RECONSTRUCTION OF MAXILLARY DEFECT WITH MUSCULOADIPOSE RECTUS FREE FLAP TSU-HUI (HUBERT) LOW, M.B.B.S. (Hons),1,2* ANDREW LINDSAY, M.B.B.S.,1 JONATHAN CLARK, M.B.B.S., M.Biostat,2,3 FRANCIS CHAI, M.B.B.S.,1 and RICHARD LEWIS, M.B.B.S.1
Background: The rectus myocutaneous free flap (RMFF) is used for medium to large maxillectomy defects. However, in patients with central obesity the inset could be difficult due to the bulk from excessive layer of adipose tissue. We describe a modification of the RMFF for patients with excessive central obesity with a flap consisting of adipose tissue with minimal rectus muscle; the musculo-adipose rectus free flap (MARF). Methods: Five cases of MARF reconstruction were performed between 2003 and 2013, with patients’ body mass indexes ranging from 29.0 to 41.2 kg/m2. All patients had sinonasal tumor, of which three were adenoid cystic carcinoma, one squamous cell carcinoma, and one melanoma. Four patients had Codeiro IIIb defects and one had Codeiro II defect. Using the MARF technique, the maxillectomy defect was obliterated with vascularized adipose tissue overlying the rectus muscle and was trimmed to fit the maxillectomy defect. The adipose tissue was allowed to granulate and mucosalize. Results: The volume of adipose tissue harvested was between 120 and 160 mL. All flaps survived with no requirement for re-exploration. Complete oro-nasal separation was achieved in all patients. The time to commencement of oral intake ranges from 5 to 15 days. One patient developed seroma and one developed wound breakdown on the donor site. The length of stay at the hospital ranges from 9 to 22 days. On follow-up ranging 7.5–32.8 months, two patients died from their malignancies. The other three patients were able to tolerate oral soft diet. Conclusion: The MARF may be considered as an alternaC 2015 Wiley tive to myocutaneous rectus free flap particularly for the reconstruction of maxillary defects in patients with central obesity. V Periodicals, Inc. Microsurgery 00:000–000, 2015.
Reconstructing
large maxillectomy defects is complicated and challenging due to its three dimensional nature.1 Various reconstructive options exist including prosthetics, local flaps, and free vascularized flaps.2 The reconstructive goals for reconstructing a large maxillectomy defects include obtaining good wound healing with separation of oral, nasal, and cranial compartments; optimized functional outcomes in speech, swallow, and dental rehabilitation, as well as acceptable cosmesis.3 An ideal free flap for this region should satisfy these goals and have a long pedicle with the option of increasing soft tissue bulk when necessary; the flap should also have the ability to reconstruct multiple oral and nasal mucosal defects with a good bone stock for osseous framework reconstruction in patients where dental rehabilitation is planned.4 Therefore, all maxillary defects need to be approached individually.5 With improved healthcare, we are now seeing increasing numbers of geriatric patients, with multiple co-morbidities presenting with advanced head and neck 1 Otolaryngology Head and Neck Department, Royal Perth Hospital, Perth, WA 6000 2 Sydney Head and Neck Cancer Institute, Royal Prince Alfred Hospital, Camperdown, NSW 3 Central Clinical School, University of Sydney, Sydney, NSW Presentation: This work was presented at the ANZHNC, Melbourne, 2013 *Correspondence to: Tsu-Hui (Hubert) Low, M.B.B.S. (Hons), Suite 5, 35 Fourth Ave Eastwood NSW 2122. E-mail: [email protected] Received 25 November 2014; Revision accepted 29 May 2015; Accepted 3 June 2015 Published online 00 Month 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/micr.22439
Ó 2015 Wiley Periodicals, Inc.
tumors.6,7 Due to concerns over prolonged anaesthetic time and donor site morbidity, soft tissue reconstructions with no osseous component are frequently used in elderly patients with maxillary defects. The rectus myocutaneous free flap (RMFF) offers a large volume of soft tissue bulk, multiple skin islands, and long pedicle length.2,8–11 However, in patients with central obesity, the excess layer of adipose tissue between the muscle and the skin could result in challenging inset and suboptimal results for speech and dental rehabilitation. The objective of this report was to describe our experience with a technical modification of the conventional RMFF, where the maxillary defect is reconstructed with a small cuff of rectus muscle and a large amount of adipose tissue with no skin component—the musculoadipose rectus free flap (MARF).
