J Korean Acad Rehabil Med.
2007 Aug;31(4):440-446.
Early Diagnosis of Sacroiliitis with Magnetic Resonance Imaging
- Affiliations
-
- 1Department of Physical Medicine and Rehabilitation, Samsung Medical Center, Sungkyunkwan University School of Medicine, Korea. dhsung@smc.samsung.co.kr
- 2Department of Radiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Korea.
- 3Department of Physical Medicine and Rehabilitation, The Armed Forces Capital Hospital, Korea.
Abstract
OBJECTIVE
To investigate the diagnostic value of magnetic resonance imaging (MRI) in early detection of sacroiliitis, to identify risk factors of early sacroiliitis, and to propose a diagnostic algorithm for early ankylosing spondylitis (AS).
METHOD: Twenty-nine consecutive patients with inflammatory back pain (IBP) and unclear sacroiliitis (unilateral grade 2> or=sacroiliitis in plain radiography (PR) based on modified New York criteria) were studied. Clinical features of spondyloarthropathy, HLA B27 positivity, and MR image set of the sacroiliac (SI) joints were obtained. Two radiologists interpreted MR images independently to diagnose definite sacroiliitis. An association between sacroiliitis in MRI and each clinical and laboratory feature was assessed with linear logistic regression analysis. Post-test probability was determined with sensitivity/specificity of clinical and laboratory features.
RESULTS
MRI showed definite sacroiliitis in sixteen patients. The most frequently noted finding was erosion and high signal intensity lesion within the joint cavity in gadolinium enhanced T1-weighted images. Unilateral grade 2> or =sacroiliitis in PR was the only significant risk factor of definite sacroiliitis in MRI. When unclear sacroiliitis in PR, more than one clinical feature of spondyloarthropathy, and HLA B27 were found, probability of AS was 83% in a proposed diagnostic algorithm.
CONCLUSION
MRI of the SI joints can detect sacroiliitis in more than half of patients with IBP and unclear sacroiliitis in PR. Unilateral grade 2> or =sacroiliitis in PR was a risk factor of definite sacroiliitis in MRI. A diagnostic algorithm for early detection of AS is proposed.