Chronic Recurrent Multifocal Osteomyelitis (CRMO) / Chronic Nonbacterial Osteomyelitis (CNO): Comprehensive Research Report

1. Disease Information

Overview. Chronic recurrent multifocal osteomyelitis (CRMO) — increasingly referred to in the broader literature as chronic nonbacterial osteomyelitis (CNO), of which CRMO represents the recurrent/multifocal end of the spectrum — is a rare, non-infectious autoinflammatory bone disease that causes sterile inflammatory bone lesions, predominantly in children and adolescents. It mimics infectious (bacterial) osteomyelitis clinically and radiographically but yields negative cultures and does not respond to antibiotics. CRMO is classified as an autoinflammatory disorder of the innate immune system rather than an autoimmune disease, since it lacks high-titer autoantibodies or clear autoreactive lymphocyte involvement (PMC10483435).

Key identifiers: - ICD-10-CM: M86.3 (Chronic multifocal osteomyelitis), M86.30 (AAPC) - Orphanet: ORPHA:324964 (also historically ORPHA:169147 for the broader syndrome group) - Disease Ontology: DOID:0060645 - MONDO: MONDO:0958177 (non-syndromic CRMO); syndromic monogenic forms carry separate MONDO/OMIM entries - OMIM (monogenic subtypes): - CRMO2 (with periostitis and pustulosis) — OMIM #612852 - CRMO3 — OMIM #259680 (autosomal dominant form) - Majeed syndrome (CRMO + congenital dyserythropoietic anemia) — OMIM #609628, caused by LPIN2 - DIRA (deficiency of IL-1 receptor antagonist), a related monogenic autoinflammatory bone disease — caused by IL1RN

Synonyms: Chronic nonbacterial osteomyelitis (CNO), chronic nonbacterial osteitis, nonbacterial osteitis (NBO), sterile multifocal osteomyelitis; the adult-spectrum manifestation overlaps with SAPHO syndrome (Synovitis, Acne, Pustulosis, Hyperostosis, Osteitis), which is now widely regarded as part of the same disease continuum in adults (Oxford Rheumatology Advances in Practice).

Evidence base: Because CRMO is rare, most disease-level knowledge derives from aggregated resources — multicenter retrospective cohorts, international registries (e.g., Eurofever), and national prospective surveillance studies (e.g., the UK/Republic of Ireland British Paediatric Surveillance Unit [BPSU] study) — rather than single-patient EHR mining, supplemented by murine genetic models.

2. Etiology

Disease Causal Factors

CRMO is understood as resulting from dysregulated innate immune/monocyte signaling causing an imbalance between pro- and anti-inflammatory cytokines, culminating in NLRP3 inflammasome hyperactivation, IL-1β-driven inflammation, and pathological osteoclastogenesis. Three converging mechanisms have been proposed: (1) imbalanced cytokine expression, (2) increased inflammasome activation, and (3) enhanced osteoclast differentiation (PMC10483435; PMC5705736).

Genetic Risk Factors

Monogenic syndromic forms (Mendelian, high-penetrance): - Majeed syndrome — autosomal recessive, caused by loss-of-function mutations in LPIN2 (encoding lipin-2, a phosphatidic acid phosphatase); presents with CRMO, congenital dyserythropoietic anemia, and often neutrophilic dermatosis. Only ~24 individuals from 10 families with genetically confirmed disease reported to date (PMC8252456; OMIM #609628). Novel LPIN2 mutations have been shown to link bone inflammation to inflammatory M2 macrophages and accelerated osteoclastogenesis (PMID:33314777). - DIRA (Deficiency of IL-1 Receptor Antagonist) — autosomal recessive, caused by loss-of-function mutations in IL1RN, producing unopposed IL-1 signaling; presents with neonatal-onset pustulosis, marked inflammatory markers, sterile multifocal osteitis, and periostitis. - PAPA syndrome — caused by mutations in PSTPIP1, driving pyrin inflammasome activation and elevated IL-1β; shares mechanistic overlap with CRMO. - CRMO3 (OMIM #259680) — autosomal dominant form, early childhood-onset bone pain/arthritis from sterile osteomyelitis. - CRMO2 (OMIM #612852) — CRMO with periostitis and pustulosis.

