CASE 6647 Published on 05.05.2008

Giant cell tumor of bone

Section

Musculoskeletal system

Case Type

Clinical Cases

Authors

Cinotti AM, Gagliano M, Scialpi M, Di Dio S, Magri C, Mannella P

Patient

76 years, male

Download as PDF Print Show related cases Notify admin

Clinical History

We describe a case of a patient who presented with localized pain of the right knee. There was no history of direct trauma.

Imaging Findings

A 76-year-old male presented with localized pain exacerbated with movement, joint effusion, soft tissue swelling and tenderness. Serum chemistries were normal. Radiography, CT and MR were performed. X-ray films showed an expansile eccentric lytic lesion in the proximal lateral tibial condyle. The margins of the lesions were largely nonsclerotic. Matrix mineralization was not present. There was not periosteal reaction. CT demonstrated a large radiolucent lesion with marked thinning of the cortex. MR showed a well defined lobulated lesion of low to moderate T1-signal intensity and heterogeneous high signal intensity on T2-weighted images. Both CT and MR demonstrated fluid levels.

Discussion

Giant cell tumour of bone (GCTOB) accounts for approximately 5% of primary bone tumours 20% of benign bone tumours. It may affect patients in all decades of life; however, the overwhelming majority (80%) of tumours occur in patients aged over 20 years. There is a slight female predominance. Giant cell tumour can arise in any bone of the skeleton, but it is most frequently detected around the knee (50%), involving the epiphyseal regions of the distal femur and proximal tibia. Most GCTOB are considered benign: however, they do have the potential for metastatic spread to the lungs and rarely mediastinal nodes. It has been estimated that the overall risk of metastatic spread from GCTOB is approximately 1-9%. Radiographically, GCTs are invariably radiolucent with no internal mineralization. The borders are usually nonsclerotic and geographical. The bone may be expanded with only a thin shell of cortex remaining and, as might be expected, with larger lesions pathological fractures can occur. Based on radiographic appearances the differentiation between an aggressive potentially metastasising GCT vs. the more common benign indolent tumour isn’t possible. CT provides a more detailed evaluation of tumour extent, which can prove helpful in surgical planning, especially in regions with complex anatomy. In particular, soft tissue involvement and articular surface involvement is better assessed with CT than with radiographs. CT may demonstrate fluid levels within GCTs. Fluid levels within bone tumours are non-specific and can be seen in aneurysmal bone cyst, chondroblastoma and telangiectactic osteosarcoma.The expanded and thinned cortex are vividly demonstrated and the presence or absence of matrix calcification can be assessed.
MR imaging is the imaging modality of choice for the assessment of intra- and extraosseous extent of tumour. The signal characteristics include low to intermediate T1-signal intensity and heterogeneous high signal intensity on T2-weighted images. The intramedullary extent of the tumour is best appreciated on T1-weighted images. A low signal rim on all pulse sequences similar to cortical bone may be identified separating the tumour from adjacent normal bone. Fluid levels may be demonstrated within the tumour mass. There are a number of conditions that may radiographically resemble GCTOB: aneurysmal bone cyst, intraosseous ganglion, chondroblastoma, osteosarcoma, and giant cell reparative granuloma (GCRG). In our case, the diagnosis of aneurysmal bone cyst was discarded because it tends to arise in a younger age group, tipically preskeletal fusion. Intraosseous ganglions are usually straightforward to differentiate from GCTOB. But they most commonly affect the carpus, are usually asymptomatic and have a sclerotic border. So this hypothesis was discarded. Chondroblastoma was rejected because it occurs in the immature skeleton of a child or adolescent and contains calcifications.The hypothesis of an osteosarcoma was discarded because there wasn’t aggressive periosteal reaction and osseous matrix. Giant-cell reparative granuloma (GCRG) was rejected because it is a reactive process that involves the small bones of hands and feet.Clinical history of the patient, physical examination and imaging features suggested a diagnosis of giant-cell tumour of bone and the biopsy confirmed our diagnosis.

Differential Diagnosis List

Giant cell tumor of bone

Final Diagnosis

Giant cell tumor of bone

References

[1] Yasko AW.
Giant cell tumor of bone.
Curr Oncol Rep. 2002 Nov;4(6):520-6. (PMID: 12354366)

[2] Lee MJ, Sallomi DF, Munk PL, Janzen DL, Connell DG, O'Connell JX, Logan PM, Masri BA.
Pictorial review: giant cell tumours of bone.
Clin Radiol. 1998 Jul;53(7):481-9. (PMID: 9714386)

[3] Alyas F, James SL, Davies AM, Saifuddin A.
The role of MR imaging in the diagnostic characterisation of appendicular bone tumours and tumour-like conditions.
Eur Radiol. 2007 Oct;17(10):2675-86. (PMID: 17342487)

[4] Van Dyck P, Vanhoenacker FM, Vogel J, Venstermans C, Kroon HM, Gielen J, Parizel PM, Bloem JL, De Schepper AM.
Prevalence, extension and characteristics of fluid-fluid levels in bone and soft tissue tumors.
Eur Radiol. 2006 Dec;16(12):2644-51. (PMID: 16612549)

[5] Resnick D, boint and joint imaging. In Resnick tumors and tumor-like lesions of bone: imaging and pathology of specific lesions. Saunders company, Philadelphia, London, Toronto, Montreal, Sydney, Tokyo, pp 1031-1036 (1996).

Case information

URL:	https://www.eurorad.org/case/6647
DOI:	10.1594/EURORAD/CASE.6647
ISSN:	1563-4086

Figure 1

Radiography

Fig. 1 X-ray films show an expansile, eccentric, lytic lesion located in the lateral tibial condyle. The borders are non-sclerotic. There is not matrix mineralization and periosteal reaction. We can see the cortical expansion.

Figure 2

Fig. 2 CT demonstrates an eccentrically located cystic lesion and cortical destruction. Fluid levels are present.

Figure 3

Fig.3b Axial T1-weighted and sagittal GE T2-weighted images show the tumour involving the proximal lateral tibial condyle.

Figure 4

Fig. 3a Coronal T1-weighted, T2-weighted and STIR images demonstrate a well defined lobulated lesion of low to moderate PD-signal intensity and heterogeneous high signal intensity on T2-weighted and STIR images

Figure 5

Fig.3c Sagittal PD, T2-weighted and STIR images show a lobulated lesion of low-intermediate SI in PD-weighted images and heterogeneous high SI on T2-weighted and STIR images.