The reconstruction of post-traumatic osseous defects of the forearm that are complicated by poorly vascularized, contracted, or deficient overlying soft tissues is challenging3,14,21. Operative procedures involving non-vascularized autogenous or allogenic bone grafts may be unsuitable because their success depends on the ability of the surrounding soft tissues both to withstand the operative manipulation and to revascularize the transplanted bone graft3,5,7,21. Infection, either active or quiescent, is also a major threat to the success of allogenic bone grafts13,29.
Ilizarov procedures have provided alternatives for the reconstruction of composite skeletal and soft-tissue defects; however, such procedures are complex and lengthy and have been associated with many complications10.
The vascularized osteoseptocutaneous fibular autogenous graft, described initially in 1983 by Chen and Yan6 as well as by Yoshimura et al.32, is particularly useful for the operative reconstruction of complex, composite defects of the forearm. Use of this graft enables the surgeon to transplant, during a single operative procedure, a diaphyseal segment of bone with dimensions similar to those of the radius or the ulna along with a vascularized soft-tissue envelope1,11,15,17-19,25,26. Both the bone and the soft-tissue envelope are supported by a readily accessible vascular pedicle2,4,22,27.
The purpose of the present report is to describe our experience with nine vascularized osteoseptocutaneous fibular autogenous grafts that were used for the reconstruction of complex, combined skeletal and soft-tissue defects of the forearm.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
†Department of Orthopaedics, WAC-527, Massachusetts General Hospital, Boston, Massachusetts 02114.
‡Division of Orthopaedic Surgery, Department of Surgery, Duke University Medical Center, Durham, North Carolina 27710.
§Centro Médico de Caracas, Caracas, Venezuela.
The study cohort comprised nine male patients who were an average of thirty-two years old (range, twenty-one to forty-two years old) (Table I). Before the injury, three patients were businessmen; one each was a student, a laborer, or a cattle-ranch supervisor; and three were unemployed. Six right, dominant limbs and three left, non-dominant limbs were involved.
Three patients sustained the injury as the result of a gunshot wound and two each, as the result of a motor-vehicle accident, a fall from a height, or a machinery-related accident. Seven patients sustained a combined skeletal and soft-tissue injury initially, whereas in the remaining two patients the soft-tissue defect developed after the operative treatment of an isolated diaphyseal fracture of the radius. All nine patients had involvement of the radius, and five had a concomitant fracture of the ulna. Four patients had disruption of the distal radio-ulnar joint, and one had associated carpal fractures. In addition, the extensor tendon was disrupted in one patient and the median, ulnar, and radial nerve was injured in one patient each (Table I).
The initial treatment of the skeletal injury consisted of external fixation alone for five patients, internal fixation alone for two patients, a combination of external and internal fixation for one patient, and pins-and-plaster immobilization alone for one patient. A split-thickness skin graft was used to achieve closure of the wound in three patients, and a split-thickness skin graft combined with a free vascularized scapular flap was used in one. Six patients had a deep osseous infection in the involved forearm; four of these patients had active drainage from the wound when they were first seen by us.
Operative reconstruction with use of a vascularized osteoseptocutaneous fibular autogenous graft was performed at our institutions three to twelve months after the initial injury. The four patients who had clinical evidence of an active osseous infection were managed with repeated débridement of the wound along with parenteral administration of antibiotics before the definitive procedure was performed. The radial defect that was observed at the time of presentation or that resulted from operative débridement measured an average of seven centimeters (range, four to eleven centimeters) in length. In four patients, the defect was associated with disruption of the distal radio-ulnar joint and shortening of the radius in relation to the ulna. The overlying soft tissues were contracted in every patient. In two patients, a split-thickness skin graft had become adherent to the underlying radius.
Arteriograms of the donor leg and the recipient forearm were made preoperatively for each patient. In two patients, the hand was found to be supplied by only one artery.
Operative Technique
The operative procedures were performed with the patient supine, the recipient upper extremity on a hand-table, and the donor lower extremity flexed at the hip. General endotracheal anesthesia was used. The initial exposures were performed simultaneously, under pneumatic tourniquet control, by two operative teams.