Non-syndromic (sporadic) CRMO susceptibility genes: - FBLIM1 (filamin-binding LIM protein 1) — recessive coding and regulatory mutations identified via whole-exome sequencing in a consanguineous family (South Asian ancestry); Fblim1 is the most differentially expressed gene (>20-fold downregulated) in bone marrow macrophages of the murine cmo (chronic multifocal osteomyelitis) model, implicating impaired RANKL regulation and osteoclast differentiation (PMID:28301468; PLOS ONE). - P2RX7 — a 2023–2024 targeted/exome sequencing study (n=190 CNO patients vs. 1,873 controls) found rare, presumably damaging P2RX7 variants in 5.8% of CNO patients vs. 1.9% of controls, and rare lower-impact variants in 32.4% vs. 4.4% of controls — the gene was "more than 10-fold more variable" among patients. P2X7 is an ATP-gated cation channel that mediates potassium efflux, a potent NLRP3 inflammasome trigger; CNO-associated variants alter inflammasome assembly and reduce pyroptosis, potentially prolonging monocyte survival and cytokine output. Patients with damaging P2RX7 variants had more GI symptoms and lymphadenopathy but less spinal/joint/skin (psoriasis) involvement (ScienceDirect 2024; Liverpool repository). - FGR (Src-family kinase) — expressed in mast cells/neutrophils, contributes to aseptic bone inflammation independently of NLRP3 (PMID:31138708). - IL-10 promoter haplotypes (GCC/ACC/ATA) modulate transcriptional output; most sporadic CRMO patients show enrichment for the lower-expressing haplotype pattern relative to controls, consistent with the impaired-IL-10 mechanistic model. - HLA-B27 positivity has been noted as a susceptibility marker in some cohorts, particularly overlapping spondyloarthropathy phenotypes.

Environmental/demographic risk factors: Female sex predominance in classic pediatric CRMO (though a "severe" phenotype subgroup skews male — see Phenotypes below); White/Caucasian ancestry is over-represented in published cohorts, though this likely reflects ascertainment bias rather than true differential susceptibility; family history of autoinflammatory/autoimmune disease (psoriasis, IBD, spondyloarthritis) increases risk.

Protective Factors

No well-established genetic or environmental protective factors are documented in the literature; this remains an evidence gap.

Gene–Environment Interactions

Not well characterized for CRMO specifically; the leading hypothesis is that in genetically susceptible individuals (e.g., carrying P2RX7 or IL-10 promoter risk variants), an as-yet-unidentified triggering stimulus (possibly microbial/microbiome-related, per murine data) precipitates monocyte/macrophage TLR4-MAPK signaling defects and downstream inflammasome activation.

3. Phenotypes

Core symptoms and signs

Bone pain — insidious onset, localized, often worse at night, with a waxing-and-waning course (suggest HP:0002653 Bone pain / HP:0006414)
Local swelling, tenderness, warmth over affected bone (suggest HP:0025378 or general inflammatory swelling terms)
Arthritis/synovitis adjacent to lesions
Bone overgrowth/hyperostosis, particularly clavicular and mandibular (suggest HP:0004422 Bone hyperostosis)
Vertebral compression fractures, kyphosis, leg-length discrepancy in spinal disease
Generally afebrile or low-grade fever, distinguishing from infectious osteomyelitis in most cases

Laboratory abnormalities

Normal-to-mildly-elevated ESR/CRP (a substantial subset show no systemic inflammatory marker elevation)
Anemia of chronic disease (common)
Elevated serum cytokines: IL-6, TNF-α (see Mechanism section for full cytokine panel)

Phenotype characteristics

Age of onset: Peak incidence 7–12 years, though reported across all pediatric age groups and increasingly recognized in adults (where it overlaps with SAPHO syndrome). Median age of symptom onset in one UK cohort was 12 years.
Severity/progression: Variable — a relapsing-remitting or progressive course is now recognized as more typical than the historically assumed "self-limited" course. Diagnosis delay averages ~15 months (range 0–92 months) due to low disease awareness (PMID:27576444).
Distinct phenotype clusters (2022 medRxiv study, "Two phenotypes of CRMO"):
Severe phenotype: predominantly male, multifocal, rare clavicular involvement, prominent inflammatory syndrome
Mild phenotype: predominantly female, unifocal, common clavicular involvement, minimal systemic inflammation, rare extraosseous lesions
Intermediate phenotype: predominantly female, multifocal, common inflammatory syndrome, some with family history and extraosseous disease

Skeletal distribution

Most commonly affected sites: metaphyses/epiphyses of long bones (femur, tibia, humerus), pelvis, clavicle, vertebrae, mandible
The clavicle and mandible are distinctively involved in CRMO but rare in bacterial osteomyelitis — involvement at these sites should raise diagnostic suspicion for CRMO specifically (Radsource; PathologyOutlines).
Periosteal reaction reported in ~33% of symptomatic sites; radiographic pattern: 50% lytic, 53% sclerotic lesions in one cohort of 36 patients (Bristol study, PMID:27576444).

Associated (extra-osseous) manifestations / comorbidities

Inflammatory bowel disease (Crohn disease, ulcerative colitis): ~10% of patients
Psoriasis / palmoplantar pustulosis: ~8–21% (estimates vary by cohort)
Severe acne: ~10%
Ankylosing spondylitis / spondyloarthropathy features: up to ~25% in some series
Pulmonary involvement: 3–8% (German cohort)

Quality of life impact

Chronic pain and activity restriction (especially with spinal involvement, given vertebral fracture risk) substantially affect quality of life; amplified musculoskeletal pain syndromes can persist even when active inflammation resolves, per longitudinal UK cohort follow-up data.

