Chronic nonbacterial osteomyelitis (CNO)/chronic recurrent multifocal osteomyelitis (CRMO)

INTRODUCTION — Chronic nonbacterial osteomyelitis (CNO)/chronic recurrent multifocal osteomyelitis (CRMO) is a chronic, sterile, inflammatory disorder in children that primarily affects the bones and can cause bone destruction if untreated [1-4]. CNO can be sporadic or occur as part of a monogenic autoinflammatory condition such as deficiency of interleukin (IL) 1 receptor antagonist (DIRA) or Majeed syndrome. SAPHO syndrome (synovitis, acne, pustulosis, hyperostosis, and osteitis) is an adult disease with similar features to CNO. It is not clear if they are different diseases or represent the same condition in a spectrum with some age-specific factors [5].

The clinical features, diagnosis, management, and prognosis of sporadic CNO/CRMO are reviewed here. SAPHO is discussed separately, as are the monogenic autoinflammatory diseases. (See "SAPHO (synovitis, acne, pustulosis, hyperostosis, osteitis) syndrome" and "The autoinflammatory diseases: An overview".)

Osteomyelitis due to infection is also discussed in greater detail separately. (See "Hematogenous osteomyelitis in children: Epidemiology, pathogenesis, and microbiology" and "Hematogenous osteomyelitis in children: Clinical features and complications" and "Hematogenous osteomyelitis in children: Evaluation and diagnosis" and "Hematogenous osteomyelitis in children: Management".)

TERMINOLOGY — While both terms, "chronic nonbacterial osteomyelitis" (CNO) and "chronic recurrent multifocal osteomyelitis" (CRMO), are used to describe this disorder, CRMO is often perceived as the more severe form, and CNO is used as the general term for the condition.

EPIDEMIOLOGY — CNO is a rare disease (MIM #259680, ORPHA:324964), with slightly over 500 existing patients reported worldwide [6]. CNO is likely underdiagnosed and underreported due to lack of awareness of CNO in the general population and medical communities, absence of validated diagnostic criteria, and difficulty of collecting data in countries without national registries [7]. CNO is reported in all races, ethnicities, and geographic areas [8]. Reported incidence is higher in Western countries, especially in Europe. As an example, a nationwide survey conducted among treating clinicians through pediatric hospitals in Germany between 2006 and 2008 estimated the annual incidence of CNO at 1 in 250,000 [9]. However, this higher reported incidence may just reflect higher rates of diagnosis in such regions.

CNO can affect children of all ages, with a peak onset between 7 and 12 years [9,10]. There is female predominance, with a female-to-male ratio ranging from 1.5:1 to 4:1 in different series [9-14]. This ratio is in contrast to infectious conditions, which are more common in males. Even though CNO is seen in children of all ages, disease onset younger than age two years should trigger an investigation of a monogenic form of autoinflammatory conditions such as Majeed syndrome or deficiency of interleukin (IL) 1 receptor antagonist (DIRA). (See "The autoinflammatory diseases: An overview" and 'Differential diagnosis' below.)

GENETICS AND PATHOGENESIS — The exact pathogenesis of CNO/CRMO is unknown. Both genetic and environmental factors are thought to contribute to the underlying inflammation.

Genetics — The underlying molecular causes of CNO are not fully understood. Several monogenic autoinflammatory disorders causing sterile bone inflammation have been reported in humans and in animals [15-18], and sporadic CNO shares many clinical features with some monogenic autoinflammatory conditions [19]. However, in multiple studies, no single gene defects were identified as important contributing factors to the pathogenesis of CNO [20-22].

Genetic predisposition in a polygenic pattern is more likely in sporadic forms of CNO, especially when the bone inflammation is associated with other polygenic diseases such as inflammatory bowel disease (IBD) [23] and psoriasis. In addition, polygenic inheritance best explains the sibling and parent/child pairs, as well as the increased prevalence of associated conditions in other family members [10]. Human leukocyte antigen (HLA) B27 was reported as more common in CNO patients than the general population in two cohorts (21 versus 8 percent) [12,13], while no difference was noted in another study [24]. However, only some of the patients were tested for HLA-B27 in these studies.

