Imaging of Chronic Recurrent Multifocal Osteomyelitis

Note: This copy is for your personal non-commercial use only. To order presentation-ready
copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.
EDUCATION EXHIBIT
1159
Imaging of Chronic
Recurrent Multifocal
Osteomyelitis1
Online-Only
CME
See www.rsna
.org/education
/rg_cme.html
LEARNING
OBJECTIVES
After reading this
article and taking
the test, the reader
will be able to:
■■Discuss
the epidemiologic and clinical
features of chronic
recurrent multifocal
osteomyelitis.
■■List
the common
sites of disease in
chronic recurrent
multifocal osteomyelitis.
■■Describe
the typical imaging findings
of chronic recurrent
multifocal osteomyelitis.
Geetika Khanna, MD, MS • Takashi S. P. Sato, MD • Polly Ferguson, MD
Chronic recurrent multifocal osteomyelitis (CRMO) is an autoinflammatory disorder of children and young adults that is characterized by
nonbacterial osteomyelitis. Patients typically present with multifocal
bone pain secondary to sterile osseous inflammation, and the disease
has a relapsing and remitting course. The cause of CRMO remains
unclear, although the results of several studies have suggested a genetic component. The typical imaging findings of CRMO include lytic
and sclerotic lesions in the metaphyses of long bones and the medial
clavicles. Other common sites of disease are the vertebral bodies, pelvis, ribs, and mandible. CRMO is often bilateral and multifocal at
presentation. Owing to the lack of a diagnostic test, CRMO remains
a diagnosis of exclusion. Although generally a self-limiting disease,
CRMO can have a prolonged course and result in significant morbidity. Radiologists can be the first to suggest this diagnosis given its
characteristic radiographic appearance and distribution of disease.
Radiologists should be familiar with the typical imaging findings of
CRMO to prevent unnecessary multiple biopsies and long-term antibiotic treatment in children with CRMO.
Introduction
TEACHING
POINTS
See last page
Chronic recurrent multifocal osteomyelitis (CRMO) is an idiopathic inflammatory
disorder of bone seen primarily in children and adolescents. The syndrome of CRMO
was originally described in 1972 in four children, under the name “subacute and
chronic recurrent osteomyelitis” (1). It is characterized by multifocal nonpyogenic
inflammatory bone lesions, a course of exacerbations and remissions, and an association with other inflammatory disorders (2). Since its original description, there
have been multiple case reports and studies of this disorder under a number of different names. These include subacute and chronic symmetric osteomyelitis, subacute
Abbreviations: CRMO = chronic recurrent multifocal osteomyelitis; SAPHO = synovitis, acne, pustulosis, hyperostosis, osteitis
RadioGraphics 2009; 29:1159–1177 • Published online 10.1148/rg.294085244 • Content Codes:
From the Mallinckrodt Institute of Radiology, Washington University School of Medicine, Campus Box 8131, St Louis, MO 63110 (G.K.); and the
Department of Pediatrics (P.F.), Carver College of Medicine (T.S.P.S.), University of Iowa, Iowa City, Iowa. Recipient of a Certificate of Merit award for
an education exhibit at the 2008 RSNA Annual Meeting. Received December 15, 2008; revision requested January 20, 2009, and received February 6;
accepted February 11. All authors have no financial relationships to disclose. Address correspondence to G.K. (e-mail: [email protected]).
1
©
RSNA, 2009 • radiographics.rsnajnls.org
1160 July-August 2009
symmetric osteomyelitis, chronic multifocal symmetric osteomyelitis, chronic multifocal cleidometaphyseal osteomyelitis of childhood, and
chronic recurrent multifocal osteomyelitis (1,3–6).
The term CRMO was first used by Probst
and colleagues (3,5) and encompasses the typical
clinical features of this entity: chronicity with a
protracted course, involvement of more than one
site, multiple exacerbations and relapses at old
and new sites, and disease at sites that are typical
of osteomyelitis in children. Although primarily a
disease of children, it has been reported in older
patients up to the age of 55 years (7,8). Since its
original description more than 3 decades ago,
several hundred cases of CRMO have been reported, likely owing to an increasing awareness
of this disease. Because CRMO is a diagnosis of
exclusion, it is likely that its actual prevalence is
higher than reported in the literature.
As imaging plays an important role in the
evaluation of this disorder, it is important that
radiologists be familiar with its various manifestations. Suggestion of the diagnosis by a radiologist could help avoid unnecessary diagnostic
procedures and antibiotic therapy and initiate an
appropriate therapy. In this article, we present
our clinical experience with CRMO and review
its epidemiologic, etiologic, clinical, and histopathologic features. We then describe the imaging approach, imaging appearances, and specific
disease sites: tubular bones, clavicle, spine, pelvis,
mandible, and hands and feet. Finally, we discuss
differential diagnosis and outcomes and compare
CRMO with SAPHO (synovitis, acne, pustulosis,
hyperostosis, osteitis) syndrome.
Clinical Experience
Between 2003 and 2008, 25 patients with CRMO
were referred to the department of pediatric rheumatology at our institution. There were 10 male
and 15 female patients with an average age at onset of symptoms of 8.3 years (range, 2.5–24 years).
The three most common sites of disease at initial
presentation were the lower extremities (39.7%),
spine (25.9%), and pelvis (20.7%). During the
course of follow-up, a total of 121 bone lesions
were seen in this patient population, with the
most common site of disease being the lower extremities (tibia, 15%; femur, 12%). Metaphyseal
lesions were the most common, accounting for
49% of all long bone lesions.
The diagnosis of CRMO in our patient population was confirmed by means of bone biopsy and
radiographics.rsnajnls.org
negative cultures in 23 of the 25 cases. Biopsy
was not performed in two cases owing to a typical
clinical picture and the presence of comorbid conditions (Crohn disease in one case and a pustular
rash in the other), which supported the diagnosis
of CRMO. The images from these cases have been
used to illustrate this pictorial review of CRMO.
Epidemiologic Features
CRMO is primarily a disease of childhood,
with most patients presenting between 9 and 14
years of age. However, it has been described in
infants as young as 6 months and in adults as
old as 55 years (7,9–11). Several studies have
demonstrated a female predominance, with a
male-to-female ratio of 1:2.1 (12–15). Although
most of the initial reports of CRMO came from
Europe, cases have been reported worldwide
(1,10,11,16–19). The true prevalence of this
condition remains unknown, with more than
250 cases of CRMO reported in the literature
(12,13). Given the fact that it is a diagnosis of
exclusion, it is likely that the actual prevalence of
CRMO is higher than realized.
Etiologic Factors
The cause of CRMO remains elusive. Initial
reports suggested an infectious origin, with suspected organisms including Staphylococcus aureus,
Mycoplasma hominis, Propionibacterium acnes, and
Chlamydia (7,9,20). However, larger studies have
shown no common infectious agent responsible
for CRMO (8,21). The coexistence of skin organisms (such as diphtheroids and coagulasenegative staphylococci) in the biopsy cultures and
the failure of CRMO symptoms to improve with
antibiotic therapy suggest that the organisms are
likely contaminants of biopsy specimens (19,22).
The association of CRMO with dermatologic
disorders (such as psoriasis) and inflammatory
bowel disease and its response to steroids have
led to the suggestion of an autoimmune cause
(23–25). More recently, a genetic origin for
CRMO has been suggested secondary to observation of disease in siblings and monozygotic
twins (10,11,26–28). By means of a family-based
association study, a susceptibility locus has been
identified at 18q21.3–22 (29).
The genetic origin of CRMO is further supported by the identification of mutations of the
LPIN2 gene in Majeed syndrome (10,11,30,31).
