Chronic Nonbacterial Osteomyelitis and Chronic Recurrent

Chronic Nonbacterial

Osteomyelitis and Chronic

Recurrent Multifocal

Osteomyelitis in Children

Yongdong Zhao, MD, PhD

, Polly J. Ferguson, MD

INTRODUCTION

Chronic nonbacterial osteomyelitis (CNO; a.k.a., chronic recurrent multifocal oste-

omyelitis) is an inflammatory disorder that presents with bone pain arising from sterile

osteomyelitis. It is primarily a pediatric disorder but can persist into adulthood or have

an adult-onset presentation. The condition is difficult to diagnose, most commonly

suspected to be infectious osteomyelitis or malignancy, with milder cases resembling

growing pains. Children may have decreased physical function and poor school

Disclosure: Dr Y. Zhao received research funding from CARRA and Bristol-Myers Squibb (IM101-691_

Zhao). Dr P.J. Ferguson’s work is supported by R01AR059703 from NIH/NIAMS.

Pediatric Rheumatology, Seattle Children’s Hospital, University of Washington, MA 7.110,

4800 Sand Point Way Northeast, Seattle, WA 98105, USA;

Department of Pediatrics, University

of Iowa Carver College of Medicine, 200 Hawkins Drive, 4038 Boyd Tower, Iowa City, IA

52242, USA

* Corresponding author.

E-mail address: [email protected]

KEYWORDS



Chronic nonbacterial osteomyelitis



Chronic recurrent multifocal osteomyelitis



DIRA



Majeed syndrome



NSAID



Whole-body MRI

KEY POINTS

 Chronic nonbacterial osteomyelitis (CNO; also known as chronic recurrent multifocal oste-

omyelitis) is an inflammatory/autoinflammatory bone disease that primarily affects chil-

dren and adolescents. It is a diagnosis of exclusion.

 Often the diagnosis of CNO in children is delayed because of a lack of awareness and the

occult nature of CNO. Prompt referral to pediatric rheumatology can help establish a diag-

nosis and determine appropriate treatment.

 Imaging studies, especially MRI with short tau inversion recovery, are essential diagnostic

tools.

 Whole-body MRI is the gold standard for disease monitoring.

 Long-term treatment and follow-up are needed to prevent complications, such as vertebral

compression fractures and leg-length discrepancies.

Pediatr Clin N Am 65 (2018) 783–800

https://doi.org/10.1016/j.pcl.2018.04.003 pediatric.theclinics.com

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

attendance when bone pain and inflammation are not controlled adequately. Delays in

diagnosis can lead to permanent skeletal damage. The cause of the disease remains

unknown for most but involves immune dysregulation resulting in inflammation of the

bone and sometimes of other tissues, including skin, joints, and the intestine.

NOMENCLATURE

The disease has gone by many names, making nomenclature complicated (Box 1). It

was first described as a symmetric multifocal osteomyelitis and later given the name

chronic recurrent multifocal osteomyelitis (CRMO).

However, because the disease

may begin or stay unifocal, CRMO may not be an accurate term for these patients.

Thus, the term CNO has been proposed as an umbrella term.

INCIDENCE AND DEMOGRAPHICS

In 2011, the annual incidence of CNO in Germany was reported to be 0.4 per 100,000

children,

as compared with the reported incidence range of infectious osteomyelitis

of 10 to 80 per 100,000 children per year.

However, during 2004 to 2014 in a single

center in Germany, of the 109 children seen for osteomyelitis, 53% were categorized

as infectious and 47% as noninfectious, unexpectedly similar proportions.

A single

center in Britain reported increased patient referral for CNO after a letter was sent

to all orthopedic centers to enhance recognition of the disease.

Although the actual

incidence of CNO is likely to vary from one region to another, these studies suggest

that it is more common than previously appreciated and underscore the importance

of raising awareness of CNO.

Age

 The average age of disease onset is 9 to 10 years.

2,4–8

Rarely, disease onset oc-

curs before 3 years of age.

Sex

 Girls are more likely to be affected, with a female to male ratio of 2:1.

2,4–8

Box 1

Reported terms of chronic nonbacterial osteomyelitis

Bone lesions of acne fulminans

Chronic multifocal cleidometaphyseal osteomyelitis

Chronic recurrent multifocal osteomyelitis

Chronic sclerosing osteitis

Chronic symmetric osteomyelitis

Clavicular hyperostosis and acne arthritis

Diffuse sclerosing osteomyelitis

Pustulotic arthro-osteitis

Sclerosing osteomyelitis of Garre

Sternocostoclavicular hyperostosis

Sternoclavicular pustulotic osteitis

Synovitis, acne, pustulosis, hyperostosis osteitis

Zhao & Ferguson

784

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

Race and Ethnicity

 Most reported cases are of European ancestry, although it has been reported in

all races.

2,4–8

 Prevalence among different ethnicities has not been described.

MAJOR DIFFERENCES FROM ADULT CHRONIC NONBACTERIAL OSTEOMYELITIS (ALSO

KNOWN AS SYNOVITIS, ACNE, PUSTULOSIS, HYPEROSTOSIS, OSTEITIS)

 Cutaneous involvement in children is not as common as in adults.