Suggested HPO terms: HP:0002653 (Bone pain), HP:0100774 (Osteomyelitis), HP:0004422 (Hyperostosis), HP:0002758 (Osteoarthritis-adjacent joint involvement), HP:0001369 (Arthritis), HP:0100255 (Long bone bowing — for structural sequelae), HP:0002650 (Scoliosis/kyphosis — vertebral involvement), HP:0004616 (Vertebral wedging/compression), HP:0100785 (Recurrent fractures).

4. Genetic/Molecular Information

Causal genes (monogenic forms)

Table (click to expand)

Gene	HGNC	Disorder	Inheritance	Mechanism
LPIN2	phosphatidic acid phosphatase	Majeed syndrome	AR	Loss of lipin-2 function → dysregulated lipid metabolism, macrophage polarization toward inflammatory M2 phenotype, accelerated osteoclastogenesis
IL1RN	IL-1 receptor antagonist	DIRA	AR	Loss of IL-1Ra function → unopposed IL-1α/β signaling
PSTPIP1	proline-serine-threonine phosphatase-interacting protein 1	PAPA syndrome	AD	Pyrin inflammasome dysregulation, elevated IL-1β
FBLIM1	filamin-binding LIM protein 1	Non-syndromic CRMO (rare)	AR (reported family)	Impaired RANKL regulation, altered osteoclast differentiation

Variant classification and population data

Most reported pathogenic variants in LPIN2 and IL1RN are classified pathogenic/likely pathogenic per ACMG/AMP criteria in ClinVar given clear loss-of-function consequences and segregation in consanguineous pedigrees. P2RX7 CNO-associated variants are predominantly rare, missense, presumed-damaging variants rather than clear loss-of-function alleles; population frequency data (gnomAD-derived control cohorts) show these variants at low but non-zero frequency (~1.9–4.4% carrying rare variants in healthy controls vs. 5.8–32.4% in CNO cohorts), consistent with a susceptibility/risk-modifier rather than fully penetrant Mendelian model.

Functional consequences

LPIN2 loss-of-function → gain of inflammatory signaling (M2 macrophage skewing, enhanced osteoclastogenesis)
IL1RN loss-of-function → gain of IL-1 pathway activity (classic loss-of-inhibitor mechanism)
P2RX7 risk variants → gain-of-function-like effect on inflammasome assembly coupled with reduced pyroptotic cell death, prolonging pro-inflammatory monocyte/macrophage activity
FBLIM1 loss-of-function → dysregulated RANKL-driven osteoclast differentiation

Modifier genes

IL-10 promoter haplotype (GCC/ACC/ATA) modulates baseline anti-inflammatory tone and may modify severity/expressivity in sporadic disease.

Epigenetic information

CRMO monocytes show reduced DNA methylation at loci controlling NLRP3 inflammasome components (NLRP3, ASC/PYCARD) and IL-1β, and reduced histone H3 serine-10 phosphorylation at the IL-10 promoter — both consistent with an epigenetically reinforced pro-inflammatory monocyte state (PMC10483435).

Chromosomal abnormalities

No recurrent aneuploidies, translocations, or copy-number syndromes are established causes of CRMO; it is not currently modeled as a chromosomal disorder.

Suggested gene/ontology annotations: HGNC gene symbols LPIN2, IL1RN, PSTPIP1, FBLIM1, P2RX7, FGR (all lowercase hgnc: CURIEs per local convention); GO:0061702 (canonical inflammasome complex), GO:0043123 (positive regulation of NF-kB signaling), GO:0045453 (bone resorption), GO:0030316 (osteoclast differentiation).

5. Environmental Information

Environmental factors: No specific toxin, chemical, or occupational exposure has been robustly linked to CRMO onset; the disease is pediatric-onset in the great majority of cases.

Lifestyle factors: Not established as causal; physical activity/trauma to affected long bones can precipitate symptomatic flares once disease is established (mechanical loading is a recognized aggravator, not an initiator).

Infectious agents: CRMO is explicitly non-infectious and cultures are sterile by definition. However, murine model data (Pstpip2-deficient mice) and some human hypotheses implicate the microbiome as a possible disease-modifying or triggering factor via innate immune signaling (TLR activation), rather than a specific causal pathogen — this remains an active but unresolved research area ("New discoveries in CRMO: IL-1β, the neutrophil, and the microbiome implicated in disease pathogenesis in Pstpip2-deficient mice," Seminars in Immunopathology).