Pathogenesis — The exact pathophysiology or mechanism of the sterile bone inflammation is poorly understood, although CNO is probably an osteoclast-mediated disease. In addition, it is thought of as an autoinflammatory disease associated with alteration of the gut microbiome, and an imbalance between pro- and anti-inflammatory cytokines is suspected to play a role [25]. Available data so far point to the interplay among genetics, environmental, and immunologic factors as the causes of CNO. Infectious etiology does not seem to play a crucial role in the pathogenesis of CNO.

Serum cytokine studies have shed light on the possible disease pathogenesis of CNO. Monocytes/macrophages in CNO are skewed towards releasing proinflammatory cytokines such as interleukin (IL) 6, tumor necrosis factor (TNF) alpha, and IL-20, while there is sparse production of anti-inflammatory molecules such as IL-10 and IL-19 [26,27]. Dysregulation of IL-10 and IL-1-beta appears to lead to bone inflammation and bone loss in mouse models and in humans with autoinflammatory syndromes such as deficiency of IL-1 receptor antagonist (DIRA) [28,29]. A decrease in IL-10, for example, causes activation of NLR family pyrin domain-containing 3 (NLRP3) inflammasome, which in turn increases IL-1-beta and induces osteoclastogenesis via the receptor activator of nuclear factor kappa-B (RANK) receptor activator of nuclear factor kappa-B ligand (RANKL) pathway.

In one study, monocyte-derived cytokines, IL-1RA, IL-2R, IL-6, IL-12, eotaxin, monocyte chemoattractant protein 1 (MCP-1), macrophage inflammatory protein 1-beta (MIP-1b), and RANTES were expressed significantly differently among children with CNO, Crohn disease, and healthy controls [30]. Similar cytokine profiles in children with juvenile idiopathic arthritis (JIA) and children with CNO suggest a partial common disease pathway. However, no autoantibodies or disease-specific T cell subsets have been identified in children with CNO.

An infectious etiology has been proposed, although antibiotic treatment, even prolonged courses, is ineffective in the majority of patients with CNO. Organisms found in some bone biopsy samples were subsequently felt to be contaminants [31]. However, some infectious burden/skewed microbiota have been implicated in triggering disease onset in Crohn disease [32], and a similar process may play a role in CNO. In a CRMO animal model, dietary changes were shown to alter the disease course [33]. A high-fat diet (HFD), which was accompanied with significant alterations in gut microbiome, protected the mice from bone inflammation [33]. Fecal transplant from HFD-fed mice protected the young mice fed with a regular diet from developing bone inflammation. However, there are no data available regarding diet and the microbiome in human CNO.

CLINICAL MANIFESTATIONS — The cardinal feature of CNO is bone pain of insidious onset. Affected children may have swelling over the painful area. There is usually focal tenderness with occasional warmth. CNO can be recurrent (episodic) or persistent (chronic). It can involve only one area (unifocal) or multiple bones (multifocal). The average number of lesions is three or four (range: two to six) [10,24,34].

CNO may affect any bone, but the metaphysis of the long bones of the lower extremities is the most common site, followed by the pelvis, vertebra, clavicle, long bones of the upper extremity, and mandible. Involvement of the skull is rare in CNO; therefore, an alternative diagnosis such as Langerhans cell histiocytosis (LCH) of bone should be sought in these patients (see 'Differential diagnosis' below). Isolated mandible disease (eg, Garré osteomyelitis, diffuse sclerosing osteomyelitis) is most likely in the spectrum of CNO [35].

In reports with long-term follow-up, most patients with unifocal presentation were found to have more lesions during further work-up (silent lesions) or have developed new lesions during the follow-up period. In one study, 54 percent of patients had unifocal disease at diagnosis, which dropped to 24 percent after further work-up and over time, and 29 percent of lesions were asymptomatic [7]. In a French study, there was progression from unifocal to multifocal disease over time. Only 7 percent remained unifocal after four years of follow-up [10]. Clavicle involvement was more common in the persistent unifocal group, and there were more males in the multifocal group.

ASSOCIATED CONDITIONS — CNO is associated with other conditions that primarily affect the skin, gut, or joints, such as psoriasis, palmoplantar pustulosis (PPP), acne, inflammatory bowel disease (IBD), spondyloarthropathy/enthesitis related arthritis (ERA), and, less commonly, pyoderma gangrenosum, Sweet syndrome, Takayasu arteritis, or granulomatosis with polyangiitis (GPA) [6,36,37]. PPP is the most common associated skin condition and usually worsens with CNO flares. PPP and psoriasis are included in one of the diagnostic criteria (table 1). Some patients have one or more of these extraosseous conditions at the time of diagnosis, whereas others develop them later in the disease course [11,12,38,39]. Rarely, other organs can be involved, such as the kidneys (eg, proteinuria) [40]. Arthritis usually develops in the adjacent joints, though it can also happen in remote joints. Pelvic region lesions are more likely to be associated with sacroiliac arthritis than spinal lesions are.

