Keywords osteosarcoma - Ewing's sarcoma - distal tibia - vascularized fibular graft - ankle
arthrodesis
Advances in neoadjuvant chemotherapy and surgical reconstruction techniques rendered
limb preservation possible in most malignant bone tumor patients.[1 ] However, limb salvage for malignant tumors around the ankle joint is still associated
with higher incidence of complications, as compared with amputation due to subcutaneous
location of the distal tibia and fibula, tendinous nature of the surrounding musculature,
and proximity of neurovascular structures. These factors lead to limitations in local
soft tissue coverage and to the difficulty to obtain safe surgical margins.[2 ] Another challenging problem is the relatively young age of patients diagnosed with
primary malignant bone tumors, the majority of which are skeletally immature, with
high functional demands and a remaining growth potential.[1 ]
Reports on intra-articular resections of malignant bone tumors of the distal tibia
and fibula in children and adolescents are limited and often comprise small number
of patients due to the low incidence of primary malignant bone tumors in this anatomic
location.[3 ] Reconstructive options include ankle arthrodesis using bone autografts or allografts,
or a mobile ankle reconstruction using endoprosthesis, each having its advantages
and disadvantages.[4 ]
[5 ]
[6 ]
The aim of this work is to retrospectively evaluate the surgical technique and functional
outcome of ankle arthrodesis using a vascularized fibular autograft in skeletally
immature bone tumor patients.
Methods
All patients diagnosed with a malignant bone tumor of the distal tibia or fibula in
the interval between 2008 and 2017 were reviewed. The inclusion criteria were patients
for whom an intra-articular resection of the distal tibia and/or fibula and reconstruction
of the resulting bone defect was done by ankle arthrodesis using a vascularized fibular
bone graft. The exclusion criteria were: (1) tumors of the distal tibia or fibula,
not reaching the articular surface, and for which an ankle joint sparing resection
was done. These were considered as intercalary tibial or fibular bone resections that
should be evaluated separate from the current analysis. (2) Tumors of the distal fibula
for which an intra-articular resection was done without reconstruction. (3) Patients
with locally advanced tumor necessitating a below knee amputation. (4) Patients with
a follow-up period of less than 12 months since limb salvage surgery.
A total of seven patients met the inclusion criteria. The mean age at presentation
was 10.6years (range, 6.7–14 years). There were three male and four female patients.
The diagnosis was osteosarcoma in five patients and Ewing's sarcoma in two patients.
All patients received neoadjuvant and adjuvant chemotherapy. None of the patients
received radiotherapy neither preoperative nor postoperative. Planning for limb salvage
was based on the evaluation of the preoperative magnetic resonance imaging (MRI) for
safe bone and soft tissue resection margins. Proximal resection margins were planned
at 2 cm away from the intramedullary extent of the tumor, while distally, the articular
cartilage of the distal tibia and/or fibula were considered as a distal safe margin.
The mean length of bone defect after resection was 13.2 cm (range, 8–24 cm). In one
patient, diagnosed with Ewing's sarcoma of the distal fibula, the soft tissue extent
of the tumor was reaching the lateral cortex of the adjacent distal tibia and an enblock
resection of the distal fibula and hemi cortex of the tibia was done.
Reconstruction was done in all patients by ankle arthrodesis using a fibular flap.
Reconstruction with ipsilateral pedicled vascularized fibular flap (centralization
of the ipsilateral fibula) : This method of reconstruction was first described by Ebeid et al.[7 ] We applied this technique in the first two patients diagnosed with a distal tibia
osteosarcoma. Through the same anterior midline approach used for the resection, the
ipsilateral fibula was dissected from the attached muscles. The peroneal vessels were
proximally dissected and mobilized with the fibular graft following proximal osteotomy
and distally the vessels were ligated to be able to mobilize and trough the graft
into talar dome after the disarticulation of the lateral malleolus from the talofibular
joint. The advantage of this technique compared with using a contralateral free vascularized
fibular graft is that it does not require microvascular reconstruction, and it avoids
the donor site morbidity in the contralateral leg.[7 ]
Reconstruction with contralateral vascularized osteocutaneous free fibular flap: In the following five patients, we changed the method of reconstruction from an ipsilateral
fibular graft to a contralateral free vascularized osteocutaneous fibular flap. The
reasons for changing the method of reconstruction were to preserve the ipsilateral
fibula and consequently facilitate proper alignment of the limb during fixation and
to provide a skin island with the bone graft to facilitate the soft tissue closure
and coverage of the defect. The average skin flap size was 7 and 3.5 cm in craniocaudal
and transverse dimensions, respectively. As in the previously described technique
of the pedicled fibular graft, the free fibular graft was troughed proximally into
medulla of the tibia and distally into talar dome. Proper adjustment of the rotation
of the graft was done in a manner so that the peroneal pedicle comes to lie in line
with the anterior tibial vessels (donor vessels). Finally, the anastomosis of the
peroneal vessels to the anterior tibial vessels was performed in an end-to-end fashion
and the skin island was used for closure of the anterior skin. At the donor site,
the skin defect at site of previously harvested skin island was left to heal by secondary
intention.
