Microsurgery 2005; 25: 538-542

Summary

The free fibula flap is the microsurgeon’s workhorse for the reconstruction of osseous or osteocutaneous defects. Donor-site morbidity of this flap is reported to occur infrequently, and is generally considered minor and transient. We present the case histories of three patients with necrosis at the fibula flap donor site to stress the risks and explain the possible mechanisms of such severe complications. The small risk of debilitating donor-site necrosis should be considered and discussed preoperatively with the patient.

Introduction

The fibula free flap has become the microsurgeon’s workhorse for the reconstruction of osseus or osteocutaneous defects of the head and neck region,(1-3) the extremities,(3-5) and the trunk.(3,6) Flap survival and receptor site function after fibula flap transplantation have been evaluated extensively, but donor-site morbidity has been assessed less often. Donor-site morbidity includes wound healing disturbance and cellulitis,(1-3) transient peroneal sensory loss or cold intolerance,(1-4,7,8) motor weakness of lower leg muscles,(1,4-9) pain or occasional cramp,(1-4,7-9) impingement of the peroneal muscle,(10) edema after prolonged standing,(2-4,8) and ugly scarring.(4) Even though valgus deformity, instability, or stiffness of the ankle,(1,2,4,7-9,11) and walking restriction or gait disturbances,(2,3,5,6) can be considered more serious donor-site complications, most of the complications mentioned above are generally rated as acceptable by patients.(1,2,4,6,8,9)

Major donor-site complications with extensive wound breakdown or necrosis are generally considered to occur infrequently after fibula free-flap transfer. To date, only five cases of such debilitating complications due to necrosis have been reported to lead to deep necrotectomy. Among these five were two cases of acute compartment syndrome,(3,12) two cases of late necrosis,(13) and one case of necrotizing infection.(4) To stress that such donor-site necrosis can lead to severe postoperative disability, we present three more patients who suffered from extensive necrosis of the lower leg, and we discuss how to prevent these complications.

Case reports Patient 1

A 36-year-old woman underwent commando resection and supraomohyoidal lymphadenectomy for T4N0M0 squamous-cell carcinoma of the floor of the mouth and left hemimandible. Her medical history included systemic lupus erythematosis, tobacco abuse, and a mild myocardial infarct sustained 2 months prior to surgery. Because preoperative angiography showed no acquired or congenital anomalies of the crural arteries, the defect was reconstructed using a fibula free flap, including a significant part of the lateral hemisoleus muscle and a skin segment of 6 x 14 cm (Fig.1). Per our routine, no tourniquet was used during surgery. The donor defect was closed primarily over a suction drain after loose approximation of the muscles of the lateral compartment. No attempt was made to approximate the lateral fascia. The lower leg and foot were splinted on a dorsal cast to prevent pes equinus, and the patient was kept on subcutaneous heparin prophylaxis. No pressure or tensor bandage was applied to the leg, but the patient was required to keep the leg elevated until mobilization, one week postoperatively.

On the first postoperative day, however, the patient complained of pain in the donor leg and sensory loss of the dorsum of the foot. On physical examination, there was functional loss of the extensor digitorum muscle, decreased capillary refill of the foot, and lack of a Dopptone signal over the dorsal tibial artery. These symptoms being indicative of compartment syndrome, the cast and skin sutures were removed to reduce tension in the calf.(14,15) Angiography subsequently showed arterial filling of both tibial arteries (Fig.2). Still, necrosis of the extensor digitorum muscle and parts of both peroneal and long flexor hallucis muscles occurred, and repeated necrotectomy had to be performed surgically, on postoperative days 11 and 22. After this, the crural wound healed except for its most distal part (Fig.3). Six months after surgery, part of the distal fibula remnant still protruded from the wound, and X-ray examination could not exclude osteomyelitis. Consequently, this part was resected, and the remaining bone was covered during a fourth and final surgical procedure. When last seen at 44 months of follow-up, the patient was free of recurrence, and the neomandible was functional. However, her gait and walking distance were severely impaired.

