biological reconstruction after resection of bone tumors of the proximal tibia using allograft shell...

BIOLOGICAL RECONSTRUCTION AFTER RESECTION OF BONETUMORS OF THE PROXIMAL TIBIA USING ALLOGRAFT SHELLAND INTRAMEDULLARY FREE VASCULARIZED FIBULAR GRAFT:LONG-TERM RESULTS

MARCO INNOCENTI, M.D.,1* YASSER Y. ABED, M.D.,2 GIOVANNI BELTRAMI, M.D.,3 LUCA DELCROIX, M.D.,1

MARCO MANFRINI, M.D.,4 and RODOLFO CAPANNA, M.D.3

Reconstruction after excision of bone tumor of the proximal tibia is a challenging issue for the reconstructive surgeon. The combined useof a free fibular flap and allograft can provide a reliable reconstructive option in this location. This article describes the authors’ long-termfollow-up using this technique. Twenty-seven patients that had resection of proximal tibia bone tumors underwent reconstruction using thistechnique. Only 21 patients that had primary reconstruction were included in this study. All patients had their surgeries performed at least24 months before the end of the study. The average age at time of operation was 18.1 years. The average follow-up time was 139.3months. Limb salvage was 82.7%. The average length of the resected tibial segment was 15.3 cm and that of the residual proximal tibiaremaining after resection was 2.7 cm. The average time of union of fibula was 5.4 months and for union of allograft was 19.1 months.Primary union of the allograft was achieved in 90.5% of cases. Full weight-bearing was achieved at an average of 21.6 months. Tenpatients (47.6%) had 14 local complications. The (MTSRS) average score at final follow-up was 27.3. Local recurrences occurred in twopatients (9.5%). Distant metastasis to the lung occurred in three patients (14.3%). One patient died of disease. This technique providesgood long-term results in reconstruction of proximal tibia. The viability of the fibula is a cornerstone in both success of reconstruction aswell as successful management of complications. VVC 2009 Wiley-Liss, Inc. Microsurgery 29:361–372, 2009.

The proximal tibia is one of the most common locations

for primary and metastatic bone tumors. Limb salvage

surgery has become an alternative to amputation in most

patients, due to development in diagnostic imaging, neo-

adjuvant chemotherapy, and improved surgical methods.1

Limb salvage procedures for tumors of the proximal tibia

must overcome several anatomic considerations that pre-

dispose reconstruction in this site for high complication

rates. The difficulties are mainly due to close relation to

neurovascular bundle, inadequate soft tissue coverage,

loss of the origins for the foot extrinsic musculature, and

the need to reconstruct the extensor mechanism of the

knee that is considered as a major determinant of the

functional outcomes.2,3

Several options are available for proximal tibia recon-

struction. They include arthrodesis with autogenous or

allogenic bone,4 osteoarticular allograft, allograft prosthe-

sis composite, and prosthetic arthroplasty. Both prosthetic

replacement and allograft has many advantages in recon-

struction of proximal tibia. However, higher rates of com-

plications are encountered more than other anatomical

locations with highest rate of secondary amputation after

limb salvage due to infection and local recurrence.5–8

The ideal reconstruction of the proximal tibia should

have biological ability and in the same time provide

stable construct at short time postoperatively. In the late

1980s, Capanna et al.9 introduced the concept of hybrid

reconstruction by combined allograft shell with free

vascularized fibula. The massive allograft provides

additional initial stability, ability to save small size

epiphyseal bone fragment, and reattachment of tendon

and ligaments of the knee. The free vascularized fibula

offers long-term biologic integration, bony hypertrophy,

and soft tissue coverage in cases of composite flap.

This article will discuss the long-term results of com-

bined use of bone allograft and free fibula in primary

reconstruction after resection of bone tumors of the proxi-

mal tibia.

PATIENTS AND METHODS

From May 1988 to June 2006, 27 patients of primary

malignant bone tumors of the proximal tibia were man-

aged primarily by vascularized fibular graft with allograft

bone shell after bone tumors resection of the proximal

tibia. Six cases were excluded from this study. Three

cases that had secondary reconstruction after failure of

previous limb salvage, two cases had knee arthrodesis,

and one case that had postoperative chemotherapy

induced osteoporosis with multiple fractures. The bio-

graphical data of the reaming 21 patients, the nature and

stage of the tumor and the operative details were

reviewed. All survivors were examined for follow-up. All

1Department of Orthopaedics, Reconstructive Microsurgery Unit, CTO,Florence, Italy2Musculoskeletal Oncology Unit, Mansoura University, Egypt3Department of Orthopedic Oncology, CTO, Florence, Italy4Department of Orthopedic Oncology, IOR, Bologna, Italy

*Correspondence to: Marco Innocenti, M.D., Department of Orthopaedics,Reconstructive Microsurgery Unit, Azienda Ospedaliera Careggi, C.T.O.,Largo Palagi, 1 50139 Florence, Italy. E-mail: [email protected]

Received 31 March 2009; Accepted 31 March 2009

Published online 16 June 2009 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/micr.20668

VVC 2009 Wiley-Liss, Inc.

patients had their surgeries performed at least 24 months

before the end of the study.