Internal fixation was used in two patients (a 3.5 reconstruction plate with proximal
screws fixed to the tibial shaft and distal screws fixed to talus). An external ring
fixator was applied in the other five patients with combination of shanz screws and
tensioned wires in the tibial shaft proximal to implanted fibula and distally into
the talus and calcaneus. All patients remained nonweight bearing until periosteal
new bone formation was radiographically observed at the proximal and distal ends of
the fibular graft, after which, partial weightbearing was allowed using two crutches
and a below knee cast. Full weight bearing without support or cast was allowed following
the radiographic evidence of complete bone union and graft hypertrophy ([Fig. 1 ]).
Fig. 1 (A ) Preoperative anteroposterior radiograph of an osteosarcoma involving the distal
tibia in a 12-year-old male patient. (B ) Postoperative lateral radiograph following the resection of the distal tibia and
ankle arthrodesis using a contralateral vascularized fibular graft and fixation by
a 4.5 reconstruction plate. Limb is immobilized in a below knee cast. (C ) Lateral radiograph 3 months postoperative showing periosteal new bone formation
at the proximal and distal junction of the fibular graft. (D ) Lateral radiograph 9 months postoperative showing hypertrophy of the graft.
Follow-up was at 6 weekly intervals in the first 6 months, 3 monthlies in the following
2 years, and 6 monthlies thereafter. Functional evaluation was done using the musculoskeletal
tumor society scoring system[8 ] and radiographic evaluation was done using anteroposterior and lateral radiographs.
To check for local recurrence, a local MRI was done at 6 months interval, in the first
5 years following surgery.
Results
The mean follow-up period was 24.5 months (range, 13–69 months).
Oncologic outcome : At diagnosis, three patients were nonmetastatic and four (two osteosarcoma and two
Ewing's sarcoma) were initially metastatic to lung. Surgical margins were negative
in all patients. All patients, with exception of one, in which tumor necrosis could
not be retrieved from medical records, had good response to chemotherapy with a tumor
necrosis above 90%. No patient had evidence of local relapse during their follow-up
period. Two patients, diagnosed with Ewing's sarcoma and lung metastases at presentation,
had complete resolution of their lung metastases during their neoadjuvant chemotherapy
and had no evidence of further lung metastases until their latest follow-up visit.
Two patients diagnosed with osteosarcoma of the distal tibia and initial lung metastases
had metastasectomy following limb salvage surgery. One patient remained free from
disease until his latest follow-up and the other patient developed subsequent inoperable
lung metastases and died from disease 22 months later. Out of the seven patients,
six remained with no evidence of disease until their latest follow-up period.
Bone union and graft hypertrophy : The mean time for radiographic evidence of periosteal new bone formation at the
proximal and distal graft-host junction was observed at 2.6 months (range, 2.3–6.5
months). Consolidation of bone union and partial graft hypertrophy was observed following
a mean time of 7.1 months (range, 4–13 months).
Functional outcome : Partial weight bearing in a below knee cast started at a mean duration of 3 months
(range, 2.3–7 months). Full weight bearing, without cast, and without support started
at a mean duration of 7.1 months (range, 4–13 months), as calculated from the date
of surgery. The mean Musculoskeletal Tumor Society (MSTS) functional score[8 ] was 84.5% (range, 73–100%). All patients had occasional pain, not requiring medication,
were able to walk unsupported with a minor limp, and had returned to their school
activity within the 1st year following surgery ([Video 1 ]).