 Common bone sites affected are long bones in children compared with the

sternum and c lavicles in adul ts.

DELAYS IN DIAGNOSIS ARE COMMON

The median time between initial symptoms and diagnosis of CNO is 2 years.

Forty-

eight percent of children were not evaluated by a pediatric rheumatologist until at least

12 months after their first symptom occurred. The delay of diagnosis was likely related

to the insidious development of pain, minimal findings on clinical examination, rela-

tively normal laboratory studies, and lack of awareness of this condition. A patient sur-

vey study discussed delays in diagnosis for 21 patients who had a single symptomatic

site and were initially misdiagnosed as bacterial osteomyelitis; 55% of these patients

did not have whole-body imaging, and this was hypothesized to contribute to the

delay in the diagnosis of CNO.

WHEN TO CONSIDER CHRONIC NONBACTERIAL OSTEOMYELITIS

CNO should be considered in a child who has intermittent or persistent focal bone or

joint pain of the lower extremities, clavicle, spine, and/or mandible. The pain is worse

at night and may interfere with sleep, and the child usually has point tenderness of the

affected site. Children with more superficially affected bones (eg, tibia, fibula, clavicle,

and mandible) can also have local swelling and warmth ( Fig. 1).

HOW CHRONIC NONBACTERIAL OSTEOMYELITIS PRESENTS

CNO typically presents as insidious bone pain with or without systemic features;

however, it can also present as acute onset of pain. Young children may stop using

an affected limb. Common features of CNO ar e shown in Box 2. Often there are min-

imal to no objective changes overlying the lesions, but point tenderness is common

when the disease is active. The pain often results in reduced physical activities;

Fig. 1. (A) Left mandibular CNO with facial swelling and asymmetry in a child. (B) Swelling

of right medial clavicle in a child with CNO.

Chronic Nonbacterial Osteomyelitis

785

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

school attendance may be affected, particula rly if lower extremities and/or axial

bones are involved.

Approximately 40% of patients present with or develop arthritis with their CNO,

which is associated with joint swelling and stiffness.

These symptoms often occur

in joints adjacent to active areas of osteomyelitis of the long bones but can occur in

areas without bone involvement. Even without arthritis, many patients have functional

limitation of the joints.

Patients may present with constitutional symptoms, such as fever (20%)

and

weight loss; however, most patients with CNO appear well. Fatigue is common in chil-

dren with CNO.

Pattern of bone involvement (Fig. 2)

 It most commonly affects the metaphyseal regions of long bones.

 Diaphyseal regions of the long bones are rarely affected.

Box 2

Clues to the diagnosis of chronic nonbacterial osteomyelitis

 Point tenderness with or without swelling or warmth

 Pain worse at night

 Limited function/use of affected limb

 Coexisting psoriasis, inflammatory bowel disease, or inflammatory arthritis

 Absence of constitutional symptoms

Fig. 2. Common sites affected in CNO. Metaphysis is indicated as the ending of long bones

and diaphysis is indicated as the middle part of long bones. (Adapted from Vecteezy.com

with free license. Accessed February 1, 2018.)

Zhao & Ferguson

786

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

 The commonly affected sites are lower extremity long bones, vertebrae, clavicles,

and mandible

 In these sites, the trabecular bone is enriched and bone turnover is higher.

 There is asymmetric involvement in approximately 60% and symmetric involve-

ment in approximately 40%.

 Unifocal bony involvement at presentation is up to 30%.

 Over time, most develop multifocal disease (93% in one study after mean of

4 years).

 Clavicular and mandibular lesions are more likely to present as unifocal lesions

and remain so throughout the disease course.

 Asymptomatic bone lesions are common.

DISEASE MORBIDITY

The most common complications of CNO are fractures in affected bones (especially of

the vertebrae) and deformities due to growth alterations. Kyphosis may occur in pa-

tients with multiple vertebral compression fractures. Pathologic fractures in the long

bones may occur when there is accelerated bone resorption. Leg-length discrepancy

may result after the growth plate is damaged by CNO (Fig. 3A) or due to bony over-

growth of the epiphysis from excess inflammation. Angulation of a joint may occur

when the growth plate is damaged asymmetrically, and often this complication re-

quires surgical intervention (Fig. 3B). Aggressive treatment of patients with physeal

damage before closure may reduce the risk of leg-length discrepancy or joint angula-

tion. Untreated or inadequately treated patients may have increasing bony expansion

(see Fig. 3B). One large study reported that 26% of children with CNO have compli-

cations, including localized deformation (particularly of the clavicle) (15%), vertebral

fractures (4%), and growth asymmetry (6%).

IMPACT ON QUALITY OF LIFE

CNO causes a significant impact on the quality of life in affected children. School

absence due to pain, fatigue, and frequent medical visits are common. Assistive devices

may be needed in some children. Inability to participate in desired sports activity

because of limited function is frequent. Based on a family survey, most parents reported

that their child with CNO was challenged to perform daily tasks or hobbies because of

pain, fatigue, and physical limitation.

Another family survey study in Germany also re-

ported a negative influence on family life in 80% of children with CNO.