There were 13 male and 8 female with an average age

at the time of surgery of 18.1 years (range, 5–52), 15

patients were skeletally immature. The pathology is shown

in Fig. 1 and Table 1. Based on the pathological diagnosis,

15 patients received preoperative chemotherapy.

The average length of the resected tibial segment was

15.3 cm (range, 7–26). The average length of the har-

vested fibula was 18.8 cm (range, 12–28). The average

length of the residual proximal tibia remaining after

resection was 2.7 cm (range, 1.5–5) sparing the joint sur-

face, and if possible the epiphyseal plate. The resection

of the tumor was transepiphyseal in 13 patients (61.9%)

and intercalary in 8 patients (38.1%). The method of

reconstruction was concentric tibial allograft around cen-

tral vascularized fibula in all patients, but one that had

central vascularized fibula and fibular allograft on both

sides. The average operative time was 7.7 hours (range,

6–12) (Table 2).

Fixation was done using diaphyseal plate and meta-

physeal screws in 13 patients (61.9%), bridge plate in

one patient (4.8%), only screws in three patients (14.3%),

double proximal and distal platting in three patients

(14.3%), and metaphyseal wires and diaphyseal screws in

one patient (4.8%). All the arterial anastomosis was done

end-to-end on the stump of the anterior tibial artery with

one or two veins (Table 2).

Partial weight bearing was allowed as soon as the

union of the vascularized fibula was radiographically

assessed and maintained until union of the allograft and

initial hypertrophy of the vascularized fibula was evident.

Then, full weight-bearing was allowed.

Regular CT-scan has been used to evaluate the morpho-

logical changes of the vascularized fibula.10 The functional

evaluation of the patients was done at the end of the follow-

up using the modified 30-point Musculoskeletal Tumor

Society Rating Score (MTSRS) for the lower limb.11Figure 1. Tumor pathology of the 21 patients.

Table 1. Preoperative Data of the 21 Patients

Case/Sex/Age (years) Pathology Enneking’s stage Previous operative procedures

Chemotherapy

Pre Post

1/F/8 Ewing – Biopsy Yes Yes

2/M/10 Os IIB Biopsy Yes Yes

3/M/17 Os IB Biopsy No No


5/M/12 MFH IIB Biopsy Yes Yes

6/F/23 FS IB Curettage and graft No No

7/M/15 Os IB Curettage and graft No No

8/F/7 Os IIB Biopsy Yes Yes

9/F/30 Ada IB Curettage and graft No No

10/M/26 FS IIA Biopsy Yes Yes

11/M/43 Ang IIB Biopsy Yes Yes

12/M/16 Ada IA Biopsy No No


14/M/12 Ewing – Biopsy Yes Yes




18/M/5 Os IIA Biopsy Yes Yes

19/F/52 Cs IA Biopsy No No



M, male; F, female; OS, osteosarcoma; CS, chondrosarcoma; MFH, malignant fibrous histocytosis; FS, fibrosarcoma; Ang, angiosarcoma; Ada,adamantenoma.

362 Innocenti et al.

Microsurgery DOI 10.1002/micr

Table 2. Operative Data of the 21 Patients

Case Type of resection

Length (cm)

Operative

time (hours) Method of fixation

Anastomosis

Defect Fibula

Remnant

bone Arterial Venous (no.)

1 Intercalary 13 15 2 10 Wires and screws Ant. tibial VC (1)

2 Intraepiphyseal 10 13 2 8 Screws Ant. tibial VC (1)

3 Intraepiphyseal 20 23 2 12 DPMS Ant. tibial VC (2)



6 Intercalary 21 26 3 10 DPMS Ant. tibial VC (1)



9 Intercalary 26 28 4 8 DP Ant. tibial VC (1)










19 Intraepiphyseal 16 18 2 10 BP Ant. tibial VC (1)



Average 15.3 18.8 2.7 7.7

DPMS, diaphyseal plate and metaphyseal screws; BP, bridge plate; DP, double plate; Ant. tibial, anterior tibial artery; VC, vena comitant.