PATIENTS AND METHODS
Five cases of MARF reconstruction for maxillary reconstruction were performed between 2003 and 2013. The ages of the patients ranged from 50 to 68 years old and the body mass indexes (BMI) ranged from 29.0 to 41.2 kg/m2. All patients had sinonasal tumor, of which three were adenoid cystic carcinoma, one squamous cell carcinoma, and one melanoma. Four patients underwent maxillectomy with concomitant orbital exenteration (Cordeiro type IIIb defects) and one patient underwent an infrastructure maxillectomy (Cordeiro type II defect). Details of the patients are shown in Table 1.
Microsurgery DOI 10.1002/micr 120mL
160mL
IIIb
160mL
120mL
120mL
IIIb
II
IIIb
IIIb
Facial artery
Facial artery
Facial artery
Facial artery
Facial artery
Flap/donor site outcomes
Common facial vein
Internal jugular vein
External jugular vein
No flap complications
Breakdown of donor site, require repeated packings.
No flap complication.
Offered brachytherapy to primary site. No flap complication. Common Seroma of facial vein donor site.
Common facial vein
Recipient vein
SCC, Squamous cell carcinoma; PPF, Pterygopalatine fossae; V2, Maxillary division of Trigeminal nerve.
59.5
68.2
51.7
Female Adenoid cystic tumour with perineural invasion left maxilla Male Adenoid cystic tumour right hard palate with perineural invasion of PPF and V2 Female Right sinonasal melanoma with invasion into pterygoid muscle, and pterygopalatine space Male Adenoid cystic tumour left hard palate with perineural invaison, PPF
48.6
Pathology
Female SCC right hard palate, perineural and vascular invasion
Sex
50.0
Age (Years)
Approximate volume Codeiro of donor Defects adipose Recipient Clasifications tissue artery
6
6
5
15
15
12
10
9
20
22
Days Length of to oral hospital intake stay
Table 1. Baseline Information, Pathology and Recipient Vessels of the Patients
Soft
Soft
Soft
Soft
Liquid
Diet type
Denture
No
No
Denture
No
32.8 months
7.2 months
9.5 months
11.0 months
11.3 months
Dental Follow-up rehabilitation duration
Evidence on persistent disease on imaging
Patient died of malignancy
No evidence of disease on imaging Evidence of persistent disease on imaging
Patient died of malignancy
Status at last follow up
2 Low et al.
Musculo-Adipose Rectus Free Flap
Figure 1. (a) Shows the maxillary defect after the ablation. (b) Shows the MARF inset after trimming of the adipose tissue. The final volume of the adipose tissue is still slightly larger than the defect to allow for adequate obturation of the defect.
3
Figure 2. (a) Shows the appearance of the fatty tissue with early granulation tissue at day 10. (b) Shows the mucosalization over the existing fatty tissue and pigmentation of the new epithelial tissue at 9 month.
Operative Technique
The initial steps of this harvest follow those of a RMFF.9,11 After identification of the perforators from the rectus sheath, the lateral aspect of the rectus sheath was incised to inspect the vascular anatomy of the inferior epigastric vessels on the undersurface of the rectus muscle. A small cuff of the rectus muscle was then harvested to preserve the integrity of the perforator. The overlying adipose tissue was subsequently harvested. The inferior epigastric vessels (artery and venae commitantes) were followed down to the junction where they join the external iliac vessels. Ties were applied close to the junction to maximize the length of the pedicle. An abdominal mesh was routinely used to re-enforce the rectus sheath and the donor site was closed over a suction drain. Inset of the flap was performed by initially tunneling the pedicle to the neck deep to the mandible. The flap was subsequently positioned with the rectus sheath facing superiorly and suspension sutures were positioned from the sheath to the remaining bony framework to prevent prolapse. At this point, the volume of the adipose tissue was assessed and the adipose tissue trimmed just beyond the mucosal defect, as shown in Figures 1a and 1b. The adipose tissue was then secured to the edges of the oral mucosa and left to epithelialize. No sutures were used on the nasal surface but a nasal stent was placed to maintain nasal patency. Vascular anastomoses were performed once satisfactory position of the free flap was achieved. A naso-gastric tube was placed on the contralateral nasal passage and patients were kept nil by mouth for 7–10 days. Patients were also advised not to blow their nose for 2 weeks. RESULTS
The volume of the harvested adipose tissue ranged from 120 to 160 mL. The facial artery served as the
donor artery for all five patients. Three patients had endto-end anastomosis of the venous outflow to the common facial vein, one patient had an end-to-end venous anastomosis to the external jugular vein, and one patient had an end-to-side venous anostomosis to the internal jugular vein. Given the long vascular pedicle, vein graft was not required in all cases. All flaps survived with no cases requiring reexploration. The hospital stay of the patients ranged from 9 to 22 days. One patient stayed an additional two weeks in the hospital for brachytherapy treatment of perineural invasion into the pterygopalatine fossae (discharge day 22); the other patient had excessive nausea and vomiting, delaying the commencement of oral intake (day 15), hence the discharge was delayed until day 20. Of the five patients, one patient developed seroma of the donor site, requiring repeated aspiration and one patient developed wound infection and breakdown of the donor site requiring wound packing. None of the patients developed an abdominal hernia. Complete oro-nasal separation was achieved in all patients with no oronasal fistula. The time to commencement of oral intake ranged from 5 to 15 days. Granulation over the fatty tissue was seen on the oral surface by the first week (Fig. 2a), followed by mucosal epithelium. By the ninth month, migration of the pigmented cells was observed on the mucosalized fatty surface (Fig. 2b). The follow-up of the patients ranged from 7.5 to 32.8 months. The midface contour was adequately maintained with minimal loss of tissue volume (Fig. 3).Two of the patients in this series died within 12 months from their malignancies; however, both of them were able to return home with good palliation of their local pain symptoms. Two patients were alive with disease and one patient was alive with no evidence of disease. All three patients that Microsurgery DOI 10.1002/micr
4
Low et al.
