1 3

Arch Orthop Trauma Surg (2014) 134:299–304DOI 10.1007/s00402-013-1909-0

ORTHOPAEDIC SURGERY

Poor prognosis and complications are common in limb salvage surgery for malignant tumors of the proximal tibia invading the fibula

Xiaodong Tang · Wei Guo · Rongli Yang · Yifei Wang

Received: 17 July 2013 / Published online: 4 January 2014 © Springer-Verlag Berlin Heidelberg 2014

acceptable postoperative functions compared with patients without fibula invasion.

Keywords Limb salvage · Complications · Neoplasm · Tibia · Fibula

Introduction

Proximal tibia is a common place for primary malignant bone tumors. When discovered early, most patients with sar-coma in proximal tibia have favorable oncological treatment results with 5-year survival rates between 62 and 81.2 % [10, 15, 16]. Although encountered with insufficient soft tissue coverage, disruption of the extensor mechanism, and loss of the origins for the foot extrinsic musculature, limb salvage can be performed in most patients with acceptable functional outcomes. In a report [14] on osteoarticular prox-imal tibia allograft reconstruction after the resection of a bone tumor, the overall allograft survival was 65 % at 5 and 10 years with an average Musculoskeletal Tumor Society (MSTS) functional score of 26 points. In another study [16] with 52 patients who received proximal tibial endopros-thetic reconstruction for tumor-related diagnosis, the overall prosthesis survival at 5, 10, 15, and 20 years was 94, 86, 66, and 37 %, respectively. The mean postoperative MSTS score at the most recent follow-up was 82 %.

However, in malignant tumors in proximal tibia invad-ing fibula, the oncological and functional outcomes of limb salvage are uncertain. These tumors usually have a large volume and are adjacent to the neurovascular bundles. Furthermore, since Type II resection of proximal fibular tumors requires resection of the peroneal nerve and has a relatively high incidence of anterior tibia vascular injury as described by Malawer [12] and other authors [3, 7], high

Abstract Background Malignant tumors in proximal tibia invading fibula usually have a large tumor volume and are adjacent to the neurovascular bundles. The prognoses and functional outcomes of limb salvage for these patients are uncertain. We, therefore, asked whether patients with limb salvage surgery for malignant tumors in proximal tibia invading fib-ula had poorer oncological prognosis, higher complication rate, and lower postoperative functional score compared with patients without fibula invasion.Methods We retrospectively reviewed 129 patients with primary malignant tumors in proximal tibia. The patients were divided into two groups, i.e., with and without proxi-mal fibula invasion. A total of 35 and 94 patients were in the group with and without fibula invasion, respectively. Data on demography, operation time, blood loss volume, complications, survival time, and postoperative function were compared between two groups.Results The patients with fibula invasion had a longer mean operative time (p = 0.011), less percentage of obtain-ing wide surgical margin (p = 0.027), lower estimated 5-year survival rate (p = 0.05), higher tumor local recur-rence rate (p = 0.042), and earlier postoperative complica-tions (p = 0.01) than the patients without fibula invasion. The difference in postoperative functions as evaluated by the Musculoskeletal Tumor Society functional scoring sys-tem was not significant (p = 0.233).Conclusion Patients with limb salvage surgery for malig-nant tumors in proximal tibia invading fibula had poorer oncological prognosis, higher complication rate, and

X. Tang · W. Guo (*) · R. Yang · Y. Wang Musculoskeletal Tumor Center, Peking University People’s Hospital, Beijing 100044, Chinae-mail: bonetumor@163.com

300 Arch Orthop Trauma Surg (2014) 134:299–304

1 3

complication rates and poor postoperative function may be encountered in patients with malignant tumors in proxi-mal tibia invading fibula. However, these patients were not studied separately in most reports [2].

Therefore, patients with limb salvage surgery for malig-nant tumors in proximal tibia invading fibula were inquired whether they had (1) poorer oncological prognosis, (2) higher complication rate, and (3) lower postoperative func-tional score than patients who had no fibula invasion.

Materials and methods

The study was performed after the approval of medical eth-ics board was obtained. From November 1997 to February 2010, 198 consecutive patients with tumors in proximal tibia were treated in our institute. The medical records, image files, pathological diagnosis, and follow-up information were reviewed retrospectively. The inclusion criteria included: (1) primary malignant tumors; (2) proximal tibia tumor with or without fibula invasion; (3) with limb salvage as primary sur-gical treatment; and (4) complete information of more than 12-month follow-up time. Patients with benign or low-grade malignant tumors, metastatic disease, tumors originated from proximal fibula, and revision surgery were excluded from the study. The remaining 129 patients were divided into two groups: with and without proximal fibula invasion.