Table 3. Reconstructive Results of the 21 Patients

Case

Time (months)

HTP% (m)*

Stress

fracture

of fibula (m)

Follow-up

(m)

Results of reconstruction

Union WB

Failure Cause of failure ManagementFibula Allograft PWB FWB

1 9 27 9 27 40 (48) No 198 No – –

2 6 34 8 36 150 (83) Yes (18) 161 No – –

3 10 22 11 23 30 (213) No 213 No – –

4 4 13 7 18 80 (140) Yes (8) 137 No – –

5 5 Amp 7 Amp Amp No 139 Yes Infection and

fracture

Amp

6 3 12 7 20 70 (36) No 136 No – –

7 5 10 7 14 Failed No 136 Yes Fracture with

nonviable fibula

Revision by

bone allograft

8 6 21 7 22 40 (150) No 134 No – –

9 5 13 5 18 15 (131) No 131 No – –

10 3 22 4 22 50 (120) No 130 No – –

11 4 Amp 5 Amp Amp No 90 Yes Local recurrence Amp

12 6 22 8 23 75 (192) No 191 No – –

13 4 22 6 23 105 (191) No 191 No – –

14 5 18 5 20 60 (189) No 189 No – –

15 5 15 6 17 20 (24) No 67 Yes Local recurrence Amp

16 4 21 5 22 110 (40) No 166 No – –

17 4 18 5 20 30 (82) No 92 No – –

18 4 14 5 15 70 (111) No 146 No – –

19 7 18 12 20 55 (121) Yes (5) 111 No – –

20 7 20 8 22 40 (26) No 36 No – –

21 5 20 6 21 38 (24) No 28 No – –

Sum 3 4 (19%)

Average 5.4 19.1 6.9 21.6 59.9 (106.7) 10.3 m 139.3

WB, full weight-bearing; FWB, full weight-bearing; PWB, partial weight-bearing; m, month; HTP, hypertrophy of fibula; AMP, amputation.*Time of measurement of hypertrophy on CT.

Biological Reconstruction of the Proximal Tibia 363


RESULTS

The average follow-up was 139.3 months (11.6 years)

(range, 28–204). The follow-up of flap viability was done

using postoperative bone scan in 17 patients (81%) and

skin paddle in four patients (19%). Two flaps failed

(9.5%). In one case, a fracture of the reconstructed seg-

ment occurred with nonunion and eventual revision of

the reconstruction using another allograft bone segment

with intramedullary bone cement. The other case devel-

oped postoperative resistant infection, and nonunion and

amputation was done (Table 3).

Time of Union

The average time of union of fibula was 5.4 months

(range, 3–10). The average time of union of allograft was

19.1 months (range, 10–34). All allograft united primarily

except two cases (90.5%); one case required bone graft

and replatting at 13 months postoperative after implant

failure to achieve union 2 months later. There was signif-

icant correlation between the time of union of fibula and

that of allograft (P 5 0.034). This reflects the biological

effect of the vascularized fibula over the allograft

(Fig. 2A). There was a significant correlation between the

time of union of allograft and the stability of method of

fixation (P 5 0.03). The allograft healed earlier with sta-

ble platting than with minimal fixation using screws and

wires (Fig. 2B).

Vascularized Fibular Graft Hypertrophy

Grafts hypertrophy was evaluated using regular CT

scans. Significant graft hypertrophy (�20%) could be

detected in 94.4% of patients, at an average of 106.7

months (range, 24–213). The mean graft hypertrophy

was 59.9% (range, 15–150%) (see Fig. 3). There was a

significant correlation between the extent of graft hyper-

trophy and time of union of allograft (P 5 0.025)

(Fig. 4A). When the allograft took longer time to unite,

there was more loading on the fibula. This led to more

graft hypertrophy. Occurrence of stress fractures of

fibula significantly affected the extent of fibular hyper-

trophy (P 5 0.001) (Fig. 4B). With minimal fixation

using wires and screws, there were more significant

graft hypertrophy than with stable platting (P 5 0.05)

(Fig. 4C).

Figure 2. (A) Relation between time of union of fibula and that of allograft. (B) Correlation between stability of fixation and time of union of

allograft.

Figure 3. CT scan showing hypertrophy of the fibula and integration

with allograft.



Weight-Bearing

Partial weight-bearing (with the use of ischial weight-

bearing knee brace) was achieved at an average of 6.9

months (range, 4–12). Full weight-bearing (without use

of brace) was achieved at an average of 21.6 months

(range, 15–36).