Video 1
Clinical outcome of a 9-year-old female patient 2 years following surgery.
Complications : Nonunion occurred in one patient. This patient was diagnosed with an osteosarcoma
of the distal tibia and was treated by intra-articular resection and ankle arthrodesis
using an ipsilateral pedicled fibular graft. Nonunion developed at the proximal graft-host
junction, 9 months following the primary surgery. Cast was removed, external fixation
was reapplied and autologous iliac crest bone graft was added to nonunion site. Bone
healing with a residual varus angulation at the junction site was evident 3 months
later ([Fig. 2 ]). Patient has been walking since unsupported with no pain and a mild limp, refusing
to do any corrective surgery for the varus deformity. A wound dehiscence developed
at the junction of the cutaneous flap and the original skin incision in one patient
3 days following surgery and was managed by secondary sutures with eventual healing.
One patient developed a stress fracture of his fibular graft 7 months following surgery.
The patient was restricted from weight bearing and complete healing of the fracture
was observed 3 months later ([Fig. 3 ]). A mean limb length discrepancy of 0.57 cm (range, 0–3 cm) was measured at the
latest clinical evaluation.
Fig. 2 (A ) A 14-year-old female patient diagnosed with an osteosarcoma of the distal tibia.
Lateral radiograph of the leg, 9 months postoperative, showing proximal nonunion at
the junction between tibia and an ipsilateral pedicled fibular graft (centralization
of the fibula). (B ) Anteroposterior radiograph following revision of fixation using ring fixator and
iliac crest bone graft. (C ) Anteroposterior radiograph 36 months following surgery showing complete union and
hypertrophy of the graft.
Fig. 3 (A ) A 13-year-old male diagnosed with an osteosarcoma of the distal tibia. Anteroposterior
radiograph of the distal leg, 4 months postoperative, showing complete proximal and
distal union of a contralateral vascularized fibular graft. (B ) Anteroposterior radiograph showing hypertrophy and a stress fracture in the fibular
graft 7 months postoperative with abundant callus surrounding the fracture. (C ) Anteroposterior radiograph 3 months following fracture showing healing and complete
remodeling of the surrounding callus.
Discussion
Mavrogenis et al[2 ] published the largest series on osteosarcoma of the distal tibia in 42 patients.
They compared amputation with limb salvage and found that, although there was higher
incidence of local recurrence and complications in the limb salvage group compared
with amputation, the survival in both groups was the same but function was superior
in limb salvage group. They concluded that it may be worthwhile to accept these higher
risks, with strict follow-up of selected patients for whom limb salvage is feasible
and reserve amputation for patients with local recurrence. Tillman[9 ] stated that in the practice of his institute, they advise patients that their function,
in the long term, is likely to be inferior with limb salvage surgery as opposed to
expected function from a modern below knee prosthesis. Most of literature on tumors
around the ankle comprises level IV evidence (case series) due to the low frequency
of malignant bone tumors in this area. Accordingly, there is no clear consensus on
the oncologic and functional benefit of limb salvage versus amputation in this anatomic
location. In the authors' institute, we explain to patients the higher risk of complications
and possibility of multiple revision surgeries associated with limb salvage surgery
as compared with amputation.
Reconstruction of the ankle joint using custom made megaprosthesis has been previously
reported with a good functional outcome.[6 ]
[10 ]
[11 ] Yang et al published the largest series and they had complications in 37% of their
patients including deep infection and loosening of the talar component.[6 ] Despite the disadvantage of the loss of ankle motion following ankle arthrodesis
as compared with a mobile ankle reconstruction using endoprosthesis, we preferred
ankle arthrodesis using vascularized fibular flap in our group of patients due to
the following factors: (1) a great majority of patients diagnosed with osteosarcoma
or Ewing's sarcoma are children or adolescents with high functional demand and relatively
longer life expectancy. A long-lasting biologic reconstruction would be more suitable
for this age group. (2) Most patients are diagnosed with malignant bone tumors and
receive chemotherapy with a higher risk for wound infection, especially for reconstructions
around the ankle with limited soft tissue coverage. Vascularized fibular flaps are
less bulky, more resistant to infection owing to their vascularity, and provide additional
soft tissue coverage when harvested with a skin island. (3) All reported series on
megaprosthetic reconstructions of the ankle describe the use of custom-made implants
with a minimum of 6 weeks for delivery of the implant and a high cost.