In the group with fibula invasion, 35 patients, 22 males and 13 females had an average age of 23 years (range 10–66 years). The diagnoses consisted of 24 osteosarco-mas, 6 chondrosarcomas, 1 malignant giant cell tumor, and 4 soft tissue sarcomas invading the bone. According to the Enneking Staging System [6], 29 patients were in stage IIB whereas 6 patients were in stage III. The group without fibula invasion has 59 males and 35 females with an average age of 22 years (range 13–67 years). The diag-noses included 76 osteosarcomas, 4 chondrosarcomas, 4 malignant giant cell tumors, 4 malignant fibrous histocyto-mas of the bone, 2 Ewing’s sarcomas, 1 hemangiosarcoma, and 4 soft tissue sarcomas invading the bone. A total of 80 patients were in stage IIB whereas 14 patients were in stage III. All the patients diagnosed as osteosarcoma and Ewing sarcoma received neoadjuvant chemotherapy (two cycles preoperatively and four cycles postoperatively, including doxorubicin, cisplatin, methotrexate, and ifosfamide for osteosarcoma, while doxorubicin, vincristine, cyclophos-phamide, ifosfamide, and etoposide for Ewing’s sarcoma).

Before operation, the X-ray and magnetic reso-nance imaging files (Figs. 1, 2) were studied thoroughly to determine the invasion of tumor in the fibula. All patients received en bloc tumor resection and reconstruc-tion. The surgery was performed using a standard tech-nique described in previous studies [10, 13]. Briefly, a

longitudinal medial incision was usually adopted in tumors without fibular invasion. The neurovascular bundle was first identified proximally in the popliteal space and traced dis-tally into the leg to insure the safety of limb salvage. The gastroc-soleus and posterior compartmental musculature were reflected posteriorly. The majority of the anterior compartmental musculature can be retained and usually the anterior tibial vessels and the deep peroneal nerve can also be retained in continuity. The knee and the proximal

Fig. 1 A preoperative plain radiograph shows destruction of the right proximal tibia in a 24-year-old female patient with osteosarcoma

Fig. 2 Enhanced T1-weighted magnetic resonance imaging shows involvement of proximal fibula

301Arch Orthop Trauma Surg (2014) 134:299–304

1 3

tibiofibular joints were opened at the joint line and disartic-ulated. After osteotomy of tibia with safe margin, the tumor was resected completely. If the tumor extended laterally into the anterior compartment and invaded the fibula, an anterior lateral incision was feasible for the proximal tibi-ofibula resection. The anterior tibial vessels were ligated, and sometimes the deep peroneal nerve or the common peroneal nerve had to be sacrificed. The proximal fibula was removed in continuity with the tumor (Fig. 3). With the proximal tibia resected, endoprosthetic reconstruction (GMRS, Stryker, USA; LINK, Germany) can be performed using conventional techniques for limb salvage surgery. The medial gastrocnemius flap was then developed for soft tissue coverage over the proximal tibia (Fig. 4). The patella tendon was reattached to the prosthesis and medial gastroc-nemius flap. In the group with fibula invasion, all patients had ligation of anterior tibial vessels. Eight patients had resection and reconstruction of posterior tibial artery for direct tumor invasion or injury during resection. Among them, three patients had vessel anastomosis and five patients had vascular autograft of the great saphenous vein. In the group without fibula invasion, eight patients had liga-tion of anterior tibial vessels and no patients had posterior tibial artery injury.

The resected specimens were evaluated by an experi-enced pathologist to determine the surgical margins. The postoperative rehabilitation procedures were standard. The patients were encouraged to do isotonic muscle exercise at the earliest possible time after the operation, and have par-tial to full weight bearing under the protection of a brace with the knee in extension position for 6 weeks. Subse-quently, passive and active ranges of motion exercise for the knee were conducted.

All patients were followed up regularly every 3 months (Fig. 5). The postoperative function was evaluated by the MSTS score [5]. Operation time, blood loss volume, com-plications, metastatic disease, and death were recorded for

Fig. 3 The gross view of resected specimen including the proximal tibia and fibula

Fig. 4 The intraoperative view shows soft tissue coverage with medial gastrocnemius muscle flap

Fig. 5 The postoperative plain radiograph taken in 3-month follow-up shows the reconstruction of proximal tibia with endoprosthesis

302 Arch Orthop Trauma Surg (2014) 134:299–304

1 3

the two groups of patients. Complications occurred within 1 year of the index operation were defined as early postop-erative complication.