Figure 3. At 33 month, the volume of the adipose tissue was preserved with adequate midface contour on the left side despite no osseous framework reconstruction.
were alive were able to tolerate soft diet during assessment on their last follow-up consultation. DISCUSSIONS
We describe a modification of the rectus abdominis flap for reconstruction of the maxillectomy defects that is simple, with acceptable functional and cosmetic results. Codeiro et al. described their experience of 100 maxillary reconstructions,5,12 where the rectus abdominis flap is recommended for most of the larger defects. In their series, the rectus abdominis flap is used as a myocutaneous flap, with skin paddles fashioned to reconstruct both the palatal and the nasal mucosal defects. They did not comment on their strategies amongst patients with excessive BMI. Obese patients often have a very thick adipose layer, resulting in difficulties in fashioning and inset of the skin paddles. Most reconstructive surgeons abandon the rectus flap in obese patients for that reason. However with the MARF, excessive fatty tissue is no longer a limitation as the fatty tissue can be trimmed to fit the size of the defect and the additional bulk is used to seal the oral defect to provide robust oro-nasal separation. In our series of patients, oral intake of these patients was commenced as early as day 5 without oronasal fistula. Any excessive tissue prolapsing into the oral cavity will be remodeled by local pressure from the tongue as the flap mucosalizes. Furthermore, the final reconstructed surface is less pliable than skin and therefore more suitable for dental rehabilitation. In this small series, four Microsurgery DOI 10.1002/micr
patients were able to tolerate a soft diet and two patients were successfully fitted with modified dentures. Unlike conventional harvest of the RMFF, where a large volume of muscle is harvested; the MARF harvest contains a small cuff of muscle with the aim to minimize volume loss from muscle atrophy and minimize donor site morbidity.10 This is important to preserve the midface contour, as shown in Figure 3. In our series, the MARF flap was only offered for patients with advanced disease, with multiple co-morbidities and excessive central obesity. These patients require a robust flap for their reconstruction that is quick to harvest and easy to inset to minimize the time under general anesthesia. The choice of free flap reconstructions for maxillary defects is predominantly determined by the volume required for replacement of the defect.5 In cases of small volume maxillary defect confined to hard palate, with mostly intact maxillary alveolar bone, reconstruction with fasciocutaneous free flap, such as radial forearm free flap,5 ulnar forearm free flap,13 or peroneal artery perforator flaps14 is adequate. Reconstruction of a medium to large maxillary defect ideally involves reconstruction of both the bony framework and the soft tissue with a composite flap, such as the vascularized iliac crest, fibula, or scapula; or soft tissue only reconstruction with good tissue bulk.1,5,15,16 The argument for bony framework reconstruction is largely twofold. Firstly, to provide sufficient bone stock for dental implants and secondly, without bony framework, the tissue may prolapse or contract making dental rehabilitation unpredictable. Whilst these are valid concerns, very few patients in our institution are candidates for dental implantation and the majority will be fitted with a denture. Therefore, the emphasis for alveolar bone stock is less important in this setting. Excessive oral bulk on the other hand, would preclude the retention of a denture and so presents a far greater functional concern in our patient group. MARF affords a greater degree of flexibility in avoiding excessive oral bulk than either the traditional rectus free flap with skin or a poorly designed osseous flap. We have found the MARF easy to contour once inset is completed to avoid excessive oral bulk. In regards to restoring facial contour, much of the concern regarding atrophy of soft tissue flaps in maxillectomy defects has been related to flaps with large muscle components.17 Vascularized fat demonstrates much lower rates of atrophy than muscle17,18 and so we find that cosmetic results with the predominantly adipose MARF to be both more predictable and permanent. A limitation to this series is the very small number of patients, due to strict selection criteria. The longest follow up so far is 32 months; hence, it is unclear whether the contour in the midface would be maintained over a longer period of time. Nevertheless, the MARF
Musculo-Adipose Rectus Free Flap
flap is easy to raise, fashion, and harvest. It allows concurrent harvest with the ablative team, with minimal donor site morbidity making MARF especially suitable for elderly patients with multiple co-morbidities. CONCLUSION
The MARF may be considered as an alternative to myocutaneous rectus free flap particularly for the reconstruction of maxillary defects in patients with central obesity.
