[

RESIDENTS CASE PROBLEM

AKIO SAKAMOTO, MD, PhDA7PK>?HEJ7D7A7"MD, PhD

J7JIKO7OEI>?:7"MDOKA?>?:;?M7CEJE" MD, PhD4

Journal of Orthopaedic & Sports Physical Therapy

Downloaded from www.jospt.org at on January 5, 2016. For personal use only. No other uses without permission.
Copyright 2008 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

Nonossifying Fibroma Accompanied

by Pathological Fracture in a
12-Year-Old Runner

onossifying broma (NOF),

or broxanthoma, is the
most common type of
benign brous lesion that
occurs in the metaphysis of the long
bones of the lower extremities.4,8
On plain radiographs, NOFs are
identied by a thin osseous border
that is usually scalloped and
slightly expansile. Histologically,
NOF consists of spindle-shaped
broblasts, multinucleated giant
cells, and foamy histiocytes.4,8
Fibrous cortical defect is thought
to be the same lesion based upon
histology; however, some people
differentiate brous cortical defect from
NOF when the lesion is less than 3 cm in
size.8 NOF can be found in approximately
30% of young individuals within the rst
and second decade.4 The lesion is asymptomatic and is typically identied incidentally. Actually, this benign lesion is
not a true neoplasm but is considered to
be a developmental bony defect.4,8 The lesion is usually self-limiting, with spontaneous resolution, and most of the lesions

disappear by the age of 20 to 25 years.1

A pathological fracture occurs when
the bone has been weakened due to a
lesion and therefore lacks its normal
biomechanical properties.14 It has been
T STUDY DESIGN: Residents case problem.
T BACKGROUND: Nonossifying broma (NOF) is
the most common brous bone lesion in children.
The lesion is usually asymptomatic, and rarely
leads to pathological fractures.

T DIAGNOSIS: We present the case of a 12-yearold boy who appeared to be normally developed
but had a pathological insufficiency fracture
associated with NOF in the distal femur. He was
a member of a track athletics club and ran more
than 5 km every day. Seven weeks prior to the
initial evaluation he felt discomfort in the left distal
thigh when running and felt pain upon knee exion.
The amount of discomfort increased gradually
and he began to experience pain while running 4
weeks prior to his initial evaluation. At the time of
the initial evaluation, he had tenderness over the
distal thigh region and there was increased pain
with weight bearing. Plain radiographs showed an
irregular, well-dened cortical bone lesion, suggesting NOF, with vague increased density in the
bone marrow across the femur and periosteal new
bone, suggesting a fracture. Computed tomography conrmed a linear fracture with increased
density across the femur leading to the cortical
lesion. In the process of differential diagnosis

reported that the distal tibia is the most

frequently involved site for a pathological fracture that passes through a NOF.
Lesions that occupy more than 50% of
the transverse diameter of the involved
osteosarcoma, or Ewing sarcoma, and bone/joint
infection were ruled out using magnetic resonance
imaging. The nal diagnosis based upon the
images and clinical course was pathological
insufficiency fracture associated with NOF. The
patient was treated with initial avoidance of weight
bearing using 2 crutches for ambulation, followed
by progressive weight bearing over a period of 5
weeks. Active range of motion of the knee joint was
allowed. Three months after onset (5 weeks after
the initial evaluation), the patient had normal gait
without pain, whereupon the patient resumed his
sport activities, beginning with jogging.

T DISCUSSION: Although pathological fractures

secondary to NOF in the femur are rare, NOF
can cause pathological insufficiency fractures
in athletes, even if the lesion is conned and
small. The current case is a reminder of such a
possibility. This case also provides a time course
as a reference for the rehabilitation of patients in
similar cases.
T LEVEL OF EVIDENCE: Diagnosis, level 4.
J Orthop Sports Phys Ther 2008;38(7):434-438.
doi:10.2519/jospt.2008.2655

T KEY WORDS: athletes, bone lesion, femur,

broxanthoma

1
Research associate, Department of Orthopaedic Surgery, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan. 2 Research associate, Department of
Orthopaedic Surgery, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan. 3 Research associate, Department of Orthopaedic Surgery, Graduate School of
Medical Sciences, Kyushu University, Fukuoka, Japan. 4 Professor, Department of Orthopaedic Surgery, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan.
The patient in this study and his family were informed that data from the case would be submitted for publication, and consent was provided. Address correspondence to Dr
Akio Sakamoto, Department of Orthopaedic Surgery, Graduate School of Medical Sciences, Kyushu University, Maidashi 3-1-1, Higashi-ku, 812-8582, Fukuoka, Japan. E-mail:
akio@ med.kyushu-u.ac.jp