Independent Mann–Whitney U test and Chi-square analysis were used to compare variables between the two groups (Table 1). The survival rates were evaluated by Kaplan–Meier survivorship analysis. The level of signifi-cance was a probability value of <0.05.

Results

No significant difference existed between the two groups in terms of age (p = 0.984), sex (p = 0.992), and stage (p = 0.754). The total blood loss volume was 215.00 ± 55.20 mL in patients with fibula invasion com-pared with 189.15 ± 27.04 mL in patients without inva-sion (p = 0.053). Patients with fibula invasion had a longer mean operative time (p = 0.011) than patients without fibula invasion. More patients (p = 0.027) in the group

without fibula invasion got wide surgical margins in the pathological evaluation.

The mean follow-up time was 36 and 40 months in the group with and without fibula invasion (p = 0.357), respectively. At the end of the follow-up, 17 of 35 patients with fibula invasion and 30 of 94 patients without inva-sion died of the disease. The estimated 5-year survival rate was 42.8 and 62.3 % in the two groups (p = 0.05), respec-tively (Fig. 6). The patients with fibula invasion had higher (p = 0.042) tumor local recurrence rate of 20.0 % (7/35 patients) compared with 7.4 % (7/94 patients) in patients without fibula invasion.

Patients with fibular invasion had more (p = 0.01) early complications compared with patients without inva-sion. In the former group, one or more early complications occurred in 18 patients, which included five cases of wound healing problems that received debridement, 13 common or deep peroneal nerve palsy including nine patients with permanent palsy who underwent peroneal nerve sacrifice and 4 patients with temporary palsy, 4 extremity ischemia, and 1 case of deep infection. Tendon transfers were per-formed for patients with permanent peroneal nerve palsy. In patients with ischemia, two could maintain their extrem-ities after clearance of hematoma and vascular transplan-tation. Unsuccessful revision surgery was carried out in patient with periprosthetic infection which led to ampu-tation. In the group without fibular invasion, 16 patients had early complications including 11 patients with wound healing problems, 3 with deep infections, and 2 common

Table 1 Comparative features between patients with and without fibula invasion

Characteristics With invasion Without invasion p value

Age (years; mean ± SD)

21.91 ± 13.16 21.86 ± 13.19 0.984

Gender 0.992

Male 22 59

Female 13 35

Stage 0.754

IIB 29 80

III 6 14

Operative time (min; mean ± SD)

215.00 ± 55.20 189.15 ± 27.04 0.011

Total blood loss (mL; mean ± SD)

574.29 ± 169.42 512.13 ± 123.42 0.053

Surgical margin 0.027

Wide 14 58

Marginal or intralesional

21 36

Local recurrence 0.042

Yes 7 7

No 28 87

Early complications 0.01

Yes 18 16

No 17 78

Late complications 0.731

Yes 2 7

No 33 87

Function (MSTS points; mean ± SD)

21.67 ± 4.55 22.81 ± 3.26 0.233

Follow-up time 36.29 ± 24.41 40.26 ± 20.63 0.357

Fig. 6 A Kaplan–Meier curve showing survivorship for patients with malignant tumors in proximal tibia involving fibula compared with patients without fibular invasion

303Arch Orthop Trauma Surg (2014) 134:299–304

1 3

peroneal nerve palsy that had fully recovered. Finally, three patients in the group with fibular invasion and two patients from the group without fibular invasion received amputa-tion for early complications. Late complications occurred in two patients with fibular invasion including one peripros-thetic fracture, and one aseptic loosening; and in seven patients without fibula invasion including three deep infec-tions and four aseptic loosening. No significant difference (p = 0.731) existed in the late complication rate between the two groups.

Postoperative functional outcomes were evaluated in patients who were still alive at the end of the follow-up. The mean MSTS score was 21.67 ± 4.55 points (72.2 %) and 22.81 ± 3.26 points (76.0 %) in 18 patients with fibula invasion and 64 patients without invasion, respectively. The difference was not statistically significant (p = 0.233).

Discussion

Proximal tibia is the second frequent site for primary malignant bone neoplasms. With the development of sur-gical techniques, an increasing number of patients with tumors in this site receive limb salvage surgery. In recent years, most studies [9, 15, 16] focused on reconstruction of proximal tibia with endoprosthesis, allograft-prosthetic composition, and allograft articular replacement, wherein optimistic outcomes were reported. However, tumor resec-tion and reconstruction in proximal tibia encountered more difficulties than other sites, especially when large-volume tumors invaded the proximal tibiofibular joint or proximal fibula. This study aims to determine whether patients with limb salvage surgery for malignant tumors in proximal tibia invading fibula had poorer oncological prognosis, higher complication rate, and lower postoperative functional score than patients who had no fibula invasion.

