v10i2p120.pdf

8/14/2019 v10i2p120.pdf

1/9

Journal of Orthopaedic Surgery 2002: 10(2): 120128

Giant cell tumour of bone with latepresentation: review of treatmentand outcome

ES Ng, A Saw, S Sengupta

Department of Orthopedic Surgery, University Malaya Medical Center, University of Malaya, Kuala Lumpur, MalaysiaAR Nazarina, M PathDepartment of Pathology, University Malaya Medical Center, University of Malaya, Kuala Lumpur, Malaysia

Address correspondence and reprint requests to: Dr A Saw, Department of Orthopedic Surgery, Faculty of Medicine, University

of Malaya, 50603 Kuala Lumpur, Malaysia. E-mail: [email protected]

ABSTRACT

Purpose. To review cases of giant cell tumour of boneor osteoclastoma managed at the University MalayaMedical Center, University of Malaya, Kuala Lumpur,from January 1990 to December 1999.Methods. Medical records of all patients withmusculoskeletal tumours were reviewed.Demographic data, clinical presentation, surgicalmanagement, and clinical outcomes were reviewedretrospectively.Results.Most of the 31 patients who were treated forgiant cell tumour of bone presented late on the basisof the duration of their symptoms and radiologicalfeatures. Five of the patients had been referred for localrecurrences. 26 patients were treated for primarytumours: 18 needed wide excision, 7 curettage, andone amputation. The joint could not be preserved andarthrodesis was performed for 11 patients. Three (12%)of the 26 patients had local recurrence during a meanfollow-up of 60 months, including one (6%) who hadrecurrence after wide excision and 2 (29%) aftercurettage. Pulmonary metastasis was noted in 4 cases,2 of which were confirmed histologically.

Conclusion. Even in an advanced stage of disease,good clinical outcomes can be achieved with adequateexcision and appropriate reconstruction. For lesionsaround the knee, autologous rotational grafting is agood alternative method of reconstruction.

Key words: bone tumor; giant cell tumor of bone;osteoclastoma

INTRODUCTION

Giant cell tumour (GCT) of bone, or osteoclastoma, isclassically described as a locally invasive tumour thatoccurs close to the joint of a mature bone. It is generallyconsidered to be a benign tumour. The incidence of

bone GCT also varies depending on the reportingcentre. In an earlier study conducted in our institution,Peh et al.1 found that 28 (21%) out of 135 biopsiesperformed for bone tumours were GCT. In ourexperience, a substantial proportion of patients seektraditional means of treatment before medicalconsultation. Infrequently, they present late withextensive involvement of soft tissue and articularsurface, making joint preservation difficult or

8/14/2019 v10i2p120.pdf

2/9

Vol. 10 No. 2, December 2002 Giant cell tumour of bone with late presentation: review of treatment and outcome 121

impossible. This study aimed to investigate theepidemiological data of bone GCT in the localMalaysian population, and to review the outcomes oftreatment among these patients.

MATERIALS AND METHODS

We reviewed medical records of all patients withmusculoskeletal tumours who were treated at theUniversity Malaya Medical Center, University ofMalaya, Kuala Lumpur, from January 1990 toDecember 1999. Of a total of 224 cases, 141 wereprimary bone tumours; 34 of these cases were GCT of

bone. Three of the cases of bone GCT were excludedfrom the study: one case was eventually managed inanother hospital, while in 2 cases, patients refused

surgery and were lost to follow-up. 31 patientsunderwent surgery and were followed up for a meanduration of 60 months (range, 20121 months). 26 ofthese patients were treated for primary tumours,whereas the remaining 5 had been referred to us forlocal recurrences. Six patients with primary tumourshad a biopsy performed before referral. All the tumourspecimens from definitive surgeries were retrievedand reviewed again by one of the authors to confirmthe diagnosis. We classified the surgical margin asone of 4 types according to Enneking et al.2:intralesional (curettage), marginal, wide, and radical

(compartmental). Demographic data, clinicalpresentation, surgical management, and clinicaloutcome were reviewed retrospectively.

RESULTS

Patient characteristics

There were 20 (59%) female and 14 (41%) male patients.

The mean age of all the patients was 30.2 years (range,1673 years), while 71% of them were aged between16 and 35 years (Fig. 1). Approximately half (52%) ofthe patients were Chinese, 27% were Malay, and 21%were Indian.