Psychosocial

support should be considered to improve the quality of life for these children and their

families. Nearly half of these families reported a desire to contact other patients/families

for mutual support.

A pamphlet for families of children with CRMO has been developed

by a parent group and can be accessed on the Web site (www.crmoawareness.org).

Other sites that allow families to seek further information and connect have been devel-

oped (https://www.facebook.com/groups/CRMOawareness, www.kailaskomfort.org).

ASSOCIATED CONDITIONS

In a minority of patients, other coexisting conditions, including psoriasis vulgaris,

palmar plantar pustulosis, and inflammatory bowel disease, may occur before,

concurrently with, or after the diagnosis of CNO. Psoriasis has been reported in 2%

to 17% (Fig. 4), palmar plantar pustulosis (PPP) in 3% to 20%, and inflammatory

bowel disease (IBD) in 3% to 7% of patients with CNO.

5–8

One proposed diagnostic

criterion considers the presence of one of these comorbid conditions to help support

Chronic Nonbacterial Osteomyelitis

787

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

Fig. 3. (A) Radiograph s of a child with CNO show a 1.5-cm discrepancy between the total

lengths of 2 legs (right > left) due to growth plate damage in left proximal tibia. (B) Radiograph

of a child with CNO shows persistent angulation of left knee a year after stapling of the medial

proximal tibia.

Fig. 4. (A) Plaque psoriasis on the scalp of a child with CNO. (B) Plantar psoriasis on heel of a

child with CNO.

Zhao & Ferguson

788

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

the diagnosis of CNO.

These comorbid conditions are also present in a high propor-

tion of close relatives of patients with CNO. Up to 50% of patients with CNO have a

first- or second-degree family member with psoriasis, PPP, Crohn disease, ulcerative

colitis, or inflammatory arthritis. Family history of these diseases increases the positive

predictive value of a diagnosis of CNO.

5–7

One category of treatments used for CNO,

tumor necrosis factor (TNF)–alpha inhibitors, has also been associated with the devel-

opment of psoriasis in patients with CNO.

DIFFERENTIAL DIAGNOSIS OR MIMICS

The differential diagnosis of CNO is broad and includes infections of the bone or joint,

malignancy, benign bone lesions, metabolic bone disease, amplified pain syndromes,

and nutritional deficiencies.

The common differential diagnosis of CNO includes the following

 Leukemia

 Lymphoma

 Langerhans cell histiocytosis

 Primary malignant bone disease

 Benign bone tumor (osteoma, endo-chondroma)

 Infectious osteomyelitis

 Septic arthritis

 Avascular necrosis (osteonecrosis)

 Vitamin C deficiency (scurvy with bony changes)

 Enthesitis-related arthritis

 Psoriatic arthritis

 Amplified musculoskeletal pain syndrome/complex regional pain syndrome

 Hypophosphatasia

 Occult fracture

 Benign limb pain of childhood (growing pains)

A thorough physical examination is essential to help differentiate CNO from malig-

nancy and infection, assessing for the presence of other areas of bone or joint swelling,

lymphadenopathy, hepatosplenomegaly, mass, or rash. Lesions in the diaphysis should

prompt evaluation for alternative diagnosis, including benign and malignant bone tu-

mors. Lactate dehydrogenase (LDH) and uric acid are useful to screen for increased

cell turnover in children with leukemia and lymphoid malignancies. Alkaline phospha-

tase and serum phosphorus are useful to screen for hypophosphatasia.

Biopsy of multiple bone sites or repeated bone biopsies may need to be performed,

especially when there is a concern of a CNO-mimicking disease, such as intraosseous

lymphoma.

Bone marrow biopsy may also be obtained in patients with strong sus-

picion of leukemia (cytopenia, episodic pain, nocturnal pain, elevated LDH, raised uric

acid). Infectious workup is routinely performed on the biopsied bone samples, and

additional blood culture is needed when the concern for infection is high.

Jansson and colleagues

developed a scoring system to guide the diagnostic

workup based on the presence or absence of 7 components (Table 1). This scoring

system was intended to distinguish nonbacterial osteomyelitis from infectious osteo-

myelitis as well as from benign and malignant bone tumor in adults and children. A to-

tal score of 28 or less had a negative predictive value of 97% for CNO, whereas a total

score of 39 or greater had a positive predictive value of 97% or greater for CNO.

The pattern of pain may also help differentiate CNO from other diseases. Pain is

often worse at night and interferes with sleep. This pattern is also seen in children

Chronic Nonbacterial Osteomyelitis

789

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

who have growing pains (also known as benign limb pain of childhood, see Jennifer E.

Weiss and Jennifer N. Stinson’s article, “Pediatric Pain Syndromes and Non-

inflammatory Musculoskeletal Pain,” in this issue). Episodic severe bone pain with

significantly elevated erythrocyte sedimentation rate (ESR) and C-reactive protein

(CRP) should raise concerns for osseous infiltration from leukemia or lymphoma.

Diffuse body pain and tenderness to touch (allodynia) may be an indication of ampli-

fied musculoskeletal pain syndrome (discussed in Chapter 10) rather than active CNO.

Some children with CNO may also have amplified musculoskeletal pain syndrome,

which can make it challenging to determine if they have active disease; in such cases,

MRI can be helpful.

