Beruflich Dokumente
Kultur Dokumente
discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/13515995
CITATIONS
READS
27
35
11 AUTHORS, INCLUDING:
Won-Hee Jee
Bo-Young Choe
SEE PROFILE
SEE PROFILE
Jin-suck Suh
Yonsei University
SEE PROFILE
SEE PROFILE
Won-Hee
Jee, MD
Bo-Young
Choe,
Heung-Sik
Kyung-Jin
Kang, MD
Suh, MD
Nonossifying
Characteristics
with
Pathologic
PhD
Jin-Suck Suh, MD
Kyung-Nam
Ryu, MD
Yeon-Soo
Lee, MD
In-Young
Jung-Man
Ok, MD
Kim, MD
Kyu-Ho
Choi,
PURPOSE:
To correlate
the findings
of nonossifying
(MR) imaging
with those at pathologic
examination.
Shinn,
MATERIALS
years) with
MD
for
signal
and
Index terms:
Bone neoplasms,
in infants and
children,
40.31 31
Bone neoplasms,
MR. 40.121411,
From
1998;
the
of Radiology
Kyungpook
National
University,
Taegu,
South
Korea
(K.jS.);
Department
of
Radiology,
Yonsei
University,
Seoul,
South Korea (J.S.S.); and Department
of Radiology,
Kyung
Hee University,
Seoul, South Korea (K.N.R.).
From the
1997
RSNA
scientific
assembly.
Received
November
1 2, 1 997; revision
requested
December
31 ; final revision
received
March
1 7, 1998; accepted
April 23. Address
reprint
requests
to
W.H.J.
were
were
correlated.
hypointense
images,
all
nonossifying
fibromas
of skeletal
muscle.
On T2-weighted
and
four
(21%)
were
hyperintense.
had
low
signal
images,
1 5 lesions
On gadolinium-
CONCLUSION:
The distinguishing
features of nonossifying
fibroma
included
hypointensity
and septation
on T2-weighted
images.
Signal
intensity
on Ti
and T2weighted
MR images
and the patterns
of contrast
enhancement
were dependent
on
the amounts
of hypercellular
fibrous
tissue,
hemosiderin,
hemorrhage,
collagen,
foamy
histiocytes,
and bone trabeculae.
-
Nonossifying
fibroma
is the most
common
benign
bone
tumor
and is typically
encountered
during
childhood
and adolescence
(1). Most
lesions
are discovered
incidentally
and
show
classic
radiographic
features
within
the
long
bones
(2,3).
There
is no specific
indication
for magnetic
resonance
(MR) imaging
in cases of nonossifying
fibroma
unless
the radiographic
appearance
is atypical.
To our knowledge,
the appearance
of nonossifying
fibroma
at MR imaging,
with
pathologic
correlation,
has not
been
fully
illustrated
previously
(4,5).
The purpose
of this study
was to describe
the findings
at MR imaging
in patients
with
nonossifying
fibroma
and to correlate
them
with pathologic
findings.
1998
AND
MATERIALS
Author
contributions:
Guarantor
of integrity
of entire study,
W.H.j.;
study
concepts,
H.S.K.;
study
design,
W.H.J.;
definition
of intellectual content,
J.S.S., K.N.R.;
literature
research, W.H.J.; clinical studies, J.M.K.,
I.Y.O.;
experimental
studies,
Y.S.L.;
data
acquisition,
H.S.K.,
K.J.S.; data
resonance
enhanced
images,
intense
contrast
enhancement
was seen throughout
1 5 lesions
(heterogeneous
pattern
in 1 2 and homogeneous
in three)
and in the margins
and
septa
in four. Extensive
hypercellular
fibrous tissue and hemosidenn
seen at pathologic
examination
were depicted
with low signal intensity
on T2-weighted
MR images.
209:197-202
Departments
biopsy
(79%)
RSNA,
at magnetic
AND METHODS:
In 1 9 patients (age range, 8-25 years; mean age, 14
pathologically
proved nonossifying
fibroma,
MR images were analyzed
intensity
and patterns of contrast
enhancement.
