The fibula free flap: advantages of the musclesparing technique

Roger G. Graham*, Marc C. Swan, Donald A. Hudson, Jacobus E. van Zyl

Department of Plastic and Reconstructive Surgery, University of Cape Town, Groote Schuur HospitalObservatory, Cape Town 7701, South Africa

Received 15 January 2003; accepted 9 May 2003

KEYWORDSFibula flap; Haematoma;

Dextran 40; Dissection

Summary The amount of muscle that should be retained on the free fibula duringharvest is unresolved. Muscle is used to protect the periosteum, but by harvesting alarge muscle cuff, the recipient and donor site morbidity increases. A retrospectivereview of 47 free fibula flaps performed between January 1997 and March 2002 wasundertaken. There was an average follow-up of 15 months. The dissection methodused for all cases was a muscle sparing technique where the peroneal vessels wereskeletonised anteromedially. Only a very thin rim of muscle (1–2 mm) was leftattached to the fibula. The recipient and donor vessels were flushed with heparinsaline solution intra-operatively and a Dextran 40 infusion was used for four days post-operatively in all cases. Of the 47 flaps, 39 were used for mandible reconstruction, sixfor maxillary reconstruction and two for long bone reconstruction following trauma.The average age was 47.7 years (range 13–82) and two-thirds (28/43) of the patientswere male. There was one post-operative death. The overall failure rate was 10.9%(5/46). Two flaps were lost as a result of arterial thrombosis, one from venousthrombosis, one from sepis and in one case the cause could not be determined. Therewere 2 (4.3%) recipient site haematomas. It is suggested that the low recipient sitehaematoma rate in this series may be related to the limited muscle bulk transferredwith the flap. Harvesting less muscle also enables easier insetting and folding of skinflap, and reduces the donor site problems of haematoma and weakness of the foot. Theblood supply to the fibula does not appear to be compromised.Q 2003 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All rightsreserved.

The fibula free flap was first used by Taylor et al. in19751 to reconstruct a large tibial defect, and it hassince become one of the workhorse flaps formandible and long bone reconstruction. Originally,the flap was dissected using a posterior approachthat commenced in the popliteal fossa.1 In 1979,Gilbert2 advocated the lateral approach, which has

subsequently become the standard method ofdissection. The next development occurred in1983 when the osteocutaneous flap was describedby Chen and Yan.3 Wei and Chen4 in 1986 furtherclarified the cutaneous anatomy by demonstratingthat the septocutaneous perforators alone providean adequate blood supply to the skin paddle. Thisknowledge allowed the technique to be refined toits present form as an osteoseptocutaneous flap.

Several detailed descriptions of the technique of

The British Association of Plastic Surgeons (2003) 56, 388–394

S0007-1226/03/$ - see front matter Q 2003 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All rights reserved.doi:10.1016/S0007-1226(03)00184-X

*Corresponding author. Tel.: þ27-21-4066231; fax: þ27-21-4486461.

E-mail address: rgraham@uctgsh1.uct.ac.za

free fibula harvest,1 –9 have been published over theyears, but despite this, one aspect remains unclear:how much muscle should be retained on the fibulaand its pedicle? There is a wide range in the amountof muscle that different surgeons harvest with afree fibula and no standard method of dissectionexists. Some surgeons1,3,9 recommend a wide cuff(1 cm) to preserve the blood supply to the fibulawhile others4 have reported retaining a thin musclecuff (1–2 mm). Taylor,1 in his original description,did not dissect out the peroneal vessels andretained a 1 cm sleeve of muscle around the fibulaand vascular pedicle. In contrast, Buncke7 onlyretained 5 mm to 1 cm of muscle. Several yearslater, Hidalgo5 described a technique in which a 1–2 mm rim of muscle was retained on the fibula whendissecting off the anterior and lateral compartmentmuscles, but a much thicker cuff of flexor hallucislongus (FHL) and tibialis posterior (TP) retained onthe posterio-medial aspect of the fibula and theperoneal vessels. Masquelet9 followed the sametechnique as Hidalgo,5 believing it is necessary toretain a protective cuff of muscle posteromediallyto ensure blood supply. Wei4 and Gilbert2 describeleaving a ‘minimal muscle cuff’ but the amount leftis difficult to determine, as it is not clear from theirdescriptions how much muscle is retained on thefibula itself. More recent reports10,11 define a thinmuscle cuff as being between 1 and 10 mm thick,highlighting the fact that, as yet, there is nostandard technique for free fibular dissection.