The limitation of this study is attributed to its retrospec-tive nature. Although the patients were from one institute, the operations were performed by different surgeons in a span of more than a decade. The criteria and surgical tech-niques for limb salvage have improved through time. There were also some other limitations. For example, the patient number in the group with fibula invasion was insufficient, which precluded strong statistical analysis. The follow-up time was not long enough for evaluation of long-term oncological and functional outcomes. Although limitations exist, the investigation on the effects of limb salvage for patients with malignant tumors in proximal tibia invading fibula is valuable.

Patients with malignant tumors in proximal tibia invad-ing fibula have poor prognosis and high local recurrence rates. Generally, in the early discovery of tumors, patients with malignant tumors in the proximal tibia have better

outcomes than patients with tumors in other sites. For pop-liteus providing extra barrier to protect the posterior tibial neurovascular bundles from the tumor, wide surgical mar-gin is expected. Some authors [10, 13] considered that proximal tibiofibula joint is easily invaded by tumors located in proximal tibia, and that proximal tibia and fibula should be resected as a whole. Other studies revealed that in patients without obvious lateral invasion, good tumor control can be obtained by preserving the fibula. In a report [14] of 58 patients with proximal tibia tumors, all resec-tions were intraarticular and intracompartmental, and local recurrence only occurred in four patients. In another study [16] of 52 patients who received endoprosthetic reconstruc-tion of the proximal tibia, the 5- and 10-year disease-spe-cific survival rates for the 43 patients diagnosed with high-grade localized disease were 81.2 and 69.9 %, respectively, after a mean follow-up of 96 months. The authors reviewed the literature on proximal tibial endoprosthetic reconstruc-tion and reported the local recurrence rate from 0.0 to 16.0 %. Few reports focused on proximal tibial tumor with fibula invasion. In a study [17] of 11 osteosarcoma patients with proximal tibiofibular joint invasion, 1 patient had local recurrence and 3 died of the disease. However, the surgical margins were not mentioned in the report. In the present study, 58 of 94 patients without fibula invasion obtained wide surgical margins. The survival and local recurrence rate were comparable with other studies. However, patients with fibula invasion had lower 5-year survival rate and higher local recurrence rate. In proximal tibia tumor invad-ing fibula, which usually indicates a large tumor volume, wide surgical margin is restricted considering the posterior tibial neurovascular bundles and peroneal nerve. Wide sur-gical margins existed in 14 of 35 patients in the group with fibula invasion.

Early postoperative complications are usually encoun-tered in patients with malignant tumors in proximal tibia invading fibula. As confronted with insufficient soft tissue coverage, disruption of the extensor mechanism, and the adjacent neurovascular bundles, the complication rate of limb salvage for proximal tibia is higher than other sites. In the study [11] on amputation following limb salvage with endoprosthesis, the proximal tibial replacements have the highest amputation risk (15.5 %) due to infection and local recurrence. In another report on reconstruction of proxi-mal tibia with allograft-prosthesis composites, in addi-tion to complications related with reconstructions, 35 % of patients experienced postoperative complications [1]. The resection of tumors in the proximal fibula may lead to peroneal nerve palsy and ligation of anterior tibial artery [7, 12]. More complications are predicted in patients with tumors in proximal tibia invading fibula. In a report [13] of seven patients with tumors in proximal tibia who received extraarticular tibiofibular joint resection, four patients had

304 Arch Orthop Trauma Surg (2014) 134:299–304

1 3

transient peroneal nerve palsy and one patient had wound healing problem. In another report [10] of 40 patients with extraarticular excision of tibiofibular joint, 9 patients had early complications including 7 patients with skin necrosis, 2 patients with extension contracture, and 2 patients with deep vein thrombosis. In the present study, patients with fibular invasion had more early complications compared with patients without invasion. This result is attributed to the fact that more than one-third of patients in the group with fibula invasion had nerve palsy. In the attempt to achieve wide surgical margin, vascular injury also contrib-uted to the high early postoperative complication rate.

Although frequently encountered with poor prognosis and complications, patients with fibular invasion who sur-vived from the disease have an acceptable postoperative function. The reconstruction after resection of proximal tibia, including prosthetic replacement, allograft, and allo-graft-prothesis composite combined with reconstruction of extension mechanism, has satisfactory postoperative func-tion. The reported functional MSTS scores were between 75 and 90.4 % [4, 8, 9, 14, 16]. In patients with fibula invasion, the postoperative function is compromised by a wide range of resections and various complications. Jeon et al. [10] reported the highest average MSTS functional score of 76 % in 27 available patients with malignant tumor who received extraarticular resection of proximal tibiofibular joint. Zhang et al. [17] achieved a mean MSTS93 score of 70 % in 11 patients reconstructed with custom prosthesis for proximal tibial osteosarcoma with tibiofibular joint invasion. The most common complication in patients with fibula inva-sion that can affect postoperative function is peroneal nerve palsy. In this study, patients who received proximal tibiofib-ular joint resection with long-term survival after rehabilita-tion and tendon transfers had acceptable functional scores compared with patients without proximal fibula resection.