REFERRAL — Children with persistent chronic bone pain and concerning imaging findings (see 'Imaging' below) should be referred to a pediatric rheumatologist for further evaluation.

INITIAL TESTING — The initial evaluation of suspected CNO/CRMO involves obtaining laboratory studies (complete blood count [CBC] with differential, C-reactive protein [CRP], and erythrocyte sedimentation rate [ESR]); imaging; and, in many patients, bone biopsy (table 1).

Laboratory studies — A CBC, CRP, and ESR should be obtained in all children with suspected CNO to evaluate for infectious osteomyelitis (see 'Differential diagnosis' below). The majority of children with CNO have a normal CBC, CRP, and ESR [10,11,39,41], although some patients present with significantly elevated CRP and ESR [7,39]. There is no apparent correlation between the number or location of bone lesions and the laboratory findings.

Other screening labs are useful to exclude some differential diagnoses (see 'Differential diagnosis' below):

●Uric acid and lactate dehydrogenase (LDH) – Elevated in children with tumor lysis syndrome from leukemia or other malignancies [42,43].

●Alkaline phosphatase (ALP), as a bone formation marker – Decreased in children with hypophosphatasia [44,45].

●Blood cultures – Collect in children with suspected infectious osteomyelitis.

●Vitamin C level – Obtain in a child with an unbalanced diet to rule out scurvy [46].

Urinary N terminal telopeptide (NTx) was reported in a small cohort study as a potential biomarker to monitor disease activity [47]. Serum cytokines, particularly interleukin (IL) 6 and C-C motif chemokine ligand 11 (CCL11), were reported as biomarkers distinguishing CNO from other diseases [48]. Further validation is needed for these biomarkers before clinical application.

Imaging — Initial imaging of patients with suspected CNO always starts with a radiograph of the painful site [1,37,49]. However, plain radiography has poor sensitivity for detection of CNO compared with magnetic resonance imaging (MRI), with only 15 percent of children having abnormal radiographic findings in one cohort [50]. Thus, MRI should be performed when radiographs are normal. In addition, clinically asymptomatic lesions, also known as "silent" lesions, are often detected by MRI. Vertebral "silent" lesions may lead to compression fracture [51]. Thus, whole-body MRI (WBMRI) (image 1) remains the gold standard for diagnostic imaging and disease monitoring in CNO [50,52-55]. In MRI, fluid-sensitive sequences are most useful to accentuate the marrow changes in CNO. Either short tau inverse recovery (STIR) or fat saturation sequence is sufficient for initial screening. The optimal MRI protocol includes coronal STIR of the entire body, sagittal STIR of the entire spine, axial STIR of the pelvis and knee, and sagittal STIR of both ankles, all without contrast. However, contrast is necessary to determine the specific pattern if there is a high suspicion of infection or malignancy.

Other imaging modalities are not reasonable alternatives to MRI because of their poor sensitivity and risk of delaying diagnosis and corresponding treatment. Computed tomography (CT) has less value due to its lack of sensitivity in distinguishing active disease from chronic damage. Bone scintigraphy and positron emission tomography (PET)-CT are sometimes used to offer a whole-body level assessment in order to identify all lesions. However, both are radiation-based techniques, sensitivity of bone scintigraphy is lower than MRI [56], and it is more difficult to interpret bone scintigraphy results in children with growing bones due to the presence of physiologic uptake at the growth plates, where CNO lesions are commonly found.