434 | july 2008 | volume 38 | number 7 | journal of orthopaedic & sports physical therapy

Journal of Orthopaedic & Sports Physical Therapy

Downloaded from www.jospt.org at on January 5, 2016. For personal use only. No other uses without permission.
Copyright 2008 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

bone are likely to lead to a fracture.4,6

However, pathological fracture of the
distal femur of children is rare, regardless of the existence of NOF.5,17 We report
a case of pathological fracture associated
with NOF that was conned to the cortex
in the distal femur of a 12-year-old runner. Furthermore, the fracture was considered to be an insufficiency fracture,
which is dened as a fracture resulting
from normal repetitive stress placed on
abnormal bone.18 The purpose of the current report is a reminder that, while rare,
such cases can occur. Awareness of the
details of the current case would be benecial for the rehabilitation of patients
with a similar condition. We emphasize
the differential diagnosis process used
for the current case.

Vascular/hematologic etiology
Sickle cell disease
Hemophilic arthritis
Infectious etiology
Osteomyelitis
Joint infection
Neoplastic etiology
Malignant tumor
Ewing sarcoma
Osteosarcoma
Leukemia
Benign tumor
Osteoid osteoma
Autoimmune/rheumatologic etiology
Simple arthritis

DIAGNOSIS

TABLE

Causes of Pain
Around the
Knee in Young
Individuals

Rheumatoid arthritis
Seronegative rheumatoid arthritis

he patient was a member of a

track athletics club and was running
more than 5 km a day since entering a junior high school 6 months before
the symptoms began. Seven weeks prior
to his initial evaluation in our institute,
he felt discomfort in his distal left thigh
region when running. The discomfort
was perceived as pain when he exed his
knee. The degree of this discomfort/pain
increased gradually over time. Four weeks
prior to his initial evaluation he began to
experience pain while running, although
he felt no pain at rest. Because the pain
was getting worse, he was evaluated at
an orthopedic hospital. An abnormality
of the distal femur was noted on plain
radiograph, and the patient was referred
to our institute. When the patient visited
our institute 4 days later, he was using 2
crutches and was non-weight bearing on
the affected side, and had been doing so
since his visit to the orthopedic hospital.
At the initial evaluation, he appeared
to be well nourished and normally developed, with height and body mass of 152
cm and 44 kg, respectively, and there
was little suspicion of nutritional deciency or calcium/vitamin D deciency.
In addition, there was no past medical,

Congenital/hereditary etiology
Gauchers disease
Mechanical etiology
Repetitive overuse/stress fracture
Endocrine/metabolic etiology
Hyperparathyroidism
Malnutrition (ie, calcium/vitamin D deciency)
Other bone/joint problems
Steroid
Alcohol

surgical, or family history of factors that

might contribute to bone fragility or
pathological fracture. He had tenderness
over the distal thigh region and pain that
worsened with weight bearing. No abnormalities such as alteration in skin temperature or color, ecchymosis, or edema
were noted. Knee range of motion was
nearly normal, with a small limitation
of exion (approximately 5) due to pain
felt when nearing full knee exion. Knee
joint edema was not observed, suggesting
little possibility of arthritis of rheumatoid
or infectious origin. Thigh circumference
was symmetrical: right/left, 36.0/36.5
cm at a point 10 cm above the superior
pole of the patella. Blood tests were not
performed because of the absence of any

