MRI Findings in CRMO Across Different Sequences
In CRMO, the characteristic MRI pattern shows hypointense (dark) signal on T1-weighted images and hyperintense (bright) signal on both T2-weighted fat-suppressed sequences and STIR sequences, representing bone marrow edema in a geographic, ill-defined pattern typically adjacent to growth plates.
T1-Weighted Imaging Findings
T1-weighted sequences are critical for identifying structural involvement in CRMO 1, 2:
- Hypointense (dark) signal compared to normal bone marrow
- Signal intensity typically isointense or hypointense to skeletal muscle
- Geographic, ill-defined lesions involving the medullary canal
- Best sequence for identifying the full extent of marrow replacement
- Useful for detecting chronic changes and structural damage
The T1 hypointensity reflects replacement of normal fatty marrow by inflammatory tissue and is present in essentially all active CRMO lesions 1.
T1 Fat-Suppressed (Post-Contrast) Imaging
While the provided evidence focuses primarily on osteomyelitis rather than CRMO specifically for contrast-enhanced sequences, T1 fat-suppressed post-gadolinium imaging in inflammatory bone conditions:
- Shows enhancement of active inflammatory lesions (osteitis)
- Helps differentiate active inflammation from chronic changes
- Can identify associated soft tissue involvement
- However, contrast is not routinely necessary for CRMO diagnosis as fluid-sensitive sequences are typically sufficient 1
STIR (Short Tau Inversion Recovery) Imaging
STIR is one of the most sensitive sequences for detecting CRMO lesions 2, 3:
- Markedly hyperintense (bright) signal in affected bone marrow
- Sensitive to free water content, making it ideal for detecting bone marrow edema
- 510 of 513 lesions (99.4%) showed significant hyperintensity compared to normal marrow in one study 2
- Excellent for whole-body screening to detect multifocal involvement
- Uniform fat suppression across the entire field of view
T2-Weighted Fat-Suppressed Imaging
T2 fat-suppressed sequences show similar findings to STIR 1, 2:
- Hyperintense (bright) signal in areas of bone marrow edema
- Corresponds to increased free water from inflammation
- Equivalent to STIR for detecting active lesions
- May be used interchangeably with STIR in most protocols
Characteristic Distribution Patterns Across All Sequences
Regardless of sequence type, CRMO demonstrates distinctive anatomic patterns 1, 2:
Location Preferences:
- Metaphyseal involvement in 86% of tubular bone lesions
- Epiphyseal involvement in 67% of cases
- Contiguous physeal relationship in 89% of lesions
- Most common sites: distal femur (21%), proximal tibia (17%), distal tibia and fibula (14% each)
Additional Features:
- Bilateral symmetric involvement in 75-85% of patients
- Periosteal reaction in 32-48% of lesions
- Multifocal involvement in essentially all patients (mean 9.7 lesions per patient)
- Pelvis, lower extremities, shoulders, and spine most frequently affected
Clinical Pearls and Pitfalls
Key diagnostic considerations:
- Normal T1 signal reliably excludes active CRMO - the sensitivity of marrow signal abnormality approaches 100% 1
- Whole-body MRI is superior to targeted imaging for revealing the true extent of multifocal disease, often detecting asymptomatic lesions 2, 3
- The combination of T1 hypointensity + STIR/T2FS hyperintensity in a metaphyseal/epiphyseal location adjacent to growth plates is highly characteristic
- Fluid-sensitive sequences alone may be sufficient for diagnosis in many cases 4
Common pitfalls to avoid:
- Don't rely on inflammatory markers (ESR, CRP) - they show no correlation with number of lesions and have only 15% sensitivity 1
- Don't assume unifocal disease based on symptoms - only 26 of 53 patients with multifocal MRI findings presented with multifocal complaints 2
- Don't overlook spinal involvement - vertebral complications may be asymptomatic and require whole-body imaging to detect 2
The diagnostic protocol should include both T1-weighted sequences (for structural assessment) and fluid-sensitive sequences like STIR or T2 fat-suppressed (for detecting active inflammation), ideally in a whole-body format to capture the multifocal nature of the disease 2, 3.