The operative technique that was used to obtain the osteoseptocutaneous fibular composite graft is based on the work of Chen and Yan6, Yoshimura et al.32, and Tonkin et al.25. The cutaneous island, which is centered with its long axis in line with the posterior margin of the fibula, is located at or about the junction of the middle and distal thirds of the leg, where the perforating blood vessels to the overlying skin are located most reliably.
After exsanguination of the leg and inflation of the pneumatic tourniquet, the anterior portion of the cutaneous paddle, along with its underlying fascia, is elevated in the subfascial plane from the muscles of the anterior compartment as well as from the peroneus longus and peroneus brevis muscles. As the posterior crural septum is identified, one begins to see small, perforating vessels passing up to the skin in the anterior aspect of the septum. At this point, the peroneal muscles can be elevated from the fibula with care being taken to preserve the periosteum of that bone. Particular care is taken to identify and to protect the superficial peroneal nerve at the proximal extent of the operative wound. The anterior crural septum is divided and the extensor hallucis longus, extensor digitorum longus, and peroneus tertius are elevated from the interosseous membrane.
Next, an incision is made through the skin, subcutaneous fat, and deep fascia along the planned posterior margin of the cutaneous paddle; the incision meets the anterior half of the paddle at its junction with the posterior crural septum. The posterior margin of the cutaneous paddle then is elevated through the subfascial plane overlying the soleus and flexor hallucis longus muscles.
Dissection through the fibers of the flexor hallucis longus muscle at the distal aspect of the operative exposure reveals the peroneal artery and vein on the posteromedial aspect of the fibula. By this time, the required length of the fibular segment will have been determined by the operative team that has exposed the defect in the forearm. The fibula is osteotomized both distally and proximally with use of either a pneumatic saw or a Gigli saw. The fibular segment then can be rotated to facilitate exposure of the peroneal vessels, the remaining muscular attachments, and the interosseous membrane.
The peroneal vessels are ligated at the site of the distal osteotomy, and the interosseous membrane is divided from distal to proximal, leaving only a small cuff of flexor hallucis longus muscle attached to the fibular segment in addition to the peroneal artery and its venae comitantes.
At the proximal extent of the fibular osteotomy, the peroneal vessels are traced to their junction with the posterior tibial vessels, leaving the fibular osteoseptocutaneous graft connected to the leg only by its vascular pedicle. The pneumatic tourniquet is deflated, and hemostasis is obtained. Before the peroneal vessels are divided from their origin, there should be visual confirmation of blood flow at the cut ends of the fibular segment and in the overlying skin flap. Once the patency of the arterial anatomy of the flap has been ensured, the peroneal vessels can be divided from their origin off the tibioperoneal trunk and the composite graft can be transferred to the forearm.
The composite graft is inserted into the defect in the forearm, with care being taken to place the peroneal vessels in an optimum position for the performance of the microvascular anastomoses to the recipient vessels. The method of skeletal fixation is based in part on the unique requirements of the particular defect, as will be described. With use of the operating microscope and microvascular techniques, the segment of the peroneal artery that is attached to the fibular autogenous graft is anastomosed to either the radial or the ulnar artery; in the present study, an end-to-end anastomosis was performed in seven patients and an end-to-side anastomosis was performed in two. After completion of the microvascular anastomoses, the septocutaneous portion of the graft is sutured into the overlying soft-tissue defect and the wound is closed over a suction drain.
Postoperative Treatment
Postoperatively, the patients were managed with anticoagulation therapy (either low-molecular-weight dextran or buffered aspirin) and parenteral administration of antibiotics. The cutaneous portion of the composite graft served as a visual monitor of the patency of the microvascular anastomosis.
The forearm was protected with either an above-the-elbow cast or a below-the-elbow splint. Immobilization was discontinued when there was radiographic evidence of union. The donor limb was supported in a posterior splint until the wound had healed and then was protected in a below-the-knee ankle-foot plastic orthosis for an additional six weeks.