REFERENCES 1. Clark JR, Vesely M, Gilbert R. Scapular angle osteomyogenous flap in postmaxillectomy reconstruction: Defect, reconstruction, shoulder function, and harvest technique. Head Neck 2008;30:10–20. 2. Shrime MG, Gilbert RW. Reconstruction of the midface and maxilla. Facial Plast Surg Clin North Am 2009;17:211–223. 3. Dalgorf D, Higgins K. Reconstruction of the midface and maxilla. Curr Opin Otolaryngol Head Neck Surg 2008;16:303–311. 4. Andrades P, Militsakh O, Hanasono MM, Rieger J, Rosenthal EL. Current strategies in reconstruction of maxillectomy defects. Arch Otolaryngol Head Neck Surg 2011;137:806–812. 5. Cordeiro PG, Chen CM. A 15-year review of midface reconstruction after total and subtotal maxillectomy. I. Algorithm and outcomes. Plast Reconstr Surg 2012;129:124–136. 6. Hayat MJ, Howlader N, Reichman ME, Edwards BK. Cancer statistics, trends, and multiple primary cancer analyses from the Surveillance, Epidemiology, and End Results (SEER) Program. Oncologist 2007;12:20–37. 7. Clayman GL, Eicher SA, Sicard MW, Razmpa E, Goepfert H. Surgical outcomes in head and neck cancer patients 80 years of age and older. Head Neck 1998;20:216–223.
5
8. Olsen KD, Meland NB, Ebersold MJ, Bartley GB, Garrity JA. Extensive defects of the sino-orbital region. Results with microvascular reconstruction. Arch Otolaryngol Head Neck Surg 1992;118: 828–833; discussion 859–860. 9. Pryor SG, Moore EJ, Kasperbauer JL. Orbital exenteration reconstruction with rectus abdominis microvascular free flap. Laryngoscope 2005;115:1912–1916. 10. Browne JD, Burke AJ. Benefits of routine maxillectomy and orbital reconstruction with the rectus abdominis free flap. Otolaryngol Head Neck Surg 1999;121:203–209. 11. Urken M. Free flaps—Rectus abdominis. In: Urken M, Cheney M, Sullivan M, Biller H, editors. Atlas of Regional and Free Flaps for Head and Neck Reconstruction. New York: Raven Press; 1995. pp 119–138. 12. Cordeiro PG, Chen CM. A 15-year review of midface reconstruction after total and subtotal maxillectomy. II. Technical modifications to maximize aesthetic and functional outcomes. Plast Reconstr Surg 2012;129:139–147. 13. Antony AK, Hootnick JL, Antony AK. Ulnar forearm free flaps in head and neck reconstruction: Systematic review of the literature and a case report. Microsurgery 2014;34:68–75. 14. Acarturk TO, Maldonado AA, Ereso A. Intraoral reconstruction with “thinned” peroneal artery perforator flaps: An alternative to classic donor areas in comorbid patients. Microsurgery 2014 [Epub ahead of print]. 15. Zaker Shahrak A, Zor F, Kanatas A, Acikel C, Sapountzis S, Nicoli F, Altuntas SH, Knobe M, Chen HC, Prescher A, Holzle F, Sonmez TT. Morphological and morphometric evaluation of the ilium, fibula, and scapula bones for oral and maxillofacial reconstruction. Microsurgery 2014;34:638–645. 16. Brown JS, Shaw RJ. Reconstruction of the maxilla and midface: Introducing a new classification. Lancet Oncol 2010;11:1001–1008. 17. Sakamoto Y, Takahara T, Ota Y, Aoki T, Yamazaki H, Otsuru M, Takahashi M, Aoyama K, Kaneko A, Kawada S, Ichikawa T, Imagawa K, Miyasaka M. MRI analysis of chronological changes in free-flap volume in head and neck reconstruction by volumetry. Tokai J Exp Clin Med 2014;39:44–50. 18. Tran NV, Chang DW, Gupta A, Kroll SS, Robb GL. Comparison of immediate and delayed free TRAM flap breast reconstruction in patients receiving postmastectomy radiation therapy. Plast Reconstr Surg 2001;108:78–82.