This rare autosomal-recessive syndrome is phenotypically characterized by a triad of CRMO,
congenital dyserythropoietic anemia, and an inflammatory dermatosis. Spontaneously occurring
RG ■ Volume 29 • Number 4
Khanna et al 1161
Figures 1, 2. (1) CRMO in a 9-year-old girl with a history of mandibular pain and swelling. Technetium 99m (99mTc) bone scan (anterior projection) shows increased uptake in the mandible. Bilaterally
symmetric radionuclide-avid lesions are present in the distal radius and ulna, distal femur, distal tibia,
and tarsal bones. Mild increased uptake is also seen in the proximal left tibial metaphysis and in the
maxilla. Bone biopsy and negative culture results led to a diagnosis of CRMO. (2) Chronic nonbacterial
osteomyelitis in a 14-year-old patient with a 3-year history of right arm pain. Radiograph (a) and sagittal reformatted computed tomographic (CT) image (b) of the distal humerus show hyperostosis secondary to chronic periosteal reaction. There is marked cortical thickening with secondary compromise
of the medullary space. Specimens from two biopsies showed chronic inflammatory changes without
isolation of any organisms. A diagnosis of chronic nonbacterial osteomyelitis was established.
mouse models of CRMO have been identified that
show an autosomal-recessive gene defect localized to the murine pstpip2 gene (27,32,33). Biopsy
samples from these mice show histologic features
similar to those seen in CRMO (27,32,33).
Clinical Features
Patients typically present with nonspecific complaints such as pain, tenderness, swelling, or limited range of motion at one or more sites. Systemic
manifestations such as fever, weight loss, and lethargy may be present but are unusual. The duration
of symptoms can vary from days to several years
(19). In the series of Mandell et al (34), children
were symptomatic for 3 days to 2 years before initial presentation. The duration of symptoms can
often be hard to assess because of the insidious
onset and vague nature of the symptoms. Laboratory findings at initial presentation are essentially
nonspecific, with the most common findings being
mildly elevated erythrocyte sedimentation rate and
C-reactive protein level with a normal white blood
cell count (26).
Although most patients present with a single
symptomatic site, other sites of disease become
apparent at imaging or during clinical followup (Fig 1). At any one time, the number of
osteomyelitis lesions can range from one to 18
(34,35). Although most patients with CRMO
have recurrent and multifocal disease, a significant number of children with nonbacterial osteomyelitis may have only a single lesion, which
exhibits the same clinical, radiologic, and histologic features as multifocal disease (Fig 2). The
term chronic nonbacterial osteomyelitis has been
suggested for this entity (21).
1162 July-August 2009
radiographics.rsnajnls.org
Pathologic Conditions Associated with CRMO
Dermatologic disorders
Palmoplantar pustulosis (16,37)
Psoriasis (23,37)
Acne fulminans (14)
Sweet syndrome (44)
Pyoderma gangrenosum (24,36,39)
Autoinflammatory disorders
Takayasu arteritis (24)
Wegener granulomatosis (41)
Gastrointestinal disorders
Crohn disease (25,38)
Ulcerative colitis (25)
Genetic syndromes
Majeed syndrome (10,11)
Other conditions
SAPHO syndrome (40)
Spondyloarthropathies (21,42,43)
Note.—Numbers in parentheses are references.
CRMO has been reported in association
with skin lesions including palmoplantar pustulosis, psoriasis vulgaris, Sweet syndrome, and
pyoderma gangrenosum (16,19,23,24,36,37)
(Table). In the review by Schultz et al (37), the
prevalence of cutaneous lesions in children with
CRMO was reported to be 25%, with pustulosis
being the most common, followed by psoriasis.
An association between CRMO and inflammatory bowel disease, especially Crohn disease, has
also been reported (Fig 3) (25,37,38).
Histopathologic Findings
Because the imaging findings of CRMO can
be nonspecific, especially early in the course of
the disease or in the presence of a solitary lesion, microbiologic and histopathologic analysis
are essential to exclude infectious osteomyelitis
and neoplastic lesions. At histologic evaluation,
CRMO bone lesions show nonspecific inflammatory changes with granulocytic infiltration (8,19).
In the acute stages, there is a predominance of
polymorphonuclear leukocytes and osteoclastic
bone resorption with or without multinucleated
giant cells. As the disease progresses, the predominant features are lymphocytes, plasma cells,
histiocytes with occasional granulomas, and increased osteoblast activity (8).
The histologic findings correlate relatively well
with the radiologic appearance; however, some
studies have shown poor correlation between
Figure 3. CRMO in an 8-year-old boy with a 2-year
history of multifocal bone pain in the lower extremities and pelvis. (a) Axial fat-saturated T2-weighted
magnetic resonance (MR) image shows scattered foci
of edema (black arrows) around the sacroiliac joints on
both sides. The edema is more extensive on the right
side than on the left. The terminal ileum is thickened
(white arrow). (b) Spot fluoroscopic image shows the
thickened terminal ileum, a finding consistent with the
patient’s history of Crohn disease.
the histologic findings and the duration of clinical disease (15,45). Biopsy specimens may show
acute, subacute, and chronic changes mixed in
the same lesion (45).
Imaging Approach
The imaging evaluation of CRMO should start
with radiographic evaluation of the symptomatic
sites. If the radiographs are negative in the presence of significant clinical symptoms, further
evaluation with MR imaging should be considered to evaluate for marrow edema. When a
diagnosis of CRMO is considered on the basis
of clinical findings and initial imaging evaluation, further evaluation of the whole body is
RG ■ Volume 29 • Number 4
Khanna et al 1163
Figures 4, 5. (4) Bilaterally symmetric
disease in a 23-year-old woman who presented with right knee pain at the age of 9
years; the diagnosis of CRMO was made.
99m
Tc bone scan obtained at the time of
relapse shows bilaterally symmetric uptake
in the distal femora and additional sites of
disease in the proximal right tibia, mid left
fibula, and distal left tibia. (5) CRMO in a
24-year-old woman with a 10-year history
of mandibular pain and swelling, for which
she had undergone multiple surgeries and
prolonged antifungal therapy at outside
hospitals. (a) Axial CT image shows lytic
areas in the right mandibular ramus (arrow) with sclerosis and bony expansion.
The associated cortical disruption was
due to prior surgery. There is associated
overlying soft-tissue swelling. (b) Postoperative chest radiograph obtained at our
institution shows expansion of the anterior
aspect of the left first rib (arrow), a finding
that aided in the diagnosis of CRMO.
suggested to identify multifocal lesions that may
be clinically asymptomatic (Fig 1). Whole-body
evaluation has traditionally been performed
with 99mTc bone scintigraphy, although wholebody MR imaging is being increasingly used for
evaluation of multifocal bone lesions (46–48).
Although MR imaging has the advantages of
avoiding radiation and demonstrating lesions
confined to the marrow, it is considerably more
expensive than bone scintigraphy.
Whole-body imaging can demonstrate clinically occult sites of disease, confirming the multifocal nature of disease. In the series of Yu et al
(49), imaging demonstrated 20 additional sites of
disease in seven children, who had clinically apparent disease at only 14 sites. In another series
of patients, scintigraphy demonstrated an average
of six lesions per patient, even though 12 of the
14 cases had a unifocal presentation (34). In addition, 64% of the cases in that series were found
to have bilateral disease at scintigraphy.