Findings
at MR imaging
RESULTS:
On Ti -weighted
intensity
compared
with that
40.12143,
40.1 21415
Bones, fibroma, 40.3131
fibroma
MD
Kyung-Sub
Radiology
Fibroma:
at MR Imaging
Correlation1
B.Y.C.,
K.H.C.,
prepara-
editing
K.S.S.
and
METHODS
From February
1993 through
March
1997,
19 patients
(10 female
and nine male patients;
age range,
8-25 years;
mean
age, 14 years)
with nonossifying
fibroma
participated
in this
study. The study
was approved
by the ethics
committee
at each
institution,
and written
informed
consent
was obtained
for each subject.
Each subject
underwent
plain
radiography, MR imaging,
and biopsy
of the nonossifying
fibroma.
Pathologic
findings
were based
on commonly
accepted
histologic
criteria
(3). MR imaging
and subsequent
biopsy
with
curettage
and
bone
grafting
were
performed
for the following
reasons:
prevention
of
pathologic
fracture
(ii =
1 1), pathologic
fracture
(ii =
3), persistent
pain (n = 3), and
atypical
findings
We reviewed
and
pattern
septation,
patterns
at radiography
the MR images
of
homogeneity;
on abnormal
of increased
signal
contrast.
(ii
and
the
2).
noted
sites
presence
intensity
Signal
of skeletal
of
involvement;
peripheral
of bone
marrow
intensity
lower
than
the
hypointense
and
or
signal
rim,
extraosseous
equal
to that
intensity
internal
areas;
and
of skeletal
197
Patient,
MR Imaging,
and Pathologic
MR Imaging
Fibroma
Findings
Enhancement
Pattern on
Signal
Patient
No./
Intensity
on
12-weighted
Age (y)/Sex
Images
1/1 5/M
Low
Fat-suppressed,
Gadolinium-enhanced
Images
Heterogeneous,
Septation
Bone
Marrow
Change
SoftTissue
Change
Present
Present
Present
Marked
Present
Hypercellular
Fibrous Tissue
Hemosiderin
Pathologic
Features
Hemorrhage
Collagen
Foamy
Histiocytes
Slight
Slight
Absent
Absent
Bone
Trabeculae
intense
2/1 1/F
Low
Homogeneous,
intense
Present
Absent
Absent
Marked
Absent
Absent
Slight
Absent
Absent
3/8/M
High
Heterogeneous,
Present
Present
Absent
Slight
Absent
Absent
Slight
Marked
Slight
Present
Present
Absent
Absent
Absent
Absent
Marked
Moderate
Absent
Absent
Slight
Absent
Slight
Slight
Absent
Absent
Slight
Marked
Present
Absent
Absent
Marked
Absent
Absent
Slight
Slight
Slight
Present
Present
Present
Slight
Absent
Slight
Slight
Moderate
Absent
Absent
Present
Absent
Moderate
Absent
Slight
Moderate
Absent
Slight
Present
Absent
Absent
Marked
Absent
Absent
Slight
Absent
Slight
Present
Absent
Absent
Moderate
Absent
Slight
Moderate
Marked
Slight
intense
4/10/F
Low
Marginal,
5/25/M
Low
Homogeneous,
intense
septal
6/1 3/M
Low
Heterogeneous,
intense
Homogeneous,
intense
Heterogeneous,
intense
Heterogeneous,
intense
Heterogeneous,
intense
Marginal,
septal
7/14/F
High
8/1 1/F
Low
9/15/F
Low
10/1 1/F
Low
1 1/1 3/F
12/1 3/M
Present
Present
Present
Marked
Present
High
Heterogeneous,
intense
Present
Absent
Absent
Slight
Absent
Slight
Moderate
Absent
Absent
Absent
Marked
Slight
Slight
1 3/1 7/M
Low
14/11/F
Low
Present
Present
Absent
Absent
Absent
Absent
Marked
Moderate
Present
Absent
Slight
Slight
Slight
Slight
Absent
Marked
Slight
Slight
1 5/16/F
16/10/M
Low
Low
Present
Present
Absent
Present
Absent
Present
Marked
Marked
Absent
Absent
Absent
Absent
Slight
Slight
Absent
Slight
Slight
Slight
1 7/20/M
Low
Marginal,
septal
Heterogeneous,
intense
Marginal,
septal
Heterogeneous,
intense
Heterogeneous,
Present
Absent
Absent
Moderate
Absent
Slight
Slight
Moderate
Slight
Present
Absent
Absent
Slight
Absent
Slight
Slight
Moderate
Slight
Present
Absent
Absent
Marked
Absent
Absent
Slight
Absent
Slight
Low
intense
18/19/M
High
19/14/F
Low
Heterogeneous,
intense
Heterogeneous,
intense
muscle
was considered
hypointense;
greater
than that of skeletal
muscle
but
less than that of fat, intermediate;
equal
to or greater
than that of fat, hyperintense.