Site of lesion

Eight (24%) tumours involved the upper limb, whereas25 (74%) were in the lower limb (Fig. 2). 19 (56%)lesions occurred around the knee joint. There was onlyone (3%) axial lesion at the T10 vertebra.Three patientshad multicentric lesions that involved 2 sides of a joint:one patient had lesions at the distal tibia and talus(Fig. 3), one had lesions at the femoral head andacetabulum, and one had lesions at the proximal tibiaand fibula.

Clinical presentation

Most patients in our series presented late and at anadvanced stage of disease, when preservation ofarticular cartilage was not possible. Their meanduration of symptoms was 8 months, while 54%

of them presented at more than 6 months after theonset of symptoms. The chief symptoms atpresentation among the 26 patients who were treated

Male Female

Number ofpatients

Age (years)

9

0

1

2

3

4

5

6

7

8

50

Figure 1 Distribution of patients

according to age and sex.

8/14/2019 v10i2p120.pdf

3/9

122 ES Ng et al. Journal of Orthopaedic Surgery

primarily in our institution were pain (n26), swelling(n20), and pathological fracture (n5). 13 (50%)patients had history of antecedent trauma. The meanduration of symptoms was 8 months (range, 7 days2 years).

Management

Tissue diagnosis was established for all patients beforedefinitive surgery. Surgical procedures were selected

according to tumour location, tumour size, and the

extent of the surrounding structures involved. Eightpatients (7 with primary tumours) underwentintralesional curettage, 19 (18 primary) underwentwide excision with limb preservation, and 4 (oneprimary) required amputation of the affected limb(Table 1).

Intralesional curettage was done by thoroughcurettage of the tumour through a wide corticalwindow. For most patients, an autologous bone graftwas used to fill the cavity. Bone cement was used in

one patient (case 1). Wide excision was performed

Figure 2 Distribution of GCT of boneaccording to the site of tumour, indicatingthe number of cases of GCT in parts ofthe body.

Figure 3 (a) and (b) X-ray of a 20-year-old Indian woman with right ankle painand swelling, showing lytic lesionsinvolving both the distal tibia and talus(case 9).(a) (b)

8/14/2019 v10i2p120.pdf

4/9


Table 2Summary of management and outcome of patients with GCT of bone

Case Location Margin of Reconstruction Local Pulmonary Duration Currentexcision recurrence metastasis of follow-up outcome

(months) (months) (months)

1 Femur Curettage Bone cement N N 30 Disease-freeproximal

2 Radius Wide Fibula graft N N 60 Disease-freedistal



5 Humerus Wide Fibula graft N N 60 Disease-freeproximal

6 Femur distal Wide Rotational tibial N N 55 Disease-freegraft, kneearthrodesis

7 Femur distal Wide Allograft, knee N N 24 Disease-freearthrodesis

8 Femur, Wide Hip arthrodesis N N 120 Disease freeacetabulum

9 Tibia distal, Wide Ankle arthrodesis N N 21 Disease freetalus

10 Tarsal Amputation Below-knee N Y, 5 Diedamputation

11 Humerus Amputation Forequarter N N 20 Disease-freeamputation

12 Radius Amputation Below-elbow N Y, 5 33 Alive withdistal amputation disease

13 Femur distal Amputation High transfemoral N N 39 Disease-freeamputation

14 T10 Curettage Fusion with rib Y, 6 N 86 Disease-freegraft

Table 1Surgical procedure and recurrence in patients with primary GCT

Treatment

Curettage Wide excision Amputation Total

No. of patients 7 18 1 26

Recurrence 2 1 0 3

Percentage recurrence 29% 6% 0% 12%

8/14/2019 v10i2p120.pdf

5/9


Table 2 Continued

Case Location Margin of Reconstruction Local Pulmonary Duration Currentexcision recurrence metastasis of follow-up outcome

(months) (months) (months)