Stress Fracture of the Graft

Stress fracture occurred in three patients within an

average of 10.3 months (range, 5–18) from the index

operation. All occurred before allograft union or after

allograft bone resorption. All patients managed primarily

by cast immobilization with eventual union in average of

2 months (range, 1–3) (see Fig. 5).

Secondary Procedures

Thirteen secondary procedures were performed for

seven patients to achieve union or control complications.

One case had four secondary operations. Plating and graft

was necessary in one patient to achieve allograft union to

recipient site and achieve healing of allograft fracture in

two patients. Three gastrocnemius rotational flaps were

done to manage wound complications. Bone lengthening

was done in one patient with proximal tibial tumor with

6-cm shortening at 63 months postoperative. Lengthening

of 6 cm was done in distal femur of same side.

Figure 4. (A) Correlation between union time of allograft and hypertrophy of fibula. (B) Demonstrating the difference of mean hypertrophy

of fibula in relation to stress fractures. (C) Relation between hypertrophy of fibula and stability of method of fixation.

Figure 5. The effect of stress fracture on hypertrophy of the fibula. (A) Stress fracture after cracking of allograft shell. (B) Healing of stress

fracture with callus formation. (C) 83rd postoperative month X-ray showing more allograft resorption and hypertrophy of the fibula to 150%

of its original width.



Limb Length Discrepancy

At last follow-up, 13 patients had limb length discrep-

ancy of average: 2.31 cm (range, 1–6). Only one case

underwent bone lengthening because of 6-cm shortening

of the reconstructed extremity.

COMPLICATIONS

Donor-Site Morbidity

Five donor site complications occurred in four

patients. Dynamic flexion deformity of the big toe

occurred in three patients. Only one patient required

lengthening of the tendon. Only one case had asymptom-

atic valgus ankle. Transient peroneal nerve palsy occurred

in one patient that spontaneously recovered in few

months (Table 4).

Local Complications

Ten patients (47.6%) had 14 local complications

(66.7%). Management of these complications succeeded

to control them in eight patients (80%). Six fractures

(28.6%) of allograft occurred in five patients at average

Table 4. Complications and Secondary Operations of the 21 Patients

Case

Donor site Local complication LLD

Complications

Management

(m)

Complications

(m)

Management

(Secondary

procedures) (m) Outcome

Value

(cm) Management

1 No – No No – 2 No

2 No – Fracture

(11,18)

Casting Healed 3 No


4 BTFD and

LPNP

No No No – 2 No

5 No – Infection (2) Debridement (4) Failed and

Amputation (11 m)

0 –

Wound

gapping (4)

MHGN (6)

Fracture (11) Hyperbaric O2 (8)

Amp (11)

6 No – No No – 0 –

7 No – Fracture (64) Casting Failed—revision by

bone allograft (67)

2 No


9 BTFD No No No – 0 –

10 BTFD No No No – 0 –

11 No – No No – 0 –

12 No – No No – 0 –

13 No – Wound

gapping (1)

MHGN (2) Healed 6 Bone

lengthening

(63)Fracture (38) Bone graft and

platting(40)

Plate removal(168)

14 No – Wound gapping (1) MHGN (2) Healed 3 No

15 No – Allograft delayed

union and implant

failure (13)

Bone graft and

replatting (14)

Healed 1 No

16 No – Traumatic fracture

(29) associated

femoral fracture

Screw removal (5) Healed 2 No

Platting of tibia and

ILN for femur (29)

17 No – Mild LPNP (1) No Partial recovery 2 No

Screws removal (52)

18 VA No No No – 4 No

19 No – No No – 0 –

20 No – Massive DVT (2) Medical treatment Healed 0 –

21 No – Wound gapping (1) MHGN (2) Healed 1 No

Sum 5 1 14 12 (operations) 1

Average 0.24 0.67 2.31

BTFD, big toe flexion deformity; LPNP, lateral popliteal nerve palsy; VA, valgus ankle; m, month; DVT, deep venous thrombosis; MHGN, medial headgastrocnemius rotational flap; LLD, limb length discrepancy.



Figure 6. Range of motion of the reconstructed left tibia and the stability of the knee.

