels or posteriorly on intercostal vessels. The vascular anatomy of the thoracodorsal axis is consistent and provides a long vascular pedicle with large-diameter vessels and minimal donor site morbidity (Rowsell et al., 1984). The flap also adapts well to the simultaneous transfer of vascularised bone, cartilage, and muscle in the reconstruction of composite defects of the mandibular ramus and condyle. In addition, for the reconstruction of extensive soft tissue defects, this flap also provides additional soft-tissue bulk with the latissimus dorsi muscle because the serratus anterior muscle alone cannot provide sufficient volume (Kim and Blackwell, 2007). In our case series, in which 18 serratus anterior rib free flaps were harvested, there were no flap failures or wound complications related to hardware infections or non-union of grafted bone. In our study, the most common donor site complication was pleural tearing, which occurred in three patients and was managed by chest tubing without further pulmonary complications. To prevent pleural tearing, the rib was carefully harvested, leaving the posterior periosteum attached to the parietal pleura, as described in a previous report (Chang and Miller, 2001). Although the inner side of the rib was exposed without the covering periosteum, the periosteal vessels remained intact after harvesting, and the blood supply to and healing of the graft were not concerns. Additionally, the bare part of rib could be used for sawing for osteotomy and to secure the graft to the host mandible directly with bone contact. Another great advantage of vascularised costochondral grafts based on the serratus anterior muscle was that two VCGs were obtainable by a unilateral or bilateral approach. The serratus anterior has a dual blood supply from the lateral thoracic artery, which arises more proximally than the subscapular artery and enters the lateral surface of the serratus anterior of the fourth and fifth slips, while the serratus braches of the thoracodorsal artery enter the lower sixth to tenth slips of the muscle. This vascular anatomy allows for two VCGs to be harvested from the fifth rib, nourished by the lateral thoracic artery, and the seventh rib, nourished by the thoracodorsal system, at the same that the lateral chest incision is made when bilateral condyles are to be reconstructed. When 2 VCGS are harvested on the same side, there is an increased possibility of scapular winging. We experienced one case of scapular winging at a 2-VCGS donor site. Although the patient did not complain of any pain or discomfort, either anchoring the remaining muscle slip to the periosteum of the adjacent ribs or harvesting the lowest two or three muscle slips was recommended to minimise such complications (Derby et al., 1997). In our studies, four cases (26%) developed postoperative ankylosis during the follow-up period. Initially, we thought that the likelihood of re-ankylosis would be very low because the cartilage cap is located between the grafted rib and the glenoid fossa, although there is no intervening meniscus or soft tissue. Of our three cases of re-ankyloses, two were bilateral, and one was a unilateral reconstruction of previous TMJ ankyloses in which the posterior facial height was largely increased with 2 VCGs for the correction of bird face and a retruded mandible. In these cases, heavy compressive force was inevitable and could have caused degeneration of the cartilage cap. Additionally, incomplete removal of bone from the medial aspect might have been a cause of reankylosis. It is well known that removal of the medial aspect is difficult due to the difficulty in providing adequate protection

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

8

K.-M. Pang et al. / Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

Fig. 5. Four-year-old girl with right-sided hemifacial microsomia. The right condyle was reconstructed with a vascularised costochondral serratus anterior composite flap. (AeC) Preoperative panoramic radiograph and clinical photograph. (D) Surgical template was fabricated to reconstruct the defect. (E, F) Immediate postoperative radiograph. To allow for maxillary growth of the right side, the mandibular angle was displaced inferiorly, coupled with elevation of the anterior mandible to bring the incisors into articular contact with the maxillary mandible. A bite block was then used in right posterior teeth to allow for eruption of the natural dentition in a controlled manner. (G) The harvested flap was contoured with screws. (HeK) Panoramic radiograph and clinical photograph at postoperative 5 months. Maxillary growth was observed, and the occlusion was improved. (LeO). Panoramic radiograph and clinical photograph at post-operative 1 year and 5 months. (PeS) Panoramic radiograph and clinical photograph at post-operative 3 years 7 months. (TeW) Panoramic radiograph and clinical photograph at post-operative 4 years 7 months. As the patient had grown, balanced growth of the grafted bone was observed and the graft volume is increased (compare arrow head at E and T).