15 Femur distal Curettage Bone graft Y, 4.5 Y, 0 91 Disease-free

16 Tibia Wide Allograft Y, 42 N 42 Disease-freeproximal

17 Tibia Wide Rotational femoral N N 84 Disease-freeproximal graft, knee

arthrodesis

18 Femur distal Wide Bone graft, K wire N N 61 Disease-free

19 Femur distal Wide Bone graft, DCS* N Y, 22 39 Disease-free

20 Ulna distal Wide None N N 120 Disease-free

21 Fibula Wide None N N 58 Disease-freeproximal


arthrodesis


arthrodesis


arthrodesis

25 Tibia Wide Bone graft, knee N N 89 Disease-freeproximal arthrodesis


arthrodesis

27 Talus Curettage None N N 24 Disease-free

28 Femur distal Curettage Bone graft N N 85 Disease-free

29 Femur distal Curettage Bone graft N N 24 Disease-free

30 Tibia Curettage Bone graft N N 20 Disease free

proximal

31 Humerus Curettage Bone graft N N 82 Disease-freeproximal

* DCS: dynamic condylar screw fixation

on 19 patients. The method of reconstruction de-pended on the size and site of the lesion (Table 2).Autologous non-vascularised fibula graft was usedfor the reconstruction of 3 distal radial tumours (cases2, 3, and 4) and one proximal humeral tumour (case

5).

Five proximal tibia lesions were reconstructed withrotational grafting (Fig. 4), while 4 distal femur lesionswere reconstructed with iliac bone grafting in 2,rotational grafting in one (case 6), and structuralallograft in one (case 7). In the case managed by

rotational grafting, the uninvolved distal femur or

8/14/2019 v10i2p120.pdf

6/9


proximal tibia was split and rotated with attached

condylar soft tissue, and collateral ligaments to bridgethe defects after wide excision; the knee was then fusedwith a long plate or intramedullary nail. Additionalcancellous iliac bone graft was deposited around the

bone ends.The 2 patients with adjacent bone involvement

across the hip and ankle joint (cases 8 and 9) weretreated by wide excision followed by hip and anklefusion. Four patients required limb amputation forextensive local disease (cases 1013). One of them hada lesion at distal femur with extensive soft tissueinvolvement, one had local recurrence at the tarsal

bones after intralesional curettage had been performedin another hospital, and one had a tumour in theproximal humerus that was treated with wide excisionand autologous non-vascularised fibula graft atanother hospital. In case 11, local recurrence after 5years involved the graft and extended into the adjacentscapula (Fig. 5); in this case, forequarter amputationwas performed. In case 12, a distal radial tumour had

been treated by wide excision followed by bonetransport using an external fixator; the patient wasreferred to us 5 months later with lung metastasis andlocal recurrence ulcerating through the skin, for which

a below-elbow amputation was performed.

Figure 5 X-ray of a 40-year-old Chinese man with a tumourin the proximal humerus managed by wide excision andautologous non-vascularised fibula graft but resulting in localrecurrence: destructive lesions involved the graft extending

into the adjacent scapula (case 11).

Figure 4 Case 17(a)lytic lesion over subchondral region of proximal tibia; (b)after wide excision and rotational femoralautograft, the knee was fused using long plate; (c)the knee was strongly fused and the plate was removed one year later.

(a) (b) (c)

8/14/2019 v10i2p120.pdf

7/9


Figure 6 Computed tomogram of a 23-year-old Chinese manwho presented with recurrent GCT at the distal radius, showingmultiple bilateral lung nodules (case 12).

Outcome

Three (12%) of the 26 patients with primary tumoursdeveloped local recurrence during a mean follow-upof 60 months (Table 2). One patient (case 14) had GCTin the T10 vertebral body, which was managed bymarginal excision and rib grafting; local recurrencedeveloped 6 months later, which was treated by furtherexcision, bone grafting, and posterior instrumentation.Five months after the second surgery, the tumourrecurred; another excision and bone graft wasperformed, and the patient remained disease-free 4.5years after the last surgery. Another patient (case 15)had been referred to us with GCT at the distal femurwith single lung metastasis; the primary lesion wasmanaged by curettage and the lung nodule was excised

by thoracotomy. Both lesions were confirmed to beGCT, but 4.5 months later, the patient again developedknee swelling that was confirmed by biopsy to be GCT;chest radiography showed multiple lung lesions. Thepatient was treated with a course of local radiotherapyand systemic chemotherapy, after which both distalfemur and lung lesions resolved; the patient wasclinically and radiologically free of disease 7 years aftertreatment. The third patient (case 16) had wide excisionof GCT at the proximal tibia followed by allograftreconstruction, but the tumour recurred at theproximal fibula 42 months after surgery. Three months

before this study, wide excision of fibu la withpreservation of allograft was performed.