Although bilateral disease is morphologically
symmetric, it typically lacks clinical and temporal
symmetry. The identification of additional sites of
disease can aid in the diagnosis of CRMO, especially when disease is present at typical locations
like the anterior chest wall or when disease is
symmetrically distributed (Figs 4, 5). In addition,
the identification of all sites of disease aids in the
follow-up of disease activity.
Teaching
Point
Imaging Appearances
The findings at imaging studies can be suggestive of
a diagnosis of CRMO but are not pathognomonic.
Common sites of skeletal involvement include the
long tubular bones and clavicle, but lesions have
been described throughout the skeleton, including the spine, pelvis, sacroiliac joint, ribs, sternum,
scapula, mandible, and hands and feet. Involvement
of the lower extremity has been reported to be three
Teaching
Point
1164 July-August 2009
radiographics.rsnajnls.org
Figure 6. CRMO in a 12-year-old girl with a 7-month history of ankle pain. (a) Anteroposterior radiograph of the ankle shows multiple pyramid-shaped lytic lesions with surrounding
sclerosis in the metaphyses of the tibia and fibula (straight arrow); the lesions extend across
the growth plate into the epiphysis (wavy arrow). Multiple apophyseal lesions are also present
in the tarsal bones (arrowheads). Also note the generalized osteopenia. (b) Photomicrograph
(original magnification, ×100; hematoxylin-eosin stain) shows fragments of bone surrounded
by fibrous stroma, in which there are scattered inflammatory cells and macrophages. Results of
microbiologic analysis were negative and thus supportive of the diagnosis of CRMO.
Figure 7. CRMO in a 5-year-old patient with right knee pain. (a) Lateral radiograph of
the knee shows a permeative lytic pattern in the patella (arrow) with overlying soft-tissue
swelling. (b) Sagittal fat-saturated T2-weighted MR image shows erosive changes of the
ossification center of the patellar apophysis (arrow), whereas the surrounding cartilage appears intact. Other sites of nonbacterial inflammation developed in both femora and tibiae
over the next 3 years. The diagnosis of CRMO was confirmed with histopathologic analysis
and negative culture results from bone biopsy.
times more common than disease in the upper extremity (34,49). The tibia has been reported as the
most common bone involved—for both unilateral
and bilateral disease (50).
The most common sites of disease are the
Teaching metaphyses or metaphyseal equivalents, accountPoint
ing for approximately 75% of all lesions in the
series of Mandell et al (34) (Figs 6, 7). The dis-
tribution of CRMO, specifically its tendency to
involve the metaphyses and metaphyseal equivalents, is reminiscent of hematogenous osteomyelitis in children. However, CRMO is unique in its
tendency to involve the clavicle—an uncommon
site for hematogenous osteomyelitis.
In the early stages, plain radiographs typically demonstrate an osteolytic lesion located
RG ■ Volume 29 • Number 4
Khanna et al 1165
Figure 8. CRMO in a 7-year-old patient with a 4-month history of leg pain.
(a) Anteroposterior radiograph of the ankle shows a small lytic lesion adjacent
to the physis. (b) Follow-up radiograph obtained 3 months later shows the enlarging lytic lesion with surrounding sclerosis that fades away from the lesion.
(c) Radiograph obtained 3 months later shows other lytic areas adjacent to
the physis. (d) On a radiograph obtained 4 months later, the lesions appear to
coalesce and heal without significant periosteal reaction or sclerosis.
adjacent to the growth plate in the metaphysis.
With time, progressive sclerosis is seen around
the lytic lesion, so that chronic lesions may be
predominantly sclerotic with associated hyperostosis. The appearance of CRMO lesions can
range between purely osteolytic, osteolytic with
a sclerotic rim, mixed lytic and sclerotic, and
purely sclerotic (Fig 8). Multiple metaphyseal
lesions may be present extending along a growth
plate (Figs 6, 8). The radiographic findings
may indicate a process of longer duration than
suggested by the patient’s clinical symptoms,
likely due to the insidious nature of the clinical
symptoms (Fig 9) (49). Radiographic findings
of exacerbations include new areas of lytic destruction and periosteal reaction, which develop
sclerosis with time, resulting in progressive hyperostosis and sclerosis.
1166 July-August 2009
Figure 9. CRMO in a 6-year-old girl
with a 1-week history of right leg pain.
Anteroposterior (a) and lateral (b) radiographs of the right tibia and fibula
show sclerosis of the proximal tibia with
cortical thickening, which extends into
the distal diaphysis. A small lytic lesion
is present in the mid fibula with mild associated periosteal reaction. Bone scintigraphy showed uptake in these lesions and
in the right femur. The histopathologic,
clinical, and culture findings were compatible with CRMO.
Figure 10. CRMO in an 11-year-old patient with multifocal bone pain. (a) Oblique
radiograph of the right ankle shows a lytic
lesion of the metaphysis that appears to extend across the growth plate (arrow). There
is a suggestion of overlying soft-tissue swelling. (b) Sagittal fat-saturated T2-weighted
MR image shows edema involving the distal
metaphysis and extending across the physis
into the epiphysis. There is associated softtissue inflammation and periosteal reaction.
radiographics.rsnajnls.org
RG ■ Volume 29 • Number 4
Khanna et al 1167
CRMO, 30% were noted to have joint involvement with synovial thickening, joint effusion, or
destruction of joint cartilage and subchondral
bone (52). MR imaging may show small fluid
collections or areas of necrotic bone, features that
have been demonstrated in histologic studies of
CRMO as well (8). However, the presence of a
large fluid collection or abscess, fistulous tract, or
sequestrum makes the diagnosis of infectious osteomyelitis more likely than CRMO (51).
MR imaging is useful for follow-up of lesions
during exacerbation of disease, especially when
radiographs show marked sclerosis or hyperostosis. Areas of new activity demonstrate high signal
intensity on T2-weighted images and contrast enhancement due to marrow edema, with the surrounding sclerotic bone appearing hypointense
on both T1- and T2-weighted images.
Note that CRMO shares many features with
chronic sclerosing osteomyelitis of Garré. Both
entities are seen in children and adolescents and
are characterized by an insidious onset, a relapsing-remitting course, negative culture results, and
lack of suppuration. Both entities tend to involve
the long tubular bones—mainly the femur or
tibia. It is likely that CRMO and chronic sclerosing osteomyelitis of Garré are different names
given to the same disease (45).
Figure 11. Soft-tissue inflammation in an 11-yearold girl with known CRMO and a 2-year history of
chest pain. (a) Posteroanterior radiograph of the chest
shows an expanded right third rib (arrow). (b) Axial
contrast material–enhanced T1-weighted MR image
of the chest shows marked soft-tissue inflammatory
changes around the rib (arrow), a finding that can
mimic a malignancy like Ewing sarcoma. Biopsy revealed sterile inflammation.
MR imaging is useful both for determining
the extent of disease and for surveillance. During the active phase of the disease, MR imaging
shows typical findings of marrow edema, which
appears hypointense on T1-weighted images and
hyperintense on T2-weighted images. MR imaging can demonstrate associated periostitis, softtissue inflammation, and transphyseal disease—
findings that are typically underestimated on
radiographs (Figs 10, 11) (39,51). The detection
of transphyseal involvement may have prognostic
significance by allowing identification of patients
at higher risk for growth deformities due to formation of physeal bars (22).
Clinically occult joint effusions adjacent to
the osseous lesions can be demonstrated with
MR imaging (28). In a series of 66 patients with
Tubular Bones
The most common location of disease in the
tubular bones is the metaphysis adjacent to
the growth plate (over 50%), followed by the
diaphysis, diaphyseal-metaphyseal region, and
metaphyseal-epiphyseal region (34). The typical finding in the long tubular bones is a round
or column-shaped osteolytic metaphyseal lesion
abutting the growth plate (Fig 6). The lytic lesion becomes surrounded by a thin, sclerotic
zone within 1–2 weeks (15). As the inflammatory process extends into the cortex, periosteal
reaction develops, which can be quite extensive
in long tubular bones.
