JBRBTR, 2007, 90: 368-376.

EWINGS SARCOMA: IMAGING FEATURES

B. Peersman1, F.M. Vanhoenacker1, S. Heyman1, B. Van Herendael1, M. Stam2, P. Brys3, K.L. Verstraete4, I. Samson5,
J. Sybers6, P. Van Dyck1, P.M. Parizel1, A.M. De Schepper1,2

Aim: To define an imaging prototype of Ewings sarcoma (ES).

Materials and methods: Sixty-four patients with a histopathologically and/or genetically proven diagnosis of ES
were analyzed for clinical parameters (age, gender and location), radiographic and CT appearance (distribution,
matrix, margins, periosteal reaction, articular extension, cortical reaction and the presence of a pathologic fracture).
Size, local extension, signal intensity, degree and pattern of enhancement, and the presence of skip metastases were
evaluated on MRI.
Distant metastases were recorded on bone scintigraphy and chest CT scan.
Results: Patients age ranged between 7-67 (mean 17.9). Male/female ratio was 2.4/1.
Location in the pelvis was most frequent (31%), followed by the femur (20%) and tibia (11%). Most tumors were
mixed lytic-sclerotic (75%), and purely lytic in 25%. Plain films and CT scan showed a spiculated periosteal reaction
in 50%. A Codmans triangle was seen in 27%.
Articular extension was difficult to assess on radiographs. Cortical permeation and destruction is seen in respectively
31 and 42%, whereas cortical thickening is seen in 20%. Pathologic fracture occurred in 7.8%.
MRI showed a large mass, with a soft tissue component of more than 50% in 67%.
Degree and pattern of enhancement pattern was variable.
Signal intensity on T1- and T2-WI was non-specific.
Joint involvement was seen in 23%. Isolated involvement of the soft tissue (extraskeletal ES) was seen in 1.5%.
Skip metastases at initial presentation were present at initial presentation in 14% and distant metastases in 22%.
Conclusions: ES occurs in young patients. On radiographs/CT, 37.5% are located in the axial skeleton and 62.5% in
the peripheral skeleton. ES is mostly mixed sclerotic-lytic. A spiculated periosteal reaction is most frequent.The most
characteristic finding on MRI is the presence of a large soft tissue mass.
Key-word: Ewing sarcoma.

Ewings sarcoma (ES), peripheral

primitive neuroectodermal tumors
(PNET), and Askin tumors (AT) are
referred to as Ewings tumors
(ETs) (1). ES is the second most
common malignant bone tumor in
children and young adults, with an
unknown histogenesis (2).
ES is a round-cell sarcoma, showing varying degrees of neuroectodermal differentiation. These tumors
are cytogenetically well described,
and in 85% of the cases, a balanced
t(11;22)(q24;12) is found (3).
Most patients complain of pain,
occasionally
accompanied
by
swelling at the affected site.
Infrequent presenting symptoms
include fever, weight loss, cough,
anemia and leukocytosis.
Aim
The aim of this article is to define
an imaging prototype of Ewings
sarcoma, based on the analysis of a
large cohort of patients, originating

from four institutions, and to compare our findings with the literature.
Materials and methods
Sixty-four patients with histologically and/or genetically proven diagnosis of ES were included. In our
series, we assembled 52 plain radiographs, 36 CT scans and 62 MRI
scans. The MR imaging protocol
consisted of at least T1-weighted, T2weighted images (with or without
fat suppression) and T1-weighted
images after intravenous administration of gadolinium chelates in different slice directions.
We analyzed all patient files for
clinical parameters, i.e. age, gender
and location. Secondly, we evaluated the available radiographs/CT
scans for lesion distribution within
the skeleton (axial skeleton, flat
bones or epi-, meta- or diaphyseal
location within the long bones),
matrix, margins, periosteal reaction
and articular extension.

From: 1. Dept. of Radiology University Hospital Antwerp, 2. Dept. of Radiology Leiden

University Medical Center, 3. Dept. of Radiology University Hospital Leuven, 4. Dept.
of Radiology University Hospital Gent, 5. Dept. of Orthopaedic Surgery University
Hospital Leuven, 6. Vision Lab, University of Antwerp.
Address for correspondence: Dr B. Peersman, MD, Department of Radiology,
University Hospital Antwerp, Wilrijkstraat 10, B-2650 Edegem, Belgium.
E-mail: benjaminpeersman@hotmail.com.