Table 5. Functional and Oncological Results of the 21 Patients

Case

Tumor

volume (cm3) LR (M)

Distant

metastasis (m)

Status at end

of study

MTSRS

(30-points)*

Failure of

technique

Follow-up

(m)

1 182 – – CDF 29 – 198

2 70 – þ (69) lobectomy NED 26 – 160

3 280 – – CDF 30 – 213

4 60 – – CDF 30 – 140

5 200 – – CDF Failed Amp (11 m) þ 139

6 260 – – CDF 28 – 140

7 370 – – CDF Failed þ 140

8 140 – – CDF 27 – 150

9 220 – – CDF 28 – 131

10 130 – – CDF 18 – 120

11 200 þ (5) – NED Amp (6 m) þ 89

12 70 – – CDF 30 – 192

13 150 – – CDF 30 – 191

14 160 – – CDF 27 – 189

15 120 þ (24) þ (42) DOD Amp (24 m) þ 62

16 180 – þ (21) lobectomy NED 26 – 165

17 210 – – CDF 28 – 90

18 70 – – CDF 27 – 121

19 50 – – CDF 27 – 146

20 125 – – CDF 25 – 36

21 168 – – CDF 28 – 28

Average 162.6 44 m 27.3 139.3

þ, yes; 2, no; m, month; Amp, amputation; CDF, continuous disease free; DOD, died of disease; NED, live with disease; LR, local recurrence; cm3, cubiccentimeter.*At final follow-up.



28.5 months (range, 11–64). Two of these fractures

healed with cast immobilization in average time of 3 and

4.5 months. Two cases healed after platting and cancel-

lous bone graft in 3 and 4 months, respectively. Two

cases failed to unite; limb salvage was achieved in one

of them by revision of reconstruction by allograft and

bone cement. One case had amputation due to associated

massive infection (4.8%). Wound gapping occurred in

four patients (19.1%) and managed successfully in three

cases by local gastrocnemius flap and skin graft. One

patient had postoperative peroneal nerve palsy that

partially recovered during follow-up (Table 4).

Functional Results

Functional results were evaluated using the modified

30-points MTSRS at final follow-up of average 139.3

months (range, 28–204) after exclusion of four cases

(two local recurrences and two failed flaps). The average

total score of these 17 patients was 27.3 (range, 18–30)

(see Fig. 6). All patients had satisfactory functional range

of motion of the knee with stable knee at final follow-up

and were able to perform sport. The limb salvage success

rate was 85.7% (Table 5).

Oncological Results

Fifteen patients received preoperative and postopera-

tive chemotherapy. Postoperative chemotherapy was

usually started within 2–3 weeks after surgery. The mean

tumor volume was 162.6 cc (range, 50–370 cc). Local

recurrence occurred in two patients (9.5%) at 5 and 24

months postoperatively, respectively. Both were managed

by amputation. One of them showed no evidence of dis-

ease at final follow-up (140 months) and the other devel-

oped chest metastasis 42 months postoperatively and died

of disease 20 months later. Distant metastasis to the lung

occurred in three patients (14.3%) at average time 44

months (range, 21–69). One patient died of disease 62

months postoperatively. The other two patients underwent

lobectomy of the metastasis and showed no evidence of

disease at final follow-up of 160 and 165 months postop-

eratively, respectively (Table 5).

Kaplan–Meier survivorship analysis showed that the 2-

year limb survival from amputation (local recurrence,

infection, and flap failure) was 85.7 6 13% (95% confi-

dence interval 78–93) with no amputation expected to

occur later (Fig. 7A). The 2-year success rate of reconstruc-

tion was 85.7 6 7.6% (95% confidence interval 78–93) and

the 5-year success rate of reconstruction was 80.4 6 8.8%

(95% confidence interval 72–89) with no failure of recon-

struction expected to occur after that (Fig. 7B).

DISCUSSION

The most popular reconstructive options after tumor

resection of the proximal tibia are massive allograft,

endoprosthesis, or combination of both. Despite the

excellent results that can be achieved with these methods,

high rates of complications after reconstruction of the

proximal tibia can not be overlooked.2,12,13

The soft tissue nature of the proximal tibia is not in

favor of the endoprosthetic reconstruction. The inadequate

soft tissue coverage increases the incidence of infection.14

Difficulties in reconstruction of the extensor mechanism

jeopardize the long-term functional outcome.15 The

expectant limb length discrepancy and the long-term sur-

vival of the prosthesis are to be considered when using

this method of reconstruction in pediatric patients.16

The biology of allograft integration into host bone is

slow and incomplete process that occurs by creeping sub-

stitution from the host bone only for a short distance

beyond the host bone-allograft junction.17 This results in

high rates of nonunion, fracture, and infection with multi-

Figure 7. (A) Probability of over all success of limb salvage. (B)

Probability of success of reconstruction.