The amount of periosteal reaction depends on
both the duration of disease and the anatomic site
involved. Although small-diameter bones like the
fibula and metacarpals-metatarsals can have extensive periosteal reaction and soft-tissue inflammation, lesions in large-diameter bones like the femur
can have a predominantly lytic-sclerotic process
with minimal surrounding reaction (Fig 12). It has
been suggested that small-diameter bones demonstrate more periosteal reaction owing to earlier
extension of the inflammatory process from the
medullary space into the cortex (22). The marked
1168 July-August 2009
radiographics.rsnajnls.org
Figure 12. CRMO of tubular bones in a 7-year-old girl with a 5-year history of multifocal bone pain. (a) Radiograph of the ankles shows multiple lytic lesions with surrounding sclerosis adjacent to the growth plate in both tibias
and fibulas. The physes and epiphyses appear to be spared. (b) Follow-up radiograph obtained 7 years later shows
bilaterally symmetric hyperostosis of the fibula without significant periosteal reaction in the tibia.
periosteal reaction and sclerosis seen with CRMO
lesions of tubular bones has also been referred to
as “tumorous osteomyelitis” (Fig 2) (20).
Chronic inflammation at the metaphysis can
Teaching result in abnormal tubulation of the long bones
Point
with metaphyseal expansion and sclerosis, a useful imaging finding in long-standing CRMO (Fig
12). Although the inflammation can extend into
the cortex, the presence of cortical disruption
should raise concern for an aggressive lesion,
such as a malignancy, and further evaluation with
a biopsy should be performed (53).
Clavicle
Teaching
Point
CRMO of the clavicle most commonly manifests
with local swelling, pain, or impairment of shoulder movement and may cause thoracic outlet
syndrome (54). Like CRMO in other parts of the
body, clavicular involvement appears to be more
common in female patients than in male patients.
CRMO is the most common nonneoplastic process involving the clavicle in patients less than
20 years of age (55). It is also the most common
disease process to involve the medial third of
the clavicle in all age groups (55). Up to 30%
of all CRMO lesions are located in the clavicle
(40). Given the preceding facts, the diagnosis of
CRMO should be considered in the differential
diagnosis of a sclerotic expanded lesion affecting
the medial third of the clavicle in children and
adolescents (Fig 13). Considering this diagnosis
can help avoid unnecessary biopsies and anxiety
for patients, secondary to a misdiagnosis of a malignancy such as Ewing sarcoma (55,56).
Figure 13. CRMO of the clavicle in a 7-year-old boy
with a 1-year history of CRMO. Plain radiograph of the
left clavicle shows sclerosis and expansion of its medial
aspect (arrow), while the lateral aspect is normal.
Clavicular CRMO is unique compared with
the other common sites of disease in CRMO, in
that hematogenous osteomyelitis of the clavicle is
extremely uncommon. CRMO of the clavicle has
been referred to by several different names, including recurrent hyperostosis of the clavicle, sternocostoclavicular hyperostosis, idiopathic cortical hyperostosis, and condensing osteitis of the clavicle
(56,57). There seems to be a higher prevalence of
clavicular disease in CRMO patients with palmoplantar pustulosis and acne fulminans (14).
Clavicular disease typically manifests as a lytic
medullary process in the medial part of the clavicle with surrounding periosteal reaction, which
may have an “onion skin” appearance. During
remissions, the lesions heal with progressive sclerosis. With each exacerbation, the disease tends
to extend laterally, although the most lateral part
of the clavicle remains unaffected (58). CRMO
RG ■ Volume 29 • Number 4
Khanna et al 1169
Figure 14. CRMO of the
spine in a 4-year-old boy with
back pain. (a) Lateral radiograph of the lumbar spine
shows sclerosis of the L3 vertebral body (arrow). (b) Axial
CT image shows sclerosis and
permeative lytic changes in
the vertebral body and extension into the right pedicle
(arrow). (c) Axial postcontrast
T1-weighted MR image shows
extension of the inflammatory
process into the adjacent paravertebral soft tissues. Biopsy
revealed chronic inflammatory
changes with no organism
identified.
in children is characterized by progressive hyperostosis and sclerosis of the medial end of the
clavicle without involvement of the sternoclavicular joint. This is in contradistinction to clavicular
involvement in adults with CRMO; such involvement occurs as a subphenotype of SAPHO
syndrome, in which clavicular disease is often accompanied by arthritic changes of the joint space
(58). Also, the ligamentous ossification and bony
bridging across the sternoclavicular joint reported
in adults have not been described in children.
Radiographic evaluation of the clavicle can be
limited due to overlapping structures. CT can
more clearly demonstrate the extent of disease
with lytic areas and surrounding sclerosis and
hyperostosis. MR imaging reveals bone marrow edema with surrounding periosteal reaction. Inflammatory changes can also be seen
extending into the surrounding soft tissues and
muscles; these changes may give the appearance
of an aggressive process such as Ewing sarcoma,
lymphoma, or histiocytosis (54). With increasing sclerosis and hyperostosis, both T1- and T2weighted images show markedly hypointense areas. New areas of activity can be identified in the
hyperostotic clavicle as hyperintense lesions on
T2-weighted images, in cases where radiographic
evaluation may be limited due to overlapping
structures and hyperostosis. Whereas the metaphyseal lesions of long bones can heal completely,
the sclerotic-hyperostotic clavicular lesions tend
to persist for several years and can cause thoracic
outlet syndrome (41).
Spine
Spinal involvement, although less common than
disease in the long tubular bones, has been reported in CRMO. In a review of 35 cases of
CRMO with 157 lesions, only 3% of lesions
were present in the vertebral bodies (12). However, CRMO of the vertebral body is the most
common site to be complicated by a pathologic
fracture (28). Spinal disease may be detected
incidentally during the radiologic work-up of suspected CRMO, or it may be the site of primary
presentation. Presenting symptoms can include
back pain, scoliosis or kyphosis, and rarely cord
compression (59).
Radiologic findings include partial or complete
loss in height of the vertebral body. A lytic lesion
with surrounding sclerosis or a purely sclerotic appearance may be seen in the vertebral body with
or without extension into the pedicle (Fig 14). The
thoracic spine is the most common site of disease
(accounting for 46 of 65 spinal lesions in one review), followed by the lumbar spine (13 of 65),
while disease in the cervical spine and sacrum is
less common (60).
1170 July-August 2009
radiographics.rsnajnls.org
Figure 15. CRMO of the spine in a 6-year-old boy with multifocal bone pain for 1 year.
(a) Anteroposterior radiograph shows marked loss in height of the T4 vertebral body (arrow)
and partial collapse of the T7 and T8 vertebral bodies with endplate irregularity (arrowhead).
(b) Sagittal T2-weighted MR image shows a vertebra plana at T4. Spondylodiskitis is seen at
T7-8 with endplate irregularity, but disk height and signal intensity are preserved.
Figure 16. CRMO of the spine in a 6-year-old girl with a history of back pain. Radiography showed some loss in height of the T9 vertebral body. (a) Sagittal T1-weighted MR image
shows loss of normal marrow signal intensity in the T2, T9, T10, and T12 vertebral bodies
with loss in height of T9. (b) T2-weighted MR image shows edema in the T9 vertebral body
with a low-signal-intensity line paralleling the inferior endplate (arrow). Note that the intervertebral disk space appears unremarkable even in the presence of contiguous vertebral body
disease.