Tumor matrix was scored on a

subjective scale as the sclerosis
compared to of osteolysis within
the osseous component of the
tumor.
The margins of the tumor on radiographs/CT were evaluated and
graded by the following criteria:
sharp demarcation, partially sharp
demarcation (less than 50%
unsharp) and unsharp demarcation
(more than 50% unsharp).
Periosteal reaction was scored as
lamellar or onion peel, interrupted
lamellar, Codmans triangle, sunburst or hair-on-end or mixed.
Codmans triangle is due to formation of reactive bone between the
elevated intact periosteum and the
underlying cortex at the zone of
transition to the extraosseous
extension of the tumor. Formation
of long, thin filiform spicules, radiating perpendicularly from the cortex,
is typical for a spicular, sunburst
periosteal reaction.
Articular extension on radiographs/CT was defined as cortical
breakthrough of the joint margins
and/or tumor extension on both
sides of the adjacent joint. Cortical
reaction was scored by following
criteria, namely cortical destruction,
cortical permeation and cortical
thickening or thinning. Also the
presence of a pathologic fracture at
initial presentation was evaluated.

EWINGS SARCOMA PEERSMAN et al

On MRI, the overall signal intensity of the tumor (intra- and

extraosseous component) on T1weighted images, T2-weighted
images and T2-weighted images
with fat suppression were recorded.
We also evaluated the degree and
pattern of enhancement of the
tumor on T1-weighted images after
intravenous
administration
of
gadolinium chelates. The tumor
volume was measured on contrast
enhanced T1-weighted images with
fat suppression, by using the
formula craniocaudal x transverse x
anteroposterior diameter x 0.5.
The percentage of bone and soft
tissue component of the tumor was
recorded on a subjective scale. Joint
involvement on MRI was evaluated,
using the same parameters as mentioned above (CT) and the presence
of concomitant joint effusion. The
presence of skip metastases within
the same bone was scored on T1weighted images.
Distant metastases were evaluated on bone scintigraphy and chest
CT-scan.

369

Fig. 1. Location of ES.

The most frequent location is the pelvis accounting for 31.25%, followed by the
femur (20.3%) and the tibia (11%). Location in humerus, fibula, scapula and rib is rare.
Only one extraskeletal ES was seen in our study (posterior aspect of the right knee
joint).

Results
The mean age of the population
is 17.9 years old with a range from 7
to 67 years. Our population demonstrates a male to female ratio of
2.4:1
(45 male
patients
and
19 female patients).
In the majority of our cases, the
lesion is located in the pelvis or
lower extremity.

The location of the lesions is

summarized in Fig. 1.
More than 90% of the lesions in
the long bones are located in the
(meta)diaphysis. Extension to the
epiphysis is rare.
A mixed sclerotic-lytic type with a
sclerotic component of 75% is most

frequently encountered (40%).

Lesions matrix is summarized in
Table I and illustrated in Fig. 2.
All tumors present as unsharp
marginated lesions on plain radiography/CT (Fig. 3A-C).
Aggressive spicular periosteal
reaction is frequently encountered

Fig. 2. Lesion matrix.

A. Plain radiograph of the pelvis shows an example of a purely lytic ES of the left iliac bone.
B. Axial CT image of a mixed sclerotic-lytic type in the right
femoral neck. Note areas of sclerosis (white arrow) combined
with areas of osteolysis (bone destruction) (black arrow).

370

JBRBTR, 2007, 90 (5)

Evidence of joint involvement is

present in 23% of cases.
Fig. 5 demonstrates joint extension on different imaging modalities.
Cortical permeation and destruction are common findings in our
study (respectively 31 and 42%),
whereas cortical thickening is rather
uncommon (20%). A pathologic
fracture is seen in 7.8% of cases.
The overall signal intensity on T1weighted images of the tumor is in
70% of cases equal to that of the
surrounding muscle. In the other
30% we notice an overall signal
intensity higher than surrounding
muscle (Fig. 6).
Seventy-three percent of the
tumors in our series show high signal intensity on T2-weighted images.
The remainder show signal intensity
equal to that of fat (Fig. 7).
All tumors show an overall signal
intensity higher than muscle but
lower than water on T2-weighted
images with fat suppression (Fig. 8).
MRI reveals skip metastases in
14% (Fig. 9). The soft tissue component occurs to be more than 50% in
67% of cases (Fig. 10). No specific
degree and pattern of enhancement
can be demonstrated in our series.
Distant metastases are seen in 22%
(14% pulmonary metastases, 5%
bone metastases on bone scintigraphy and 3% both).
Discussion