In summary, this study indicates that patients with limb salvage surgery for malignant tumors in proximal tibia invading fibula had poorer oncological prognoses, higher complication rates, and acceptable postoperative functions than patients who had no fibula invasion. Surgeons and patients should be aware of these before undergoing limb salvage surgery.

Acknowledgments No benefits have been received from a commer-cial party related directly or indirectly to the subject of this article. This study complies with all current relevant laws in China, where this report was produced.

References

1. Biau DJ, Dumaine V, Babinet A, Tomeno B, Anract P (2006) Allograft-prosthesis composites after bone tumor resection at the proximal tibia. Clin Orthop Relat Res 456:211–217

2. Clohisy DR, Mankin HJ (1994) Osteoarticular allografts for reconstruction after resection of a musculoskeletal tumor in the proximal end of the tibia. J Bone Jt Surg Am 76(4):549–554

3. Dieckmann R, Gebert C, Streitburger A, Henrichs MP, Dirksen U, Rodl R, Gosheger G, Hardes J (2011) Proximal fibula resection in the treatment of bone tumours. Int Orthop 35(11):1689–1694

4. Donati D, Colangeli M, Golangeli S, Bella CD, Mercuri M (2008) Allograft-prosthetic composite in the proximal tibia after bone tumor resection. Clin Orthop Relat Res 466:459–465

5. Enneking WF, Dunham W, Gebhardt MC, Malawar M, Pritchard DJ (1993) A system for the functional evaluation of reconstruc-tive reocedures after surgical treatment of tumours of the muscu-loskeletal system. Clin Orthop Relat Res 286:241–245

6. Enneking WF, Spanier SS, Goodman MA (1980) A system for the surgical staging of musculoskeletal sarcoma. Clin Orthop Relat Res 153:106–120

7. Erler K, Demiralp B, Ozdemir MT, Basbozkurt M (2004) Treat-ment of proximal fibular tumors with en bloc resection. Knee 11:489–496

8. Flint MN, Griffin AM, Bell RS, Ferguson PC, Wunder JS (2006) Aseptic loosening is uncommon with uncemented proximal tibia tumor prostheses. Clin Orthop Relat Res 450:52–59

9. Gilbert NF, Yasko AW, Oates SD, Lewis VO, Cannon CP, Lin PP (2009) Allograft-Prosthetic composite reconstruction of the prox-imal part of the tibia. J Bone Jt Surg Am 91:1646–1656

10. Jeon DG, Kawai A, Boland P, Healey JH (1999) Algorithm for the surgical treatment of malignant lesions of the proximal tibia. Clin Orthop Relat Res 358:15–26

11. Jeys LM, Grimer RJ, Carter SR, Tillman RM (2003) Risk of amputation following limb salvage surgery with endoprosthetic replacement, in a consecutive series of 1261 patients. Int Orthop 27:160–163

12. Malawer MM (1983) Surgical management of aggressive and malignant tumors of the proximal fibula. Clin Orthop Relat Res 186:172–181

13. Malawer MM, Mchale KA (1989) Limb-sparing surgery for high-grade malignant tumors of the proximal tibia. Clin Orthop Relat Res 239:231–248

14. Muscolo DL, Ayerza MA, Farfalli G, Aponte-Tinao LA (2010) Proximal tibia osteoarticular allografts in tumor limb salvage sur-gery. Clin Orthop Relat Res 468:1396–1404

15. Natarajan MV, Sivaseelam A, Rajkumar G, Hussain SHJ (2003) Custom megaprosthetic replacement for proximal tibial tumours. Int Orthop 27:334–337

16. Schwartz AJ, Kabo JM, Eilber FC, Eilber FR, Eckardt J (2010) Cemented endoprosthetic reconstruction of the proximal tibia: how long do they last? Clin Orthop Relat Res 468:2875–2884

17. Zhang Y, Yang Z, Li X, Chen Y, Zhang S, Du M, Li J (2008) Cus-tomprosthetic reconstruction for proximal tibial osteosarcoma with proximal tibiofibular joint involved. Surg Oncol 17:87–95