RG ■ Volume 29 • Number 4
Figure 17. CRMO of the pelvis in a 9-year-old girl
with multifocal bone pain since the age of 2 years. Anteroposterior radiograph of the pelvis shows multiple
lytic lesions with sclerotic margins at both ischiopubic
synchondroses (arrows); the lesions extend into the
ischial tuberosities.
Khanna et al 1171
MR imaging demonstrates altered signal
intensity of the vertebral marrow and partial or
complete loss in height of the vertebral body. The
adjacent disk may show some loss in height or altered signal intensity; however, to our knowledge,
none of the reported cases have shown extension
of disease across the intervertebral disk space—a
feature that can help differentiate CRMO from
infectious osteomyelitis (Fig 15) (60,61). A subchondral fracture-like line has also been reported
in CRMO patients with vertebral collapse, but
there are no data on the specificity of this finding
for a diagnosis of CRMO (Fig 16) (60). MR imaging often demonstrates additional sites of disease
in the spine that are occult at clinical examination,
radiography, or scintigraphy (Fig 16) (61,62).
Involvement of multiple noncontiguous vertebral bodies with sparing of the intervening
disks helps differentiate CRMO from infectious osteomyelitis. Although histiocytosis X
remains the most common cause of vertebra
plana in children, CRMO is another disease that
can cause vertebra plana (49). Since the initial
description in 1989, several cases of vertebra
plana secondary to CRMO have been reported
(49,51,63–65). Unlike in vertebra plana caused
by eosinophilic granuloma, restitution of vertebral body height has not been observed in vertebra plana caused by CRMO (49).
Pelvis
Figure 18. CRMO of the pelvis in a 7-year-old boy
with biopsy-proved CRMO who presented with back
pain. (a) Pelvic radiograph shows widening of the left
sacroiliac joint (arrow). (b) Axial CT image shows
marked erosive changes along the iliac aspect of the
joint (arrow) with joint space widening, findings consistent with unilateral sacroiliitis.
CRMO of the pelvis is less common. The sites of
predilection in the pelvis include the metaphyseal
equivalents—such as the ischiopubic synchondrosis and the sacroiliac joints—sites that are prone
to hematogenous osteomyelitis as well (Fig 17).
These lesions may be subtle on radiographs; however, cross-sectional imaging with CT will clearly
demonstrate the osteolysis with surrounding periosteal reaction and sclerosis. MR imaging shows
edema in active lesions with surrounding periostitis and associated soft-tissue inflammation.
Because the synchondroses have avidity for
99m
Tc, bone scans should be evaluated carefully
at single-photon-emission CT for any asymmetry
of activity, to identify pathologic sites. CRMO of
the pelvic synchondrosis tends to heal without
significant sequalae (66). Pelvic disease can also
manifest as sclerosis of bones such as the iliac
wings. During evaluation of the pelvis, close attention should be paid to the sacroiliac joints,
as patients with CRMO are at increased risk of
developing spondyloarthropathy with unilateral
sacroiliitis (Fig 18) (42).
1172 July-August 2009
radiographics.rsnajnls.org
Figure 19. CRMO of the mandible in an 8-year-old girl with a 10-week history of mandibular pain and swelling. (a) Axial maxillofacial CT image shows marked sclerosis and
expansion of the mandible with multiple lytic foci. (b) Axial maxillofacial CT image shows
that the sclerosis and expansion also involve the right maxilla and zygoma (arrow).
Mandible
CRMO can involve the mandible in about 5% of
cases (40). Mandibular CRMO has also been referred to as diffuse sclerosing osteomyelitis of the
mandible (67–69). It typically manifests as chronic
mandibular pain, although associated jaw swelling may also occur. Mandibular disease can be
isolated or accompanied by disease at other sites
in the body. In a review of 12 adult cases of diffuse
sclerosing osteomyelitis of the mandible, 50% were
found to have disease at other sites such as the
anterior chest and spine (68). Mandibular disease
has also been shown to be associated with lesions
in other skull or facial bones such as the maxilla
and zygoma (Figs 1, 19) (69,70).
Radiographic findings are similar to those in
other sites of disease, with initial stages demonstrating a lytic process with associated variable
amounts of sclerosis. Cross-sectional imaging
often reveals associated soft-tissue inflammation as well, without evidence of discrete abscess
formation. As the mandibular disease progresses,
increasing sclerosis with hyperostosis and progressive enlargement of the mandible develop.
Biopsy results reveal a culture-negative chronic
osteomyelitis. Awareness of this entity, especially
among oral surgeons, is essential to avoid unnecessary antibiotic therapy and multiple invasive
procedures (Fig 5).
Hands and Feet
CRMO is more common in the small bones of
the feet than in the hands. It can involve the tarsal bones such as the calcaneus and talus, which
are metaphyseal equivalents, or the short tubular
bones including the metatarsals and phalanges
(Fig 20). The short tubular bones typically demonstrate lytic lesions with surrounding sclerosis,
periosteal reaction, and associated soft-tissue
inflammation (Fig 21). CRMO of the metatarsals
and phalanges has been reported to cause premature physeal closure (22).
Differential Diagnosis
The differential diagnosis of CRMO includes
subacute and chronic infectious osteomyelitis,
histiocytosis, hypophosphatasia, and malignancies
like leukemia, lymphoma, and Ewing sarcoma
(71). Although the typical imaging findings of
CRMO can aid in suggesting the correct diagnosis, they are not pathognomonic.
Clinical and radiographic features that help in
differentiating CRMO from infectious osteomyelitis and tumors include the following: (a) a prolonged clinical course, with most patients being
RG ■ Volume 29 • Number 4
Khanna et al 1173
Figures 20, 21. (20) CRMO of the foot in an
8-year-old patient with a 1-year history of biopsyproved CRMO who presented with new-onset heel
pain. (a) Lateral radiograph of the ankle shows sclerosis of the calcaneus (arrow) adjacent to the physis with
overlying soft-tissue swelling. (b) Sagittal T1-weighted
MR image shows edema of the posterior calcaneal
apophysis (arrow) with mild surrounding soft-tissue
inflammation. The diagnosis of nonbacterial osteomyelitis was confirmed with biopsy. (21) CRMO of the
foot in a 12-year-old girl with a 1-year history of ankle
and foot pain. Anteroposterior radiograph of the foot
shows generalized osteopenia. A subphyseal lytic lesion
is seen in the first proximal phalanx (arrow). Multiple
other oval, peripherally sclerotic lesions (arrowheads)
are present in the metatarsal and tarsal bones. The
clinical features, histopathologic findings, and culture
results were compatible with a diagnosis of CRMO.
Outcomes
healthy between recurrent episodes; (b) an unusual location of lesions in comparison with those
of infectious osteomyelitis, such as involvement
of the clavicle, bilateral symmetry, and frequent
multifocality; (c) radiographs showing multiple
foci of osteolysis with associated sclerosis or hyperostosis; (d) lack of abscess formation, fistulas,
or sequestra; (e) lack of response to antibiotics;
and (f) comorbid inflammatory disorders such as
psoriasis, palmoplantar pustulosis, or inflammatory bowel disease (19,51). Because the imaging
findings of CRMO can be nonspecific, especially
early in the course of the disease, evaluation of
the osseous lesions with biopsy and culture is required to establish a diagnosis of CRMO.
The disease course of CRMO is unpredictable.