6. Mechanism / Pathophysiology

Causal chain overview

Trigger (unknown in sporadic disease; monogenic loss-of-function in syndromic forms) →
Defective TLR4/MAPK/ERK1-2 signaling in monocytes → failure to phosphorylate ERK1/2 → failure to activate Sp-1 transcription factor →
Impaired expression of anti-inflammatory cytokines IL-10 and IL-19 (chromatin/epigenetically mediated) →
Disinhibited NLRP3 inflammasome assembly (elevated NLRP3, ASC, caspase-1) →
Increased IL-1β and IL-18 secretion →
RANKL upregulation on osteoblasts/bone marrow macrophages, OPG downregulation →
Enhanced osteoclast differentiation and activation →
Bone resorption, sterile inflammatory bone lesions, and secondary reparative hyperostosis/sclerosis

Molecular pathways

TLR4 → MAPK/ERK → Sp-1 → IL-10/IL-19 transcription (defective in CRMO monocytes; PMID:22940633)
NF-κB and MAPK signaling downstream of IL-1β, TNF-α, IL-6, IL-8, IL-18 — all converge on osteoclastogenic and pro-inflammatory transcriptional programs (KEGG hsa04621 NOD-like receptor signaling; KEGG hsa04064 NF-kB signaling)
JAK-STAT3 signaling downstream of IL-6, driving Th17 differentiation and IL-17 production
RANK/RANKL/OPG axis — central convergence point for bone resorption

Cytokine profile (imbalance model)

Table (click to expand)

Elevated (pro-inflammatory)	Reduced (anti-inflammatory)
IL-1β, TNF-α, IL-6, IL-8, IL-18, IL-17, IL-23, IL-20	IL-10, IL-19, IL-9

Elevated serum IL-6 (>17 pg/mL) and eotaxin (>110 pg/mL) have been proposed as a minimal diagnostic biomarker panel with 93% sensitivity and 97% specificity in preliminary ROC analyses, alongside candidate markers S100A8, collagen Iα, RANTES, and soluble IL-2 receptor (JBMR 2024 review); these require validation in larger independent cohorts.

Cellular processes

Monocyte/macrophage dysfunction is the central cellular lesion — attenuated TLR4/MAPK signaling, chromatin-level failure of IL-10/IL-19 induction
Osteoclast hyperactivation — RANKL-driven differentiation from monocyte/macrophage precursors
Mast cell involvement — a preclinical study ("Mast Cells Enhance Sterile Inflammation in Chronic Nonbacterial Osteomyelitis," PMC6737947) demonstrated mast cells amplify sterile bone inflammation in the murine model
Th17 cell differentiation — chronic pro-inflammatory cytokine release (IL-6, IL-23) may drive IL-17-expressing effector T cells, further amplifying osteoclastogenesis (adaptive immune contribution to a primarily innate disease)
Reduced pyroptosis in P2RX7 variant carriers — paradoxically prolongs pro-inflammatory monocyte survival rather than clearing them via cell death

Tissue damage mechanisms

Cycles of osteolysis (active inflammatory phase) followed by reparative sclerosis/hyperostosis (chronic phase) — this destruction-repair cycle underlies the mixed lytic/sclerotic radiographic appearance and progressive bone overgrowth seen at sites like the clavicle and mandible.

Immune system involvement

Predominantly innate immune dysregulation (monocyte/macrophage/neutrophil axis, inflammasome), with secondary adaptive immune (Th17) amplification; not classically autoimmune (no dominant autoantibody or autoreactive T-cell clone identified).

Molecular profiling data

Genomics: Whole-exome sequencing has identified FBLIM1 (familial) and P2RX7 (cohort-level risk variant) associations
Transcriptomics: Single-cell RNA-sequencing in the Pstpip2⁻/⁻ mouse model identified co-expression of the lncRNA Morrbid with Pstpip2 in mature myeloid cells (neutrophils, eosinophils, classical monocytes); Morrbid knockout significantly inhibited CRMO initiation and progression in this model by reducing inflammatory myeloid cell lifespan and cytokine output (2025, Disease Models & Mechanisms, PMID:40503910) — a promising mechanistic and therapeutic lead published very recently.
Proteomics/serum biomarker panels: As above (IL-6, eotaxin, S100A8, RANTES).

Advanced technologies

Single-cell RNA-seq has been applied in the murine Pstpip2-deficient model (Morrbid study) to define myeloid-lineage-specific contributions; equivalent human single-cell/spatial transcriptomic data in bone biopsy tissue remains an area for future research — not yet reported at scale for CRMO.

Suggested GO terms: GO:0032640 (TNF production), GO:0032611 (IL-1β production), GO:0032693 (negative regulation of IL-10 production — CRMO shows failure of this), GO:0045453 (bone resorption), GO:0030316 (osteoclast differentiation), GO:0002534 (cytokine production involved in inflammatory response), GO:0061702 (inflammasome complex). Suggested CL terms: CL:0000576 (monocyte), CL:0000235 (macrophage), CL:0000775 (neutrophil), CL:0000097 (mast cell), CL:0000092 (osteoclast), CL:0001051 (CD4-positive, IL-17-secreting Th17 cell).