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

K.-M. Pang et al. / Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

medially for the maxillary artery as well as adequate visualisation (Norman, 1978). Furthermore, regarding interpositioning material, use of a rotated flap of the deep portion of the temporalis muscle, instead of a silastic sheet or temporal fascia, has been recommended (Holmlund et al., 2013). These experiences suggest that a temporalis flap as an interpositioning material to block excessive bone formation and sufficient bone removal in medial fossa are critical. The one remaining case of re-ankylosis of VCG was a resection of the condyle and ramus, including the meniscus, due to chondrosarcoma with primary reconstruction of the joint without substituting the meniscus. In this case, no pressure was loaded at the grafted joint, unlike in other re-ankylosis cases, and the association between the degree of growth of a graft and ankylosis was considered. Many studies have shown that overgrowth of the graft commonly leads to ankylosis (Politis et al., 1987; Perrott et al., 1994; Svensson and Adell, 1998). Thus, much attention has been directed towards the size of the cartilage on the graft, which is responsible for the osteogenic potential (Perrott et al., 1994; Kaban et al., 2009). According to Perrott et al. (1994), overgrowth of the costochondral graft was not observed in their patients when they harvested 2e4 mm of cartilage. Furthermore, Kaban et al. (2009) indicated that a cartilage cap length ranging from 1 to 2 mm was optimal. In our study, all of the patients were underwent reconstruction with a VCG, using approximately 3 mm of cartilage, and there were no cases of VCG overgrowth. The limitations in mouth opening due to ankyloses were relieved by additional surgery with gap arthroplasty and/or coronoidectomy with interpositional grafts in all of the re-ankylosis cases. The growth potential of costochondral grafts for condylar replacement has been documented in children. The growth of these grafts has been unpredictable, ranging from resorption to overgrowth (Ware and Brown, 1981; Ortiz-Monastario and Fuente del Campo, 1985; Munro et al., 1989; Mulliken et al., 1989). Fukuta et al. (1992) reported a case of a 2-year-old boy who showed overgrowth of the VCG; the elongation was 1.4 cm in the costochondral grafted hemimandible and 0.4 cm in the unoperated hemimandible over a follow-up of 2 years 9 months based on three-dimensional CT scans. In addition, subsequent observations with panoramic radiographs showed continuous uncontrolled overgrowth. Among our five growing patients who received VCGs, only two patients showed overgrowth of the vascularised graft. Over follow-up of 5 years and 3 months, the elongation in one patient was 20% greater on the nonoperated side, comparable with the case reported by Fukuta et al. However, a 4-year-old girl with hemifacial microsomia showed nearly balanced growth without any overgrowth of the VCG. Richards et al. (1985) also had three patients with hemifacial microsomia treated with the serratus anterior/rib composite flap. They stated that the growth of the “nonvascularised” costochondral struts of the condylar reconstruction was unpredictable and that perhaps vascularised transfer would allow more “normal” growth. Two children in our study showed deficient growth of the grafts. Our results were somewhat different from previous studies that showed that overgrowth was frequently noted in growing children who received costochondral grafts (Guyuron and Lasa, 1992; Svensson and Adell, 1998). Although we could not clearly explain the reasons for undergrowth in these cases, we could speculate that the growth deficiency of the grafts was due to inadequate or poor vascularity at the cartilage cap, because the serratus anterior muscle and intercostal muscles are attached only to the rib and the blood supply does not reach the cartilage, simulating the nonvascularised costochondral graft at the cartilage portion. Although it is expected that vascularised costochondral grafts might provide more predictable growth than nonvascularised free grafts, this potential advantage must still be demonstrated. Moreover, we should

9

consider that the use of nonvascularised costochondral grafts has been regarded as the gold standard since Norman's publications (Norman, 1978, 1982). However, the vascularised costochondral graft might provide a more predictable growth potential, thereby avoiding insufficient growth due to poor vascularity. 5. Conclusion The serratus/rib composite flap reliably provided vascularised bone, cartilage, and soft tissues for uni- and bilateral reconstruction of complex defects of the mandibular ramus and condyle. Although the TMJ function after reconstruction was within the normal limits in most cases, it showed a tendency towards ankylosis of the joint in 26% of cases. Special care should be taken to prevent re-ankylosis between the VCG and glenoid fossa. VCGs have growth potential; however, the growth predictability is not definite. Careful follow-up for abnormal growth of VCGs is necessary. Although the serratus/ rib composite flap provided improved mouth opening, facial contour and facial growth in our cases, this flap must be selected in specific cases, considering some of the complications with flap harvesting. Conflicts of interest The authors declare no conflicts of interest. Acknowledgement This study was supported by a grant from the Ministry for Trade, Industry, and Energy of the Republic of Korea (10047615). References Ariyan S: The viability of rib grafts transplanted with the periosteal blood supply. Plast Reconstr Surg 65: 140e151, 1980 Chang DW, Miller MJ: A subperiosteal approach to harvesting the free serratus anterior and rib myo-osseous composite flap. Plast Reconstr Surg 108: 1300e1304, 2001 Chen ZW, Yan W: The study and clinical application of the osteocutaneous flap of fibula. Microsurgery 4: 11e16, 1983 Derby LD, Bartlett SP, Low DW: Serratus anterior free-tissue transfer: harvestrelated morbidity in 34 consecutive cases and a review of the literature. J Reconstr Microsurg 13: 397e403, 1997 Fukuta K, Jackson IT, Topf JS: Facial lawn mower injury treated by a vascularized costochondral graft. J Oral Maxillofac Surg 50: 194e198, 1992 Guyot L, Richard O, Layoun W, Cheynet F, Bellot-Samson V, Chossegros C, et al: Long-term radiological findings following reconstruction of the condyle with fibular free flaps. J Craniomaxillofac Surg 32: 98e102, 2004 Guyuron B, Lasa Jr CI: Unpredictable growth pattern of costochondral graft. Plast Reconstr Surg 90: 880e886, 1992 discussion 887-889 Guzel MZ, Arslan H, Sarac M: Mandibular condyle reconstruction with inlay application of autogenous costochondral graft after condylectomy: Cerrahpasa's technique. J Oral Maxillofac Surg 65: 615e620, 2007 Harii K, Yamada A, Ishihara K, Miki Y, Itoh M: A free transfer of both latissimus dorsi and serratus anterior flaps with thoracodorsal vessel anastomoses. Plast Reconstr Surg 70: 620e629, 1982 Hayden RE, Mullin DP, Patel AK: Reconstruction of the segmental mandibular defect: current state of the art. Curr Opin Otolaryngol Head Neck Surg 20: 231e236, 2012 Holmlund A, Lund B, Weiner CK: Mandibular condylectomy with osteoarthrectomy with and without transfer of the temporalis muscle. Br J Oral Maxillofac Surg 51: 206e210, 2013 Kaban LB, Bouchard C, Troulis MJ: A protocol for management of temporomandibular joint ankylosis in children. J Oral Maxillofac Surg 67: 1966e1978, 2009 Kim PD, Blackwell KE: Latissimus-serratus-rib free flap for oromandibular and maxillary reconstruction. Arch Otolaryngol Head Neck Surg 133: 791e795, 2007 Maruyama Y, Urita Y, Ohnishi K: Rib-latissimus dorsi osteomyocutaneous flap in reconstruction of a mandibular defect. Br J Plast Surg 38: 234e237, 1985 Mulliken JB, Ferraro NF, Vento AR: A retrospective analysis of growth of the constructed condyle-ramus in children with hemifacial microsomia. Cleft Palate J 26: 312, 1989 Munro IR, Phillips JH, Griffin G: Growth after construction of the temporomandibular joint in children with hemifacial microsomia. Cleft Palate J 26: 303, 1989 Netscher D, Alford EL, Wigoda P, Cohen V: Free composite myo-osseous flap with serratus anterior and rib: Indications in head and neck reconstruction. Head Neck 20: 106e112, 1998