D
C
Fig. 3. Local tumor extension.
A. Plain radiograph of the forearm. Codmans triangle and spicular periosteal reaction at the ulnar side of the radius. Cortex permeation and subtle moth-eaten pattern
of the neighbouring bone. The intra- or extraosseous extension of the lesion is hardly
to demonstrate on plain film.
B-C. CT scan also shows the periosteal reaction and cortical permeation. Extension
of a soft tissue mass towards the ulna (black arrows).
D. MRI of the same lesion. T1-weighted image after intravenous administration of
gadolinium contrast. Note the sharp margins of the intraosseous extent, in contrast
with the normal bone marrow. The extraosseous component of the tumor is well
demonstrated (white arrows).

in our study and presents in more

than 50% of cases. Laminated
periosteal reaction is not frequently
demonstrated (only 14%). The differ-

ent types of periosteal reaction are

demonstrated
in
Fig. 4
and
described in order of occurrence in
Table II.

Ewings sarcoma (ES) is an

aggressive, highly malignant bone
tumor that belongs to the group of
small round (blue) cell tumors and
occurs predominantly in the second
decade of life. Extraskeletal ES is
extremely rare and occurs in a
slightly older age group. The mean
age in our study is 17.9 years old.
The male to female ratio in our population is slightly higher than
described in literature, i.e. 2.4/1 versus 3/2 in most studies (3-6).
ES has a predilection for the
pelvis and lower extremities (4-7).
The most frequent location in our
study is the pelvis, followed by the
femur, tibia and humerus. In long
bones, lesions are located in the
(meta)diaphysis and rarely extend
into the epiphysis (8).
Reinus et al. (6) reviewed the
radiographs of 373 patients and
described the radiographic features
of ES. Findings were divided into
three categories, depending on their
frequency of occurrence: common
findings (> 30%), uncommon findings (> 10%, < 30%) and rare findings
(< 10%). Poor margination, soft tis-

EWINGS SARCOMA PEERSMAN et al

Table I. Percentage of sclerotic component in lesion matrix

An exclusively sclerotic lesion is infrequent (6%). A mixed sclerotic-lytic type

with a sclerotic component of 75% occurs in 40% of cases. A 50% sclerotic50% lytic type is seen in 20% of the cases. Only 11% present as a lesion with
a sclerotic component of 25%. 23% of ESs in our study present as a purely
lytic lesion (sclerotic component of 0%).

Table II. Type of periosteal reaction frequency

Frequency of occurrence of the different types of periosteal reaction. The

onion peel reaction occurs in 14%, the interrupted lamellar form in 18%,
Codmans triangle in 27% and the spicular periosteal reaction in 52% of
cases.

sue involvement, permeative component, laminated periosteal reaction and sclerotic matrix were
described as common findings.

Spiculated periosteal reaction, cortical thickening and violation, purely

lytic matrix, pathologic fracture, cystic component and bone expansion

371

were described as uncommon findings. Soft tissue calcification,

saucerization, honeycombing, sharp
margins and vertebra plana were
described as rare findings. Poor
margination, extensive soft tissue
component, sclerotic matrix and
permeation are also common findings in our study. On the other hand,
laminated periosteal reaction is an
uncommon finding in our study
(14%). Although Reinus et al.
described spiculated periosteal
reaction as an uncommon finding ,
it is more than common (> 50%) in
our series. The milder forms of
periosteal reaction, namely laminated periosteal reaction and onion
skin were described also by Resnick
to be common manifestations
(57%), and the spiculated periosteal
reaction as a less frequent manifestation (28%) (8).
Cortical permeation and destruction are common findings in our
study (respectively 31 and 42%),
whereas cortical thickening is rather
uncommon (20%). A pathologic
fracture (7.8%), as well as soft tissue
calcification and sharp margins are
rare findings in our study.
Pathologic fracture was described
as an uncommon finding (14%) by
Reinus et al.
CT scan is a good imaging
modality to evaluate joint extension,
periosteal reaction and matrix of the
lesion. A large soft-tissue mass is
well illustrated by CT-scan, especially after intravenous administration
of contrast material.
MRI is accurate in the assessment
of the intramedullary tumor extent
in patients with bone sarcoma
which is documented by the excellent correlation between longitudinal T1-weighted images and identical macrosections of the surgical
specimens. Transverse TSE T2weighted images best display the
interface between tumor and adjacent soft tissues and the anatomical
relationship with the neurovascular
structures, allowing differentiation
between intracompartmental and
extracompartmental
disease.
Contrast between tumor and normal
tissue, especially fat-containing tissue, is greatly enhanced by combining TSE with fat-selective presaturation. T1-weighted images after
administration of contrast material
can be successfully combined with
fat-selective
presaturation
to
enhance contrast resolution (8-23).
Non-homogeneous signal intensity
is seen on all pulse sequences. No
specific degree and pattern of
enhancement on MRI is seen. Skip

372

JBRBTR, 2007, 90 (5)

Fig. 4. Variable periosteal reaction in different patients.