7. Anatomical Structures Affected

Organ level

Primary: Skeletal system — long bone metaphyses (femur, tibia, humerus), clavicle, mandible, pelvis, vertebrae
Secondary: Joints (adjacent arthritis/synovitis), skin (psoriasis, palmoplantar pustulosis, acne), gastrointestinal tract (IBD), lungs (rare, 3-8%)
Body systems involved: Musculoskeletal (primary), integumentary, gastrointestinal, and — in SAPHO overlap — the axial skeleton/sacroiliac joints

Tissue and cell level

Bone tissue: metaphyseal/epiphyseal trabecular and cortical bone; periosteum (periosteal reaction/new bone formation)
Cell populations: monocytes, macrophages, osteoclasts, neutrophils, mast cells, and — in chronic lesions — lymphocytes, plasma cells, histiocytes

Subcellular level

Implicated: inflammasome assembly at the cytoplasmic level (NLRP3/ASC/caspase-1 complex — GO Cellular Component: inflammasome complex), plasma membrane P2X7 channel activity.

Localization

UBERON terms (suggested): UBERON:0002481 (clavicle), UBERON:0002397 (mandible), UBERON:0001474 (bone element), UBERON:0002228 (vertebral column), UBERON:0004538 (long bone metaphysis is not a distinct UBERON term but metaphysis-adjacent structures apply)
Lateralization: Typically asymmetric/multifocal, non-contiguous lesions distributed across multiple bones simultaneously — a hallmark distinguishing feature from unifocal bacterial osteomyelitis.

8. Temporal Development

Onset

Typical age: Peak 7–12 years; can present at any pediatric age and, less commonly, in adults (where overlap with SAPHO is emphasized)
Pattern: Insidious, gradually worsening bone pain/swelling rather than acute high-fever presentation typical of bacterial osteomyelitis

Progression

Disease course pattern: Historically viewed as self-limited, but increasingly recognized as chronic with a relapsing-remitting OR progressive course — a key revision in current understanding (JBMR 2024 review; UK longitudinal cohort data)
Stages: acute/active inflammatory (lytic) phase → reparative/chronic (sclerotic, hyperostotic) phase, often cycling at the same or different sites over time
Duration: Chronic, potentially lifelong disease activity in a subset; median follow-up studies show many but not all patients eventually reach durable remission

Patterns

Remission: Can be spontaneous or treatment-induced (NSAID, bisphosphonate, or biologic-associated); >50% of NSAID responders experience a flare at a median of ~29 months despite initial response
Relapse: A UK tertiary center cohort (17 patients, 1999–2015) demonstrated ongoing relapse risk into adolescence/young adulthood requiring long-term rheumatology follow-up
Critical periods: Early diagnosis (median delay currently ~15 months) is emphasized as a window to prevent structural bone damage (vertebral fracture, growth disturbance) before it occurs.

9. Inheritance and Population

Epidemiology

Incidence (most authoritative recent estimate): 0.65 per 100,000 person-years in children <16 in the UK/Republic of Ireland (2024 BPSU national prospective surveillance study, October 2020–November 2022; 288 patients reported, 165 confirmed + 20 probable cases analyzed) (PMC11962910; Rheumatology Oxford 2025).
Other national estimates: 0.4 per 100,000 children in Germany; 2.3 per 100,000 children in one large catchment area of the northwestern United States; a pooled pediatric estimate of ~0.605 CNO cases per 100,000 person-years has also been cited (JBMR 2024).
Historic literature-review estimate: as low as 4 per million children, though rising with increased clinical recognition — likely reflecting under-ascertainment rather than a true rising incidence.

For genetic etiology

Inheritance pattern: Sporadic/non-syndromic CRMO is generally considered complex/multifactorial (polygenic risk-variant model, e.g., P2RX7); the monogenic syndromic forms (Majeed, DIRA, PAPA) follow classic autosomal recessive (Majeed, DIRA) or autosomal dominant (PAPA, CRMO3) Mendelian inheritance.
Penetrance: Complete/high for monogenic LPIN2/IL1RN loss-of-function alleles; incomplete/variable for P2RX7 and IL-10 haplotype risk variants (risk-modifier model, not deterministic).
Consanguinity: A recognized risk factor for the recessive monogenic forms — the original FBLIM1 and several LPIN2 pedigrees were identified in consanguineous families.
Carrier frequency: Not systematically established for LPIN2/IL1RN given extreme rarity (~24 genetically confirmed Majeed syndrome individuals worldwide reported to date).