Pulmonary metastasis

Pulmonary metastasis developed in 4 patients (Table2): one presented with a lesion in the lung, whereas 3developed lung nodules after surgical treatment. Twoof these patients underwent excision of pulmonarynodules and had histological confirmation of GCT;

both remained disease-free after surgery. Diagnosis inthe other 2 patients were based on clinical andradiological evidence: one subsequently died of

respiratory disease while the other developedrecurrent distal radial tumour and haemoptysis (Fig.6), for whom below-elbow amputation was performed.On last review, 33 months after presentation, thispatient was well and had only occasional bouts of drycough, although X-ray findings of the lungs remainedunchanged.

Complications

There were 19 complications in 12 patients. Twopatients had postoperative wound breakdown that

healed after secondary suturing. Four patients

developed deep infection: one had a tumour of thedistal radius, which was reconstructed with a fibulagraft, and 3 had tumours of proximal tibia, and weretreated with knee fusion and plate fixation. Three ofthe 4 patients responded to debridement andintravenous antibiotic, but the fourth developedchronic osteomyelitis of the upper tibia with persistentsinus discharge.

Common peroneal nerve palsy occurred

postoperatively in 4 patients: 3 underwent wideexcision for tumours around the knee, and onedeveloped footdrop after harvesting of fibula graft.Three of these 4 patients recovered spontaneously.

Non-union with implant failure occurred in 6patients (one with a broken K nail, 2 with broken plates,2 with implant loosening for knee fusion, and one with

broken screws for hip fusion). Another patient had afracture proximal to the fused knee following a fall.All these complications occurred in patients with wideexcision.

DISCUSSION

Bone GCT is not a common tumour. It makes up 5% to7% of all primary bone tumours in several majorwestern series.3534 (24%) of 141 primary bone tumourstreated in our institution were GCT. Similar pro-portions of bone tumours were identified to be GCTin 2 separate studies in our country (21% and 28%).1,6

Likewise, another study conducted in Hong Kong andChina showed that GCT of bone contributed 15% to20% of all primary bone tumours. 7 The higher

percentage of GCT in bone tumours in Asian

8/14/2019 v10i2p120.pdf

8/9


populations is clear. There were more females (59%)than males (41%), which is in agreement with westernseries4,8although Campanacci et al.3and Larsson et al.9

could not demonstrate any sex predilection in theirseries. The majority of cases described occurred

between second to fourth decades,35,79

and 59% of thepatients in our series were aged between 20 and 40years. As in most series,35,79the most common siteswere proximal tibia and distal femur, followed by distalradius. More than half (56%) of the patients in ourseries had lesions around the knee. Lesions at the spineare not common, and multicentric GCT is also rare(occurrence,1%).3,810Cummins et al.10reviewed 29cases of multicentric GCT in the literature andobserved that they were more common in atypicalsites, and among younger patients. About 80% ofpatients published were younger than 25 years. Two

of the 3 patients with multicentric lesions were youngerthan 25 years.

Most patients in our series presented late, andnearly two-thirds (64%) had a grade III tumouraccording to radiological grading by Campanacci.3

This proportion is higher than that found in other seriesof 27% to 50%.3,5Moreover, 4 cases either presentedwith or eventually developed lung metastasis, whichmay reflect delay in presentation. One in 5 admittedthat they had visited a traditional healer before medicalconsultation, which may have contributed to the delay.

The local recurrence rate after surgery reported in

the literature ranges from 15% to 26%.3,5,7

Many authorshave tried to determine the factors that predict the riskof local recurrence. Histological grading by Jaffe et al.11

and radiological grading by Campanacci3have beenshown to be unreliable.4,8,9The most important factorthat predicts prognosis is the adequacy of tumourresection. It is generally accepted that recurrence rate

after intralesional curettage is higher (27%41%) thanthat after wide excision (0%7%).3,5,7Most of our caseshad soft tissue extension and articular surfaceinvolvement that rendered joint preservationimpossible. This accounted for more wide excision

procedures (n19; 61%) being performed comparedwith intralesional curettage (n8; 26%). Nearly half(n9; 47%) of the wide excisions (n19) were followed

by arthrodesis. Three (12%) of 26 patients developedlocal recurrence after a mean follow-up of 17.5 months.There was only one (6%) local recurrence among 18patients who underwent wide excision, whereas 2(29%) of the 7 patients who had intralesional curettagedeveloped local recurrence. These results showed thatin spite of their more advanced stage of disease, theoverall clinical outcomes of our series are comparableto those of other series.