Although most cases undergo spontaneous resolution in several months to several years, patients
have been found to be symptomatic for as long as
25 years after the initial diagnosis (72). In a study
of 13 patients, Jurik et al (66) found all cases to be
symptomatic after a follow-up period of 16 years,
although in most cases symptoms related to the
initial lesions had resolved. In contrast, long-term
follow-up of a German cohort revealed all cases
to be in remission (19). In a follow-up study of 23
patients, Huber et al (50) showed that 25% of subjects had persistent signs of disease activity when
evaluated at a median time of 12.4 years after diagnosis. In the follow-up study of Duffy et al (72),
the shortest duration of symptoms was 2.5 years
and the longest duration was 20 years.
1174 July-August 2009
Although most CRMO lesions heal to become
occult at radiography, others can demonstrate
persistent sclerosis. Persistent sclerosis appears
to be the pattern in older patients, in whom the
normal modeling and growth process of bone is
already complete. Lesions occurring in skeletally
immature children may undergo restitution to
normal bone. This might not be true in bones
that have developed marked hyperostosis, with
sclerosis extending into the medullary cavity. One
example of this is the clavicle, where the hyperostosis can produce a palpable bump and be a
source of thoracic outlet obstruction (41). Another residual deformity is thoracic kyphosis secondary to vertebral body compression (37,66).
CRMO of long bones can result in orthopedic complications like bony overgrowth, angular
deformities, and limb-length discrepancy at maturity (72). Owing to its tendency to occur near
the physes, CRMO can cause premature physeal
closure, resulting in growth arrest. Manson et al
(22) described six sites of premature physeal fusion among 29 sites of disease in seven patients
with CRMO. Premature physeal fusion has been
reported most often in the metatarsals, followed
by the tibia and femur (22,41). Limb-length
discrepancy has been reported to be the most
common physical deformity in patients with a
history of CRMO, with the affected limb being
shorter in most cases (72). The hyperemia caused
by chronic inflammation may result in diffuse demineralization, predisposing to fractures (73).
An increased risk of chronic spondyloarthropathy, unilateral sacroiliitis, and enthesitis has
also been suggested in children with a history of
CRMO in the absence of a human leukocyte antigen (HLA)–B27 association (21,42,43). In the
study of Vittecoq et al (42), 12 of 15 patients with
CRMO developed spondyloarthropathy that met
the criteria of the European Spondyloarthropathy
Study Group, with spine involvement developing
after a median follow-up of 5.63 years.
CRMO versus SAPHO Syndrome
SAPHO syndrome is the adult equivalent of
CRMO. Whereas CRMO typically manifests in
the first decade of life, the mean age of onset for
SAPHO syndrome is 28 years (74). However,
adults with CRMO and children with SAPHO
syndrome have been reported (40,75). The prevalence of skin lesions like palmoplantar pustulosis
is much higher in SAPHO syndrome than in
radiographics.rsnajnls.org
CRMO. Furthermore, the skin lesions may manifest several years after the first bone lesion in children with CRMO (43).
Whereas chest wall and pelvic disease predominates in SAPHO syndrome, the long bones
are the most common sites of disease in CRMO.
This is likely related to the presence of open
physes in children predisposing to a higher prevalence of metaphyseal lesions, as is seen in hematogenous osteomyelitis. CRMO and SAPHO
syndrome differ in the type of disease seen at the
anterior chest wall. Although involvement of the
sternoclavicular and first sternocostal joint and
ossification of the costoclavicular ligaments are
frequent in SAPHO syndrome, these findings
are not reported in CRMO.
Adults with SAPHO syndrome can have arthritis of peripheral joints and ossification of ligaments
and entheses, features not typically seen with
CRMO (66). Children with CRMO have been reported to have arthritis at sites distant from the osteomyelitis lesions. Whether CRMO and SAPHO
syndrome are distinct diseases or the same disease
process with differences in phenotype based on
age of onset remains to be determined.
Summary
CRMO is an established clinicopathologic entity,
although its cause remains elusive. Along with the
variability in its initial presentation, there is great
variability in the numbers of sites affected, recurrence rates, and prognosis. The typical clinical
picture of CRMO is a child or young adult with a
history of chronic multifocal bone pain, in whom
biopsy of osseous lesions shows osteomyelitis with
failure to culture any organisms. The typical radiographic appearance is a lytic lesion at a metaphysis
or metaphyseal equivalent that develops progressive sclerosis and hyperostosis over time.
With the exception of CRMO in the context of
Majeed syndrome, there is no diagnostic test for
CRMO and it remains a diagnosis of exclusion.
The presence of disease in typical sites such as the
medial clavicle, the presence of bilateral disease in
a symmetric distribution, and the presence of skin
findings such as palmoplantar pustulosis support
the diagnosis. Diagnosis of CRMO requires a team
approach between the pediatric rheumatologist,
orthopedic surgeon, radiologist, and pathologist.
Radiologists need to be familiar with the imaging
findings because they may be the first to suggest
this diagnosis. This can help minimize the number
of unnecessary interventions in the form of repeated biopsies, surgeries, and antibiotic therapy.
RG ■ Volume 29 • Number 4
References
1.Giedion A, Holthusen W, Masel LF, Vischer D. Subacute and chronic “symmetrical” osteomyelitis [in
French]. Ann Radiol (Paris) 1972;15:329–342.
2.Ferguson PJ, El-Shanti HI. Autoinflammatory bone
disorders. Curr Opin Rheumatol 2007;19:492–498.
3.Probst FP, Bjorksten B, Gustavson KH. Radiological aspect of chronic recurrent multifocal osteomyelitis. Ann Radiol (Paris) 1978;21:115–125.
4.Solheim LF, Paus B, Liverud K, Stoen E. Chronic
recurrent multifocal osteomyelitis: a new clinicalradiological syndrome. Acta Orthop Scand 1980;51:
37–41.
5.Probst FP. Chronic multifocal cleido-metaphyseal
osteomyelitis of childhood: report of a case. Acta
Radiol Diagn (Stockh) 1976;17:531–537.
6.Willert HG, Enderle A. Multifocal symmetrical
chronic osteomyelitis [in German]. Arch Orthop
Unfallchir 1977;89:109–113.
7.Marino C, McDonald E, Megna D, Brennessel D,
Reddy KS, Jain PC. Chronic recurrent multifocal
osteomyelitis in adult women. N Y State J Med
1992;92:360–362.
8.Bj0rksten B, Boquist L. Histopathological aspects
of chronic recurrent multifocal osteomyelitis. J Bone
Joint Surg Br 1980;62:376–380.
9.Cyrlak D, Pais MJ. Chronic recurrent multifocal osteomyelitis. Skeletal Radiol 1986;15:32–39.
10.Majeed HA, Al-Tarawna M, El-Shanti H, Kamel B,
Al-Khalaileh F. The syndrome of chronic recurrent
multifocal osteomyelitis and congenital dyserythropoietic anaemia: report of a new family and a review.
Eur J Pediatr 2001;160:705–710.
11.Majeed HA, El-Shanti H, Al-Rimawi H, Al-Masri
N. On mice and men: an autosomal recessive syndrome of chronic recurrent multifocal osteomyelitis
and congenital dyserythropoietic anemia. J Pediatr
2000;137:441–442.
12.Jurriaans E, Singh NP, Finlay K, Friedman L. Imaging of chronic recurrent multifocal osteomyelitis.
Radiol Clin North Am 2001;39:305–327.
13.Jurik AG. Chronic recurrent multifocal osteomyelitis. Semin Musculoskelet Radiol 2004;8:243–253.
14.Laasonen LS, Karvonen SL, Reunala TL. Bone disease in adolescents with acne fulminans and severe
cystic acne: radiologic and scintigraphic findings.