MR imaging
was performed
with i.5-T
(Signa
Advantage,
GE Medical
Systems,
Milwaukee,
Wis; Magnetom
SP 4000, Siemens
Medical
Systems,
Isebin, NJ) and
1 .0-T (Magnetom;
Siemens
Medical
Systems, Erbangen,
Germany)
imagers,
with
the general-purpose,
circular,
14-cm-diameter
surface
coil
and
the
body
coil.
In all
cases, Ti-weighted
MR imaging
(repetition time msec/echo
time msec
350650/i 1-30) was performed.
In six cases,
=
proton-density-weighted
(1,800-2,700/
19-40)
and T2-weighted
(1,800-2,700/60100) imaging
were performed.
In 13 cases,
a fast spin-echo
pulse sequence
was used
to perform
proton-density-weighted
(2,500-4,000/i
(2,500-4,000/76-108)
5-40)
and
T2-weighted
imaging.
In
cases, imaging
was performed
suppression.
In all cases, axial,
198
Radiology
#{149}
October1998
#{149}
10
with fat
coronal,
and sagittab
fat-suppressed
Ti-weighted
imaging
was performed
after infusion
of
0.1 mmol per kilogram
of body weight of
gadopentetate
dimeglumine
(Magnevist;
Schering,
Berlin,
Germany).
Typical
MR
imaging
parameters
included
the following: field of view, 10-20 cm; two signals
acquired;
matrix
size, 256 x 192; section
thickness,
2.0 mm;
Histopathobogic
by a bone pathologist
of MR imaging
and
slides
were
(Y.S.L.).
pathologic
gap,
eight.
1.5-
reviewed
Correlation
findings
was performed
by a radiologist
(W.Hj.)
and the pathologist.
Correlation
was general rather than site to site because
curettage,
not complete
resection,
was the
method
of tumor removal.
Histologic
sections were evaluated
for relative
amounts
of hypercellular
fibrous
tissue, hemosidenin, hemorrhage,
collagen,
foamy histiocytes, and bone trabeculae.
Correlations
were performed
between
pathologic
and
MR imaging
findings
and radiographic
and MR imaging
findings.
RESULTS
All lesions
involved
the metaphysis
and
revealed
intramedulbary
extension,
with
the following
skeletal distribution:
proximal tibia (n
8), distal tibia (n
4), distal
femur
(n
3), proximal
fibula (n = 2),
proximal
femur (n
1), and distal fibula
(n = 1). Multiple
lesions
were not observed
in any case. Ti-weighted
MR images showed
homogeneous
decreased
signal intensity
throughout
the lesion; other
findings
are listed in the Table. On T2weighted
images,
iS of the 19 lesions
=
(79%)
four
were
(2i%)
hypointense
(Figs
1, 2) and
were
hypenntense
(Fig 3). The
hypointensity
on T2-weighted
images corresponded
to extensive
hypercellular
fibrous
tissue
at histologic
examination
(Fig 1) in all 15 lesions.