Pulmonary metastasis of benign GCT is very rare.So far, 69 cases have been reported in the literature.Most cases were single case reports and only a fewlarger series have been reported recently.4,1219 Four(13%) of the 31 patients in our series developed lunglesions, including 2 whose diagnosis were confirmedhistologically. There has been no standard treatmentfor GCT metastasis described in the literature and theresults of those described are variable. One of thepatients in our series who had multiple pulmonarymetastasis was still alive 2.5 years after the diagnosiswas made. Nodule excision has been reported to yield

good results,4,12,14,15,18

and the role of chemotherapy isstill debatable.20

In conclusion, even in an advanced stage ofdisease, good clinical outcomes can be achieved withadequate excision and appropriate reconstruction. Forlesions around the knee, autologous rotational graftingis a good alternative method of reconstruction.

REFERENCES

1. Peh SC, Cheah PL, Sengupta S. The pathology of tumour and tumour-like lesions of bone in the University Hospital KualaLumpur. Malays J Pathol 1988;10:4550.

2. Enneking WF, Spanier SS, Goodman MA. A system for the surgical staging of musculoskeletal sarcoma. Clin Orthop1980;153:10620.

3. Campanacci M, Baldini N, Boriani S, Sudanese A. Giant cell tumour of bone. J Bone Joint Surg Am 1987;69:10614.4. Goldenberg RR, Campbell CJ, Bonfiglio M. Giant cell tumour of bone. An analysis of two hundred and eighteen cases.

J Bone Joint Surg Am 1970;52:61964.5. McDonald DJ, Sim FH, McLeod RA, Dahlin DC. Giant cell tumour of bone. J Bone Joint Surg 1986;68:23542.6. Sundari SI, Razak MA. Bone and soft tissue sarcoma: an epidemiological study at the National University of Malaysia,1992

97. J ASEAN Orthop Assoc 2000;13:313.7. Sung HW, Kuo DP, Shu WP, Chai YB, Liu CC, Li SM. Giant cell tumour of bone: analysis of two hundred and eight cases in

Chinese patients. J Bone Joint Surg Am 1982;64:75561.8. Dahlin DC, Cupps RE, Johnson EW Jr. Giant cell tumour: a study of 195 cases. Cancer 1970;25:106170.9. Larsson R, Lorentzon L, Boquist L. Giant cell tumour of bone. J Bone Joint Surg 1975;57:16773.

10. Cummins CA, Scarborough MT, Enneking WF. Multicentric giant cell tumour of bone. Clin Orthop 1996;332:24552.

8/14/2019 v10i2p120.pdf

9/9


11. Jaffe HL, Lichtenstein L, Portis RB. Giant cell tumour of bone. Its pathologic appearance, grading, supposed variants andtreatment. Arch Pathol 1940;30:9931031.

12. Bertoni F, Present D, Sudanese A, Baldini N, Bacchini P, Campanacci M. Giant cell tumour of bone with pulmonary metastasis.Six case reports and a review of the literature. Clin Orthop 1998;237:27585.

13. Bertoni F, Present D, Enneking WF. Giant cell tumour of bone with pulmonary metastasis. J Bone Joint Surg Am 1985;67:890900.

14. Cheng JC, Johnston JO. Giant cell tumour of bone. Prognosis and treatment of pulmonary metastases. Clin Orthop1997;338:20514.15. Kay RM, Eckardt JJ, Seeger LL, Mirra JM, Hak DJ. Pulmonary metastasis of benign giant cell tumor of bone. Six histologically

confirmed cases, including one of spontaneous regression. Clin Orthop 1994;302:21930.16. Rock MG, Pritchard DJ, Unni KK. Metastasis from histologically benign giant-cell tumour of bone. J Bone Joint Surg Am

1984;66:26974.17. Szyfelbein WM, Schiller AL. Cytologic diagnosis of giant cell tumour of bone metastasis to lung. A case report. Acta Cytologica

1979;23:4604.18. Vanel D, Contesso G, Rebibo G, Zafrani B, Masselot J. Benign giant-cell tumour of bone with pulmonary metastasis and

favorable prognosis. Report on two cases and review of the literature. Skeletal Radiol 1983;10:2216.19. Maloney WJ, Vaughan LM, Jones HH, Ross J, Nagel DA. Benign metastasizing giant-cell tumour of bone. Report of three

cases and review of the literature. Clin Orthop 1989:243:20815.20. Ladanyi M, Traganos F, Huvos AG. Benign metastazing giant cell tumours of bone. A DNA flow cytometric study. Cancer

1989;64:15216.

Documents

v10i2p120.pdf