Population demographics

Sex ratio: Overall female predominance in classic pediatric CRMO cohorts, though the "severe/multifocal" phenotype subgroup skews male (see Phenotypes section) — UK longitudinal cohort: 10 female : 7 male.
Ethnic/geographic distribution: White/Caucasian populations most frequently reported in the literature, though this likely reflects study-site bias; global epidemiologic data remain limited. FBLIM1-associated CRMO was identified in a South Asian consanguineous family, indicating the disease is not confined to any single ancestry.
Age distribution: Concentrated in the pediatric/adolescent range (peak 7–12 years), with adult-onset presentations increasingly reported (overlapping SAPHO nomenclature).

10. Diagnostics

Clinical/laboratory tests

Blood tests: ESR, CRP (often normal-to-mildly elevated — a distinguishing feature versus bacterial osteomyelitis), CBC (anemia of chronic disease)
Biomarkers (investigational): Serum IL-6 + eotaxin panel (93% sensitivity/97% specificity in preliminary studies), S100A8, RANTES, soluble IL-2 receptor — not yet standard of care
Bone biopsy/histopathology: Traditionally required to exclude infection/malignancy; early lesions show neutrophils, lymphocyte clusters, occasional eosinophils; chronic lesions show lymphocytes, plasma cells, histiocytes; established lesions show necrotic bone fragments, fibrosis, increased osteoblasts, and dilated vessels. Cultures are sterile.

Imaging

Plain radiographs: May be normal early; later show lytic and/or sclerotic lesions with periosteal reaction (Bristol cohort, n=36: 50% lytic, 53% sclerotic, 33% periosteal reaction)
Whole-body MRI (WB-MRI): the diagnostic "gold standard." Detects bone marrow edema on fat-saturated T2/STIR sequences before structural change is visible, and — critically — reveals clinically silent, radiographically occult multifocal lesions, establishing the multifocal distribution pattern central to diagnosis and helping exclude alternative diagnoses (Insights into Imaging 2022; PMID:36114435).
Bone biopsy may be avoided when WB-MRI shows the characteristic multifocal pattern at typical sites (clavicle, mandible, metaphyses) with no systemic infectious signs.

Diagnostic criteria

Two named clinical criteria sets exist, neither prospectively validated at scale: - Jansson criteria — found to be more sensitive than Bristol criteria (OR 3.94, P<0.001) - Bristol criteria — use by an experienced clinician may obviate the need for biopsy in some patients - ACR/EULAR candidate classification criteria, derived from ~450 international cases, are in development/pending final dissemination as of the 2024 JBMR review — representing an important near-term advance in standardized diagnosis.

Genetic testing

Not routine for sporadic CRMO; recommended when syndromic features suggest Majeed syndrome (congenital dyserythropoietic anemia, neutrophilic dermatosis — test LPIN2), DIRA (neonatal pustulosis, extreme inflammatory markers — test IL1RN), or PAPA syndrome (pyoderma gangrenosum, cystic acne — test PSTPIP1).

Differential diagnosis

Bacterial/fungal/mycobacterial osteomyelitis, malignancy (Ewing sarcoma, Langerhans cell histiocytosis, leukemia/lymphoma bone involvement), benign bone tumors (osteoid osteoma, bone cysts), other monogenic autoinflammatory disorders (PAPA, DIRA, Majeed syndrome), metabolic bone disease (hypophosphatasia), osteonecrosis, osteopetrosis, juvenile idiopathic arthritis.

11. Outcome/Prognosis

Mortality: CRMO is not associated with excess mortality; it is a morbidity-driving, not life-threatening, condition.
Disease course: The historically assumed "self-limiting" natural history is now understood to be an oversimplification — CRMO frequently follows a relapsing-remitting or progressive course, with a substantial proportion of patients experiencing relapses even after apparent remission.
Complications: Vertebral compression fractures, kyphosis/scoliosis, leg-length discrepancy (from growth-plate involvement), progressive bone overgrowth/hyperostosis (clavicle, mandible), and — independent of active inflammation — amplified musculoskeletal pain syndromes that can persist and are treatment-refractory.
Prognostic factors: Multifocal/severe phenotype (male predominant subgroup), spinal involvement, and delayed diagnosis (median 15 months) are associated with greater risk of structural damage.
Quality of life: Chronic pain and activity restriction substantially affect daily functioning; long-term rheumatology follow-up into adulthood is recommended given persistent relapse risk documented in UK tertiary-center cohorts.