ple reoperations and sometimes amputation.12,18 In case

of use of osteoarticular allograft if it survived the previ-

ously mentioned complications, subchondral bone col-

lapse and subsequent degenerative osteoarthritis jeopard-

izes the long-term results and requires revision by another

method.19 Because of high rate of complications and revi-

sion surgeries, some authors believed that their use in the

management of malignant bone tumors should, at best, be

considered a temporary solution.20

The use of free vascularized fibular graft provides

biologic reconstructions of the proximal tibia that is

expected to avoid the problems encountered with the use

of bone allograft alone or endprosthesis reconstruction of the

proximal tibia. Despite the high rates of union of the vascular-

Figure 8. Case 3 A: (A, B) Preoperative X-ray and MRI of osteosarcoma of proximal tibia. (C, D) 20 cm resected segment. (E, F) Immedi-

ate postoperative X-ray, showing 1.5 cm ruminant of the epiphysis. (G, H) 30th month postoperative showing solid union of allograft and

hypertrophy of fibula. (I) 213th month (18 years) X-ray showing hypertrophy of fibula, growth of the remnant epiphysis, and integration of

both fibula and allograft. (J) Axial CT scan at 213th month showing hypertrophy of 30% of fibula and integration between fibula and allo-

graft. (K) Clinical limb alignment. (L) Rang of motion of the reconstructed limb. (M) Stability and function of the reconstructed limb. (N)

Function of the reconstructed limb that scored 30/30 on the MTSS score.



ized fibula, it has mechanical weakness and the long time

needed to hypertrophy requires long time of protected weight-

bearing and immobilization. Despite the long protection of the

fibula, fractures were common.21,22 In case of intraepiphyseal

knee joint sparing resection of the proximal tibia, it is difficult

to apply the vascularized fibula alone to support the articular

surface even with the use of double barrel fibula.23 Technical

difficulty in bone fixation and reattachment of the patellar

tendon are encountered. In such conditions, arthrodesis of the

knee has been advocated as the only solution in order to over-

come these difficulties.24–26

Therefore, the ideal biological reconstruction should

retain the host articular surface if oncologically possible

and to take advantage of the combination of both the

strong bone stock provided by the allograft and the bio-

logical potentials of the vascularized fibular graft. This

method of reconstruction would be expected to provide a

long lasting reconstruction (see Fig. 8).9 This technique was

first introduced in the late 1980s,9 since then it gained

popularity and multiple studies report good long-term

results with excellent functional outcomes, high rates of

union, and low rates of infection and fracture.27–32 The

reliability of this technique expanded its indications to

use the vascularized fibula to treat allograft nonunion

with good results to achieve union.33

The long-term results of this study indicate that com-

bined use of fibula and allograft provide a reliable solu-

tion for bone reconstruction of the proximal tibia. The

high primary union of allograft 90.5% in this study com-

pared to 73–83% in other studies.34,35 The correlation

between union time of both the allograft and the fibula

and the extent of fibular hypertrophy reflect the mecha-

nical role of the fibula in the weight-bearing beside its

biological role in union of allograft.

Functional score at final follow-up of average 139.3

months (ranged from 28 to 204 months) was 27.3/30

(91%) (range, 18–30) after exclusion of four cases (two

local recurrences and two failed flaps). Of the four

Figure 8. Continued.



excluded cases, three cases had amputations and one had

revision of the reconstruction after fracture of the recon-

struction that occurred at 64 months postoperative with

final functional score of 21/30. Three cases had amputa-

tions, two local recurrences, and one deep infection. The

limb salvage success rate was 85.7%.

Kaplan–Meier survivorship analysis showed that failure

of limb salvage occurred in the first 2 years and that failure

of reconstruction occurred in the first 5 years; with no

change expected to occur after that. And these strongly

support the favorable long-term result of the method of

reconstruction in comparison to other methods.35,36

Although the resection was intraepiphyseal in 13 patients

(61.9%), only two cases had local recurrence (9.5%) and both

had intercalary resection. This percentage is in the range of

published rate of local recurrence (5–10%).37

Despite the high rate of local complications, their man-

agement resulted to be successful in 80% of cases.

Although of allograft fractured in five patients six times

(28.6%), union occurred in four of them (66.7%) reflecting

the biological effect of the vascularized fibula. The cases

that failed to unite all had failed vascularization of the

fibula. Only one case had deep infection (4.8%), which is

very low incidence compared to 10–20% reported in pub-

lished series of allograft reconstruction.35,38 The incidence

of wound gapping may be decreased by routine use of

osteocutaneous fibula to have tension-free closure of

wound. However, the intramedullary location of the fibula

makes it difficult to orient the septal perforator to the skin

paddle with kink or compression.