signs or symptoms suggestive of inammation, such as swelling, redness, or

warmth over the involved region, the absence of signs and symptoms suggestive
of systemic disease such as fever, chills,
sweats, or nausea/vomiting, and the absence of a recent history of the presence
of an open wound or dental work. Leukemia or endocrine disorder, such as hyperparathyroidism, can present as bone/joint
pain; however, the absence of systemic
symptoms, including weight loss, malaise, history of easy bruising/bleeding,
or fever, made it unlikely for these conditions to be present. It is important that
the presence of a malignant neoplasm
such as osteosarcoma, Ewing sarcoma,
or leukemia, as well as bone/joint infection is considered in young patients who
present with knee pain and no history of
specic trauma. A list of possible tumors
and diseases for differential diagnosis is
presented in the TABLE. In this case, to assess bone abnormality, plain radiographs
were taken in our institute approximately
7 weeks after the onset of discomfort with
running.
The plain radiographs showed a
well-dened radiolucent lesion with an
increased density margin in the medial
posterior cortex, and the cortical lesion
was compatible with NOF. Plain radiographs also showed a vague linear lesion
with increased density across the distal
femur, obliquely from superior-lateral to
inferior-medial, and the medullary lesion
suggested bone callus formation such as
that seen in secondary bone healing. Thin
and multilayered periosteal new bone
along the anterior and posterior cortex
was seen in the distal femur (FIGURE 1). Periosteal new bone, or periosteal reaction,
is actually periosteal lifting with new
bone formation or reactive bone formation secondary to cortical disruption due
to fracture or malignant bone neoplasm.
Although periosteal reaction is less common in NOF, based on these ndings,
both NOF (well-dened cortical radiolucent lesion) and pathological fracture
(the vague linear medullary lesion with
periosteal reaction) were suspected. In

journal of orthopaedic & sports physical therapy | volume 38 | number 7 | july 2008 | 435

RESIDENT S CASE PROBLEM

Journal of Orthopaedic & Sports Physical Therapy

Downloaded from www.jospt.org at on January 5, 2016. For personal use only. No other uses without permission.
Copyright 2008 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

FIGURE 2. Computed tomography (CT) shows a

decreased density lesion in the cortex with periosteal
new bone (A). Coronal reconstruction CT shows a
linear increased density lesion of fracture, which
passes through the cortical lesion (B).
FIGURE 1. Antero-posterior plain radiographs show a
well-dened radiolucent lesion in the medial posterior
cortex (arrows). A vague increased density lesion in
the bone marrow can also be seen (A). Lateral plain
radiographs show periosteal new bone at the anterior
and posterior cortex (B).

the differential diagnosis process, a radiolucent cortical lesion with surrounding

increased bone density could be considered to be osteoid osteoma. However, periosteal reaction is also less common in
cases of osteoid osteoma, and night pain
and rest pain, which are typical in cases
of osteoid osteoma, were not present in
the current case.
Computed tomography (CT) was or-

dered to further examine the extent and

morphology of bony abnormality. CT
depicts bone structure in detail, and is
useful for determining the presence of
cortical integrity or fracture. In this case,
CT imaging showed the radiolucent cortical lesion seen on plain radiographs as
a well-dened area of decreased density
in the cortex, consistent with a diagnosis
of NOF, with periosteal new bone (FIGURE
2). The medullary lesion with vague increased density observed on plain radiographs across the distal femur was seen
as a linear lesion of increased density in
the distal femur on CT and was compat-

FIGURE 3. Magnetic resonance imaging (MRI) shows a cortical lesion with low signal intensity on T1-weighted
(A, E) and T2-weighted (B, F) images (arrows). Slight enhancement by Gadolinium is seen on T1-weighted fat
suppression (C) and T1-weighted images (G). A linear fracture with low signal intensity on both T1- and T2-weighted
and short-tau inversion recovery (STIR) images can be seen (A, B, C). MRI shows diffuse low signal intensity of
T1-weighted images (A, B) and high signal intensity of STIR (D) images of the distal femoral metaphysis, suggesting
bone marrow edema (arrows) (A-D, coronal; E-G, axial).

ible with a healing fracture with mineralization and consistent with the fact
that 7 weeks had passed since onset. The
medullary linear lesion of the fracture
reached the NOF cortical lesion, thereby
suggesting a pathological fracture associated with NOF.
In the current case, fracture in the
later stages with increased bone density
on a plain radiograph and CT in the bone
marrow needs to be differentiated from
osteoblastic primary or metastatic bone
tumor. Magnetic resonance imaging
(MRI) is useful for assessing bone marrow and soft-tissue abnormality, such as
infection and overall neoplasm, regardless of the abnormality of the bone structure. Accordingly, MRI was performed
to further evaluate the lesion in this case
(FIGURE 3). MRI demonstrated the cortical radiolucent lesion seen on plain radiographs and CT as a lesion with low
signal intensity on T1- and T2-weighted
images in the medial-posterior cortex
of the distal femur. The cortical lesion
was slightly heterogeneously enhanced
by Gadolinium on T1-weighted imaging
(FIGURE 3). The low signal intensity on
both T1- and T2-weighted images within
the lesion could reect hemosiderin and
brous tissue elements, being compatible to features of NOF. MRI revealed the
medullary linear lesion with increased
density seen on plain radiographs and
CT as low signal intensity on both T1and T2-weighted images. The low signal
intensity medullary band noted on MRI
could correspond to calcium in bone
healing, suggesting that the medullary
linear lesion is a fracture line. High signal intensity on both T2-weighted and
short-tau inversion recovery (STIR) images could be representative of a uid
nature. Therefore a diffuse area with low
signal intensity on T1-weighted images
and high signal intensity on STIR images, suggesting bone marrow edema,
was seen surrounding the medullary
linear lesion on CT and MRI (FIGURE 3).
The degree and size of the bone marrow
edema were small. This nding may be
due to the fact that the fracture was not