The patients were followed throughout the course of treatment by the surgeon who had performed the operation. The follow-up evaluation consisted of subjective and objective assessments of functional capacity as well as the documentation of any problems associated with either the reconstructed forearm or the donor limb. Radiographs of the forearm were made at the time of the most recent follow-up examination.
CASE 9. A twenty-three-year-old student sustained a severe skeletal and soft-tissue injury of the left, non-dominant forearm and elbow in a motor-vehicle accident. The skin and subcutaneous tissue on the dorsal aspect of the forearm were degloved. The skeletal injuries included open segmental fractures of both bones of the forearm as well as disruption of the radiocapitellar and proximal and distal radio-ulnar joints (Fig. 1-A). There also was a neurovascular deficit involving the muscles innervated by the posterior interosseous nerve.
After extensive skeletal and soft-tissue débridement, the fractures were stabilized with intramedullary rods. The soft-tissue defect was closed secondarily with a free vascularized scapular flap and a split-thickness skin graft (Fig. 1-B). The development of a deep osseous infection necessitated sequential débridements, removal of the intramedullary rods, and external fixation. Ultimately, this resulted in a 6.5-centimeter segmental defect of the radius (Figs. 1-C and 1-D), which was filled with a spacer of antibiotic-impregnated polymethylmethacrylate.
Six months after the injury, the wound was without drainage and the function of the posterior interosseous nerve had returned. The segmental defect of the radius and the overlying contracted soft-tissue envelope were reconstructed with a vascularized osteoseptocutaneous fibular graft (Fig. 1-E). The skeletal component of the graft was seven centimeters long, and the cutaneous component measured fourteen by 7.5 centimeters. The transferred fibular segment was stabilized with a ten-hole plate, with three screws placed at both the proximal and the distal graft-host bone junction. The donor site in the leg was closed with a split-thickness skin graft.
Postoperatively, the development of venous thrombosis necessitated the restoration of venous outflow with an autogenous vein graft. No additional complications developed.
After two years of follow-up, the transferred fibular segment was well healed to the radius (Fig. 1-F). A proximal radio-ulnar synostosis had developed, with the forearm in approximately 30 degrees of pronation. The range of motion of the wrist was 40 degrees of extension to 45 degrees of flexion. The grip strength was forty-two pounds (nineteen kilograms) on the involved side, compared with 115 pounds (fifty-two kilograms) on the contralateral side (Fig. 1-G). The patient had no symptoms referable to the donor leg.
In four of the nine patients, the transferred fibular segment was fixed with a single long plate that bridged over the graft and was secured to the recipient bone both proximally and distally. In the other five, individual plates and screws—either alone or in combination with Kirschner wires—were used to secure the fibular segment to the recipient bone both proximally and distally. Two patients (Cases 1 and 2) had a resection of the distal part of the ulna (a Darrach procedure) in addition to the reconstruction of the radius. Two patients (Cases 2 and 5) had a concomitant total arthrodesis of the wrist.
The septocutaneous portion of the composite graft measured an average of 11.8 by 5.9 centimeters (range, seven by four centimeters to twenty by ten centimeters). Primary closure of the defect at the donor site was possible in only three of the nine patients; in the remaining six patients, a split-thickness skin graft was used (Table II).
Two patients (Cases 4 and 9) had venous thrombosis in the immediate postoperative period (Table III). In both patients, the problem was identified at the site of the venous reconstruction and the anastomosis was revised with use of an autogenous interpositional vein graft obtained from the lower limb. One other patient (Case 8) had partial loss of a split-thickness skin graft that had been used to cover the donor site in the leg. The site healed with changes of the dressing and observation; no additional operative procedure was necessary. No patient had any loss of the cutaneous portion of the fibular osteoseptocutaneous graft.
One patient (Case 8) had a second operative procedure six months postoperatively to correct displacement of the graft as well as an angular deformity that had developed secondary to a non-union at the graft-host bone junction (Table III).
The patients were followed for an average of twenty-four months (range, twelve to thirty-six months) after the reconstructive procedure. Six patients returned to their pre-injury occupation. No patient noted any symptoms suggestive of recurrent infection. All but one of the patients had radiographic evidence of osseous union at both the proximal and the distal graft-host bone junction. The remaining patient (Case 1) had a non-union of the distal junction; however, no symptoms were referable to the site and no operative intervention was necessary. None of the fibular autogenous grafts fractured, and no patient had radiographic evidence of osseous resorption.