CONCLUSION

The use of allograft and vascularized fibular graft in

reconstruction of proximal tibia provides good long-term

results. The results of this study show the superior out-

come of this method of reconstruction when compared to

other reconstructive options for proximal tibia, especially

in skeletally immature. Despite high rate of complications

of this life long reconstruction, management of these

complications did not affect the long-term results of

majority of patients. The viability of the fibula is a

cornerstone in both success of reconstruction as well as

successful management of complications. The stability of

osteosynthesis of allograft is essential to achieve early

union and avoid complications.

REFERENCES

1. Ilyas I, Younge D, Pant R, Moreau P. Limb salvage for proximaltibial tumors using a modular prosthesis. Int Orthop 2000;24:208–211.

2. Grimer RJ, Carter SR, Tillman RM, Sneath RS, Walker PS, UnwinPS, Shewell PC. Endoprosthetic replacement of the proximal tibia.J Bone Joint Surg 1999;81B:488–494.

3. Bickels J, Witting JC, Kollender Y, Neff RS, Kellar-Graney K,Meller I, Malawer MM. Reconstruction of the extensor mechanismafter proximal tibia endoprosthetic replacement. J Arthroplasty2001;16:856–862.

4. Enneking WF, Shirley PD. Resection arthrodesis for malignant andpotentially malignant lesions about the knee using intramedullaryrod and local bone graft. J Bone Joint Surg 1977;59A:223–236.

5. Jeys LM, Grimer RJ, Carter SR, Tillman RM. Risk of amputationfollowing limb salvage surgery with endoprosthetic replacement, in aconsecutive series of 1261 patients. Int Orthop 2003;27:160–163.

6. Jeys LM, Grimer RJ, Carter SR, Tillman RM. Periprothetic infectionin patients treated for an orthopaedic oncological condition. J BoneJoint Surg 2005;87-A:842–849.

7. Unwin PS, Cannon SR, Grimer RJ, Kemp HB, Sneath RS, WalkerPS. Aseptic loosening in cemented custom-made prosthetic replace-ments for bone tumours of the lower limb. J Bone Joint Surg Br1996;78:5–13.

8. Brien EW, Terek RM, Healey H, Lane JM. Allograft reconstructionafter proximal tibial resection for bone tumors. Clin Orthop 1994;303:116–127.

9. Capanna R, Bufalini C, Campanacci M. A new technique for recon-structions of large metadiaphyseal bone defects: A combined graft(allograft shell plus vascularized fibula). Orthop Traumatol 1993;2:159–177.

10. Manfrini M, Vanel D, De Paolis M, Malaguti C, Innocenti M,Ceruso M, Capanna R, Mercuri M. Imaging of vascularized fibulaautograft placed inside a massive allograft in reconstruction of lowerlimb bone tumors. Am J Roentgenol 2004;182:963–970.

11. Enneking WF, Dunham W, Gebhardt MC, Malawar M, Pritchard DJ.A system for the functional evaluation of econstructive proceduresafter surgical treatment of tumors of the musculoskeletal system.Clin Orthop 1993;286:241–246.

12. Donati D, Di Liddo M, Zavatta M, Manfrini M, Bacci G, Picci P,Capanna R, Mercuri M. Massive bone allograft reconstruction inhigh-grade osteosarcoma. Clin Orthop 2000;377:186–194.

13. Hejna MJ, Gitelis S. Allograft prosthetic composite replacement forbone tumors. Semin Surg Oncol 1997;13:18–24.

14. Horowitz SM, Lane JM, Healey JH. Soft-tissue management withprosthetic replacement for sarcomas around the knee. Clin Orthop1992;275:226–231.

15. Shimose S, Sugita T, Kubo T, Matsuo T, Ochi M. Reconstructedpatellar tendon length after proximal tibia prosthetic replacement.Clin Orthop 2005;439:176–180.

16. Eckardt JJ, Kabo JM, Kelley CM, Ward WG Sr, AsavamongkolkulA, Wirganowicz PZ, Yang RS, Eilber FR. Expandable endoprosthe-sis reconstruction in skeletally immature patients with tumors. ClinOrthop 2000;373:51–61.

17. Enneking WF, Campanacci DA. Retrieved human allografts: A clini-copathological study. J Bone Joint Surg Am 2001;83-A:971–986.

18. Muscolo DL, Ayerza MA, Aponte-Tinao L, Ranalletta M, Abalo E.Intercalary femur and tibia segmental allografts provide an accepta-ble alternative in reconstructing tumor resections. Clin Orthop 2004;426:97–102.

