Imaging for Sacroiliitis
Initial Imaging Approach
Start with conventional radiography of the sacroiliac joints as the first-line imaging modality for suspected sacroiliitis, and proceed to MRI if radiographs are negative or equivocal but clinical suspicion remains high. 1, 2
Radiography as First-Line
- Obtain anteroposterior pelvis radiographs to evaluate the sacroiliac joints and hips, as both can be involved in axial spondyloarthritis 1
- Radiographs demonstrate chronic structural changes including erosions, sclerosis, and ankylosis, but cannot show active inflammation 1
- Major limitation: Radiographs have low sensitivity (19%-72%) for early disease and may lag behind symptom onset by 3-7 years 1, 2
- Radiographs miss more than half of patients with structural changes when compared to CT as reference standard 1
- Interobserver agreement is only fair to moderate 1, 2
Complementary Spine Imaging
- Add cervical and lumbar spine radiographs alongside sacroiliac joint films to assess for syndesmophytes, erosions, shiny corners, vertebral squaring, and ankylosis 1, 2
- Thoracic spine radiographs are not broadly useful due to overlying structures, unless symptoms specifically referable to that region 1
Second-Line Imaging When Radiographs Are Negative or Equivocal
MRI of the sacroiliac joints without contrast is the preferred second-line investigation, with sensitivity of 79% and specificity of 89% for axial spondyloarthritis. 2
MRI Protocol Requirements
- Essential sequences include: 2, 3
- Coronal oblique T1-weighted sequences
- Coronal oblique fluid-sensitive sequences (T2 fat-saturated or STIR)
- Perpendicular axial oblique sequence
- Joint-line-specific sequence for optimal bone-cartilage interface evaluation 3
MRI Findings to Report
- Active inflammatory lesions: Bone marrow edema/osteitis (the essential finding), synovitis, enthesitis, and capsulitis 1, 4
- Structural damage lesions: Erosions, sclerosis, fat deposition, and ankylosis 1, 2
- Bone marrow edema must be clearly present and ideally extend at least 1 cm deep to the articular surface to be specific for axial spondyloarthritis 2
Contrast Enhancement Considerations
- MRI without contrast is generally sufficient and should be the default approach 2
- Gadolinium contrast has not been shown to significantly increase diagnostic accuracy 1
- Contrast may be considered in select initial evaluations to improve diagnostic confidence in early disease, but adds cost, requires IV access, and carries risk of nephrogenic systemic fibrosis 1, 5
Alternative Imaging When MRI Unavailable
- CT without contrast of the sacroiliac joints is a reasonable alternative when MRI is contraindicated or unavailable, particularly for young patients where radiation exposure is a concern 1, 2, 3
- Low-dose CT protocols should be used when possible 3
- CT excels at detecting subtle erosions and structural changes but cannot demonstrate active inflammation 1
- Dual-energy CT with virtual non-calcium images can depict bone marrow edema 3
Modalities NOT Routinely Recommended
- Bone scintigraphy with SPECT/SPECT-CT: Not routinely obtained as initial imaging, with no relevant literature supporting its use 1
- FDG-PET/CT: Not routinely obtained as initial imaging modality 1
- Ultrasound: Not recommended for diagnosis of sacroiliitis 1, 2
- MRI of the spine: Not generally recommended for initial diagnosis unless sacroiliac joints are normal but clinical suspicion remains high, as isolated spine involvement occurs in only 6%-23% of cases 1
Critical Pitfalls to Avoid
- Do not rely solely on radiographs in patients with short symptom duration (<3 years), as inflammatory changes on MRI precede radiographic findings by 3-7 years 1, 2
- Be aware that bone marrow edema can occur in up to 30% of healthy controls, but deep lesions extending ≥1 cm from the articular surface are more specific for disease 2
- In pediatric patients, normal developmental features (flaring, blurring, irregular articular surfaces) can simulate disease and must be distinguished from pathology 3
- Avoid ordering oblique sacroiliac joint views, as they show no benefit over standard anteroposterior views 1
- Do not routinely order contrast-enhanced MRI unless specifically needed for diagnostic confidence in equivocal early cases, as it does not significantly improve accuracy 1, 2