IMAGING, LABORATORY, AND BIOPSY STUDIES FOR DIAGNOSIS IN CARE

Imaging

Whole-body MRI is considered the gold standard imaging modality by experts. Pa-

tients almost always have a radiograph taken initially. However, MRI is preferred for

its sensitivity and lack of radiation. Typical findings of CNO from each imaging modal-

ity are listed next.

Radiographs

 Most common findings

 Lytic lesion during early phase (Fig. 5A)

 Sclerosis, bony expansion, or mixed picture during later stage (Fig. 5B)

 Normal in 80% of patients

 Other findings

 Pathologic fracture during acute lytic phase (rare)

 Compression fracture of vertebrae

- Can lead to kyphosis or vertebra plana (Fig. 5C)

 Advantages

 Quick

 Least expensive

 Disadvantages

 Least sensitive with a high false-negative rate

 Radiation

Table 1

Diagnostic scoring system of chronic nonbacterial osteomyelitis by Jansson and colleagues

Clinical, Laboratory, and Imaging Findings Points

Normal blood cell count 13

Symmetric lesions 10

Lesions with marginal sclerosis 10

Normal body temperature 9

Vertebral, clavicular, or sternal lesions 8

Radiologically proven lesions 27

C-reactive protein 1 mg/dL 6

Total possible score is 63. This scoring system was intended to distinguish nonbacterial osteomye-

litis from infectious osteomyelitis as well as from benign and malignant bone tumor in adults and

children. A total score of 28 or less had a negative predictive value of 97% for CNO, whereas a total

score of 39 or greater had a positive predictive value of 97% or greater for CNO.

Data from Jansson AF, Muller TH, Gliera L, et al. Clinical score for nonbacterial osteitis in children

and adults. Arthritis Rheum 2009;60:1152–9.

Zhao & Ferguson

790

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

Computed tomography

 Common findings

 Lytic lesions, sclerosis, bony expansion, or mixed pattern

 Advantages

 Three-dimensional rendering of affected bone to guide the biopsy

 Disadvantages

 Limited use in determining disease activity

 Radiation

Bone scintigraphy

 Common findings

 Increased uptake at affected sites (Fig. 6)

 Advantages

 Whole-body level scanning when whole-body MRI is not available

 Disadvantages

 Not as sensitive as MRI

 Challenging to distinguish inflamed sites from the physiologic increased up-

take at the growth plates

 Radiation

PET–computed tomography

 Common findings

 Increased tracer uptake

Fig. 5. (A) CNO lytic lesion (arrow) in fibula. (B) Sclerosis (arrow) of tibia. (C) Height loss of

T3 to T6 and plana of T7 due to CNO on MRI, which resulted in 47



of kyphosis.

Chronic Nonbacterial Osteomyelitis

791

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

 Advantages

 Correlates with disease activity based on increased metabolism

 Detailed localization of inflamed sites

 Disadvantages

 Radiation

MRI

 Common findings

 Hyperintensity within bone marrow in short tau inversion recovery (STIR) or T2

fat saturation sequences

 Hyperintensity within surrounding soft tissue in STIR or T2 fat saturation

sequences

 Bony expansion during late stages

 Other findings

 Physis irregularity or complete bony bar

Fig. 6. Bone scintigraphy of a child with CNO reveals increased uptake of radioisotope

(arrows) in the left third and eleventh ribs, left ilium, and left ischium indicating active

inflammation.

Zhao & Ferguson

792

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

 Compression vertebral fracture

 Advantages

 Most sensitive in detecting inflamed sites (Figs. 7 and 8)

 Assesses both disease activity and skeletal damage

 Disadvantages

 Requires sedation in young children

Whole body MRI

 Useful protocol of whole-body MRI (Fig. 9)

 Coronal STIR of total body in 4 to 5 stations

 Sagittal STIR of entire spine

 Sagittal STIR of feet

 Axial STIR of pelvis and knee

 Advantages

 Whole-body screening

 Most sensitive in detecting inflamed sites

 Assesses both disease activity and skeletal damage

 Disadvantages

 Requires sedation in young children

Frequency of imaging monitoring

There is no consensus on how often imaging should be performed to monitor disease

activity of CNO. In North America, about half of the surveyed pediatric rheumatologists

use imaging regularly, with 54% of them repeating imaging every 6 months and 25%

every 12 months.

Because of the occult nature of the disease and lack of reliable

biomarkers or physical findings, more frequent radiation-free imaging may be neces-

sary to provide an accurate estimate of disease activity in order to guide the treatment.

In general, early in the disease, repeating MRI 3 to 6 months after initiation of the treat-

ment is reasonable. In patients who have a completely normal MRI and remain on sta-

ble medications or are off medications, repeating MRI every 6 to 12 months is

appropriate.

Fig. 7. (A) Normal bone scintigraphy of a child. (B) Abnormal STIR signal (arrow) on MRI

within left distal tibia of the same child. (C) Abnormal STIR signal (arrow) on MRI within

right distal fibula of the same child.

Chronic Nonbacterial Osteomyelitis

793

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

Laboratory Evaluation

There is no diagnostic test for CNO. Most children have a normal complete blood

cell count (CBC) at presentation.