Various types of bone grafts have been used for ankle arthrodesis. These include intercalary
allografts,[5 ]
[12 ]
[13 ] contralateral autologous tibial graft,[5 ]
[14 ] pedicled ipsilateral autologous fibular flap,[7 ]
[15 ]
[16 ] and free vascularized contralateral autologous fibular flap.[14 ] The advantages and disadvantages of allograft and autografts are well documented
in literature.[17 ]
[18 ] Size-matched allografts provide initial strength and stability for the construct,
while vascularized flaps provide a faster and higher rate of bone healing and progressive
increase in strength as the graft hypertrophies.[19 ]
[20 ] The Capanna's technique, combining allograft and vascularized fibula, has been described
for reconstruction of several long-bone defects but, to our knowledge, it has not
been used for ankle arthrodesis.[21 ] This may be related to the fact that the reconstruction is too bulky for the ankle
as suggested by Scaglioni et al.[14 ] Moore et al published their results on ankle arthrodesis using allograft in nine
tumor patients.[12 ] Six of nine patients required additional surgery including two cases of graft fracture
that were managed by iliac crest bone grafting and revision of fixation. Comparing
our results, we had one case of stress fracture of the fibular graft which healed
conservatively without the need for a revision surgery. Fractures of large segment
structural bone grafts are major concern in musculoskeletal reconstructions and vascularized
grafts have the advantage over allografts for the ability to heal spontaneously.[19 ]
Several authors preferred the technique of centralization of the ipsilateral fibular
graft into tibial defect[7 ]
[15 ]
[16 ] while Scaglioni et al[14 ] published their results on the use of contralateral free vascularized fibular flap.
Although we used both techniques, the small number of patients in both groups does
not allow for a statistical comparison of the outcome. However, a better cosmetic
appearance (preserved contour) of the ankle was evident in patients treated by contralateral
free osteocutaneous fibular flap and this may be attributed to the preserved ipsilateral
lateral malleolus and the skin flap used for wound closure. Preserving the ipsilateral
fibula also act as a guide for proper alignment of the limb during the implantation
of the free fibula. This was in accordance with the series reported by Scaglioni et
al.[14 ] They used the same technique in five patients. The main differences from current
study were: the average age of their patients which was 33.2 years compared with 10.6
in the current analyses, and the two-stage technique they used whereby they implanted
the free fibular flap in a second surgery following a primary resection and spacer
implantation. Their argument for this technique was the need to assess the surgical
safety margins and avoid delay in chemotherapy after primary resection. We did not
find delaying reconstruction to a second surgery necessary in our patients, as we
assessed marrow margins intraoperatively using frozen section pathology and none of
our patients had wound problems that would have delayed the postoperative chemotherapy.
The type of fixation used for ankle fusion varied in literature, between external
as described by Shalaby et al[16 ] and minimal internal fixation and cast as described by Ebeid et al.[7 ] We found that external fixation offers a more rigid compression of the fibular graft
at the proximal and distal osteotomy sites and facilitates the postoperative wound
care as compared with internal fixation and casting.
Limb length discrepancy is a major concern in resections of malignant bone tumors
around the knee in skeletally immature patients.[1 ] In distal tibial tumors, due to the limited share in growth by distal tibial physis,
the expected limb length difference is usually minimal, as shown by Stéphane et al.[22 ] Although a mean inequality of less than 1 cm was recorded in our patients, one of
the limitations in the current study is the relatively short follow-up period. A longer
follow-up study is needed for assessment of the final limb length difference at skeletal
maturity.
Limitations
The major limitation in the current analysis was the small number of patients. Other
case series on resections of distal tibial tumors had similar limitations. This makes
a statistically representative comparison between various types of reconstruction
difficult. A prospective multicenter cooperative study is strongly needed.
Conclusion
In conclusion, ankle arthrodesis using a vascularized fibular flap provides a rigid
reconstruction of the ankle joint area and young patients are able to return to their
daily life activity within the first year following surgery. The use of the contralateral
osteocutaneous fibular flap provides soft tissue coverage and better shape of the
ankle joint as compared with ipsilateral fibular graft.