436 | july 2008 | volume 38 | number 7 | journal of orthopaedic & sports physical therapy

recent and was consistent to a healing

fracture, with mineralization on plain
radiographs and CT. There were no features of aggressive neoplasm or infection, such as bone marrow abnormality,
or extraosseous extension or adjacent
soft tissue abnormality. The MRI ndings were consistent with CT ndings of
NOF with fracture.

Journal of Orthopaedic & Sports Physical Therapy

Downloaded from www.jospt.org at on January 5, 2016. For personal use only. No other uses without permission.
Copyright 2008 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

Outcomes
The patient was treated by avoidance of
weight bearing, while active motion of
the knee joint was allowed. The pain resolved gradually, and partial weight bearing was allowed as long as he experienced
no pain. In summary, the initial evaluation was 7 weeks after onset of discomfort and the patient had been non-weight
bearing using 2 crutches for the previous
4 days. Nonweight-bearing ambulation
using 2 crutches was continued for 1 more
week followed by gradual application of
weight bearing initially using 2 crutches,
followed by 1 crutch. The pain resolved
approximately 5 weeks after the initial
evaluation (3 months after its onset), and
the patient had normal gait without pain,
whereupon the patient resumed his sport
activities, beginning with jogging. During rehabilitation, instruction on weight
bearing was provided at each visit to our
institute. The clinical course was consistent with the diagnosis of pathological
fracture associated with NOF.

DISCUSSION

diagnosis of NOF can be made

based on the patients history,
physical examination, and ndings
on plain radiographs.1,9 Given the diagnostic radiographic appearance, these lesions have been classied as dont touch
lesions, namely biopsy, including open,
core, and ne needle, is considered to be
both unnecessary and excessively invasive.8 Furthermore when a biopsy is carried out, a biopsy specimen sometimes
can be misdiagnosed as giant cell tumor
of bone.4
MRI provides a more detailed as-

sessment of the local osseous and soft

tissue response to the lesion.1 Typically,
NOF shows low signal intensity on both
T1- and T2-weighted images within the
lesion, as seen in the current case, with
correlation of hemosiderin and brous
tissue elements. In a series of 19 histologically conrmed NOF cases, they were
all reported to have low signal intensity
on T1-weighted images, but the signal intensity of T2-weighted images varied in
that some NOF cases had higher signal
intensity than others.10 The overall appearance of the NOF depends upon the
relative amounts of collagen, foamy histiocytes, hemorrhage, hemosiderin, and
bone trabeculae.8,10 NOF can be divided
into 3 phasesactive, healing, and inactivebased on the intensity of the uptake
on bone scan.2 A healing lesion on NOF
shows decreased signal intensity on both
T1- and T2-weighted images.1
A pathological fracture, by denition,
occurs through abnormal bone because
the bone is weak and lacks its normal biomechanical properties.14 In the current
case, the fracture occurred through the
NOF. The fracture was thought to have
occurred due to weakness of the bone
at the site of the NOF, even though the
NOF was conned to the cortex and its
size was small. The fracture in the current case was classied as pathological,
even though most of the area of the fracture was outside the NOF. Pathological
fracture through the NOF is rare in the
femur. In the English literature, a small
number of cases similar to the current
patient have been reported.12,13,21 In the
current case, load associated with running seemed to have contributed to the
pathological fracture.
The pathological fracture in the current case was also classied as insufficiency fracture. Stress fractures are dened
when a local load on bone exceeds the
mechanical resistance and are classied
into 2 types: namely, fatigue fractures and
insufficiency fractures. Fatigue fractures
occur when abnormal mechanical stress
is loaded on a normal bone, while insufficiency fractures occur when moderate or