Four of these 15
lesions
were depicted
with
very low signab intensity
on T2-weighted
images, and
three of the four lesions
were found
to
contain
hemosidenin
at histologic
examination
(Fig 2). The hyperintensity
on
Jee
et
al
b.
::
#{149}
#{149}.
C.
.:
_\.
-$
.-:
:
:::,.
..
I:
d.
Figure
1. Patient
(arrows).
(3,000/84;
4.
Nonossifying
(b)
Sagittal
echo-train
TI-weighted
length,
uniform,
tibia
benign-appearing,
reveals
of the distal
(Hematoxylin-eosin
reactive
sclerosis
(arrows)
stain;
that
Number
#{149}
original
surrounds
T2-weigh
ted
i mages
corresponded
to
scanty
hypercellular
fibrous
tissue
and
massive
aggregation
of foamy
histiocytes
(Fig 3) in the four remaining
lesions.
At
histologic
examination,
nonossifying
fibroma
showed
variable
amounts
of hemo-
VoIume2O9
tibia.
(a) Lateral
radiograph
image
(450/16)
shows
a hypointense
demonstrates
a hypointense
mass
fat-suppressed
Ti-weighted
MR image
(350/16)
of nonossifying
fibroma
that
involves
the distal
tibia
spindle-shaped
fibroblasts
arranged
in intersecting
(d) Gadolinium-enhanced,
(e) Photomicrograph
cells do not appear.
fibroma
MR
eight)
magnification,
the
tumor
component.
x 150.)
shows
an eccentric,
radiolucent
lesion
with
a sclerotic
margin
mass
(arrows).
(c) Sagittal,
fast spin-echo,
T2-weighted
MR image
with
marginal
(arrows)
and
septal
(arrowheads)
hyperintensity.
reveals
marginal
(arrows)
and
septal
(arrowheads)
enhancement.
shows
a large area of hypercellular
fibrotic
lesions
( * ) composed
of
fascicles.
Multinucleated
giant
cells (arrows)
are scattered,
hut foam
(f) Photomicrograph
(Hematoxylin-eosin
siderin,
hemorrhage,
collagen,
foamy
histiocytes,
and bone tnabecubae.
In all 19 cases,
a peripheral
hypointense rim on MR images
corresponded
to
marginal
sclerosis
on plain
radiographs
and
to peripheral
reactive
sclerosis
at
of nonossifying
stain;
original
fibroma
magnification,
that
involves
the distal
x40.)
(Fig
If). In 18
internal
septation
was seen
on MR images
that
correlated
with
trabeculation
on plain radiographs
(Figs 1-3).
All five lesions
(26%)
with
hyperintense
septa on TI-weighted
images
were hyperhistologic
cases
examination
(95%),
Nonossifying
Fibroma
at MR Imaging
199
#{149}
1i
-;t
I
b.
a.
C.
,,
,
#{149}1
r::
d.
Figure
2. Patient
1.
..
f.
e.
Nonossifying
fibroma
of the proximal
fibular
metaphysis.
(a) Anteroposterior
radiograph
shows
a centrally
located
osteolytic
erosion
and a pathologic
fracture.
(b) Sagittal
Ti-weighted
MR image
(400/i
1) shows
a hypointense
mass (arrows)
with
marrow
(arrowheads).
(c) Sagittal,
fat-suppressed,
fast spin-echo,
T2-weighted
MR image
(3,000/84;
echo-train
length,
hypointense
mass
(black
arrows)
with
adjacent
abnormal
high
signal
intensity
of the bone
marrow
(white
arrows).
(d) Sagittal gadolinium-enhanced,
fat-suppressed
Ti-weighted
MR image (350/11)
shows
heterogeneously
intense
enhancement
(arrows).
(e) Axial
gadolinium-enhanced,
fat-suppressed
Ti-weighted
MR image
(550/i2)
demonstrates
adjacent,
extraosseous
hyperintensity
(arrows).
(f) Photomicrograph of nonossifying
fibroma
in the proximal
fibula reveals fibrotic stromal
cells in which
hemosiderin
pigment
(arrows)
is deposited.