12. Treatment

First-line

NSAIDs (typically naproxen) — inhibit cyclooxygenase, reducing prostaglandin E–mediated osteoclast activation. Effective in >60% of patients over 12–18 months, but >50% experience a flare at a median of ~29 months. (Suggested NCIT: NCIT:C15986 Pharmacotherapy; therapeutic_agent CHEBI-bound NSAID)

Second-line / escalation

Corticosteroids: Short courses (e.g., prednisone-equivalent 2 mg/kg/day for 5–10 days, or bridging 0.1–0.2 mg/kg/day); mechanism via phospholipase A2 and NF-κB-regulated cytokine (IL-1, IL-6, TNF-α) inhibition; long-term use limited by adverse-effect burden.
Bisphosphonates (pamidronate, zoledronic acid):
Pamidronate: 1 mg/kg/dose (max 60 mg) monthly, or 3 consecutive days every 3 months, for 9–12 months
Zoledronic acid: 0.0125–0.025 mg/kg/dose (max 4 mg) every 6–12 months
Mechanism: osteoclast apoptosis + reduction of pro-inflammatory cytokine expression
Particularly effective for vertebral involvement; mean MRI resolution of inflammation ~6 months (range 2–12 months)
The only published RCT in this disease is pamidronate vs. placebo in adults, underscoring the overall paucity of controlled trial evidence
(NCIT:C15986 Pharmacotherapy; therapeutic_agent CHEBI-bound pamidronate/zoledronic acid)
Conventional DMARDs (methotrexate, sulfasalazine, leflunomide): increasingly used but with sparse supporting evidence

Biologics / emerging therapies

TNF inhibitors (adalimumab, infliximab, etanercept): 40–50% remission rates in European registry data; beneficial for vertebral lesions refractory to pamidronate; paradoxical psoriasis is a recognized adverse effect, arguably more common than with other indications; not currently licensed for CNO/CRMO. A comparative international multicenter retrospective study (n=91: pamidronate=47, TNFi=22, both sequentially=22) found both therapies associated with clinical remission at 6 months and MRI lesion reduction at 12 months; pamidronate trended toward faster MRI resolution (not statistically significant), while TNF inhibitors were associated with fewer flares (PMID:35460903).
IL-1 blockade (anakinra, canakinumab, rilonacept): beneficial for osteitis/arthritis with variable mucocutaneous response; canakinumab produced rapid response in a refractory CRMO-with-pyoderma-gangrenosum case, though efficacy for bone disease waned over time in some reports (PMID:36004431). Majeed syndrome and DIRA patients respond well to IL-1 blockade, supporting mechanistic centrality of this pathway; canakinumab has produced long-lasting remission in Majeed syndrome.
IL-17/IL-23 inhibition: Secukinumab (anti-IL-17A) successfully used in SAPHO patients in some reports but led to recurrence of osteomyelitis despite psoriasis non-response in at least one case — efficacy is inconsistent. Ustekinumab (anti-IL-12/23) has shown case-report success for CNO, including resolution of back pain.
JAK inhibitors: Deucravacitinib (TYK2 inhibitor) showed effectiveness in a 2024/2025 case report for CRMO with concomitant psoriasis (PMC12138201); tofacitinib shows promise in SAPHO with limited pediatric CNO data.
IL-6 blockade (tocilizumab): mixed results despite IL-6 being a consistent serum biomarker.
RANKL inhibition (denosumab): theoretical potential given RANK/RANKL centrality to mechanism, but no published CNO reports to date.
Investigational/preclinical targets: P2X7 antagonism, NLRP3 inflammasome small-molecule inhibitors (e.g., MCC950), ASC inhibitors (e.g., MM01), IL-18-selective inhibitors, and — from the newest (2025) murine mechanistic work — targeting the lncRNA Morrbid to shorten inflammatory myeloid cell lifespan.

Surgical/other

Surgery is generally reserved for structural complications (e.g., severe vertebral deformity) rather than as primary disease-modifying treatment.

Treatment strategy / trial design gaps

No treatments are currently FDA/EMA-licensed specifically for CNO/CRMO. Expert consensus prioritizes IL-1 and IL-17 blockade as the leading candidate interventions for future controlled trials, with pamidronate as an active comparator. ACR/EULAR classification criteria and OMERACT core outcome measures are anticipated as near-term advances that should enable better-powered trials (JBMR 2024 review).

Suggested NCIT terms: NCIT:C15986 (Pharmacotherapy), NCIT:C15632 (Chemotherapy — n/a here), NCIT:C49236 (Therapeutic Procedure), NCIT:C15329 (Surgical Procedure — for structural complications), NCIT:C20401 (Monoclonal Antibody — for biologics), NCIT:C2986 (drug class terms per specific agent).

13. Prevention

Primary prevention: None established — no known modifiable risk factor or vaccination strategy exists, since etiology in sporadic disease remains incompletely defined.
Secondary prevention (early detection): The chief actionable lever is reducing diagnostic delay (currently median ~15 months) through greater clinical awareness and appropriate use of whole-body MRI, since earlier treatment initiation is presumed (though not RCT-proven) to reduce risk of structural damage (vertebral fracture, growth disturbance, permanent hyperostosis).
Genetic counseling: Relevant for families with confirmed monogenic disease (Majeed syndrome, DIRA, PAPA) given autosomal recessive/dominant inheritance patterns and availability of single-gene testing; carrier screening/prenatal diagnosis can be discussed in consanguineous families with a confirmed proband mutation.
Tertiary prevention: Activity modification (particularly avoiding high-impact loading with active spinal lesions) to reduce fracture risk; routine surveillance imaging and bloodwork during treatment to catch flares early.