Figure 1. Design of fibula free flap used on patient 1. Flap included significant part of lateral hemisoleus muscle, and skin segment measuring 6 x 14 cm.

Figure 2. Postoperative angiography of patient 1 showed arterial filling of both tibial arteries.

Figure 3. Six months after fibula free-flap transplantation in patient 1, distal part of crural wound had not closed, and protruding part of distal fibula remnant was resected.

Patient 2

A 74-year-old woman with a history of alcohol and tobacco abuse underwent commando resection and modified radical lymphadenectomy for a T3N0M0 squamous-cell carcinoma of the floor of the mouth extending to the left hemimandible. As the preoperative angiography showed no vascular anomalies, the defect was reconstructed using an osteocutaneous free fibula flap with a skin paddle of 5 x 12 cm.

As no tourniquet was used intraoperatively, it could be confirmed that the manipulated muscles remained well-vascularized throughout the entire procedure. Following approximation of the donor-site muscles, the lower leg skin defect was covered with a split-thickness skin graft to avoid undue pressure, and the lower leg and foot were splinted to prevent pes equinus. Again, no tensor bandage was applied, and the patient was required to keep the leg elevated until mobilization, 1 week postoperatively.

The postoperative course was uneventful up to day 15, when edema occurred in the donor leg. Even though wound cultures remained negative for necrotizing species, subsequent necrosis of the skin graft and surrounding skin and subcutaneous tissue necessitated necrotectomies on postoperative days 23 and 37 (Fig.4). The resulting skin defect measured 10 x 25 cm, and skin grafts were applied secondarily. Using a vacuum-assisted closure system, the wound completely healed by 10 weeks after primary surgery. At 24 months of follow-up, the patient was free of recurrence, and her neomandible functioned adequately. No remaining functional loss of the donor leg was observed.

Figure 4. In patient 2, skin defect resulting from initial surgery, postoperative ‘‘pseudo’’-compartment syndrome, and subsequent necrotectomy measured 10 x 25 cm.

Patient 3

A 56-year-old diabetic man underwent segmental mandibulectomy for residual ameloblastoma, after local resections had twice been attempted elsewhere. Because the preoperative angiography showed no atherosclerosis or vascular anomalies, the left hemimandible and gingiva defect was reconstructed with a fibula osteocutaneous free flap. The skin paddle measured 6 x 11 cm, and was raised on one mid-lower leg muscular perforator, as no adequate septocutanous vessels were found. This perforator had a long intramuscular course and a peroneal origin near the crural trifurcation, and several muscular branches originating from the peroneal artery had to be clipped to allow for its dissection. Again, no tourniquet was used, allowing for proper hemostasis and assessment of the muscles that were manipulated during flap dissection. Following microsurgical flap transfer, the lateral lower leg muscles were loosely approximated, and no attempt was made to close the fascia. The skin was partly closed primarily, and partly by use of a skin graft. A splint was applied to prevent pes equinus, and the patient was kept in bed and on heparin prophylaxis for 1 week. No tensor bandage was applied.

The postoperative course was uneventful until the patient became febrile and complained of pain in the donor leg on postoperative day 11. Wound culture was positive for beta-hemolytic Streptococcus, and antibiotic therapy was initiated. At surgical exploration, the long peroneal muscle was found to have necrosed. Following necrotectomy, it took 10 weeks of extensive conservative treatment, including application of a vacuum-assisted closure system, for the wound to heal. When last seen at 24 months of follow-up, the patient was free of recurrence of osteoblastoma, and his neomandible was functional, but exorotation and plantar flexion of the foot were compromised.

Discussion

Fibula flap donor-site morbidity and its consequences for daily life function are generally considered to be rare and minor. Still, we observed severe donor-site necrosis in 3 of 74 patients in whom a fibula flap transplantation was performed in the Netherlands Cancer Institute from January 1985 until December 2003. The mechanism of necrosis may differ. Our first patient represented an obvious case of acute compartment syndrome, whereas a delayed ‘‘pseudo’’-compartment syndrome and a necrotizing infection were the probable causes in our second and third patient, respectively.

