Use of dynamic magnetic resonance imaging to detect sacroiliitis in HLA-B27 positive and negative children with juvenile arthritides.

OBJECTIVE

Involvement of the sacroiliac (SI) joints is a hallmark of the spondyloarthropathies (SpA), especially, in early and later stages of ankylosing spondylitis in adults. The significance of sacroiliitis in juvenile SpA is less clear and the diagnosis of juvenile SpA is difficult due to the mostly nonspecific or absent history of back pain in children and the time delay associated with the diagnosis of sacroiliitis by conventional radiographs. Our aim was to evaluate dynamic magnetic resonance imaging (MRI) of the SI joints in children and to assess the frequency and the determinants of SI joint involvement in juvenile SpA and other juvenile arthritides.

METHODS

Clinical examinations and dynamic MRI were performed in 130 children < 16 years of age with joint complaints, 100 with probable SpA, and 30 controls. The degree of back pain was assessed by a visual analog scale (VAS) (0 = no pain, 10 = very severe pain). The following groups were defined before MRI investigation according to modified European Spondylarthropathy Study Group (ESSG) criteria for SpA: Group 1: undifferentiated SpA (uSpA, n = 41, 88% B27+); Group 2: differentiated SpA (n = 29, 97% B27+), comprising reactive arthritis (n = 16), ankylosing spondylitis (n = 9), psoriatic arthritis (n = 3), and arthritis in inflammatory bowel disease (n = 1); Group 3: patients with no signs of SpA other than oligoarthritis, here named juvenile chronic arthritis (JCA) II (n = 30, 93% B27+); Group 4: HLA-B27+ controls without arthritis (n = 12); and Group 5: HLA-B27-controls with various other non-SpA diagnoses (n = 18). MRI was evaluated according to published criteria allowing for differentiation between acute and chronic changes in SI joints.

RESULTS

Acute sacroiliitis without chronic changes could only be detected by dynamic MRI: in 17 patients (11 in Group 1, 3 in Group 2, 3 in Group 3) together in 14/70 (20%) patients with SpA. All these 17 patients had normal pelvic radiographs. Using MRI acute and/or chronic sacroiliitis were found in 35 patients: 17/41 in Group 1 (41%), 15/29 in Group 2 (52%), and 3/30 (10%) patients in Group 3, but in no patients in Groups 4 and 5. Chronic SI joint changes > grade 1 were detected by MRI in 18/70 patients with SpA (25.7%). In comparison, radiographic changes > grade 1 were less often detected in 14/70 patients with SpA (20%) or 23/210 SI joints examined (11 %), compared to 29/210 SI joints found in the MRI examinations (14%) (p = 0.05). Among the 70 patients with SpA, those with MRI diagnosis of acute sacroiliitis had a significantly longer disease duration (62+/-34 vs 28+/-16 months; p = 0.01) and higher C-reactive protein (12+/-12 vs 9+/-14; p = 0.01), and also reported more back pain on VAS (4.3+/-3.6 vs 1.2+/-2.2; p = 0.001) than those without sacroiliitis.

CONCLUSION

Dynamic MRI and MRI are useful to detect acute and chronic sacroiliitis in children. The main advantages in comparison with conventional radiographs are the ability to detect acute changes in the SI joints, the higher sensitivity to detect chronic changes, and clearly, the lack of radiation exposure; while the disadvantages are the high costs and the duration of the procedure. Sacroiliitis is fairly common in juvenile SpA and seems to be associated with disease intensity and duration.