Inflammatory back pain-like symptoms and MRI mimics of sacroiliitis in patients with anatomical variants of the sacroiliac joint: a cross-sectional study.

backgroundSacroiliac joint (SIJ) anatomical variants are common on MRI and may complicate interpretation by mimicking inflammatory sacroiliitis; however, their relationship with inflammatory back pain (IBP)-like symptom patterns is not well defined in patients without axial spondyloarthritis (axSpA). To characterize inflammatory back pain (IBP)-like symptom patterns and accompanying MRI findings that may mimic sacroiliitis in patients with chronic low back pain (CLBP) and sacroiliac joint (SIJ) anatomical variants who did not fulfill classification criteria for axial spondyloarthritis (axSpA). In this cross-sectional study, consecutive patients younger than 45 years at the time of SIJ MRI/clinical evaluation, with chronic low back pain lasting at least 3 months, who underwent SIJ MRI between September 2023 and September 2025 were retrospectively screened for predefined SIJ anatomical variants and lumbosacral transitional anomalies. Patients with lumbosacral transitional anomalies were excluded at the screening stage. Those with at least one SIJ anatomical variant were invited for a standardized clinical assessment in which IBP features were captured using a single questionnaire covering the Calin, Rudwaleit (Berlin), and ASAS IBP criteria. All participants were systematically assessed against the ASAS classification criteria for axial spondyloarthritis (axSpA), and those who fulfilled these criteria were excluded from the final analysis. SIJ MRIs were read independently by two radiologists blinded to clinical data; agreement for variant detection was almost perfect (κ = 0.90). Bone marrow edema (BME) was evaluated according to the ASAS/OMERACT definition of MRI findings highly suggestive of active sacroiliitis, and structural changes were also recorded. Of 260 screened patients with CLBP, SIJ anatomical variants were identified in 108/260 (41.5%). After exclusions (n = 20), 88 patients were included in the final analysis; 75/88 (85.2%) were female, with a median age of 36 years (Q1-Q3: 28-39) and a median symptom duration of 3.0 years (Q1-Q3: 1.6-9.8). IBP positivity varied by criteria set: 51.1% (Calin), 34.1% (Berlin), and 27.3% (ASAS). Variants were predominantly bilateral (76/88, 86.4%). Among 98 recorded variant findings, accessory joint (n = 27) and iliosacral complex (n = 25) were the most frequent. ASAS/OMERACT-positive BME was present in 10/88 patients (11.4%); structural lesions included sclerosis in 22/88 (25.0%), erosions in 7/88 (8.0%), and fat metaplasia in 5/88 (5.7%). HLA-B27 positivity was observed in 4/69 tested patients (5.8%), and median CRP was 3.2 mg/L (Q1-Q3: 1.0-7.0). In a non-axSpA, variant-positive cohort, IBP-like symptom patterns were common, whereas ASAS/OMERACT-positive BME was infrequent. SIJ variants may therefore complicate clinical judgment because of symptom overlap in young patients within the typical age range for axSpA onset. Although infrequent, imaging features that mimic inflammatory disease highlight the need for careful clinical correlation to avoid misclassification as axSpA.