5–7

Because of the broad differential diagnoses

for CNO, laboratory tests are most useful to rule out alternative diagnoses. In-

flammatory markers, such as ESR and CRP, are elevated in some patients with

CNO, but most have normal levels at presentation.

4,7,8,13,18,19

The use of urinary

N-terminal telope ptide (NTx) as a disease-monitoring tool in CNO has been re-

ported in a small cohort,

with the test found able to identify some children

with disease flares. More study is needed to determine the normal range of

NTx in healthy children and evaluate the sensitivity a nd specificity of NTx for

screening children for CNO. HLA-B27 positivity and low titer antinuclear antibody

positivity have been reported in only a small fraction of these patients.

5,7

Serum

TNF-a was increased in 66% of patients in one study.

Total immunoglobulin (Ig)

G and its subclasses, IgG1, IgG2, IgG3, as well as IgD were elevated in 7% to

33% in one cohort.

Fig. 8. ( A) Normal leg radiograph of a child. (B) Abnormal STIR signal (arrow) on MRI within

distal tibia and fibula of the same child.

Zhao & Ferguson

794

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

Initial screening laboratory tests in a child with typical chronic nonbacterial

osteomyelitis usually show

 Normal CBC; sometimes anemia from chronic disease

 Normal or mildly elevated ESR, CRP

 Normal LDH, uric acid

 Normal serum calcium, phosphorus, alkaline phosphatase

Bone Biopsy

It may be necessary to obtain a bone specimen to exclude infection and malignancy.

Most rheumatologists request a bone biopsy when there are constitutional symptoms

or a single bone lesion or an atypical presentation. Conversely, patients with the

following characteristics may not need a bone biopsy because of the higher confi-

dence in the likelihood of CNO:

 Bone lesion at a typical site (clavicle, metaphysis of long bones, vertebral body)

with normal laboratory test results and no constitutional symptoms

 Multiple bone lesions

 CNO-associated conditions (ie, psoriasis or Crohn disease)

4,17

When indicated, bone biopsy can be obtained via an open biopsy or needle biopsy

depending on the affected sites. Usually a decision of how to obtain biopsy is left to

Fig. 9. Whole body MRI of a child with CNO includes coronal STIR of total body in 4 to 5 sta-

tions (A)withstitchedimage(B), sagittal STIR of entire spine (C), axial STIR of pelvis (D),

sagittal STIR of feet (E), and axial STIR of the knee (F). It demonstrates active inflammation

in right proximal tibia and left talus.

Chronic Nonbacterial Osteomyelitis

795

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

the discretion of the surgeon. Common histologic findings include acute and/or

chronic inflammation, marrow fibrosis, osteonecrosis or normal bone.

6,7,17,21,22

specific staining can confirm the diagnosis of CNO but may exclude other diagnoses.

Adequate sample should be obtained to allow for pathologic review as well as for an

infectious workup including stains and cultures for bacteria, fungi and mycobacteria.

PATHOGENESIS OF CHRONIC NONBACTERIAL OSTEOMYELITIS

CNO is considered an autoinflammatory disease. Components of the innate immune

system, including neutrophils, macrophages, monocytes, and associated cytokines,

contribute to disease pathogenesis. Increased proinflammatory cytokines, such as

TNF-a and interleukin (IL)-6, and decreased antiinflammatory cytokines, especially

IL-10, were reported in children with CNO.

IL-6 and C-C motif chemokine 11/eotaxin

have been shown to sufficiently differentiate patients with CNO from healthy children

and those with other inflammatory diseases in a German cohort.

Recently, a serum

cytokine profile has been proposed as a marker for CNO.

Persistently elevated mono-

cyte chemoattractant protein-1, IL-12, and soluble interleukin-2 receptor were associ-

ated with refractory CNO in one cohort; but their use as biomarkers needs to be

validated.

Imbalanced cytokines cause increased osteoclast activity that results in

accelerated bone breakdown during the early phase of disease. Overall bone turnover

is increased, and excessive bone formation occurs as a response during the later stage.

There is evidence of a genetic component to CNO, including reports of clustering in

some families and a reported association of CNO with a rare allele of marker D18S60

on chromosome 18.

For a small minority of patients, CNO may be genetically driven.

There are 2 monogenic autoinflammatory bone diseases called deficiency of IL-1 re-

ceptor antagonist (DIRA) and Majeed syndrome in which sterile osteomyelitis is a

prominent phenotype. These two diseases have distinct clinical features, which can

aid in diagnosis. DIRA presents as neonatal onset of sterile multifocal osteomyelitis,

periostitis, and pustulosis. Clinical symptoms typically start within the first month of

life.

DIRA is due to recessive mutations in IL1RN, which encodes the IL-1 receptor

antagonist, a regulatory protein that binds IL-1 receptors on cells and blocks the bind-

ing and activity of IL-1. Affected children respond only partially to glucocorticoids and

disease-modifying antirheumatic drugs (DMARDs) but respond very well to replace-

ment treatment with recombinant human IL-1 receptor antagonist, anakinra.