19. Clohisy DR, Mankin HJ. Osteoarticular allografts for reconstructionafter resection of a musculoskeletal tumor in the proximal end of thetibia. J Bone Joint Surg Am 1994;76:549–554.

20. Rodl RW, Ozaki T, Hoffmann C, Bottner F, Lindner N, WinkelmannW. Osteoarticular allograft in surgery for high-grade malignanttumours of bone. J Bone Joint Surg Br 2000;82:1006–1010.

21. Hsu RW, Wood MB, Sim FH, Chao EY. Free vascularised fibulargrafting for reconstruction after tumour resection. J Bone Joint SurgBr 1997;79:36–42.

22. Lee KS, Han SB, Baek JR. Free vascularized osteocutaneous fibulargraft to the tibia in 51 consecutive cases. J Reconstr Microsurg2004;20:277–284.

23. Jones NF, Swartz WM, Mears DC, Jupiter JB, Grossman A. The‘‘double barrel’’ free vascularized fibular bone graft. Plast ReconstrSurg 1988;81:378–385.

24. Usui M, Ishii S, Naito T, Wada T, Nagoya A, Takahashi T,Tsuchida Y. Arthrodesis of knee joint by vascularized fibular graft.Microsurgery 1996;17:2–8.



25. Rasmussen MR, Bishop AT, Wood MB. Arthrodesis of the kneewith a vascularized fibular rotatory graft. J Bone Joint Surg Am1995;77:751–759.

26. Amr SM, El-Mofty AO, Amin SN, Morsy AM, El-Malt OM, Abdel-Aal HA. Reconstruction after resection of tumors around the knee:Role of the free vascularized fibular graft. Microsurgery 2000;20:233–251.

27. Capanna R, Campanacci DA, Belot N, Beltrami G, Manfrini M,Innocenti M, Ceruso M. A new reconstructive technique for inter-calary defects of long bones: The association of massive allograft withvascularized fibular autograft. Long-term results and comparison withalternative techniques. Orthop Clin North Am 2007;38:51–60.

28. Chang DW, Weber KL. Segmental femur reconstruction using anintercalary allograft with an intramedullary vascularized fibula boneflap. J Reconstr Microsurg 2004;20:195–199.

29. Chang DW, Weber KL. Use of a vascularized fibula bone flap andintercalary allograft for diaphyseal reconstruction after resection ofprimary extremity bone sarcomas. Plast Reconstr Surg 2005;116:1918–1925.

30. Moran SL, Shin AY, Bishop AT. The use of massive bone allograftwith intramedullary free fibular flap for limb salvage in a pediatricand adolescent population. Plast Reconstr Surg 2006;118:413–419.

31. Zaretski A, Amir A, Meller I, Leshem D, Kollender Y, Barnea Y,Bickels J, Shpitzer T, Ad-El D, Gur E. Free fibula long bone recon-

struction in orthopedic oncology: A surgical algorithm for recon-structive options. Plast Reconstr Surg 2004;113:1989–2000.

32. German M, Mascard E, Dubousset J, Nguefack M. Free vascularizedfibula and reconstruction of long bones in the child-our evolution.Microsurgery 2007;27:415–419.

33. Bae D, Waters P, Gebhardt M. Results of free vascularized fibulagrafting for allograft nonunion after limb salvage surgery for malig-nant bone tumors. J Pediatr Orthop 2006;26:809–814.

34. Hornicek FJ, Gebhardt MC, Tomford WW, Sorger JI, Zavatta M,Menzner JP, Mankin HJ. Factors affecting nonunion of the allograft-host junction. Clin Orthop 2001;382:87–98.

35. Mankin HJ, Gebhardt MC, Jennings LC, Springfield DS, TomfordWW. Long-term results of allograft replacement in the managementof bone tumors. Clin Orthop 1996;324:86–97.

36. Biau D, Faure F, Katsahian S, Jeanrot C, Tomeno B, Anract P.Survival of total knee replacement with a megaprosthesis after bonetumor resection. J Bone Joint Surg Am 2006;88:1285–1293.

37. Picci P, Sangiorgi L, Rougraff BT, Neff JR, Casadei R, CampanacciM. Relationship of chemotherapy induced necrosis and surgicalmargins to local recurrence in osteosarcoma. J Clin Oncol 1994;12:2699–2705.

38. Hornicek FJ Jr, Mnaymneh W, Lackman RD, Exner GU, MalininTI. Limb salvage with osteoarticular allografts after resection ofproximal tibia bone tumors. Clin Orthop 1998;352:179–186.



biological reconstruction after resection of bone tumors of the proximal tibia using allograft shell...

Documents