Typical findings of CNO in each type of imaging are listed below:

●Radiographs (and CT)

•Lytic lesions (image 2)

•Sclerosis and cortical irregularity (image 3)

•Hyperostosis (image 4)

•Kyphosis

•Vertebral body height loss or plana (image 5)

•Growth plate bony bar (also called a physeal bar; a bony or fibrous bridge between the epiphysis and metaphysis that forms due to interruption of the normal growth plate cartilage)

●MRI, including WBMRI

•Increased intensity of STIR signal within bone marrow (image 6 and image 1 and image 7)

•Increased intensity of STIR signal within surrounding tissue (image 6 and image 1 and image 7)

•Bony expansion (image 8)

●Bone scintigraphy

•Increased uptake (image 9)

●PET-CT

•Increased uptake within a localized region

Biopsy of bone and other sites — Microbiologic and pathologic studies from bone biopsy are important because infectious osteomyelitis and malignancy are the two major categories of diseases with similar presentation (see 'Differential diagnosis' below). However, children with a classic presentation that includes a well appearance, normal laboratory findings, symmetrical lesions in typical sites such as the metaphysis of long bones, or associated conditions (psoriasis, palmoplantar pustulosis [PPP], inflammatory bowel disease [IBD], spondyloarthropathy/enthesitis related arthritis [ERA]) may be spared from a bone biopsy prior to treatment (table 1) [7,37,41]. In children with a unifocal lesion involving a bone other than the clavicle, even if it is in a typical location, a bone biopsy is recommended. A diaphyseal lesion is rare in CNO and more commonly seen in osteosarcoma, so a biopsy is necessary to exclude primary bone malignancies [57]. Bone marrow and/or lymph node biopsy are indicated when clinical findings suggest leukemia or lymphoma.

In most series, biopsy rates reach 70 to 80 percent, in some patients more than once. The Jansson scoring system (table 1) was tested retrospectively in one study, and it was found that 27 percent of biopsies could have been prevented or delayed [10]. In another study, 12 patients were referred to a rheumatology clinic before the biopsy, and nine of them did not need biopsy because multiple lesions were present (found after bone scan or MRI), there were siblings with CRMO, or the lesion was in the clavicle [7].

Sites of bone biopsy are often determined by surgeons and interventional radiologists based on the feasibility of accessing the lesion [37]. Biopsy of multiple sites may be performed to enhance the sensitivity of bone biopsy when indicated. Repeated biopsies may be required in patients with an unusual presentation or atypical clinical course. Special staining such as CD1a and S100 are commonly performed to rule out Langerhans cell histiocytosis (LCH).

Although there is no distinct diagnostic feature in bone pathology for CNO, common findings include:

●Destruction of normal bone structure and presence of neutrophils, monocytes, lymphocytes, and plasma cells during early phases [1]

●A predominance of fibrosis and sparse immune cells during later phases [1]

A mixture of fibrosis and inflammation, as well as normal bone, is also often reported [37].

DIAGNOSIS — CNO/CRMO should be suspected in any child with insidious-onset bone pain based upon clinical criteria. The Jansson [41] and Bristol [7] criteria, which include typical features of CNO (table 1), are most commonly used.

Biopsy of the lesion is generally needed to make the diagnosis (see 'Initial testing' above). The diagnosis remains a "diagnosis of exclusion" because of the nonspecific features in every aspect of CNO (clinical, laboratory, imaging, and pathology). The lack of specific features also usually results in a delay in the diagnosis of CNO, with a median time to diagnosis of approximately 15 months in many series [7,58]. Patients, particularly those with solitary lesions, may be misdiagnosed as having bacterial osteomyelitis and treated with long-term antibiotics or undergo unnecessary procedures even after culture-negative bone biopsies since pathologic findings of CNO are indistinguishable from infectious osteomyelitis [59].

DIFFERENTIAL DIAGNOSIS — The most common alternative diagnoses include infectious osteomyelitis and malignancy.

●Leukemia – Leukemia (most commonly acute lymphoblastic leukemia/lymphoblastic lymphoma [ALL/ABL], but acute myeloid leukemia [AML] also possible) may present as bone pain with abnormal short tau inverse recovery (STIR) signal within bone marrow on magnetic resonance imaging (MRI). Plain radiographs often show metaphyseal bands and a periosteal reaction. Bone pain is mostly due to aseptic osteonecrosis, typically waking the child from sleep. The peak age of onset is two to five years, and it mostly affects males. Greater than 60 percent children with leukemia have hepatosplenomegaly, and 50 percent have lymphadenopathy. Fever, weight loss, and pallor are common. Complete blood count (CBC) with differential may show blasts and/or cytopenia. Thrombocytopenia is present in 75 percent of cases. Bone marrow biopsy confirms the diagnosis. (See "Overview of the clinical presentation and diagnosis of acute lymphoblastic leukemia/lymphoma in children".)