(Prussian
blue
stain; original
magnification,
X400.)
lesion
(arrows)
intratumoral
eight)
reveals
intense
showed
with endosteal
remaining
bone
a septate,
very
on
T2-weighted
contrast
enhancement
binium-enhanced
1). Among
ointense
200
TI-weighted
the
septa
Radiology
#{149}
images
on
14 lesions
on
TI-weighted
October
#{149}
(74%)
1998
and
gado-
images
(Fig
with
hyp-
images,
three
(16%)
were
depicted
septa on T2-weighted
contrast
enhancement
enhanced
Ti-weighted
1 1 (58%) were depicted
as hyperintense
images
and showed
on gadoliniumimages
(Fig 2) and
as hypointense
septa
on T2-weighted
images
and showed
no
contrast
enhancement
on gadoliniumenhanced
TI-weighted
images
(Fig 3). In
four cases (21%),
abnormal
signal
intensity
was
depicted
in
both
bone
marrow
Jeeet
al
b.
a.
c.
#{231}
..-..
I#{149}.(
Figure
#{149}
.#{149}
:.
.#{149}
p.
#{149} ,
k
#{149}
.#{149}.
#{149}
.:.#{149}:
. .1
:.
12.
Nonossifying
fibroma
3. Patient
of the proximal
#{149}
shows
a heterogeneously
(arrows).
(c) Coronal,
echo,
T2-weighted
train
length,
hyperintense
linium-enhanced,
hypointense
fat-suppressed,
MR image
eight)
mass
reveals
mass
fast spin-
(3,000/108;
echo-
a heterogeneously
(arrows).
(d) Sagittal,
gadofat-suppressed
Ti -weighted
1) shows
heterogeneously
MR image
(400/i
intense
enhancement
(arrows).
(e) Photomicrograph
of nonossifying
fibroma in the proximal tibia demonstrates
a massive aggregation
of benign-appearing
foam cells (arrows).
(Hematoxylin-eosin
stain; original magnification,
xiOO.)
d.
e.
and extraosseous
areas,
with a pathologic
fracture
depicted
in three
(16%)
(Fig 2).
In two cases (1 1%), abnormal
signal
inten-
DISCUSSION
sity
brous
cortical
fibrous
lesions
nonossifying
tinct
in most
was depicted
in only bone
marrow.
After infusion
of gadopentetate
dimeglumine,
intense
contrast
enhancement
was seen in 15 cases (79%),
in a heterogeneous
(ii
=
12) (Figs 2, 3) or homogeneous
(ti = 3) pattern.
In the remaining
four
cases
(21%),
marginal
and
septab
enhancement
(Fig 1) were
depicted.
Penipheral
margins
of all lesions
showed
enhancement,
consistent
with thin hypervascular
zones
at the periphery.
Volume2O9
Number
Nonossifying
intracortical
and subpeniosteal
lesions
(6).
Nonossifying
fibroma
is an evolutionary
form
of fibrous
cortical
defect
that
extends
from the cortex
into the medullary
substance
of the
bone
(7). The
terms
fibroma
defect
are
of bones.
fibroma
is
cases.
In
and
the
related
fi-
common
benign
In the long bones,
radiologically
disa child
or adoles-
cent,
an eccentric,
well-defined,
radiolucent lesion
that arises
in the metaphysis
of a tubular
bone,
a short
distance
from
the physis,
is almost
diagnostic
of nonossifying
fibroma
or fibrous
cortical
defect
(1-3).
Nonossifying
fibroma
should
be
included
in the differential
diagnosis
of
nonossifying
fibroma
and
thoma
are sometimes
used
ably (4). Most
nonossifying
cur in the
especially
bones
about
nonossifying
pain
and
may
fracture
(1 1).
tion,
interlacing
make
up most
larly distributed
and hemosiderin
Nonossifylng
fibroxan-
interchangefibromas
oc-
of the lower
extremity,
the knee
(8-10).