14. Other Species / Natural Disease

No naturally occurring veterinary counterpart of CRMO in companion animals or wildlife is well documented in the literature surveyed (unlike many Mendelian diseases with OMIA veterinary entries); the disease's key animal correlates are engineered/spontaneous mouse mutants (below) rather than natural disease in other species.

15. Model Organisms

Murine genetic models (the field's principal disease models)

cmo (chronic multifocal osteomyelitis) mice — spontaneous homozygous missense mutation in Pstpip2 (PMID:16122996); develop sterile osteomyelitis with severe systemic cytokine/chemokine dysregulation, extramedullary hematopoiesis, and inflammatory skin lesions.
lupo mice — chemically induced (ENU) homozygous Pstpip2 mutation with a very similar phenotype to cmo mice.
Mechanistic insight from these models: Bone inflammation in Pstpip2-deficient mice is IL-1β-mediated but NLRP3-inflammasome- and caspase-1-independent — notably different from most other IL-1-mediated autoinflammatory diseases, and an important nuance for interpreting IL-1 pathway targeting (PNAS; PMC3903222).
Mast cell contribution: Mast cells enhance sterile inflammation in this model (PMC6737947/biorxiv preprint), implicating a non-myeloid innate immune cell type beyond monocytes/macrophages.
Morrbid lncRNA (2025): Single-cell RNA-seq in Pstpip2⁻/⁻ mice identified co-expression of the myeloid-restricted long non-coding RNA Morrbid with Pstpip2; genetic disruption of Morrbid significantly inhibited CRMO initiation and progression, mitigating myeloid cell activation and excessive cytokine release — proposed as a novel therapeutic strategy of shortening inflammatory myeloid cell lifespan (Disease Models & Mechanisms, PMID:40503910; PMC12309896).
FBLIM1-related insight originated from microarray analysis of the cmo mouse bone marrow macrophage transcriptome, which nominated Fblim1 before the human familial mutation was identified — a model-to-human translational discovery pathway.

Phenotype recapitulation and limitations

The Pstpip2-mutant models recapitulate multifocal sterile osteomyelitis, systemic inflammatory cytokine dysregulation, and skin inflammation, closely mirroring human CRMO/Majeed-spectrum disease. However, they model the syndromic/PSTPIP-pathway disease axis specifically; they do not directly model the P2RX7-variant or IL-10-haplotype-driven sporadic human CRMO subtype, and — as with most mouse inflammatory models — species differences in innate immune receptor biology and inflammasome regulation limit direct translational certainty (a candidate HUMAN_MODEL_MISMATCH consideration for any KB entry: the NLRP3-independence of murine IL-1β-driven disease contrasts with the NLRP3-centric human mechanistic literature).

Other model systems

Zebrafish: No CRMO-specific zebrafish model was identified in the literature searched, but zebrafish are an established platform for skeletal disorder and innate-immune/notochord-infection modeling generally, and represent a plausible future avenue (real-time imaging of innate immune cells, ease of genetic manipulation) — not yet applied to CRMO specifically per available sources.
iPSC/organoid models: No CRMO-specific iPSC-derived or organoid models were identified in the current literature — a research gap.

Resources

MGI (Mouse Genome Informatics) for Pstpip2 allele records (cmo, lupo); no dedicated CRMO patient-derived cell biobank was identified in this search.

Summary of Key Evidence Gaps (for curation prioritization)

No licensed treatment exists for CNO/CRMO; only one RCT (pamidronate vs. placebo, adults) has been conducted.
ACR/EULAR classification criteria are in development but not yet finalized/published in final form.
Sporadic (non-syndromic) CRMO genetics remain largely unresolved — P2RX7 explains only a minority of cases.
The NLRP3-independence of the leading murine model (Pstpip2-deficient mice) versus the NLRP3-centric human mechanistic literature is a notable model-translatability nuance.
Biomarker panels (IL-6/eotaxin) require independent validation before clinical adoption.
Long-term natural history / outcome data beyond adolescence remain sparse — most cohorts are pediatric-center-based with limited adult follow-up.

Sources

Reference Validation

Checked with linkml-reference-validator 0.2.1.

Table (click to expand)

Outcome	Count
References checked	31
Resolved	31
Unresolved (possible confabulation)	0
Unverifiable	0
References weighed for topical relevance	31
On topic	26
Off topic	0

All extracted references resolved successfully.