AJR Am J Roentgenol 1994;162:1161–1165.
15.Mortensson W, Edeburn G, Fries M, Nilsson R.
Chronic recurrent multifocal osteomyelitis in children: a roentgenologic and scintigraphic investigation. Acta Radiol 1988;29:565–570.
16.Bjorksten B, Gustavson KH, Eriksson B, Lindholm
A, Nordstrom S. Chronic recurrent multifocal osteomyelitis and pustulosis palmoplantaris. J Pediatr
1978;93:227–231.
17.Meller Y, Yagupsky P, Elitsur Y, Inbar-Ianay I, BarZiv J. Chronic multifocal symmetrical osteomyelitis: report of two cases in Bedouin infants. Am J
Dis Child 1984;138:349–351.
18.Murray SD, Kehl DK. Chronic recurrent multifocal
osteomyelitis: a case report. J Bone Joint Surg Am
1984;66:1110–1112.
19.King SM, Laxer RM, Manson D, Gold R. Chronic
recurrent multifocal osteomyelitis: a noninfectious
inflammatory process. Pediatr Infect Dis J 1987;6:
907–911.
Khanna et al 1175
20.Kozlowski K, Anderson RJ, Hochberger O, Sacher
M. Tumorous osteomyelitis: report of two cases.
Pediatr Radiol 1984;14:404–406.
21.Girschick HJ, Raab P, Surbaum S, et al. Chronic
non-bacterial osteomyelitis in children. Ann Rheum
Dis 2005;64:279–285.
22.Manson D, Wilmot DM, King S, Laxer RM. Physeal involvement in chronic recurrent multifocal
osteomyelitis. Pediatr Radiol 1989;20:76–79.
23.Laxer RM, Shore AD, Manson D, King S, Silverman ED, Wilmot DM. Chronic recurrent multifocal osteomyelitis and psoriasis: a report of a new
association and review of related disorders. Semin
Arthritis Rheum 1988;17:260–270.
24.Dagan O, Barak Y, Metzker A. Pyoderma gangrenosum and sterile multifocal osteomyelitis preceding
the appearance of Takayasu arteritis. Pediatr Dermatol 1995;12:39–42.
25.Bousvaros A, Marcon M, Treem W, et al. Chronic
recurrent multifocal osteomyelitis associated with
chronic inflammatory bowel disease in children.
Dig Dis Sci 1999;44:2500–2507.
26.El-Shanti HI, Ferguson PJ. Chronic recurrent multifocal osteomyelitis: a concise review and genetic
update. Clin Orthop Relat Res 2007;462:11–19.
27.Grosse J, Chitu V, Marquardt A, et al. Mutation of
mouse Mayp/Pstpip2 causes a macrophage autoinflammatory disease. Blood 2006;107:3350–3358.
28.Jansson A, Renner ED, Ramser J, et al. Classification of non-bacterial osteitis: retrospective study of
clinical, immunological and genetic aspects in 89
patients. Rheumatology (Oxford) 2007;46:154–160.
29.Golla A, Jansson A, Ramser J, et al. Chronic recurrent
multifocal osteomyelitis (CRMO): evidence for a
susceptibility gene located on chromosome 18q21.318q22. Eur J Hum Genet 2002;10:217–221.
30.Ferguson PJ, Chen S, Tayeh MK, et al. Homozygous
mutations in LPIN2 are responsible for the syndrome of chronic recurrent multifocal osteomyelitis
and congenital dyserythropoietic anaemia (Majeed
syndrome). J Med Genet 2005;42:551–557.
31.Al-Mosawi ZS, Al-Saad KK, Ijadi-Maghsoodi R,
El-Shanti HI, Ferguson PJ. A splice site mutation
confirms the role of LPIN2 in Majeed syndrome.
Arthritis Rheum 2007;56:960–964.
32.Ferguson PJ, Bing X, Vasef MA, et al. A missense
mutation in pstpip2 is associated with the murine
autoinflammatory disorder chronic multifocal osteomyelitis. Bone 2006;38:41–47.
33.Byrd L, Grossmann M, Potter M, Shen-Ong GL.
Chronic multifocal osteomyelitis, a new recessive
mutation on chromosome 18 of the mouse. Genomics 1991;11:794–798.
34.Mandell GA, Contreras SJ, Conard K, Harcke HT,
Maas KW. Bone scintigraphy in the detection of
chronic recurrent multifocal osteomyelitis. J Nucl
Med 1998;39:1778–1783.
35.Van Howe RS, Starshak RJ, Chusid MJ. Chronic,
recurrent multifocal osteomyelitis: case report and
review of the literature. Clin Pediatr (Phila) 1989;
28:54–59.
1176 July-August 2009
36.Schaen L, Sheth AP. Skin ulcers associated with a
tender and swollen arm: pyoderma gangrenosum
(PG) in association with chronic recurrent multifocal osteomyelitis (CRMO). Arch Dermatol 1998;
134:1146–1147, 1149–1150.
37.Schultz C, Holterhus PM, Seidel A, et al. Chronic
recurrent multifocal osteomyelitis in children. Pediatr Infect Dis J 1999;18:1008–1013.
38.Bognar M, Blake W, Agudelo C. Chronic recurrent
multifocal osteomyelitis associated with Crohn’s disease. Am J Med Sci 1998;315:133–135.
39.Sundaram M, McDonald D, Engel E, Rotman M,
Siegfried EC. Chronic recurrent multifocal osteomyelitis: an evolving clinical and radiological spectrum. Skeletal Radiol 1996;25:333–336.
40.Beretta-Piccoli BC, Sauvain MJ, Gal I, et al. Synovitis, acne, pustulosis, hyperostosis, osteitis
(SAPHO) syndrome in childhood: a report of ten
cases and review of the literature. Eur J Pediatr
2000; 159:594–601.
41.Pelkonen P, Ryoppy S, Jaaskelainen J, Rapola J,
Repo H, Kaitila I. Chronic osteomyelitislike disease
with negative bacterial cultures. Am J Dis Child
1988;142:1167–1173.
42.Vittecoq O, Said LA, Michot C, et al. Evolution of
chronic recurrent multifocal osteitis toward spondylarthropathy over the long term. Arthritis Rheum
2000;43:109–119.
43.Job-Deslandre C, Krebs S, Kahan A. Chronic recurrent multifocal osteomyelitis: 5-year outcomes in 14
pediatric cases. Joint Bone Spine 2001;68:245–251.
44.Majeed HA, Kalaawi M, Mohanty D, et al. Congenital dyserythropoietic anemia and chronic recurrent
multifocal osteomyelitis in three related children and
the association with Sweet syndrome in two siblings.
J Pediatr 1989;115:730–734.
45.Chow LT, Griffith JF, Kumta SM, Leung PC.
Chronic recurrent multifocal osteomyelitis: a great
clinical and radiologic mimic in need of recognition by the pathologist. APMIS 1999;107:369–379.
radiographics.rsnajnls.org
46.Goo HW, Yang DH, Ra YS, et al. Whole-body MRI
of Langerhans cell histiocytosis: comparison with
radiography and bone scintigraphy. Pediatr Radiol
2006;36:1019–1031.
47.Steinborn M, Wortler K, Nathrath M, Schoniger
M, Hahn H, Rummeny EJ. Whole-body MRI in
children with Langerhans cell histiocytosis for the
evaluation of the skeletal system [in German]. Rofo
2008;180:646–653.