Large
fibromas
frequently
cause
predispose
to pathologic
At microscopic
examinawhorls
of fibrous
tissue
of the lesion,
with
irregumultinucleated
giant cells
of variable
degree.
Non-
Fibroma
at MR Imaging
201
#{149}
ossifying
fibroma
is usually
cellular,
with
(4,12).
fibroma
is
is in doubt
involve
the entire
often
appear
atypical,
width
and
of the
must
be
sampled
at biopsy
to rube out malignant
disease
(13). Lesions
more
than
33 mm
long that involve
more
than
50% of the
transverse
diameter
of the bone should
be
observed
carefully.
Prophylactic
curettage
and
autobogous
bone
grafting
of barge
lesions
may
be considered
if there
is a
reasonable
chance
of impending
fracture
(14).
Fibrous
tissue was known
to be hypointense
on both
Ti- and T2-weighted
MR
images
(1). Fibrous
dysplasia,
one of the
benign
fibrous
bone
lesions,
recently
was
reported
to be hypenintense
60% of cases on T2-weighted
(15,16).
scnibed
weighted
intensity
the
in more than
MR images
Nonossifying
fibroma
was
dewith
low signal
intensity
on TiMR images
and
mixed
signal
on T2-weighted
images
(5). On
basis
of our
experience
with
fibrous
dyspbasia,
nonossifying
pected
to be depicted
fibroma
was
as hypenintense
rather than
MR images.
on T2-weighted
however,
79% of
hypointense
In our study,
ex-
lesions
were hypointense
on Ti- and T2weighted
images
and 21% were hypointense
on Ti-weighted
MR images
and
hypenintense
on T2-weighted
images.
These
results
are consistent
with those
of
Kransdorf
et al (4), who reported
hypointense
nonossifying
fibroma
on
T2weighted
images
in eight
of 10 (80%)
patients.
In the study
by Kransdorf
et al (4), only
nonenhanced
MR images
were reviewed
of 10 nonossifying
fibromas
in nine
patients,
and
pathologic
results
were
obtamed
by means
of biopsy
in only
four
cases.
In our
study,
however,
nonen-
hanced
and
images
Radiology
#{149}
reviewed
MR
of 19 biopsy-proved
nonossifying
fibromas.
Among
the four
lesions
with findings
confirmed
at biopsy
in their
study,
two had unusually
barge
amounts
of hemosidenin
pigment
present
and
one
had
an increased
amount
of
collagen.
In our study,
all 15 lesions
depicted
with diffuse
hypointensity
on T2weighted
MR images
contained
extensive
hypercellular
fibrous
tissue.
Three
of our
four lesions
depicted
with very low signal
intensity
on T2-weighted
images
contamed
hemosidenin
pigment.
In a study
of another
hemosidenin-baden
bone
tumor
such
as giant
cell tumor,
hemosiderin was depicted
in 63#{176}A
at MR imaging
202
were
gadolinium-enhanced
October1998
#{149}
(1 7). In our
fibrous tissue
foamy
and
study,
scanty
with massive
histiocytes
was
was depicted
noted
in four
as hypenintense
weighted
MR
and trabeculation
were correlated
tense
rim and
tively,
intense
hyperceblubar
aggregation
of
enin,
cytes,
images.
Marginal
sclerosis
on plain
nadiographs
with a peripheral
hypoininternal
septation,
nespec-
with
peripheral
reactive
1.
2.
tiated
been
with
a thick
fibroma
sclerotic
rim,
be diffenen-
should
from
fibrous
dysplasia,
associated
with a peripheral
tense
rim on Tiimages
in 77% of
rounding
layer of
(16). On the basis
nal septation
MR images
of our
dysplasia
19
Abnormal
row and
which
has
hypoin-
and
T2-weighted
MR
cases owing
to a sunsclerotic
reactive
bone
of our findings,
inter-
is more frequently
seen on
of nonossifying
fibroma
(i8
cases
(three
[95%])
than
of fibrous
of 13 cases
[23%])
(16).
signal intensity
extnaosseous
areas
and
intraosseous
ganglion
hypenintense.