A muscle-sparing technique for dissection of thefibula free flap is described, which has beenemployed in this unit since 1997. The peronealvessels are skeletonised and only a very thin (1–2 mm) cuff of muscle is retained on the fibula itself(Fig. 1). This retrospective review of 47 free fibulaflaps performed between 1997 and March 2002 wasundertaken to evaluate the results of thistechnique.

Technique (Figs. 1–6)

The patient is positioned in the decubitus positionwith a sandbag under the ipsilateral hip. Dissectionis under tourniquet control using loupe magnifi-cation, and starts with the standard lateralapproach. The skin island is raised first withdissection along the lateral intermuscular septumproceeding to the posterolateral border of thefibula. The lateral compartment muscles areretracted anteromedially and electrocautery isthen used to dissect these muscles from theattachment to the lateral border of the fibula

leaving at most a 1–2 mm cuff (Fig. 2). The septumbetween the lateral and anterior leg compartmentsis divided close to the fibula, and the anteriorcompartment muscles are retracted anterome-dially with a Langenbeck retractor. Extensor hallu-cis longus and extensor digitorum longus are thenseparated from the fibula using electrocautery,again leaving a 1–2 mm cuff of muscle. Havingidentified the anterior tibial vessels and deepperoneal nerve, these structures are protected byretracting them medially with a Langenbeck retrac-tor. The interosseous membrane is then dividedlongitudinally close to the fibula to expose the TPmuscle (Fig. 3).

The distal osteotomy is performed with anoscillating saw usually 5 cm above the lateralmalleolus. The peroneal vessels are identifieddistally, ligated and then transected. The fibula isthen retracted laterally and slightly posteriorly atthe distal osteotomy site using a blunt bone hook. Aflat metal retractor can then be placed in theavascular plane between TP, FHL and the peronealvessels. This retractor lies on the vessels andprotects the pedicle while the overlying muscle iscompletely dissected off using a combination ofelectrocautery and scissors (Fig. 4). Muscle per-forators are identified and divided after clipping.After a variable distance, when vision becomesrestricted, the proximal osteotomy is performed.This is usually about 5 cm below the fibula head.The vessels can then be traced upwards to theirjunction with the posterior tibial artery and vein(Fig. 5).

The posterior dissection is then performed andstarts with blunt mobilisation of the soleus off theFHL. This exposes the FHL, which is then dissectedoff the posterior aspect of the fibula leaving a 2–3 mm muscle cuff. The dissection continues medi-ally with skeletonisation of the pedicle. Thus, thepedicle is skeletonised anteriorly and medially andonly a 2–3 mm rim of muscle is retained posterior tothe fibula (Fig. 6). The pedicle is left intact whilethe fibula is cut, shaped and rigidly fixed with platesif a mandible reconstruction is intended.

The recipient and donor vessels are flushed withheparin saline solution (100 u/ml) intra-operativelyand a Dextran 40 infusion is used for four days post-operatively in all cases. Patients are usuallyreturned to the ward post-operatively.

Method

A retrospective review was undertaken of all freefibula flaps performed at Groote Schuur Hospital,

The fibula free flap: advantages of the muscle sparing technique 389

and two other private hospitals in Cape Town, SouthAfrica, using the technique described above. Allpatients operated between January 1997 and March2002 were included in the study. Patient dataevaluated the age and sex of the patient as well asthe indication for the free fibula transfer. Thesuccess of the technique was analysed. Thecomplications were divided into those whichoccurred at the donor site and those at therecipient site.

All the free fibula dissections and microanasto-moses were performed by the senior plastic surgeonin the head and neck unit (van Zyl).

Results

Forty-seven free fibula flaps were performed in 43patients. The average age was 47.7 years (range 13–82 years) and 28 (65.1%) were men. One patientdied 5 days post-operatively and was excludedfrom the study. There was an 89.1% (41/46) flapsurvival rate at an average follow-up of 15 months.

Two flaps failed as a result of arterial thrombosis,one secondary to venous thrombosis and one as aresult of severe sepsis. The cause of failure in oneflap could not be determined.