No patient had pain or functional disability involving the donor leg at the time of the most recent follow-up examination, although it should be noted that all patients noted discomfort with exertion for some time after the operative procedure. No patient noted any problems with the great toe secondary to the operative dissection of the flexor hallucis longus. The patient (Case 8) who had had partial loss of a split-thickness skin graft noted no additional sequelae or symptoms referable to the problem.
When the patient who had a proximal radio-ulnar synostosis is excluded, the average pronation of the reconstructed forearm was 56 degrees (range, 10 to 70 degrees) and the average supination was 38 degrees (range, 5 to 70 degrees). When the two patients who had a total arthrodesis of the wrist are excluded, the average extension of the wrist was 44 degrees (range, 20 to 80 degrees) and the average flexion was 41 degrees (range, 30 to 80 degrees). The grip strength on the involved side, measured for four patients, averaged 70 per cent of that on the contralateral side.
The injuries described in the present study involved not only the loss of skeletal continuity but also the contraction of overlying soft tissues and the distortion of the contours of the forearm. Although a number of methods for the skeletal reconstruction of the forearm have been described3,5,7,8,14,20,21,30,31, these would have been of limited value for the patients in the present series because of the associated soft-tissue deficiencies.
Operative approaches that address each component of the problem independently can be successful, but there are some potential disadvantages. For example, although resurfacing the forearm with either an axial-pattern groin flap or a revascularized free muscle transfer offers the possibility of enhancing the vascularity of the wound and eliminating the soft-tissue deficit, the need for subsequent elevation of the flap in order to obtain sufficient space for the skeletal reconstruction may expose the vascular pedicle of the flap to injury.
A number of investigators have studied and defined the vascular anatomy of the fibular osteoseptocutaneous graft. Wei et al.27, in a study of twenty cadaveric specimens, identified four to seven cutaneous vascular branches arising from the peroneal artery, which passes through the posterior crural septum to supply the skin over the lateral aspect of the distal part of the leg. The presence of these perforating vessels subsequently was confirmed by other investigators2,4,11,26. Beppu et al.2 suggested that the most reliable septocutaneous perforating vessels are located in the distal two-thirds of the leg and recommended that the cutaneous portion of the composite graft be obtained from that region.
Use of the fibula for the skeletal component of the composite graft allows transfer of a segment of diaphyseal bone that is structurally similar to the radius and that is of sufficient length for the reconstruction of most skeletal defects in the forearm23,28-31. In addition, the peroneal vessels are of sufficiently large diameter to facilitate the microvascular repairs. The morbidity at the donor site subsequent to removal of the middle segment of the fibula was relatively limited in our series, and our findings are consistent with those of other authors who have reported on the use of vascularized fibular grafts9,12,16,18,24.
The fact that overlying skin was attached to the fibular segment only by the thin, mobile posterior crural septum contributed greatly to the achievement of stable wound coverage, even when the soft-tissue defect was not directly over the skeletal defect. Although some authors have included the soleus muscle in the transfer1, the bulk of the muscle decreases the mobility of the overlying skin and limits the ability of the cutaneous flap to cover some complex wounds27. One additional advantage of the composite osteoseptocutaneous graft is that the skin serves as a visible monitor of the adequacy of the circulation, thereby reducing the need for sophisticated monitoring devices.
Given the complex nature of the injuries in the present series, it would be difficult to compare the functional outcomes for our patients with those for patients in previous studies of non-united fractures of the forearm. The fact that few patients regained nearly normal function reflects the complexity of the forearm articulation, which is dependent not only on the two bones of the forearm but also on the proximal and distal radio-ulnar joints and the interosseous membrane. Several previous studies in which the fibular osteoseptocutaneous graft was used for the reconstruction of complex injuries of the forearm also demonstrated high rates of union with successful coverage of associated soft-tissue defects6,11,25,27, but those studies included little information regarding the functional outcome.