●Hodgkin lymphoma (HL) – HL commonly affects children aged between 15 and 19 years. Systemic complaints such as fatigue and weight loss are common in addition to chronic bone pain with similar imaging findings as in CNO. Lymphadenopathy and mediastinal mass may prompt lymph node biopsy for final confirmation of lymphoma. Cases of lymphoma misdiagnosed as CNO have been reported [60]. (See "Overview of Hodgkin lymphoma in children and adolescents".)

●Langerhans cell histiocytosis (LCH) of bone – LCH (eosinophilic granuloma) of bone most commonly occurs in children between 5 to 10 years of age. Most patients complain of pain in a localized area of bone, and examination usually reveals a raised, soft, tender spot. Radiographs may show osteolytic lesions as a "punch-out" pattern. Usually, only one bone is involved, most commonly the skull, and there is a single lesion. CD1a staining on bone biopsy or other affected tissue biopsy reveals the presence of histiocytosis. (See "Clinical manifestations, pathologic features, and diagnosis of Langerhans cell histiocytosis", section on 'Lytic bone lesions'.)

●Non-LCH histiocytosis – Erdheim-Chester disease (ECD) and Rosai-Dorfman disease (RDD) may mimic CNO. ECD is rare in children and is associated with arginine vasopressin deficiency (previously called central diabetes insipidus). Bone involvement in ECD most commonly manifests as mild but persistent juxta-articular pain, particularly in the lower extremities. Bilateral and symmetric osteosclerosis of the diaphysis of the long bones is nearly universal. RDD is a macrophage-related disease that often affects lymph nodes but can also result in lytic bone lesions. Histology from bone biopsy, lymph nodes, or other affected areas is most useful in making the diagnosis. Genetic testing from bone biopsy can confirm ECD. (See "Erdheim-Chester disease" and "Peripheral lymphadenopathy in children: Etiology", section on 'Rosai-Dorfman disease'.)

●Primary malignant bone disease – Primary bone malignancies include osteosarcoma, chondrosarcoma, Ewing sarcoma, and primary non-Hodgkin lymphoma of bone (PLB). Clinical symptoms of these diseases may mimic CNO, but MRI findings usually show distinct patterns with extensive bone destruction in all except PLB. Radiographs may show disorganized periosteal reaction, new bone formation, cortical bone destruction, or soft tissue mass. Osteosarcoma and Ewing sarcoma commonly affect diaphysis or metadiaphysis of the long bones. (See "Osteosarcoma: Epidemiology, pathology, clinical presentation, and diagnosis" and "Chondrosarcoma" and "Clinical presentation, staging, and prognostic factors of Ewing sarcoma".)

●Benign bone tumor – Benign tumors of the bone include osteoid osteoma, enchondroma, osteoblastoma and chondroblastoma, and cystic bone tumors. The majority have distinct features on radiographs and can be diagnosed with such. (See "Nonmalignant bone lesions in children and adolescents".)

●Chronic infectious osteomyelitis – Hematogenous osteomyelitis may present with focal bone pain and limitation of function. Constitutional symptoms and elevation of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are common. Fever may or may not be present. Blood and bone cultures are positive in >80 percent of the cases. Radiographs may show lytic lesions or periosteal reactions. MRI findings may show subperiosteal or soft tissue collections of pus. (See "Hematogenous osteomyelitis in children: Evaluation and diagnosis".)

●Avascular necrosis (AVN) – Osteonecrosis may affect the long bones at the metaphysis and/or epiphysis and present with focal pain worsening after weight bearing. MRI shows a single-density line separating normal and ischemic bone on T1-weighted images. Risk factors such as chronic glucocorticoid use often exist in these patients. (See "Treatment of nontraumatic hip osteonecrosis (avascular necrosis of the femoral head) in adults".)

●Vitamin C deficiency – Vitamin C deficiency is commonly seen in picky eaters and children with malnutrition. Common manifestations are gingivitis, bruising, and musculoskeletal pain. Characteristic findings on MRI are sclerotic and lucent metaphyseal bands, with periosteal reaction and adjacent soft-tissue edema. Vitamin C level is low in these children. (See "Overview of water-soluble vitamins", section on 'Vitamin C (ascorbic acid)'.)