Microsurgery DOI 10.1002/micr
Background: The rectus myocutaneous free flap (RMFF) is used for medium to large maxillectomy defects. However, in patients with central obesity the inset could be difficult due to the bulk from excessive layer of adipose tissue. We describe a modification of the RMFF for patients with excessive central obesity with a flap consisting of adipose tissue with minimal rectus muscle; the musculo-adipose rectus free flap (MARF). Methods: Five cases of MARF reconstruction were performed between 2003 and 2013, with patients’ body mass indexes ranging from 29.0 to 41.2 kg/m2. All patients had sinonasal tumor, of which three were adenoid cystic carcinoma, one squamous cell carcinoma, and one melanoma. Four patients had Codeiro IIIb defects and one had Codeiro II defect. Using the MARF technique, the maxillectomy defect was obliterated with vascularized adipose tissue overlying the rectus muscle and was trimmed to fit the maxillectomy defect. The adipose tissue was allowed to granulate and mucosalize. Results: The volume of adipose tissue harvested was between 120 and 160 mL. All flaps survived with no requirement for re-exploration. Complete oro-nasal separation was achieved in all patients. The time to commencement of oral intake ranges from 5 to 15 days. One patient developed seroma and one developed wound breakdown on the donor site. The length of stay at the hospital ranges from 9 to 22 days. On follow-up ranging 7.5–32.8 months, two patients died from their malignancies. The other three patients were able to tolerate oral soft diet. Conclusion: The MARF may be considered as an alternaC 2015 Wiley tive to myocutaneous rectus free flap particularly for the reconstruction of maxillary defects in patients with central obesity. V Periodicals, Inc. Microsurgery 00:000–000, 2015.
Reconstructing
large maxillectomy defects is complicated and challenging due to its three dimensional nature.1 Various reconstructive options exist including prosthetics, local flaps, and free vascularized flaps.2 The reconstructive goals for reconstructing a large maxillectomy defects include obtaining good wound healing with separation of oral, nasal, and cranial compartments; optimized functional outcomes in speech, swallow, and dental rehabilitation, as well as acceptable cosmesis.3 An ideal free flap for this region should satisfy these goals and have a long pedicle with the option of increasing soft tissue bulk when necessary; the flap should also have the ability to reconstruct multiple oral and nasal mucosal defects with a good bone stock for osseous framework reconstruction in patients where dental rehabilitation is planned.4 Therefore, all maxillary defects need to be approached individually.5 With improved healthcare, we are now seeing increasing numbers of geriatric patients, with multiple co-morbidities presenting with advanced head and neck 1 Otolaryngology Head and Neck Department, Royal Perth Hospital, Perth, WA 6000 2 Sydney Head and Neck Cancer Institute, Royal Prince Alfred Hospital, Camperdown, NSW 3 Central Clinical School, University of Sydney, Sydney, NSW Presentation: This work was presented at the ANZHNC, Melbourne, 2013 *Correspondence to: Tsu-Hui (Hubert) Low, M.B.B.S. (Hons), Suite 5, 35 Fourth Ave Eastwood NSW 2122. E-mail: [email protected] Received 25 November 2014; Revision accepted 29 May 2015; Accepted 3 June 2015 Published online 00 Month 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/micr.22439
Ó 2015 Wiley Periodicals, Inc.
tumors.6,7 Due to concerns over prolonged anaesthetic time and donor site morbidity, soft tissue reconstructions with no osseous component are frequently used in elderly patients with maxillary defects. The rectus myocutaneous free flap (RMFF) offers a large volume of soft tissue bulk, multiple skin islands, and long pedicle length.2,8–11 However, in patients with central obesity, the excess layer of adipose tissue between the muscle and the skin could result in challenging inset and suboptimal results for speech and dental rehabilitation. The objective of this report was to describe our experience with a technical modification of the conventional RMFF, where the maxillary defect is reconstructed with a small cuff of rectus muscle and a large amount of adipose tissue with no skin component—the musculoadipose rectus free flap (MARF).
PATIENTS AND METHODS
Five cases of MARF reconstruction for maxillary reconstruction were performed between 2003 and 2013. The ages of the patients ranged from 50 to 68 years old and the body mass indexes (BMI) ranged from 29.0 to 41.2 kg/m2. All patients had sinonasal tumor, of which three were adenoid cystic carcinoma, one squamous cell carcinoma, and one melanoma. Four patients underwent maxillectomy with concomitant orbital exenteration (Cordeiro type IIIb defects) and one patient underwent an infrastructure maxillectomy (Cordeiro type II defect). Details of the patients are shown in Table 1.