Plain radiograph of the femur. Lamellar periosteal reaction.
CT scan of the same patient. Lamellar periosteal reaction.
Plain radiograph of the distal ulnar diaphysis. Note the interrupted (arrow) lamellar periosteal reaction, depicting regions of more
aggressive tumor extension into the soft tissue.
D-E. Plain radiograph of the femur and sagittal reformatted CT scan of the forearm in 2 different patients. Codmans triangle or
spur (arrows).
F-G. Radiographs and CT scan of the femur. Aggressive spicular, sunburst periosteal reaction.

EWINGS SARCOMA PEERSMAN et al

373

Fig. 5. Evaluation of local tumor extension (bone and soft tissue component).Evaluation of joint extension by different imaging modalities.
A. Plain radiograph of the pelvis. Sclerotic lesion of the left iliac bone. Evaluation of soft tissue component and articular extension is not possible.
B. Axial CT scan. Sclerosis is mainly located at the left iliac bone (black arrow) but minor sclerosis on the sacral side of the sacroiliac joint suggests articular extension (white arrow).
C. Axial T2-weighted image with fat saturation. We can now clearly demonstrate the articular extension of the tumor and invasion
of the sacrum (white arrow). Note also huge soft tissue component (black arrows).
D. Axial T1-weighted image after intravenous administration of gadolinium chelates. This sequence also demonstrates the articular
extension of the pelvic tumor and tumoral invasion of the sacrum (white arrow). Soft tissue extension is indicated by black arrows.

Fig. 6. Signal intensity on T1-weighted images.

A. Coronal T1-weighted image of the sacrum. ES of the sacrum (arrows), with a signal intensity equal to
that of the surrounding muscles.
B. Coronal T1-weighted image of the pelvis. Ewings sarcoma of the left iliac bone with a large soft-tissue
component (arrows). The overall signal intensity is higher than that of surrounding muscles, but lower than fat.

374

JBRBTR, 2007, 90 (5)

Fig. 7. Signal intensity on T2-weighted images.

Axial T2-weighted image of the pelvis. ES of the right iliac
bone, with a large soft tissue component (arrows). The overall
signal intensity is higher than the surrounding muscles, but
lower than fat.

Fig. 9. Skip metastases.

Coronal T1-weighted image of the left femur. Ewings sarcoma at the middiaphysis. Note the skip metastases (arrows) in
the distal portion of the femur.

Fig. 8. Signal intensity on fat suppressed T2-weighted

images.
Axial T2-weighted image with fat suppression. ES of the left
iliac bone, with a large soft-tissue component. The overall signal intensity is higher than that of the surrounding muscles, but
lower than water (arrows).

metastases are present in 14% of

cases. Jiya et al. described the
occurrence of skip lesion in ES to be
rare (24). Most tumors in our series
have a large soft tissue component,
65% presented with a soft tissue
component of more than 50%.
MRI is very sensitive, but less
specific for the determination of epiphyseal involvement. It is highly
sensitive for excluding joint involvement although false-positive results
may occur secondary to synovial
inflammatory reactions (25, 26).
With the recent developments of
multiplanar imaging, CT scan is a
good alternative in local staging of
the tumor. Bloem et al. and also
many other authors in the 1980s
described that MRI was significantly
superior to CT and scintigraphy in

defining intraosseous tumor length

and was as accurate as CT in
demonstrating cortical bone and
joint involvement (27). But MRI was
definitely superior to CT in demonstrating involvement of muscle
compartments. They concluded that
MRI is the modality of choice for
local staging of primary bone sarcoma. Panicek et al. concluded in 1997
that CT and MR imaging are equally
accurate in the local staging of
malignant bone and soft-tissue neoplasms in specific anatomic sites
studied (28). In our study, MRI is the
imaging modality of choice because
of its high contrast resolution and
the ability to define the margins of
the soft-tissue component, especially on T2-weighted images and T1weighted imaging after intravenous