normal pressure is loaded on a bone with

a decreased resistance.18 The typical location of a stress or fatigue fracture on normal bone is the diaphyseal region, mainly
the mid-third of the shaft or the transition
of the mid-proximal or mid-distal thirds
of the femur or tibia.22 Whereas, the typical location of NOF is the metaphyseal
region, which is somewhat different from
the location of fatigue fracture. Moreover, fatigue fracture is rare in children.13
Generally, in a diagnosis of stress fracture, including fatigue and insufficiency
fractures, serial radiographs are necessary during a period of several weeks not
only to detect healing of the fracture, but
also to rule out neoplasm. However, it is
sometimes difficult for parents to accept
a wait-and-see approach for treatment
of their child.3 Therefore, as was done in
the current case, examinations with CT
and MRI would seem to be a reasonable
approach, after suspicion of a pathological insufficiency fracture associated with
NOF arises from plain radiographs.
An early stress fracture, including fatigue and insufficiency fractures, shows
normal or decreased density on plain
radiographs, in which there is noncalcied hematoma. Therefore, plain radiography is insensitive to the early phase of
a stress fracture, and the lesion can easily
be overlooked11,16; however, plain radiograph will show evidence of stress fracture healing at 10 to 14 days, even though
the stress fracture itself is not visible.13
The age of the stress fracture should
correlate with observed ndings, and a
healing fracture will show bone callus of
varying density depending on its maturity on plain radiographs. Stress fracture
needs to be differentiated from neoplasm
because the radiographic appearances
of these 2 entities sometimes mimic
each other. The radiographic appearance of neoplasm is described as either
osteoblastic or osteolytic. An osteoblastic lesion is characterized by increased
density associated with bone formation,
whereas, osteolytic lesion refers to an
active resorption or dissolution of bone
associated with a disease process. In the

journal of orthopaedic & sports physical therapy | volume 38 | number 7 | july 2008 | 437

Journal of Orthopaedic & Sports Physical Therapy

Downloaded from www.jospt.org at on January 5, 2016. For personal use only. No other uses without permission.
Copyright 2008 Journal of Orthopaedic & Sports Physical Therapy. All rights reserved.

RESIDENT S CASE PROBLEM

later stages of stress fracture with increased bone density, a plain radiograph
may give rise to a diagnosis of osteoblastic primary or metastatic bone tumor.7
While osteosarcoma is an example of
primary malignant osteoblastic tumors
and osteoid osteoma and osteoblastoma
are its benign counterparts,4 osteoblastic
metastatic tumors are often of prostate
or breast origin. Open biopsy should be
avoided in the case of stress fractures.
However, when it is carried out, it should
be kept in mind that a biopsy taken of a
healing fracture can be misleading, suggesting osteosarcoma or chondrosarcoma, because the histological features of
these 2 conditions look similar.19
MRI is a useful imaging technique in
the diagnosis of stress fractures in that
it can differentiate them from neoplastic or infectious processes.12,16,20 In consideration of the patients age (12 years
old), malignant bone tumor, osteosarcoma, and Ewing sarcoma in particular,
needed to be ruled out. In the current
case, the stress or insufficiency fracture
was identied as a medullary linear lesion with low signal intensity on both
T1- and T2-weighted images.15 The low
signal intensity medullary band (a fracture line) noted on MRI demonstrates a
relatively old fracture because calcium
is being laid down. In the case of a recent fracture, MRI only shows diffuse
bone marrow edema with low signal intensity on T1-weighted images and high
signal intensity on T2-weighted images,
with a less prominent fracture line with
low signal intensity on both T1- and
T2-weighted images. In the current
case, STIR image, which is much more
uid sensitive, shows bone marrow
edema both proximal and distal to the
low-signal band. The degree and size
of the bone marrow edema were small,
which may be because the fracture was

not recent, consisting of a fracture line

characterized by calcium deposition on
plain radiographs and CT. These MRI
features were useful not only for making
a diagnosis, but also for ruling out other
pathologies.
In this report, we present a case of
pathological fracture associated with
NOF. CT and MRI were useful for detecting the pathological insufficiency fracture. Pathological fracture in the femur
is rare in children, regardless of the existence of NOF. Nevertheless, the possibility of pathological insufficiency fracture
should be kept in mind when the patient
is an athlete. Moreover awareness of the
time course involved in the current case
would be benecial for the rehabilitation
of patients in similar cases. T