Microsurgery DOI 10.1002/micr 120mL
160mL
IIIb
160mL
120mL
120mL
IIIb
II
IIIb
IIIb
Facial artery
Facial artery
Facial artery
Facial artery
Facial artery
Flap/donor site outcomes
Common facial vein
Internal jugular vein
External jugular vein
No flap complications
Breakdown of donor site, require repeated packings.
No flap complication.
Offered brachytherapy to primary site. No flap complication. Common Seroma of facial vein donor site.
Common facial vein
Recipient vein
SCC, Squamous cell carcinoma; PPF, Pterygopalatine fossae; V2, Maxillary division of Trigeminal nerve.
59.5
68.2
51.7
Female Adenoid cystic tumour with perineural invasion left maxilla Male Adenoid cystic tumour right hard palate with perineural invasion of PPF and V2 Female Right sinonasal melanoma with invasion into pterygoid muscle, and pterygopalatine space Male Adenoid cystic tumour left hard palate with perineural invaison, PPF
48.6
Pathology
Female SCC right hard palate, perineural and vascular invasion
Sex
50.0
Age (Years)
Approximate volume Codeiro of donor Defects adipose Recipient Clasifications tissue artery
6
6
5
15
15
12
10
9
20
22
Days Length of to oral hospital intake stay
Table 1. Baseline Information, Pathology and Recipient Vessels of the Patients
Soft
Soft
Soft
Soft
Liquid
Diet type
Denture
No
No
Denture
No
32.8 months
7.2 months
9.5 months
11.0 months
11.3 months
Dental Follow-up rehabilitation duration
Evidence on persistent disease on imaging
Patient died of malignancy
No evidence of disease on imaging Evidence of persistent disease on imaging
Patient died of malignancy
Status at last follow up
2 Low et al.
Musculo-Adipose Rectus Free Flap
Figure 1. (a) Shows the maxillary defect after the ablation. (b) Shows the MARF inset after trimming of the adipose tissue. The final volume of the adipose tissue is still slightly larger than the defect to allow for adequate obturation of the defect.
3
Figure 2. (a) Shows the appearance of the fatty tissue with early granulation tissue at day 10. (b) Shows the mucosalization over the existing fatty tissue and pigmentation of the new epithelial tissue at 9 month.
Operative Technique
The initial steps of this harvest follow those of a RMFF.9,11 After identification of the perforators from the rectus sheath, the lateral aspect of the rectus sheath was incised to inspect the vascular anatomy of the inferior epigastric vessels on the undersurface of the rectus muscle. A small cuff of the rectus muscle was then harvested to preserve the integrity of the perforator. The overlying adipose tissue was subsequently harvested. The inferior epigastric vessels (artery and venae commitantes) were followed down to the junction where they join the external iliac vessels. Ties were applied close to the junction to maximize the length of the pedicle. An abdominal mesh was routinely used to re-enforce the rectus sheath and the donor site was closed over a suction drain. Inset of the flap was performed by initially tunneling the pedicle to the neck deep to the mandible. The flap was subsequently positioned with the rectus sheath facing superiorly and suspension sutures were positioned from the sheath to the remaining bony framework to prevent prolapse. At this point, the volume of the adipose tissue was assessed and the adipose tissue trimmed just beyond the mucosal defect, as shown in Figures 1a and 1b. The adipose tissue was then secured to the edges of the oral mucosa and left to epithelialize. No sutures were used on the nasal surface but a nasal stent was placed to maintain nasal patency. Vascular anastomoses were performed once satisfactory position of the free flap was achieved. A naso-gastric tube was placed on the contralateral nasal passage and patients were kept nil by mouth for 7–10 days. Patients were also advised not to blow their nose for 2 weeks. RESULTS
The volume of the harvested adipose tissue ranged from 120 to 160 mL. The facial artery served as the
donor artery for all five patients. Three patients had endto-end anastomosis of the venous outflow to the common facial vein, one patient had an end-to-end venous anastomosis to the external jugular vein, and one patient had an end-to-side venous anostomosis to the internal jugular vein. Given the long vascular pedicle, vein graft was not required in all cases. All flaps survived with no cases requiring reexploration. The hospital stay of the patients ranged from 9 to 22 days. One patient stayed an additional two weeks in the hospital for brachytherapy treatment of perineural invasion into the pterygopalatine fossae (discharge day 22); the other patient had excessive nausea and vomiting, delaying the commencement of oral intake (day 15), hence the discharge was delayed until day 20. Of the five patients, one patient developed seroma of the donor site, requiring repeated aspiration and one patient developed wound infection and breakdown of the donor site requiring wound packing. None of the patients developed an abdominal hernia. Complete oro-nasal separation was achieved in all patients with no oronasal fistula. The time to commencement of oral intake ranged from 5 to 15 days. Granulation over the fatty tissue was seen on the oral surface by the first week (Fig. 2a), followed by mucosal epithelium. By the ninth month, migration of the pigmented cells was observed on the mucosalized fatty surface (Fig. 2b). The follow-up of the patients ranged from 7.5 to 32.8 months. The midface contour was adequately maintained with minimal loss of tissue volume (Fig. 3).Two of the patients in this series died within 12 months from their malignancies; however, both of them were able to return home with good palliation of their local pain symptoms. Two patients were alive with disease and one patient was alive with no evidence of disease. All three patients that Microsurgery DOI 10.1002/micr
4
Low et al.