●Chronic arthritis – Psoriatic arthritis, inflammatory bowel disease (IBD) associated arthritis, and enthesitis related arthritis (ERA) may present with bone pain adjacent to the affected joints. Inflammatory bony changes may be seen in the sacrum or ilium adjacent to the joint when the sacroiliac joint is affected in ERA. Children with psoriatic arthritis and psoriasis may develop coexisting CNO. Dactylitis and signs of chronic arthritis are key features. (See "Psoriatic juvenile idiopathic arthritis: Pathogenesis, clinical manifestations, and diagnosis" and "Spondyloarthritis in children" and "Clinical manifestations and diagnosis of arthritis associated with inflammatory bowel disease and other gastrointestinal diseases".)

●Amplified pain syndromes (AMPS) including complex regional pain syndrome (CRPS) – AMPS are often associated with chronic diseases and may coexist with CNO. Typically, there is allodynia (pain to touch) of the involved body part(s), with or without focal swelling, discoloration, or temperature changes. MRI can show bony edema and small joint effusions, which helps distinguish AMPS/CRPS from CNO in the appropriate clinical setting. (See "Complex regional pain syndrome in children".)

●Hypophosphatasia – Hypophosphatasia is a hereditary metabolic bone disease caused by pathogenic variants of the tissue-nonspecific alkaline phosphatase (TNSALP) gene on chromosome 1p36.1. Reduced enzyme activity causes IBD that mimics CNO. MRI findings are identical [44,45,61]. Low alkaline phosphatase, elevated phosphorus, elevated pyridoxal-50-phosphate and phosphoethanolamine, and genetic testing are diagnostic.

MANAGEMENT

●Initial therapy – The initial therapy for most children with CNO is daily nonsteroidal antiinflammatory drugs (NSAIDs) at standard doses [1,12,37,55,62,63]. Up to 50 percent of patients achieve clinical remission within the first year of starting an NSAID, but relapse is common, and new lesions may develop. For those who respond, there is no clear guideline on total duration of NSAID treatment. Most clinicians trial patients off NSAIDs if there are no clinical symptoms and magnetic resonance imaging (MRI) shows no or minimum disease activity after one to two years of treatment. Close follow-up is critical in order to capture children with recurrence of the disease. Children with episodic flares may only require retreatment with NSAIDs for a few months. However, children with frequent flares may require longer treatment with NSAIDs or additional medications.

In a prospective, observational study of 37 children and adolescents with CNO, 43 percent (16 of 37) had achieved clinical remission after six months of naproxen treatment [63]. After 12 months of therapy, 62 percent were in clinical remission, no patient reported morning stiffness or functional impairment of the legs, only five patients (14 percent) noted local swelling, and disease activity and severity scores had decreased. However, only 14 percent had complete resolution of MRI findings by 6 months and 27 percent by 12 months. In addition, 41 percent had radiologic relapses, although 67 percent were asymptomatic. In contrast, a retrospective series of 56 patients reported clinical remission in all patients after three months of NSAID therapy, although 50 percent relapsed after a median of 2.4 years [12]. In a small cohort study (n = 9) that used MRI scoring as the main outcome measurement, children with CNO showed no improvement in either the total number of lesions or the maximum bone edema score from MRI after a median of 7.3 months on NSAIDs, but soft tissue inflammation and periosteal reaction significantly decreased [62].

●Therapy for patients with spinal lesions – In children with active spinal lesions, infliximab with or without bisphosphonate [62] and bisphosphonate alone [64,65] in addition to NSAIDs at the start of therapy showed favorable outcomes in small cohorts.

●Refractory disease – In children with CNO who have persistently active symptoms and abnormal MRI findings despite a four-to-six-week trial of NSAID therapy, additional treatments including disease-modifying antirheumatic drugs (DMARDs), tumor necrosis factor (TNF) inhibitors, and/or bisphosphonates are necessary to induce remission and reduce skeletal damage [1,66]. There are no head-to-head comparisons of efficacy of these treatments. However, retrospective studies suggest that TNF inhibitors and bisphosphonates have better response rates than DMARDs and glucocorticoids [10-12].

●Monitoring – Disease activity monitoring includes regular imaging (preferably MRI) of discrete lesions, particularly symptomatic ones, and physical assessment by a pediatric rheumatologist in order to determine treatment options and duration.

●Supportive care – Management of CNO also includes supportive care such as appropriate physical and occupation therapy, as well as periodic laboratory screening for medication toxicity.