Compartment syndrome is allegedly a rare consequence of free fibula flaps, and has been reported only twice to date.(3,12) This syndrome occurs when intra-compartment pressure builds up in cases where the fascia surrounding the compartment is intact. Obviously, this is no longer true in cases where the complete release of the wound may be expected to alleviate the ‘‘skin compartment syndrome.’’

In our first patient, the width of the skin paddle was restricted to the 6-cm maximum. Moreover, part of the lateral hemisoleus muscle was harvested along with the flap, and this may be expected to reduce the circumference of the leg and hence the closure tension. Even after adequate response to symptoms presenting on the first postoperative day,(14,15,17) necrosis of some muscles of the anterior, lateral, and deep compartments occurred, and hence we feel that avoidance of primary skin closure might have prevented the compartment syndrome.(12) More careful patient selection might not have prevented the occurrence of compartment syndrome in this patient. Such occurrence has been associated with age under 35 years, male sex, diabetes mellitus, obesity, hypertension, traumatic rather than surgical injury, bleeding disorders or the intake of anticoagulant drugs, low preoperative ankle-arm index, extended duration of surgery or surgery in the lithotomy position, intraoperative hypothermia or hypotension, and the postoperative use of regional nerve blockades.(18-22) Furthermore, patients suffering from peripheral vascular disease were reported to run a higher risk of compartment syndrome when a tourniquet was used during surgery.(18,20) Except for the duration of surgery, none of these predisposing factors applied to this patient.

As for our second patient, we feel that the tension that developed in the lower leg after some delay was too high, notwithstanding the use of a skin graft. The physical mechanism underlying the development of the delayed necrosis seems to have been a vicious cycle of excessive tension caused by edema leading to tissue damage and increased capillary permeability.(13) The resulting intercellular pressure and soft-tissue edema led to more tissue damage and an increase of venous, lymphatic, and arterial vessel compression, thus perpetuating the cycle. This cycle of gradually increasing ischemia has been termed ‘‘pseudo’’-compartment syndrome, and may explain why the symptoms in this patient only presented after 15 days.(13)

Paradoxically, fibulectomy used to be the treatment modality for lower leg compartment syndrome.(14,17) Still, fibulectomy may not prevent occurrence of the complication. Rather than closing the wound primarily, we currently apply a skin graft, irrespective of the width of the skin paddle included in the flap. Often, this graft can safely be resected after the period of edema has passed. Still, a delayed

‘‘pseudo’’-compartment syndrome may develop, even in cases where a skin graft was used.(13) To minimize the damage secondary to (‘‘pseudo’’-) compartment syndrome, we advise the surgeon to be postoperatively alert to clinical symptoms such as pain that seems disproportionate to the clinical situation, pain on passive stretch of the muscles, or decreased sensibility in the distribution of the nerves running through the corresponding compartment.

The necrosis at the donor site in our third patient was likely caused by the combined occurrence of beta-hemolytic Streptococcus and long-term muscle ischemia as a result of the obligatory clipping of multiple vascular branches to the long peroneal muscle. Additional pre- disposing risk factors for such local ischemia include atherosclerosis or vasculitis, endothelial damage, abnormal blood constituents due to coagulation, fibrinolytic changes or hyperviscosity, and decreased blood flow as a result of immobilization or vascular compression.(23) The risk factors for atherosclerosis are age over 65 years, diabetes mellitus, obesity, tobacco abuse, hypertension, hypercholesterolemia, and a sedentary lifestyle.(24) Our patient was a smoker, suffered from diabetes mellitus, and was expected to be immobile for some days after surgery, but the need for extensive devascularization of the peroneal muscle could not have been foreseen preoperatively. As was true for our other two patients, more careful patient selection might not have prevented the donor-site necrosis in this patient.