Majeed

syndrome presents with sterile multifocal osteomyelitis with onset in the first 2 years of

life and is associated with congenital dyserythropoietic anemia; a few patients with

Majeed syndrome also have a neutrophilic dermatosis resembling Sweet syn-

drome.

28,29

Majeed syndrome is due to recessive mutations in LPIN2 . Recently a ho-

mozygous missense mutation in the gene FBLIM1 was reported by Cox and

colleagues

as the cause of CNO and psoriasis in 2 unrelated children.

HOW CHRONIC NONBACTERIAL OSTEOMYELITIS IS TREATED

Nonsteroidal antiinflammatory drugs (NSAIDs) are often used as the first-line treatment of

children with CNO.

8,17,31

Naproxen at 10 mg/kg (maximum 500 mg) twice daily is the

most commonly used NSAID. Other NSAIDs used for treatment are indomethacin and

meloxicam. Based on the study of Beck and colleagues,

responders have significant

pain relief and a decrease in the number of bone lesions on MRI by as early as 3 months.

Patients who have persistent bone pain and hyperintense signal within bone marrow on

STIR imaging after 3 months of NSAID treatment are considered NSAID treatment fail-

ures. These patients need treatment with a second-line agent.

31,32

Zhao & Ferguson

796

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

Second-line treatments include methotrexate, TNF inhibitors (TNFi; most commonly

monoclonal antibodies), and bisphosphonates.

5–8,17,19,20,22,32–34

Depending on the

severity of disease, one or more of the aforementioned medications may be used

sequentially or concurrently after a child fails to achieve a favorable response to

NSAIDs. Comparative effectiveness studies have not been done to determine relative

efficacy. Retrospective studies suggest that nonbiological DMARDs (such as metho-

trexate or sulfasalazine) have lower efficacy than TNFi and bisphosphonates.

When

other associated conditions, such as IBD or enthesitis-related arthritis, are present,

TNFi with or without DMARDs are more appropriate to treat both CNO and

these coexisting conditions. When spinal lesions are present, bisphosphonates may

prevent further compression and allow some recovery of vertebral body height.

18,35

Combining TNFi and bisphosphonates has been reported to provide substantial dis-

ease control in children with CNO and a poor response to NSAIDs.

There have been limited reports of the use of other biological medications. The IL-1 in-

hibitor anakinra has beenreported to be effective in 2 small cohortsof patients, associated

with a decrease in the CRP after 6 to 8 months of treatment.

36,37

Pardeo and colleagues

reported a favorable response in 5 of 9 patients who had refractory CNO (failing NSAIDs,

glucocorticoids, pamidronate). After 6 months of anakinra at a median dosage of

2 mg/kg/d, 5 of 9 patients had normalized CRP and ESR. The total number of bone lesions

detected by bone scintigraphy at baseline decreased from 77 to 35, although 20 new

asymptomatic lesions were identified. At a median of 1.7 years of follow-up, 6 of 9 patients

maintained a zero or minimal physician global assessment score.

Of note, glucocorticoids are not recommended during the initial workup or as long-

term treatment due to adverse long-term side effects and the potential for harm should

the child turn out to have a lymphoid malignancy. Short-term use (up to 6 weeks) in

patients with a well-established diagnosis of CNO as a transitional medication may

be appropriate.

Without treatment guidelines, there is variability in which medications are used for

NSAID failures. The selection and dosing of second-line medications differs among

rheumatologists. However, consensus treatment plans (CTPs) have been developed

based on the best available evidence and current treatment practices of North Amer-

ican pediatric rheumatologists for the treatment of pediatric CNO refractory to NSAIDs

and/or with active spinal lesions.

These CTPs will allow future comparative effective-

ness studies to identify the most effective therapies.

In a CNO cohort, aggressive combination treatment with a TNFi plus methotrexate

with or without bisphosphonate induced clinical remission along with significantly

decreased active lesions on MRI compared to treatment with NSAIDs.

Further study

is needed to determine if early aggressive treatment of children with CNO results in

better long-term outcomes

and to determine the optimal treatment duration.

CLINICAL MONITORING

Different criteria have been reported to define clinical responses to treatments in CNO.

Most of these criteria included 3 main components: pain due to CNO, inflammatory

markers (ESR and CRP), and imaging findings. Effective treatments may lead to com-

plete resolution of pain and normalized ESR and CRP, which may precede the com-

plete resolution of lesions on MRI. A composite score, PedsCNO score, includes ESR,

number of lesions on MRI, physician global assessment of disease activity, patient/

parent global assessment of disease activity, and childhood health assessment ques-

tionnaire.

PedsCNO 30, 50, and 70 were defined as 30%, 50%, and 70% improve-

ment in at least 3 of 5 variables with no more than one of the remaining variables

Chronic Nonbacterial Osteomyelitis

797

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

deteriorating by more than 30%, 50%, and 70%. Recently, proposed criteria of treat-

ment failure were suggested

as no improvement in more than 50% of available

criteria or at least 4 of the following 6 criteria: patient pain; total number of clinically

active lesions; total number of radiological lesions by whole body MRI or bone scintig-

raphy; size and degree of marrow edema of CNO lesions; and/or presence of soft tis-

sue swelling/inflammation related to CNO lesion on imaging, physician global

assessment, abnormal ESR, and/or CRP after exclusion of other potential causes.