PROGNOSIS — The disease course can be variable, from mild, intermittent episodes to an unrelenting course with uncontrolled inflammation causing significant morbidity including bone deformities, fractures, growth failure, arthritis, neurologic symptoms (particularly from spinal involvement), refractory and persistent pain syndromes, and psychosocial dysfunction resulting in poor quality of life (QOL) [10,52,62,64,67]. Leg-length discrepancy and angulation secondary to growth plate damage also may occur. Early diagnosis and treatment may prevent or decrease complications of the disease.

In a small, longitudinal, follow-up study, the outcomes of 17 children with CNO varied from complete remission to persistent, active inflammation revealed by magnetic resonance imaging (MRI) even after a median of 15 years [52]. In the largest cohort (n = 178), after a median of four years, only 43 percent of patients achieved remission, while 40 percent (27 of 67) remained on therapy [10]. Among those with active disease, 72 percent (70 of 97) were continuing treatment. These results underscore the importance of close monitoring by a pediatric rheumatologist during childhood and continuing follow-up with a rheumatologist into adulthood.

Factors associated with poorer prognosis (disease that is more difficult to control, with lower remission rates and higher usage of tumor necrosis factor [TNF] inhibitors and/or bisphosphonate, and increased risk of arthritis) include [6,12]:

●Multifocal bone pain, particularly involvement of lower-extremity bones or spine; also vertebral fractures

●Longstanding, untreated inflammation and higher inflammatory markers

●Male sex

●Extraosseous manifestations including arthritis, as well as acne, inflammatory bowel disease (IBD), and palmoplantar pustulosis (PPP)

RESOURCES — Clinicians and family representatives have aimed to increase the awareness of CNO in the general and medical communities through online resources and caregiver handouts.

SUMMARY AND RECOMMENDATIONS

●Terminology – Chronic nonbacterial osteomyelitis (CNO)/chronic recurrent multifocal osteomyelitis (CRMO) is a chronic inflammatory bone disorder affecting primarily children. Both terms are used, although CRMO is often perceived as the more severe form, and CNO is used as the general term for the condition. (See 'Introduction' above and 'Terminology' above.)

●Epidemiology and pathogenesis – CNO has been reported worldwide with female predominance. It is a rare disease, but lack of awareness of the condition has contributed to underdiagnosis and reporting. The exact pathogenesis is unknown. Both genetic and environmental factors are thought to contribute to the underlying inflammation. (See 'Epidemiology' above and 'Genetics and pathogenesis' above.)

●Clinical manifestation – The cardinal feature of CNO is bone pain of insidious onset. CNO is also associated with other conditions such as psoriasis, palmoplantar pustulosis (PPP), inflammatory bowel disease (IBD), and spondyloarthropathy. (See 'Clinical manifestations' above and 'Associated conditions' above.)

●Initial testing – The diagnosis of suspected CNO/CRMO involves obtaining laboratory studies (complete blood count [CBC], C-reactive protein [CRP], and/or erythrocyte sedimentation rate [ESR]); imaging; and, in many patients, bone biopsy. Whole-body magnetic resonance imaging (WBMRI) is the gold-standard imaging modality for diagnostic and staging purposes. (See 'Initial testing' above.)

●Diagnosis – CNO/CRMO should be suspected in any child with insidious-onset bone pain. Two commonly used diagnostic criteria, Jansson and Bristol, that include typical features of CNO are shown in the table (table 1). Biopsy of the lesion is generally needed to make the diagnosis. The lack of features specific to CNO usually results in a delay in the diagnosis. (See 'Diagnosis' above.)

●Differential diagnosis – The differential diagnosis includes infectious osteomyelitis, Langerhans cell histiocytosis (LCH), and malignancies that involve the bone (eg, leukemia, lymphoma, osteosarcoma). (See 'Differential diagnosis' above.)

●Management – Nonsteroidal antiinflammatory drugs (NSAIDs) are the first line of treatment for most patients. Additional treatment options for patients with active spinal lesions or in those who have persistently active symptoms and abnormal magnetic resonance imaging (MRI) findings despite NSAID therapy include tumor necrosis factor (TNF) inhibitors, bisphosphonates, and disease-modifying antirheumatic drugs (DMARDs). Patients are monitored with focal MRI studies and by physical assessment. (See 'Management' above.)

●Prognosis – The disease course can be variable, from mild, intermittent episodes to an unrelenting course with uncontrolled inflammation causing significant morbidity. Early diagnosis and treatment may prevent or decrease complications of the disease. (See 'Prognosis' above.)