Thirty-nine patients underwent free fibula trans-fer for mandible reconstruction. The majority(37/39) of the mandibular defects were the resultof resections for malignancy. One patient requiredreconstruction following trauma and another afterdebridement for mandibular osteomyelitis. Therewere four flap failures in this group (10.3%), two ofwhich occurred in the same patient. Subsequentsalvage was achieved in this patient with a free iliaccrest transfer.

Eight patients required a free fibula flap toreconstruct other defects. Six patients underwentreconstruction of their maxillae following resectionfor malignancy, while two patients had long bonereconstructions following trauma. One flap failureoccurred in a patient undergoing maxillary recon-struction, the cause of which could not be deter-mined. Salvage was achieved with a repeat freefibula flap.

Fig. 1 Dissection of the pedicle with anterior skeletonisation of the peroneal vessels.

R.G. Graham et al.390

Early complications related to the donor sitewere few. Minor wound infections occurred in sevenpatients (15.2%), which resulted in partial skin graftloss in four patients. Repeat split skin grafting wasnecessary in one of the four patients.

There were also few major recipient sitecomplications. There were 2 (4.3%) post-operative

haematomas that required evacuation in theatreand both flaps survived. Two recipient site woundinfections occurred, one of which was minor andresponded to antibiotics. The other infectionresulted in loss of the flap. Plate exposureoccurred in 17.5% of patients. Exposed plateswere removed.

Fig. 2 Dissection of lateral and anterior compartment muscles leaving a 1–2 mm cuff.

Fig. 3 Division of the interosseous membrane to expose tibialis posterior.

The fibula free flap: advantages of the muscle sparing technique 391

Discussion

There are many potential advantages to limitingmuscle dissection, but these can only be justified ifthe blood supply to the fibula is not compromised.This study suggests that skeletonising the pedicle

does not compromise the blood supply to the fibulaso long as the periosteum remains intact. This isbased on evidence that there were no cases ofosteitis or sequestrum (at an average follow-up of15 months) and only one case of severesepsis occurred in a patient having a mandible

Fig. 4 Dissection of tibialis posterior off the pedicle. Tip of the scissors is in the avascular plane between tibialisposterior and the pedicle.

Fig. 5 Tracing the pedicle upwards to the junction with the posterior tibial artery and vein.

R.G. Graham et al.392

reconstruction, which is performed in a contami-nated field. The fact that so many mandiblereconstructions were undertaken without sepsisoccurring suggests that the blood supply to theperiosteum was maintained. This concurs with Weiet al.4 who also recommended minimising theamount of muscle harvested.

There are several advantages to limiting the sizeof the muscle cuff. There is less bleeding from thesmaller muscle bulk thus reducing post-operativehaematoma formation at the recipient site. Thehaematoma rate of 4.4% (despite using Dextran 40)compares favourably with that of Coghlan andTownsend14 who reported a 25% incidence of post-operative haematoma formation at the recipientsite. It must be noted, however, that all their freefibula flaps were performed for long bone recon-structions and were not used in head and neckreconstructions. Also, less muscle harvest reducesthe dead space at the donor site, which decreasesthe risk of haematoma formation at the donor site.In this study, there were no donor site haematomas.

There may also be less risk of damage to theproximal part of the pedicle (i.e. where the pediclepasses from the fibula and crosses to join thetrifurcation) when compared to other techniqueswhere more muscle bulk is harvested posteriorly.Hidalgo5 describes a technique whereby the TP isdivided in the median raphe (thus retaining half thismuscle with the fibula) and notes that the pedicle is

at risk of injury proximally, where it crosses frommedial to lateral. However, by dissecting TP off thefibula, and thereby directly seeing and skeletonisingthe vessels, this danger point is avoided and anypotential damage to the posterior tibial artery andnerve is averted. Additionally, skeletonisation ofthe pedicle facilitates the identification of musclebranches, which are then simply ligated, and theproblems of vessel avulsion from the pedicle orvessel retraction into the muscle mass are mini-mised, thus reducing the potential for blood lossand haematoma formation.

Harvesting less muscle bulk with the flapincreases the relative length of the lateral cruralseptum and facilitates folding of the skin paddlearound the fibula. This is particularly useful inmandible reconstruction where a large amount ofbone with multiple osteotomies is required, thepedicle is often short and skin is required to closethe defect in the floor of mouth. In this situation theskin paddle may be passed either superior orinferior to the fibula to reach the floor of mouth.