Figure 3. At 33 month, the volume of the adipose tissue was preserved with adequate midface contour on the left side despite no osseous framework reconstruction.
were alive were able to tolerate soft diet during assessment on their last follow-up consultation. DISCUSSIONS
We describe a modification of the rectus abdominis flap for reconstruction of the maxillectomy defects that is simple, with acceptable functional and cosmetic results. Codeiro et al. described their experience of 100 maxillary reconstructions,5,12 where the rectus abdominis flap is recommended for most of the larger defects. In their series, the rectus abdominis flap is used as a myocutaneous flap, with skin paddles fashioned to reconstruct both the palatal and the nasal mucosal defects. They did not comment on their strategies amongst patients with excessive BMI. Obese patients often have a very thick adipose layer, resulting in difficulties in fashioning and inset of the skin paddles. Most reconstructive surgeons abandon the rectus flap in obese patients for that reason. However with the MARF, excessive fatty tissue is no longer a limitation as the fatty tissue can be trimmed to fit the size of the defect and the additional bulk is used to seal the oral defect to provide robust oro-nasal separation. In our series of patients, oral intake of these patients was commenced as early as day 5 without oronasal fistula. Any excessive tissue prolapsing into the oral cavity will be remodeled by local pressure from the tongue as the flap mucosalizes. Furthermore, the final reconstructed surface is less pliable than skin and therefore more suitable for dental rehabilitation. In this small series, four Microsurgery DOI 10.1002/micr
patients were able to tolerate a soft diet and two patients were successfully fitted with modified dentures. Unlike conventional harvest of the RMFF, where a large volume of muscle is harvested; the MARF harvest contains a small cuff of muscle with the aim to minimize volume loss from muscle atrophy and minimize donor site morbidity.10 This is important to preserve the midface contour, as shown in Figure 3. In our series, the MARF flap was only offered for patients with advanced disease, with multiple co-morbidities and excessive central obesity. These patients require a robust flap for their reconstruction that is quick to harvest and easy to inset to minimize the time under general anesthesia. The choice of free flap reconstructions for maxillary defects is predominantly determined by the volume required for replacement of the defect.5 In cases of small volume maxillary defect confined to hard palate, with mostly intact maxillary alveolar bone, reconstruction with fasciocutaneous free flap, such as radial forearm free flap,5 ulnar forearm free flap,13 or peroneal artery perforator flaps14 is adequate. Reconstruction of a medium to large maxillary defect ideally involves reconstruction of both the bony framework and the soft tissue with a composite flap, such as the vascularized iliac crest, fibula, or scapula; or soft tissue only reconstruction with good tissue bulk.1,5,15,16 The argument for bony framework reconstruction is largely twofold. Firstly, to provide sufficient bone stock for dental implants and secondly, without bony framework, the tissue may prolapse or contract making dental rehabilitation unpredictable. Whilst these are valid concerns, very few patients in our institution are candidates for dental implantation and the majority will be fitted with a denture. Therefore, the emphasis for alveolar bone stock is less important in this setting. Excessive oral bulk on the other hand, would preclude the retention of a denture and so presents a far greater functional concern in our patient group. MARF affords a greater degree of flexibility in avoiding excessive oral bulk than either the traditional rectus free flap with skin or a poorly designed osseous flap. We have found the MARF easy to contour once inset is completed to avoid excessive oral bulk. In regards to restoring facial contour, much of the concern regarding atrophy of soft tissue flaps in maxillectomy defects has been related to flaps with large muscle components.17 Vascularized fat demonstrates much lower rates of atrophy than muscle17,18 and so we find that cosmetic results with the predominantly adipose MARF to be both more predictable and permanent. A limitation to this series is the very small number of patients, due to strict selection criteria. The longest follow up so far is 32 months; hence, it is unclear whether the contour in the midface would be maintained over a longer period of time. Nevertheless, the MARF
Musculo-Adipose Rectus Free Flap
flap is easy to raise, fashion, and harvest. It allows concurrent harvest with the ablative team, with minimal donor site morbidity making MARF especially suitable for elderly patients with multiple co-morbidities. CONCLUSION
The MARF may be considered as an alternative to myocutaneous rectus free flap particularly for the reconstruction of maxillary defects in patients with central obesity.