MANAGEMENT OF CLINICALLY ASYMPTOMATIC LESIONS SHOWN IN MRI

Asymptomatic lesions have often been reported, and their clinical significance has not

been fully elucidated.

8,14,37

Currently, most physicians do not make clinical decisions

based on asymptomatic lesions unless vertebral bodies are affected, which potentially

poses a high risk of spinal fracture.

WHAT IS THE PROGNOSIS FOR CHRONIC NONBACTERIAL OSTEOMYELITIS?

Long-term observational studies have reported an average 40% rate of clinical remis-

sion (defined as absence of bone pain) in children with CNO after 1 to 5 years of follow-

up.

5–8,34

The only long-term follow-up study done on adult patients with childhood

onset of CNO showed a persistent presence of active bone lesions, defined as

increased signal intensity on STIR images, in 10 of 17 patients who were available

for a median of 15-year follow-up.

Among these 10 patients, 6 had ongoing clinical

symptoms, whereas the other 4 patients were completely asymptomatic. This study

underscores the importance of regular imaging monitoring in patients with CNO and

demonstrates that, for some, the disease may persist into adulthood. Pediatric rheu-

matologists manage these patients in collaboration with other specialists (eg, orthope-

dic surgeon, oral surgeon, neurosurgeon, dermatologist) and assist these patients in

their transition into adult rheumatologists for the optimal care.

Recurrence of disease is very common, with 50% of patients relapsing after a me-

dian of 2.4 years in a large German cohort study.

The relapse rate was even higher

(83%) in a large cohort from North America after a median of 1.8 years of follow-up.

In a large cohort study, factors associated with severe disease and a poor outcome

include male sex, multifocal disease, extrarheumatologic manifestations, family his-

tory of associated disease, and CRP greater than 1 mg/dL.

These patients were

more likely to receive bisphosphonates and/or TNFi as a result of refractory disease.

REFERENCES

1. Giedion A, Holthusen W, Masel LF, et al. Subacute and chronic “symmetrical”

osteomyelitis. Ann Radiol (Paris) 1972;15:329–42.

2. Jansson AF, Grote V. Nonbacterial osteitis in children: data of a German Inci-

dence Surveillance Study. Acta Paediatr 2011;100:1150–7.

3. Schnabel A, Range U, Hahn G, et al. Unexpectedly high incidences of chronic

non-bacterial as compared to bacterial osteomyelitis in children. Rheumatol Int

2016;36:1737–45.

4. Roderick MR, Shah R, Rogers V, et al. Chronic recurrent multifocal osteomyelitis

(CRMO) – advancing the diagnosis. Pediatr Rheumatol 2016;14(1):47.

5. Wipff J, Costantino F, Lemelle I, et al. A large national cohort of French patients

with chronic recurrent multifocal osteitis. Arthritis Rheumatol 2015;67:1128–37.

6. Borzutzky A, Stern S, Reiff A, et al. Pediatric chronic nonbacterial osteomyelitis.

Pediatrics 2012;130:e1190–7.

Zhao & Ferguson

798

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

7. 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 2007;46:154–60.

8. Beck C, Morbach H, Beer M, et al. Chronic nonbacterial osteomyelitis in child-

hood: prospective follow-up during the first year of anti-inflammatory treatment.

Arthritis Res Ther 2010;12:R74.

9. Oliver M, Lee T, Halpern-Felsher B, et al. Disease burden and social impact of

chronic nonbacterial osteomyelitis on affected children and young adults [ab-

stract#82]. Arthritis Rheumatol 2017;69(suppl 4).

10. Silier CCG, Greschik J, Gesell S, et al. Chronic non-bacterial osteitis from the pa-

tient perspective: a health services research through data collected from patient

conferences. BMJ Open 2017;7:e017599.

11. Zhao Y, Foster SK, Murdock TJ, et al. A rare case of chronic recurrent multifocal

osteomyelitis with undifferentiated juvenile idiopathic arthritis, uveitis, and psori-

asis. Case Rep Clin Med 2016;5. https://doi.org/10.4236/crcm.2016.57041.

12. Haque SA, Shad A, Ozdemirli M, et al. A thirteen year old female with primary

T-cell rich B-cell lymphoma of bone masquerading as chronic recurrent multifocal

osteomyelitis. Pediatr Rep 2009;1:e3.

13. Jansson AF, Muller TH, Gliera L, et al. Clinical score for nonbacterial osteitis in

children and adults. Arthritis Rheum 2009;60:1152–9.

14. Voit AM, Arnoldi AP, Douis H, et al. Whole-body magnetic resonance imaging in

chronic recurrent multifocal osteomyelitis: clinical long-term assessment may un-

derestimate activity. J Rheumatol 2015;42:1455–62.

15. Morbach H, Schneider P, Schwarz T, et al. Comparison of magnetic resonance

imaging and technetium-labelled methylene diphosphonate bone scintigraphy

in the initial assessment of chronic non-bacterial osteomyelitis of childhood and

adolescents. Clin Exp Rheumatol 2012;30:578–82.