Another benefit of harvesting less muscle is thatdonor limb strength is preserved. This aspect wasnot adequately assessed in this study, but there area number of studies10 –13 demonstrating that thedegree of foot weakness is directly related to theamount of muscle harvested. Shpitzer et al.10

evaluated 41 donor limbs in which the fibular flapswere harvested with a thin (1–2 mm) muscular cuff.

Fig. 6 View of posterior dissection of fibula showing skeletonisation of the pedicle with 2–3 mm rim of muscle on thefibula.

The fibula free flap: advantages of the muscle sparing technique 393

They found that 10% of patients had weakness ofextension and flexion when compared with theunoperated leg and that an additional 12% ofpatients had weakness of flexion of the big toe.Vail and Urbaniak12 performed an extraperiostealdissection leaving only a thin muscle cuff in 247fibular dissections and reported a 10% incidence ofextensor and FHL muscle weakness at long-termfollow-up. In contrast, Anthony et al.11 evaluated11 patients with isokinetic testing and comparedthe strength of the normal leg with that of thedonor leg. Their technique of dissection included athick muscle cuff of about 1 cm. In 100% of theirpatients, the strength of the donor leg was muchlower than that of the unoperated control leg. Tanget al.13 used a dissection technique similar to thatof Hidalgo5 and Masquelet9 where a thin cuff ofmuscle is left anteriorly and a thicker 1 cm cuff isleft on the fibular posteromedially. They found a43% incidence of weakness in big toe extension anda 29% incidence of weakness of flexion of the bigtoe.

There is a wide variation in the amount of musclethat surgeons harvest with a free fibula flap with thereported thickness varying between 1–2 mm andabout 1 cm. A dissection method, which leaves avery thin cuff of muscle on the fibula andskeletonises the pedicle, offers a number ofadvantages. These include a reduced recipientand donor site haematoma rate, safer dissectionof the pedicle, easier folding of the skin paddlearound the fibula and greater donor leg strength. Itappears that harvesting less muscle does notcompromise the blood supply as long as all theperiosteum is left intact.

References

1. Taylor GI, Miller GDH, Ham FJ. The free vascularized bonegraft. A clinical extension of microvascular techniques. PlastReconstr Surg 1975;55:533—44.

2. Gilbert A. Free transfer of the fibular shaft. Int J Microsurg1979;1:100—5.

3. Chen ZW, Yan W. The study and clinical application of theosteocutaneous flap of fibula. Microsurgery 1983;4:11—16.

4. Wei FC, Chen HC, Chuang CC, et al. Fibular osteoseptocu-taneous flap: anatomic study and clinical application. PlastReconstr Surg 1986;78:191—9.

5. Hidalgo DA. Fibula free flap: a new method of mandiblereconstruction. Plast Reconstr Surg 1989;84:71—9.

6. Wei FC, Seah CS, Tsai YC, et al. Fibula osteoseptocutaneousflap for reconstruction of composite mandibular defects.Plast Reconstr Surg 1994;93:294—306.

7. Buncke HJ. Fibular free flaps. Microsurgery: transplan-tation—replantation: an atlas/text, Philadelphia: Lea andFebiger; 1992. p. 328—44.

8. Flemming AFS, Brough MD, Evans ND, et al. Mandibularreconstruction using vascularised fibula. Br J Plast Surg1990;43:403—9.

9. Masquelet AC, Gilbert A. An atlas of flaps of the muscu-loskeletal system. London: Martin Dunitz; 2001. p. 188—96.

10. Shpitzer T, Neligan P, Boyd B, Gullane P, Gur E, Freeman J.Leg morbidity and function following fibular free flapharvest. Ann Plast Surg 1997;38:460—4.

11. Anthony JP, Rawnsley JD, Benhaim P, et al. Donor legmorbidity and function after free flap mandible reconstruc-tion. Plast Reconstr Surg 1995;96:146—52.

12. Vail TP, Urbaniak JR. Donor-site morbidity with use ofvascularized autogenous fibular grafts. J Bone Joint Surg1996;78A:204—11.

13. Tang CL, Mahoney JL, McKee MD, Richards RR, Waddell JP,Louie B, Yoo D. Donor site morbidity following vascularizedfibular grafting. Microsurgery 1998;18:383—6.

14. Coghlan BA, Townsend PLG. The morbidity of the freevascularized fibula flap. Br J Plast Surg 1993;46:466—9.

R.G. Graham et al.394