REFERENCES 1. Clark JR, Vesely M, Gilbert R. Scapular angle osteomyogenous flap in postmaxillectomy reconstruction: Defect, reconstruction, shoulder function, and harvest technique. Head Neck 2008;30:10–20. 2. Shrime MG, Gilbert RW. Reconstruction of the midface and maxilla. Facial Plast Surg Clin North Am 2009;17:211–223. 3. Dalgorf D, Higgins K. Reconstruction of the midface and maxilla. Curr Opin Otolaryngol Head Neck Surg 2008;16:303–311. 4. Andrades P, Militsakh O, Hanasono MM, Rieger J, Rosenthal EL. Current strategies in reconstruction of maxillectomy defects. Arch Otolaryngol Head Neck Surg 2011;137:806–812. 5. Cordeiro PG, Chen CM. A 15-year review of midface reconstruction after total and subtotal maxillectomy. I. Algorithm and outcomes. Plast Reconstr Surg 2012;129:124–136. 6. Hayat MJ, Howlader N, Reichman ME, Edwards BK. Cancer statistics, trends, and multiple primary cancer analyses from the Surveillance, Epidemiology, and End Results (SEER) Program. Oncologist 2007;12:20–37. 7. Clayman GL, Eicher SA, Sicard MW, Razmpa E, Goepfert H. Surgical outcomes in head and neck cancer patients 80 years of age and older. Head Neck 1998;20:216–223.
5
8. Olsen KD, Meland NB, Ebersold MJ, Bartley GB, Garrity JA. Extensive defects of the sino-orbital region. Results with microvascular reconstruction. Arch Otolaryngol Head Neck Surg 1992;118: 828–833; discussion 859–860. 9. Pryor SG, Moore EJ, Kasperbauer JL. Orbital exenteration reconstruction with rectus abdominis microvascular free flap. Laryngoscope 2005;115:1912–1916. 10. Browne JD, Burke AJ. Benefits of routine maxillectomy and orbital reconstruction with the rectus abdominis free flap. Otolaryngol Head Neck Surg 1999;121:203–209. 11. Urken M. Free flaps—Rectus abdominis. In: Urken M, Cheney M, Sullivan M, Biller H, editors. Atlas of Regional and Free Flaps for Head and Neck Reconstruction. New York: Raven Press; 1995. pp 119–138. 12. Cordeiro PG, Chen CM. A 15-year review of midface reconstruction after total and subtotal maxillectomy. II. Technical modifications to maximize aesthetic and functional outcomes. Plast Reconstr Surg 2012;129:139–147. 13. Antony AK, Hootnick JL, Antony AK. Ulnar forearm free flaps in head and neck reconstruction: Systematic review of the literature and a case report. Microsurgery 2014;34:68–75. 14. Acarturk TO, Maldonado AA, Ereso A. Intraoral reconstruction with “thinned” peroneal artery perforator flaps: An alternative to classic donor areas in comorbid patients. Microsurgery 2014 [Epub ahead of print]. 15. Zaker Shahrak A, Zor F, Kanatas A, Acikel C, Sapountzis S, Nicoli F, Altuntas SH, Knobe M, Chen HC, Prescher A, Holzle F, Sonmez TT. Morphological and morphometric evaluation of the ilium, fibula, and scapula bones for oral and maxillofacial reconstruction. Microsurgery 2014;34:638–645. 16. Brown JS, Shaw RJ. Reconstruction of the maxilla and midface: Introducing a new classification. Lancet Oncol 2010;11:1001–1008. 17. Sakamoto Y, Takahara T, Ota Y, Aoki T, Yamazaki H, Otsuru M, Takahashi M, Aoyama K, Kaneko A, Kawada S, Ichikawa T, Imagawa K, Miyasaka M. MRI analysis of chronological changes in free-flap volume in head and neck reconstruction by volumetry. Tokai J Exp Clin Med 2014;39:44–50. 18. Tran NV, Chang DW, Gupta A, Kroll SS, Robb GL. Comparison of immediate and delayed free TRAM flap breast reconstruction in patients receiving postmastectomy radiation therapy. Plast Reconstr Surg 2001;108:78–82.
Microsurgery DOI 10.1002/micr