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

10

K.-M. Pang et al. / Journal of Cranio-Maxillo-Facial Surgery xxx (2015) 1e10

Norman JE: Ankylosis of the temporomandibular joint. Aust Dent J 23: 56e66, 1978 Norman JE: Post-traumatic disorders of the jaw joint. Ann R Coll Surg Engl 64: 29e36, 1982 Ortiz-Monastario F, Fuente del Campo A: Early skeletal correction of hemifacial microsomia. In: Caronni E (ed.), Craniofacial surgery. Boston: Little Brown & Co., 401, 1985 Penfold CN, Davies HT, Cole RP, Evans BT, Hobby JA: Combined latissimus dorsiserratus anterior/rib composite free flap in mandibular reconstruction. Int J Oral Maxillofac Surg 21: 92e96, 1992 Perrott DH, Umeda H, Kaban LB: Costochondral graft construction/reconstruction of the ramus/condyle unit: long-term follow-up. Int J Oral Maxillofac Surg 23: 321e328, 1994 Politis C, Fossion E, Bossuyt M: The use of costochondral grafts in arthroplasty of the temporomandibular joint. J Craniomaxillofac Surg 15: 345e354, 1987 Richards MA, Poole MD, Godfrey AM: The serratus anterior/rib composite flap in mandibular reconstruction. Br J Plast Surg 38: 466e477, 1985

Rowsell AR, Davies DM, Eisenberg N, Taylor GI: The anatomy of the subscapularthoracodorsal arterial system: study of 100 cadaver dissections. Br J Plast Surg 37: 574e576, 1984 Svensson B, Adell R: Costochondral grafts to replace mandibular condyles in juvenile chronic arthritis patients: long-term effects on facial growth. J Craniomaxillofac Surg 26: 275e285, 1998 Takayanagi S, Ohtsuka M, Tsukie T: Use of the latissimus dorsi and the serratus anterior muscles as a combined flap. Ann Plast Surg 20: 333e339, 1988 Urken ML, Weinberg H, Vickery C, Buchbinder D, Lawson W, Biller HF: The internal oblique-iliac crest free flap in composite defects of the oral cavity involving bone, skin, and mucosa. Laryngoscope 101: 257e270, 1991 Wang KH, Inman JC, Hayden RE: Modern concepts in mandibular reconstruction in oral and oropharyngeal cancer. Curr Opin Otolaryngol Head Neck Surg 19: 119e124, 2011 Ware WH, Brown SL: Growth center transplantation to replace mandibular condyle. J Maxillofac Surg 9: 50, 1981

Please cite this article in press as: Pang K-M, et al., Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap, Journal of Cranio-Maxillo-Facial Surgery (2015), http://dx.doi.org/10.1016/j.jcms.2015.04.014

Mandibular condylar-ramal reconstruction using vascularised costochondral graft based on the serratus anterior composite flap.

Nonvascularised autogenous costochondral rib grafts are the gold standard for replacement of the mandibular ramus and condyle. However, condylar defec...
3MB Sizes 0 Downloads 8 Views