16. Khanna G, Sato TSP, Ferguson P. Imaging of chronic recurrent multifocal. Radio-

graphics 2009;29:1159–77.

17. Zhao Y, Dedeoglu F, Ferguson PJ, et al. Physicians’ perspectives on the diag-

nosis and treatment of chronic nonbacterial osteomyelitis. Int J Rheumatol

2017. https://doi.org/10.1155/2017/7694942.

18. Hospach T, Langendoerfer M, von Kalle T, et al. Spinal involvement in chronic

recurrent multifocal osteomyelitis (CRMO) in childhood and effect of pamidro-

nate. Eur J Pediatr 2010;169:1105–11.

19. Zhao Y, Chauvin NA, Jaramillo D, et al. Aggressive therapy reduces disease ac-

tivity without skeletal damage progression in chronic nonbacterial osteomyelitis.

J Rheumatol 2015;42:1245–51.

20. Miettunen PM, Wei X, Kaura D, et al. Dramati c pain relief and resolution o f bone

inflammation following pamidronate in 9 pediatric patients wi th persistent

chronic recurrent multifocal osteomyelitis ( CRMO). Pediatr Rheumatol Online

J 2009;7:2.

21. Girschick HJ, Krauspe R, Tschammler A, et al. Chronic recurrent osteomyelitis

with clavicular involvement in children: diagnostic value of different imaging tech-

niques and therapy with non-steroidal anti-inflammatory drugs. Eur J Pediatr

1998;157:28–33.

22. Girschick HJ, Raab P, Surbaum S, et al. Chronic non-bacterial osteomyelitis in

children. Ann Rheum Dis 2005;64:279–85.

23. Hofmann SR, Morbach H, Schwarz T, et al. Attenuated TLR4/MAPK signaling in

monocytes from patients with CRMO results in impaired IL-10 expression. Clin

Immunol 2012;145:69–76.

Chronic Nonbacterial Osteomyelitis

799

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,

24. Hofmann SR, Bo

ttger F, Range U, et al. Serum interleukin-6 and CCL11/eotaxin

may be suitable biomarkers for the diagnosis of chronic nonbacterial osteomye-

litis. Front Pediatr 2017;5:1–11. Available at: http://journal.frontiersin.org/article/

10.3389/fped.2017.00256/full.

25. Hofmann SR, Kubasch AS, Range U, et al. Serum biomarkers for the diagnosis

and monitoring of chronic recurrent multifocal osteomyelitis (CRMO). Rheumatol

Int 2016;36:769–79.

26. Golla A, Jansson A, Ramser J, et al. Chronic recurrent multifocal osteomyelitis

(CRMO): evidence for a susceptibility gene located on chromosome 18q21. 3-

18q22. Eur J Hum Genet 2002;10:217–21.

27. Aksentijevich I, Masters SL, Ferguson PJ, et al. An autoinflammatory disease with

deficiency of the interleukin-1-receptor antagonist. N Engl J Med 2009;360:

2426–37.

28. 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;

42:551–7.

29. Majeed H, Kalaawi A, Mohanty M, et al. Congenital dyserythropoietic anemia and

chronic recurrent multifocal osteomyelitis in three related children and the asso-

ciation with Sweet syndrome in two siblings. J Pediatr 1989;115:730–4.

30. Cox AJ, Darbro BW, Laxer RM, et al. Correction: recessive coding and regulatory

mutations in FBLIM1 underlie the pathogenesis of chronic recurrent multifocal

osteomyelitis (CRMO). PLoS One 2017;12(7):e0169687. PLoS One 2017;12:1–22.

31. Hedrich CM, Hofmann SR, Pablik J, et al. Autoinflammatory bone disorders with

special focus on chronic recurrent multifocal osteomyelitis (CRMO ). Pediatr

Rheumatol 2013;11:47.

32. Zhao Y, Wu EY, Oliver MS, et al. Consensus treatment plans for chronic nonbac-

terial osteomyelitis refractory to nonsteroidal anti-inflammatory drugs and/or with

active spinal lesions. Arthritis Care Res (Hoboken) 2017. https://doi.org/10.1002/

acr.23462.

33. Simm P, Allen R, Zacharin M. Bisphosphonate treatment in chronic recurrent

multifocal osteomyelitis. J Pediatr 2008;152:571–5.

34. Schnabel A, Range U, Hahn G, et al. Treatment response and long-term out-

comes in children with chronic nonbacterial osteomyelitis. J Rheumatol 2017;

44(7):1058–65.

35. Gleeson H, Wiltshire E, Briody J, et al. Childhood chronic recurrent multifocal

osteomyelitis: pamidronate therapy decreases pain and improves ver tebral

shape. J Rheumatol 2008;35:707–12.

36. Wendling D, Prati C, Aubin F. Anakinra treatment of SAPHO syndrome: short-term

results of an open study. Ann Rheum Dis 2012;71:1098–100.

37. Pardeo M, Marafon DP, Messia V, et al. Anakinra in a cohort of children with

chronic nonbacterial osteomyelitis. J Rheumatol 2017;44:1231–8.

Zhao & Ferguson

800

Downloaded for Anonymous User (n/a) at Stony Brook University from ClinicalKey.com by Elsevier on April 06,