administration
of
gadolinium
chelates. This is accentuated in difficult anatomic regions (e.g. pelvis).
We couldnt compare the value of
MRI and CT scan because the number of CT-scans available was limited, many scans contained only
images in axial plane and not all
scans were performed after intravenous contrast administration.
Other limitations in our study are
due to the retrospective design.
There were indeed slight differences
in the MR protocol between the different institutions and not all
images were available in a full digital format (DICOM) rendering more
accurate volume measurements
impossible.
Conclusion
Ewings sarcoma is a highly
malignant bone tumor that predominantly occurs in young patients. The
most characteristic finding is the
presence of a large soft tissue mass,
which is best demonstrated on MRI.
Seventy percent of these tumors
arise in the pelvis or lower extremity. In contradistinction to the literature, a spiculated periosteal reaction
is seen in more than half of our
patients. Ewings sarcoma presents
most frequently as a mixed sclerotic-lytic type of tumor, and most

EWINGS SARCOMA PEERSMAN et al

375

B
A

tumors occur in the (meta)diaphysis

of the long bones. Cortical destruction and permeation is a frequently
encountered feature. Plain radiography and CT scan are good imaging
modalities to describe periosteal
reaction, lesions matrix and cortical
changes. MR imaging remains the
imaging modality of choice for local
tumor evaluation.
Acknowledgments
We are grateful to R. Forsyth
(University of Ghent) and A. Van
Erkel (Leiden University Medical
Center) for their contribution in the
preparation of this manuscript.

Fig. 10. Local tumor extension.

A-B. CT scan of the lower leg. Aggressive tumoral lesion of the fibula with cortical
permeation and destruction. Spicular periosteal reaction. Presence of a large soft tissue component (arrows).
C-D. Sagittal T1-weighted image after intravenous administration of gadolinium
chelates (C) and an axial T2-weighted image (D). ES of the fibula with the presence of
a large soft tissue mass (arrows). The soft tissue extension of the tumor is better appreciated on MR images in comparison to the CT-images.

References
1. Lazar A., Abruzzo L.V., Pollock R.E.,
Lee S., Czerniak B.: Molecular
Diagnosis of Sarcomas, Chromosomal Translocations in Sarcomas.
Arch Pathol Lab Med, 2006, 130:
1199-1207.
2. Udayakumar A.M., Sundareshan T.S.,
Goud T.M., et al.: Cytogenetic characterization of Ewing tumors using fine
needle aspiration samples: a 10-year
experience and review of the literature. Cancer Genet Cytogenet, 2001,
127: 42-48.
3. Szuhai K., IJszenga M, Tanke H.J.,
Rosenberg C., Hogendoorn P.C.:
Molecular cytogenetic characterization of four previously established
and two newly established Ewing

4.

5.

6.

7.

sarcoma cell lines. Cancer Genet

Cytogenet, 2006, 166: 173-179.
Schajowicz F.: Marrow tumors. In:
Tumors and tumorlike lesions of the
bone: Pathology, radiology and treatment. Edited by Schajowicz F.,
Second edition. Printed by SpringerVerlag, Berlin, 1994, pp 301-325.
Dahlin D.C., Unni K.K.: Ewings tumor.
In: Bone tumors. General aspect and
data on 8,542 cases. Edited by
Dahlin D.C and Unni K.K. Printed by
C.C. Thomas Springfield, Fourth edition, Illinois, 1986, pp 322-336.
Reinus W.R., Gilula L.A., IESS
committee: Radiology of Ewings
sarcoma: Intergroup ES study (IESS).
Radiographics, 1984, 4: 929-944.
Ilaslan H., Sundaram M., Unni K.K.,
Dekutoski M.B.: Primary Ewings

376

8.

9.

10.

11.

12.

13.

14.

JBRBTR, 2007, 90 (5)