On gadolinium-enhanced
tense
contrast
enhancement
79%
of our cases
and
images,
inwas seen in
marginal
septab
enhancement
in 21%. These
contradictory
to those
et al (5) that
not depicted
sity
on
weighted
is also
findings
reported
nonossifying
with increased
gadolinium-enhanced
MR images.
We
were
3.
were
inten-
supposed
Tithis
contradiction
could
fat-suppressed
fat suppression
sequence
used in our study;
was not employed
in their
study.
In
conspicuous
our
be partly
due
6.
and
the patterns
in nonossifying
on the amounts
tissue,
hemosid-
MirraJM.
Fibrohistiocytic
origin.
tumors
In:
Mirra
JM,
Pa:
of intraPiero
P,
I, Hajek
PC, Pechmann
U. Fibrous
metaphyseal
defects: magnetic
resonance
imaging appearances.
Skeletal Radiol 1989;
18:253-259.
Jackson RP, Reckling FW. Intracortical
and
subperiosteal
lesion of unknown
etiology.
Clin Orthop
1978; 130:260-262.
Blau PA, Zwick
DL, Westphal
RA. Multiple
nonossifying
fibroma.
J Bone Joint Surg
[Am] 1988;
70:299-304.
8.
Kumar
R, Madewell
JE, Lindell
MM,
Swischuk
LE. Fibrous
lesions
of bones.
RadioGraphics
1990; 10:237-256.
Gross ML, Soberman
N, Dorfman
HD,
Seimon LP. Case report 556: multiple
nonossifying
fibromas
of long bones
in a
patient
with neurofibromatosis.
Skeletal
9.
10.
1 1.
12.
13.
14.
15.
S, Saltzstein
SL, Daniel
DM.
Non-
ossifying
fibroma:
report of an intact lesion. AmJ Clin Pathol 1974; 61:697-701.
Berkin CR. Nonossifying
fibroma
of bone.
BrJ Radiol 1966; 39:469-471.
Arata MA, Peterson HA, Dahlin DC. Pathological fractures through nonossifying
fibromas: review of the Mayo Clinic experience.J
BoneJoint
Surg [Ami 1981; 63:980-988.
UtzJA, Kransdorf
MJ, JelinekJS,
Moser RP
Jr, Berry BM. MR appearance
of fibrous
dysplasia.
J Comput
Assist Tomogr
1989;
13:845-85i.
16.
17.
19.
MR images
GD.
7.
that
logic
of contrast
enhancement
fibroma
were dependent
of hyperceblulan
fibrous
Greenway
Ritschl
18.
T2-weighted
M,
5.
to the
Kyriakos
4.
study,
the lesion
became
when
the signal
intensity
on
well with
the
Signal
intensity
D,
fat-suppressed
images was rescaled
(19).
In conclusion,
nonossifying
fibroma
has characteristic
findings
at MR imaging
correlate
features.
Resnick
medublary
by Ritschl
fibromas
signal
joint
disorders.
3rd ed. Philadelphia,
Saunders, 1995; 3628-3938.
in four
of our 19 cases
(21%)
and was
thought
to result
from fracture
in all but
one case. In contrast
to the hypointensity
of nonossifying
fibroma
on T2-weighted
images,
chondromyxoid
fibroma
is hyperintense,
aneunysmal
bone cyst is heterogeneously
hypenintense
with
a fluid-fluid
level,
histio-
Tumors
and tumor-like
lesions of bone:
imaging
and pathology
of specific lesions.
In: Resnick D, ed. Diagnosis
of bone and
foamy
References
sclero-
sis, which
may be absent
in larger lesions
(18).
In cases of nonossifying
fibroma
with
atypical
radiographic
features,
fibrous
dysplasia, chondromyxoid
fibroma,
aneunys-
nosis. In lesions
collagen,
trabeculae.
cases
on T2-
on MR images.
A peripheral
hyporim was identified
in all cases and
correlated
hemorrhage,
and bone
Jee et al