We conclude that donor-site morbidity after fibula transfer is generally minor, but there is a small risk of donor-site necrosis as a result of acute compartment syndrome, delayed ‘‘pseudo’’-compartment syndrome, or necrotizing infection. This risk and its predisposing factors should be preoperatively discussed with the patient, as donor-site necrosis may result in debilitating long-term morbidity.

Perioperatively, the risk of such necrosis should be minimized by closure of the wound with a skin graft, and postoperatively, necrosis should be considered upon presentation of any disproportional local symptom.

References

1 Anthony JP, Rawnsley JD, Benhaim P, Ritter EF, Sadowsky SH, Singer MI. Donor leg morbidity and function after fibula free flap mandibula reconstruction. Plast Reconstr Surg 1995;96:146-152.

2 Hidalgo DA, Rekow A. A review of 60 consecutive fibula free flap mandible reconstructions. Plast

5 Babovic S, Johnson CH, Finical SJ. Free fibula donor site morbidity: the Mayo experience with 100 consecutive harvests. J Reconstr Microsurg 2000;16:107-110.

6 Youdas JW, Wood MB, Cahalan TD, Chao EYS. A quantitative analysis of donor site morbidity after vascularized fibula transfer. J Orthop Res 1988;6:621-629.

7 Parker VT. Donor-site morbidity with use of vascularized autogenous fibular grafts. J Bone Joint Surg [Am] 1996;78:204-211.

8 Papadopulos NA, Schaff J, Bucher H, Groener R, Geishauser M, Biemer E. Donor site morbidity after harvest of free fibular flaps with an extended skin island. Aesthetic Plast Surg 2002;49:138-144.

9 Gore DR, Gardner GM, Sepic SB, Mollinger LA, Murray MP. Function following partial fibulectomy. intramuscular pressure in the human leg. J Bone Joint Surg [Am] 1994;76:1476-1481.

16 Hayden RE. Harvest and clinical application of the fibula flap - editorial review. Curr Opin Otolaryngol Head Neck Surg 1995;3:257-260.

17 Hyde GL, Peck D, Powell DC. Compartment syndromes. Early diagnosis and a bedside operation.

Am Surg 1983;49:563-568.

18 Ullrich W, Biermann E, Kienzle F, Krier C. Lagerungsschäden in Anästhesie und operativer Medizin.

Anasth Intensivmed Notfallmed Schmertzther 1997;32:4-20.

19 McQueen MM, Gaston P, Court-Brown CM. Acute compartment syndrome. Who is at risk? J Bone Joint Surg [Br] 2000;82:200-203.

20 Tison C, Perigaud C, Vrignaud S, Capelli M, Lehur PA. Syndrome bilatéral des loges de jambe aprés chirurgie colorectale en position àdouble équipe. Ann Chir 2002;127:535-538.

21 Thonse R, Ashford RU, Williams TI, Harrington P. Differences in attitudes to analgesia in postoperative limb surgery put patients at risk of compartment syndrome. Injury 2004;35:290-295 22 Modrall JG, Sadjadi J, Ali AT, Anthony T, Welborn MB 3rd, Valentine RJ, Hynan LS, Clagett GP.

Deep vein harvest: Predicting need for fasciotomy. J Vasc Surg 2004;39:387-394.

23 Cotton DWK. Ischaemia, infarction and shock. In: Underwood JCE, editor. General and systematic pathology. 2nd ed. New York: Churchill Livingstone; 1996. p 165-177.

24 Hooi JD, Kester ADM, Stoffers HEJH, Overdijk MM, van Ree JW, Knottnerus JA. Incidence of and risk factors for asymptomatic peripheral arterial occlusive disease: a longitudinal study. Am J Epidemiol 2001;153:666-672.

Chapter 4

Review:

An overview of methods for vascular

In document University of Groningen The significance of preoperative vascular mapping of donor- and acceptor vessels in free flap surgery Klein, Steven (Page 26-38)