Avoiding Secondary Skin Graft Donor Site Morbidity in the Fibula Free Flap Harvest

Objective  To compare donor site morbidity in patients who have undergone fibula free flap reconstruction in which the skin graft was taken from the expected cutaneous paddle of the fibula with the known complications of the popular technique of obtaining a split-thickness skin graft (STSG) from a secondary donor site.

Design  Cohort study.

Setting  The tertiary care centers at Loma Linda University Medical Center and University of California, Los Angeles, Medical Center.

Patients  From September 1, 2006, to March 30, 2007, 30 patients underwent fibula free flap harvest by 2 surgeons at separate tertiary care centers. Twenty-one of those procedures took place at the University of California, Los Angeles, and 9 at Loma Linda University. Patients included 15 men (50%) and 15 women (50%), with a mean age of 58 (range, 19-88) years. All 30 patients underwent fibula free flap harvest with a split-thickness skin graft (graft thickness, 0.04 cm), obtained from osteocutaneous paddle using a 5.1-cm-wide dermatome, as well as oral cavity and oropharyngeal reconstruction with the de-epithelialized skin paddle.

Main Outcome Measures  Measures of donor site morbidity, including graft failure and wound breakdown, and measures of recipient site morbidity, including flap failure, hardware complications, intraoral complications, and the need for additional surgery.

Results  Of the 30 patients who underwent this procedure, 4 had partial skin graft failures, for a complete skin graft survival of 87%. There were no complete skin graft losses. Regarding the fibula osteocutaneous free flap, there were no complete flap losses, 1 skin paddle necrosis that required debridement, 2 postoperative orocutaneous fistulas, 1 case of infected/extruded hardware, and 1 adhesion formation that required additional surgery for lysis of adhesion and placement of the split-thickness skin graft.

Conclusions  The outlined novel technique has similar rates of free flap survival and skin graft take compared with previously described methods. Harvesting the skin graft over the expected osteocutaneous paddle results in decreased lower extremity morbidity by providing equivalent graft tissue for donor site closure and eliminating the morbidity of a secondary donor site. As long as there will not be 2 large opposing grafted surfaces, this technique should be considered when harvesting fibula free flaps for reconstruction of oromandibular resection defects, especially in cancers of the alveolar ridge and floor of the mouth.

Cancer involving the oral cavity is frequently treated with mandibular resection and microvascular reconstruction. Use of the fibula osteocutaneous flap has become one of the most versatile methods for mandibular reconstruction in head and neck surgery. There are several advantages of this flap, including the ability to have a 2-team simultaneous resection and flap harvest; the ease with which the bone is contoured with osteotomies; the availability of a long length of bone, adequate vessel caliber, ample skin, and soft tissue volume; the potential for sensory reinnervation; the option for dental implants; and, most often, few donor site morbidities.^1-4 Exact methods for primary donor site closure are still somewhat controversial, with primary closure, or, more often, use of a split-thickness skin graft (STSG) from a secondary donor site.^2,5-7 The STSG donor site adds to patient pain during the postoperative period and poses a potential source for additional morbidity, including infection, bleeding, and delayed healing. Several recent critical studies^2,7 have addressed donor site morbidity; however, these studies concentrate on complications at the primary donor site and compare primary closure with closure using an STSG. To our knowledge, no large reconstruction studies to date involving the use of fibula free flap harvest have explored secondary donor site morbidities or options to decrease or eliminate the secondary donor site. More specifically, no studies have attempted to show a benefit from obtaining the skin graft from the expected cutaneous paddle of the fibula flap before harvest. Therefore, the purposes of this cohort study were to evaluate donor site morbidity in patients who have undergone fibula free flap reconstruction in which the skin graft was taken from the expected cutaneous paddle of the fibula and compare this finding with the known complications of the popular technique of obtaining an STSG from a secondary donor site.

From September 1, 2006, to March 30, 2007, 30 patients underwent fibula free flap harvest. Twenty-one of those procedures were performed at the University of California, Los Angeles, and 9 at Loma Linda University. Patients included 15 men (50%) and 15 women (50%), with a mean age of 58 (range, 19-88) years. All 30 patients underwent fibula free flap harvest with the STSG (graft thickness, 0.04 cm) obtained from the expected osteocutaneous paddle using a 5.1-cm-wide dermatome, as well as oral and oropharyngeal reconstruction with the de-epithelialized skin paddle (Figures 1, 2, and 3). Main outcome measures of donor site morbidity include graft failure and wound breakdown. Main outcome measures of recipient site morbidity include flap failure, hardware complications, intraoral complications, and the need for additional surgery. Patients underwent assessment in the postoperative period and at follow-up visits for donor and recipient site complications (Figure 4).