REFERENCES
1. Betsy M, Kupersmith LM, Springeld DS.
Metaphyseal brous defects. J Am Acad Orthop
Surg. 2004;12:89-95.
2. Brenner RJ, Hattner RS, Lillien DL. Scintigraphic
features of nonosteogenic broma. Radiology.
1979;131:727-730.
3. Buckley SL, Robertson WW, Jr., Shalaby-Rana
E. Stress fractures of the femoral diaphysis in
young children. A report of 2 cases. Clin Orthop
Relat Res. 1995;165-169.
4. Dorfman HD, Czerniak B. Bone Tumors. St Louis,
MO: Mosby; 1998.
5. Drubach LA, Connolly SA, Treves ST, Connolly LP.
Stress-induced fracture involving a nonossifying
broma. Clin Nucl Med. 2004;29:41-42. http://
dx.doi.org/10.1097/01.rlu.0000102987.43780.32
6. Easley ME, Kneisl JS. Pathologic fractures
through nonossifying bromas: is prophylactic treatment warranted? J Pediatr Orthop.
1997;17:808-813.
7. Fu AL, Greven KM, Maruyama Y. Radiation
osteitis and insufficiency fractures after pelvic
irradiation for gynecologic malignancies. Am J
Clin Oncol. 1994;17:248-254.
8. Hetts SW, Hilchey SD, Wilson R, Franc B.
Case 110: Nonossifying broma. Radiology.
2007;243:288-292. http://dx.doi.org/10.1148/
radiol.2431040427
9. Hod N, Levi Y, Fire G, et al. Scintigraphic char-

438 | july 2008 | volume 38 | number 7 | journal of orthopaedic & sports physical therapy

10.

11.

12.

13.
14.

15.

16.

17.

18.

19.

20.

21.

22.

acteristics of non-ossifying broma in military

recruits undergoing bone scintigraphy for
suspected stress fractures and lower limb pains.
Nucl Med Commun. 2007;28:25-33. http://
dx.doi.org/10.1097/MNM.0b013e328012e3de
Jee WH, Choe BY, Kang HS, et al. Nonossifying broma: characteristics at MR imaging with pathologic correlation. Radiology.
1998;209:197-202.
Lassus J, Tulikoura I, Konttinen YT, Salo J,
Santavirta S. Bone stress injuries of the
lower extremity: a review. Acta Orthop
Scand. 2002;73:359-368. http://dx.doi.
org/10.1080/000164702320155392
Lee SH, Baek JR, Han SB, Park SW. Stress
fractures of the femoral diaphysis in children:
a report of 5 cases and review of literature. J
Pediatr Orthop. 2005;25:734-738.
Meaney JE, Carty H. Femoral stress fractures in
children. Skeletal Radiol. 1992;21:173-176.
Ortiz EJ, Isler MH, Navia JE, Canosa R. Pathologic fractures in children. Clin Orthop Relat Res.
2005;116-126.
Otte MT, Helms CA, Fritz RC. MR imaging of
supra-acetabular insufficiency fractures. Skeletal Radiol. 1997;26:279-283.
Peris P. Stress fractures. Best Pract Res Clin
Rheumatol. 2003;17:1043-1061. http://dx.doi.
org/10.1016/j.berh.2003.09.006
Peterson HA, Fitzgerald EM. Fractures through
nonossifying bromata in children. Minn Med.
1980;63:139-144.
Soubrier M, Dubost JJ, Boisgard S, et al.
Insufficiency fracture. A survey of 60 cases
and review of the literature. Joint Bone Spine.
2003;70:209-218.
Theodorou SJ, Theodorou DJ, Schweitzer ME,
Kakitsubata Y, Resnick D. Magnetic resonance
imaging of para-acetabular insufficiency fractures in patients with malignancy. Clin Radiol.
2006;61:181-190. http://dx.doi.org/10.1016/j.
crad.2005.10.003
Tyrrell PN, Davies AM. Magnetic resonance
imaging appearances of fatigue fractures of
the long bones of the lower limb. Br J Radiol.
1994;67:332-338.
Walker RN, Green NE, Spindler KP. Stress fractures in skeletally immature patients. J Pediatr
Orthop. 1996;16:578-584.
Zwas ST, Elkanovitch R, Frank G. Interpretation
and classication of bone scintigraphic ndings
in stress fractures. J Nucl Med. 1987;28:452-457.

MORE INFORMATION
WWW.JOSPT.ORG