48.Fritz J, Tzaribachev N, Kummerl-Deschner J,
Claussen CD, Carrino JA, Horger M. Whole-body
magnetic resonance imaging of children with newonset chronic recurrent multifocal osteomyelitis:
description and comparison with laboratory and
clinical data [abstr]. In: Radiological Society of
North America scientific assembly and annual meeting program. Oak Brook, Ill: Radiological Society of
North America, 2008; 1034.
49.Yu L, Kasser JR, O’Rourke E, Kozakewich H.
Chronic recurrent multifocal osteomyelitis: association with vertebra plana. J Bone Joint Surg Am
1989;71:105–112.
50.Huber AM, Lam PY, Duffy CM, et al. Chronic recurrent multifocal osteomyelitis: clinical outcomes
after more than 5 years of follow-up. J Pediatr 2002;
141:198–203.
51.Jurik AG, Egund N. MRI in chronic recurrent multifocal osteomyelitis. Skeletal Radiol 1997;26:230–238.
52.Wang A, Babyn P. Joint lesions in children with
chronic recurrent multifocal osteomyelitis [abstr]. In:
Radiological Society of North America scientific assembly and annual meeting program. Oak Brook, Ill:
Radiological Society of North America, 2008; 1035.
53.Sato TS, Ferguson PJ, Khanna G. Primary multifocal osseous lymphoma in a child. Pediatr Radiol
2008;38:1338–1341.
54.Girschick HJ, Krauspe R, Tschammler A, Huppertz
HI. Chronic recurrent osteomyelitis with clavicular
involvement in children: diagnostic value of different
imaging techniques and therapy with non-steroidal
anti-inflammatory drugs. Eur J Pediatr 1998;157:
28–33.
RG ■ Volume 29 • Number 4
55.Suresh S, Saifuddin A. Unveiling the “unique
bone”: a study of the distribution of focal clavicular
lesions. Skeletal Radiol 2008;37:749–756.
56.Appell RG, Oppermann HC, Becker W, Kratzat R,
Brandeis WE, Willich E. Condensing osteitis of the
clavicle in childhood: a rare sclerotic bone lesion—
review of literature and report of seven patients.
Pediatr Radiol 1983;13:301–306.
57.Azouz EM, Jurik AG, Bernard C. Sternocostoclavicular hyperostosis in children: a report of eight cases.
AJR Am J Roentgenol 1998;171:461–466.
58.Jurik AG, Madsen EH, Nordentoft AM, Jensen
OM, Sneppen O. Chronic sclerosing osteitis of the
clavicle: a manifestation of chronic recurrent multifocal osteitis [in Danish]. Ugeskr Laeger 1987;149:
1260–1262.
59.Baulot E, Bouillien D, Giroux EA, Grammont PM.
Chronic recurrent multifocal osteomyelitis causing spinal cord compression. Eur Spine J 1998;7:
340–343.
60.Anderson SE, Heini P, Sauvain MJ, et al. Imaging of
chronic recurrent multifocal osteomyelitis of childhood first presenting with isolated primary spinal
involvement. Skeletal Radiol 2003;32:328–336.
61.Martin JC, Desoysa R, O’Sullivan MM, Silverstone
E, Williams H. Chronic recurrent multifocal osteomyelitis: spinal involvement and radiological appearances. Br J Rheumatol 1996;35:1019–1021.
62.Dawson JS, Webb JK, Preston BJ. Case report:
chronic recurrent multifocal osteomyelitis with
magnetic resonance imaging. Clin Radiol 1994;49:
133–136.
63.Brown T, Wilkinson RH. Chronic recurrent multifocal osteomyelitis. Radiology 1988;166:493–496.
64.Carr AJ, Cole WG, Roberton DM, Chow CW.
Chronic multifocal osteomyelitis. J Bone Joint Surg
Br 1993;75:582–591.
65.Demharter J, Bohndorf K, Michl W, Vogt H. Chronic
recurrent multifocal osteomyelitis: a radiological and
clinical investigation of five cases. Skeletal Radiol
1997;26:579–588.
66.Jurik AG, Helmig O, Ternowitz T, Moller BN.
Chronic recurrent multifocal osteomyelitis: a followup study. J Pediatr Orthop 1988;8:49–58.
Khanna et al 1177
67.Kahn MF, Hayem F, Hayem G, Grossin M. Is diffuse sclerosing osteomyelitis of the mandible part of
the synovitis, acne, pustulosis, hyperostosis, osteitis
(SAPHO) syndrome? Analysis of seven cases. Oral
Surg Oral Med Oral Pathol 1994;78:594–598.
68.Suei Y, Taguchi A, Tanimoto K. Diffuse sclerosing
osteomyelitis of the mandible: its characteristics and
possible relationship to synovitis, acne, pustulosis,
hyperostosis, osteitis (SAPHO) syndrome. J Oral
Maxillofac Surg 1996;54:1194–1199; discussion
1199–1200.
69.van Merkesteyn JP, Groot RH, Bras J, McCarroll
RS, Bakker DJ. Diffuse sclerosing osteomyelitis of
the mandible: a new concept of its etiology. Oral
Surg Oral Med Oral Pathol 1990;70:414–419.
70.Flygare L, Norderyd J, Kubista J, Ohlsson J, ValloChristiansen J, Magnusson B. Chronic recurrent
multifocal osteomyelitis involving both jaws: report
of a case including magnetic resonance correlation.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod
1997;83:300–305.
71.Girschick HJ, Mornet E, Beer M, Warmuth-Metz
M, Schneider P. Chronic multifocal non-bacterial
osteomyelitis in hypophosphatasia mimicking malignancy. BMC Pediatr 2007;7:3.
72.Duffy CM, Lam PY, Ditchfield M, Allen R, Graham
HK. Chronic recurrent multifocal osteomyelitis:
review of orthopaedic complications at maturity. J
Pediatr Orthop 2002;22:501–505.
73.Prose NS, Fahrner LJ, Miller CR, Layfield L. Pustular psoriasis with chronic recurrent multifocal osteomyelitis and spontaneous fractures. J Am Acad
Dermatol 1994;31:376–379.
74.Hayem G, Bouchaud-Chabot A, Benali K, et al.
SAPHO syndrome: a long-term follow-up study of
120 cases. Semin Arthritis Rheum 1999;29:159–171.
75.Schilling F. Chronic recurrent multifocal osteomyelitis (CRMO) [in German]. Rofo 1998;168:115–127.
This article meets the criteria for 1.0 AMA PRA Category 1 Credit TM. To obtain credit, see www.rsna.org/education
/rg_cme.html.
RG
Volume 29 • Number 4 • July-August 2009
Khanna et al
Imaging of Chronic Recurrent Multifocal Osteomyelitis
Geetika Khanna, MD, MS, et al
RadioGraphics 2009; 29:1159–1177 • Published online 10.1148/rg.294085244 • Content Codes:
Page 1163
The identification of additional sites of disease can aid in the diagnosis of CRMO, especially when
disease is present at typical locations like the anterior chest wall or when disease is symmetrically
distributed (Figs 4, 5).
Page 1163
Common sites of skeletal involvement include the long tubular bones and clavicle, but lesions have
been described throughout the skeleton, including the spine, pelvis, sacroiliac joint, ribs, sternum,
scapula, mandible, and hands and feet.
Page 1164
The most common sites of disease are the metaphyses or metaphyseal equivalents, accounting for
approximately 75% of all lesions in the series of Mandell et al (34) (Figs 6, 7).
Page 1168
Chronic inflammation at the metaphysis can result in abnormal tubulation of the long bones with
metaphyseal expansion and sclerosis, a useful imaging finding in long-standing CRMO (Fig 12).
Page 1168
It is also the most common disease process to involve the medial third of the clavicle in all age
groups (55).