In this joint-institution cohort study, 30 patients underwent fibula free flap harvest for head and neck cancer with fibular osteocutaneous microvascular reconstruction. Fourteen donor sites were from the right leg and 16 from the left leg. The average STSG thickness was 0.04 cm, with an area of 16 × 6 cm (96 cm²). Twenty-six of the skin grafts (87%) were viable at follow-up, 4 (13%) had partial loss (range, 15%-50% of the graft), and none were lost completely. Primary tumor types included 20 squamous cell carcinoma (67%), 6 mucoepidermoid carcinoma (20%), 2 osteosarcoma (7%), 1 mandibular osteoradionecrosis (3%), and 1 osteomyelitis (3%). Seven patients (23%) had undergone prior head and neck surgery; 6 patients (20%), prior chemotherapy; and 4 (13%), prior lymphadenectomy. Tumor location was broken into 6 major sites, with the most common being the floor of the mouth (12 tumors [40%]), followed by the mandibular alveolar ridge (8 [27%]), base of the tongue (3 [10%]), parotid (3 [10%]), oral tongue (2 [7%]), and pharynx (2 [7%]).

The average flap ischemia time was 2 hours 38 minutes (range, 1 hour 45 minutes to 4 hours). This is a large range; however, differences in these operative variables did not contribute significantly to outcome measures.

There were no complete flap losses. One skin paddle necrosis required surgical debridement. This patient was seropositive for the human immunodeficiency virus and had a postoperative streptococcus infection in the floor of the mouth. Two patients developed early postoperative orocutaneous fistulas that were managed conservatively with dressing changes and healed without further intervention. One patient developed an adhesion formation between the base of the tongue and a tonsil defect (Figure 5) that required additional surgery for adhesion lysis and placement of the STSG.

Complications unrelated to the flap and skin graft include 1 postoperative cerebrovascular accident, 1 postoperative lower extremity deep vein thrombosis, 1 postoperative aspiration pneumonia, and 1 case of postoperative delirium tremens.

Although the fibula free flap continues to be one of the most versatile and most often used options for mandibular reconstruction, there have not been any large head and neck studies, to our knowledge, examining techniques aimed at eliminating a secondary donor site. The main objective of this study was to show a benefit from obtaining the STSG from the anticipated osteocutaneous paddle. The main outcome measures for success included at least equivalent primary donor site healing, with an overall donor site complication rate of less than 28%, which is the rate previously described, and equivalent flap survival.²

It is possible to close the lower extremity donor site primarily if a small (width, ≤5 cm) cutaneous paddle is harvested or if the fibula alone is harvested.² In fact, many of the donor sites in this study were partially closed in primary fashion, but that was limited to the proximal portion, where the elasticity of the remaining skin was adequate to close with minimal tension. However, because of the need for wide cutaneous paddles in all of the patients in this study, we decided that the donor site would be better closed with a skin graft.

The postoperative minor fistula rate was 7% (n = 2). There were no major fistulas that needed another flap or other surgical intervention. This demonstrates the ability of the de-epithelialized skin paddle to heal to the recipient mucosa regardless of preoperative radiation status. Previous studies have demonstrated fistula rates of as much as 6.8%, which is similar to our results.⁸ Seventeen of our patients (57%) received preoperative radiation therapy, which included 1 of the 2 patients who developed fistulas.

In this study, 40% of our patient sample presented with floor of the mouth cancer, and 26% with alveolar ridge cancer. Oral cavity and oropharyngeal defects that would benefit from this method would likely include these 2 subsites as well as others that do not have large opposing surfaces.

One complication in particular may help to direct boundaries for this procedure. A patient who underwent resection for cancer of the right side of the base of the tongue, which had left a large defect involving a tonsil, the floor of the mouth, and the base of the tongue, developed a postoperative adhesion between the base of the tongue and the tonsil. This patient underwent additional surgery at 4 months after the primary procedure for lysis of adhesions and additional STSG. This particular complication was most likely due to the large opposing surfaces that were first grafted using de-epithelialized tissue. We therefore propose that, if the anticipated resection will create large opposing surfaces, this technique should be substituted with the use of epithelialized tissue.

The outlined technique shows no difference in the outcome of the free flap survival, nor does it change survival rates for skin grafts at the donor site. Harvesting the skin graft over the expected osteocutaneous paddle results in decreased lower extremity morbidity by providing equivalent graft tissue for donor site closure and eliminating the morbidity of a secondary donor site. As long as there will not be 2 large opposing grafted surfaces, this technique should be considered when harvesting fibula free flaps for oral cancer resection, especially for cancers of the alveolar ridge, floor of the mouth, parotid, pharynx, oral tongue, or base of the tongue.

Correspondence: Paul D. Kim, MD, Division of Otolaryngology, Loma Linda University Medical Center, 11234 Anderson St, Loma Linda, CA 92354 (kimpaul46@gmail.com).

Submitted for Publication: September 20, 2007; final revision received February 20, 2008; accepted February 25, 2008.

Author Contributions: Drs Kim, Fleck, and Blackwell had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Kim, Heffelfinger, and Blackwell. Acquisition of data: Blackwell. Analysis and interpretation of data: Fleck and Blackwell. Drafting of the manuscript: Fleck. Critical revision of the manuscript for important intellectual content: Kim, Heffelfinger, and Blackwell. Statistical analysis: Kim. Administrative, technical, and material support: Heffelfinger and Blackwell. Study supervision: Kim and Blackwell.

Financial Disclosure: None reported.

Previous Presentation: This study was presented as a poster at the Combined Otolaryngology Spring Meeting; April 28, 2007; San Diego, California.