gms | German Medical Science

Background: Osteitis condensans ilii (OCI) is regarded as a non-inflammatory disorder that is believed to be induced, for example, by mechanical stress and mechanical instability of the sacroiliac joints (SIJ) related to pregnancy/delivery. OCI is being increasingly recognized as an important differential diagnosis for axial spondyloarthritis (axSpA), due to onset at young age, possible inflammatory character of back pain and recently described presence of subchondral bone marrow edema on magnetic resonance imaging (MRI) of the SIJ. So far, no systematic comparison of MRI changes in the sacroiliac joints in patients with OCI and axSpA has been performed.

The aim of the study was to compare active and chronic inflammatory lesions of the SIJ as detected by MRI in patients with OCI and axSpA.

Methods: Using medical database search we identified n=103 patients aged ≥18 years who were diagnosed with OCI upon presentation with chronic back pain in the Early Spondyloarthritis Clinic of the rheumatology department in the Charité University Hospital between January 2010 and May 2015. These patients were contacted in order to obtain an informed consent and to complete a survey on the disease-related history. Out of 60 OCI patients who provided an informed consent and completed the survey, 27 patients had evaluable MRIs of the SIJ in STIR and T1-weighted sequences that were used for the current study. These patients were matched to 27 patients with definite axSpA according to the back pain duration. MRIs were scored according to the Berlin scoring system for osteitis, fatty degeneration, erosions, sclerosis and ankylosis independently by 3 trained and calibrated readers who were blinded for all clinical data including diagnosis. In addition, the preferential localization of lesions (ventral, mid, or dorsal part of the SIJ) was recorded.

Results: There was a good agreement between readers for all types of changes: the intra-class correlation coefficient was 0.824 for osteitis, 0.942 for fatty degeneration, 0.960 for erosions, 1.0 for ankylosis and 0.841 for sclerosis.

There were no differences either in the osteitis score or in the proportion of patients with presence of osteitis on MRI of the SIJ between OCI and axSpA patients (Table). The fatty degeneration score was significantly lower in OCI as compared to axSpA, although the difference in the prevalence of the fatty lesions did not reach the level of statistical significance. There was a non-significant trend towards a higher sclerosis score in OCI patients. Importantly, there was a highly significant difference in the erosion score and in the prevalence of erosions: only 2 (7.4%) OCI vs. 18 (66.7%) axSpA patients had at least one erosion (Table 1 [Tab. 1]). Importantly, none of the OCI patients had high-grade (>5 erosions) erosive changes.

There were substantial differences concerning localization of the lesions: in OCI, ventral localization was recorded in 96% of the cases for osteitis, in 100% for fatty degeneration, and in 96% for sclerosis, while in axSpA, osteitis was preferentially localized in the ventral part only in 29% of the cases, fatty degeneration in 25%, sclerosis in 29%. Ankylosis and erosions were localized in the mid part in almost all cases.

Conclusion: MRI of sacroiliac joints in OCI is characterized by preferential ventral localization of lesions (osteitis, fatty degeneration, sclerosis), absence of ankylosis and absence of extended erosive changes. Such a findings constellation should be taken into account as suggestive of OCI for the differential diagnosis of axSpA in clinical practice.