sarcoma of the vertebral column.
Skeletal Radiol, 2004, 33: 506-513.
Resnick D., Kransdorf M.J.: Tumors
and tumorlike diseases. In: Bone and
Joint imaging. Edited by Resnick D.
and Kransdorf M.J., third edition.
Printed by Elsevier, Philadelphia,
2005, pp 1192-1198.
Pommersheim W. J, Chew F.S.:
Imaging, diagnosis, and staging of
bone tumors: A primer. Semin
Roentgenol, 2004, 39: 361-372.
Hanna S.L., Fletcher B.D., Kaste S.C.,
Fairclough D.L., Parham D.M.:
Increased confidence of diagnosis of
Ewing sarcoma using T2-weighted
MR images. Magn Reson Imaging,
1994, 12: 559-568.
Van der Woude H.J., Bloem J.L.,
Hogendoorn P.C.: Preoperative evaluation and monitoring chemotherapy
in patients with high grade
osteogenic and Ewings sarcoma:
review of current imaging modalities. Skeletal Radiol, 1998, 27: 57-71.
Wilner D.: Ewing sarcoma. In:
Radiology of bone tumors and allied
disorders. Edited by Wilner D.
Printed
by
WB
Saunders,
Philadelphia, 1982, pp 2462-2573.
Boyko O.B., Cory D.A., Cohen M.D.,
Provisor A., Mirkin D., DeRosa G.P.:
MR Imaging of osteogenic and
Ewings sarcoma. AJR Am J
Roentgenol, 1987, 148: 317-322.
Bloem
J.L.,
Bluemm
R.G.,
Taminiau A.H., Van Oosterom A.T.,
Stolk J., Doornbos J.: Magnetic resonance imaging of primary malignant
bone tumors. Radiographics, 1987, 7:
425-445.

15. Dwyer
A.J.,
Glaubiger
D.L.,
Ecker
J.G.,
Doppman
J.L.,
Prewitt J.M., Plunkett J.: The radiographic follow-up with Ewing
Sarcoma: A demonstration of a general method. Radiology, 1982, 145:
327-331.
16. Redmond
O.M.,
Stack
J.P.,
Dervan P.A., Hurson B.J., Carney D.N.,
Ennis J.T.: Osteosarcoma: Use of MR
imaging and MR spectroscopy in clinical decision making. Radiology, 1989,
172: 811-815.
17. Ehman
R.L.,
Berquist
T.H.,
McLeod R.A.: MR Imaging of the musculoskeletal system: a 5-year appraisal. Radiology, 1988, 166: 313-320.
18. Vogler J.B., Murphy W.A.: Bone
marrow imaging. Radiology, 1988,
168: 679-693.
19. Pettersson
H.,
Gillespy
T.,
Hamlin D.J., et al.: Primary musculoskeletal tumors : examination with
MR imaging compared with conventional modalities. Radiology, 1987,
164: 237-241.
20. Mirowitz
S.A.,
Apicella
P.,
Reinus W.R., Hammerman A.M.: MR
imaging of bone marrow lesions: relative conspicuousness on T1-weighted, fat-suppressed T2-weighted, and
STIR images. AJR Am J Roentgenol,
1994, 162: 215-221.
21. Gillespy T., Manfrini M., Ruggieri P.,
Spanier
S.S.,
Pettersson
H.,
Springfield
D.S.:
Staging
of
intraosseous extent of osteosarcoma: correlation of preoperative CT
and MR imaging with pathologic
macroslides. Radiology, 1988, 167:
765-767.

22. Gorospe L., Fernndez-Gil M.A.,

Garcia-Raya P., Royo A., LopezBarea F., Garcia-Miguel P.: Ewings
sarcoma of the mandible: Radiologic
features with emphasis on magnetic
resonance appearance. Oral Surg
Oral Med Oral Pathol Oral Radiol
Endod, 2001, 91: 728-34.
23. Singh P., Jain M., Singh D.P., Kalra N.,
Khandelwal N., Suri S.: MR findings
of primary Ewings sarcoma of
greater wing of sphenoid. Australas
Radiol, 2002, 46: 409-411.
24. Jiya T.U., Wuisman P.I.: Long-term follow-up of 15 patients with nonmetastatic Ewings sarcoma and a
skip lesion. Acta Orthop, 2005, 76:
899-903.
25. Norton
K.I.,
Hermann
G.,
Abdelwahab
I.F.,
Klein
M.J.,
Granowetter L.F., Rabinowitz J.G.:
Epiphyseal involvement in osteosarcoma. Radiology, 1991, 180: 813-816.
26. Leung J.C., Dalinka M.K.: Magnetic
Resonance Imaging in Primary Bone
Tumors. Semin Roentgenol., 2000,
35: 297-305.
27. Bloem
J.L.,
Taminiau
A.H.,
Eulderink
F.,
Hermans
J.,
Pauwels E.K.: Radiologic staging of
primary bone sarcoma: MR imaging,
scintigraphy, angiography, and CT
correlated with pathologic examination. Radiology, 1988, 169: 805-810.
28. Panicek
D.M.,
Gatsonis
C.,
Rosenthal D.I., et al.: CT and MR
imaging in the local staging of
primary malignant musculoskeletal
neoplasms: report of the radiology
diagnostic
oncology
group.
Radiology, 1997, 202: 237-246.