Differences in Topographical Location of Sacroiliac Joint MRI Lesions in Patients With Early Axial Spondyloarthritis and Mechanical Back Pain

Rosa Marie Kiil; Clara E. Mistegaard; Anne Gitte Loft; Anna Zejden; Oliver Hendricks; Anne Grethe Jurik

Disclosures

Arthritis Res Ther. 2022;24(75) 

In This Article

Results

Description of the Study Sample and Interreader Agreement

Table 1 presents the demographics, clinical, paraclinical and biochemical data on the 84 patients included in the study divided into axSpA patients (n = 25) and patients with MBP (n = 59). These data, along with further clinical and biochemical data, have been presented in detail in a recent publication.[21]

For BME sumscores, disc degenerative changes and the global axSpA MRI confidence score, there were very good agreements with ICC ≥ 0.91 between the two readers whereas ICCs. The kappa values for the presence of BME depth, FMD, FMD depth and atypical morphologies were ≥ 0.81, and thereby almost perfect whereas the kappa values for sclerosis and erosion were ≥ 0.39 and ≥ 0.45, fair and moderate agreements, respectively. The kappa values for BME, FMD and sclerosis in relation to anatomical variations were ≥ 0.39, ≥ 0.23 and ≥ 0.66, respectively.

Global axSpA MRI Confidence Score

Table 2 provides the results of the mean confidence scores for the likelihood of axSpA based on the MRI appearance. Eight patients (32%) with the final diagnosis of axSpA had a mean score between − 0.5 and − 4.5, whereas seven (12%) patients with MBP scored between 0.5 and 5.0.

BME, FMD, Sclerosis and Erosions

The overall occurrence of BME (> 0 in sumscore in at least one location) was high in both groups being 100% in the axSpA group and 95% in the MBP group. The axSpA group had a higher total sumscore of 25.1 compared to 6.8 (p < 0.005) in the MBP group.

Figure 1A, B presents the prevalence and sumscores of BME in the 14 different locations divided into axSpA (A) and MBP (B) patients. The MBP group had the highest prevalence (66%) and sumscore (5.7) in the middle anterior sacrum, which was also true for the axSpA group. However, the BME distribution in the axSpA group was widespread and equally distributed in the anterior and posterior areas (Figure 1A). Compared to the MBP group, the axSpA group had significantly higher prevalence and sumscores in all locations, except in three anterior sacral locations.

Figure 1.

A, B Prevalence and sumscores of BME in the 12 cartilaginous and two ligamentous joint portions in axSpA (A) and MBP (B) patients divided in the iliac (left) and sacral (right) sides. C, D BME depth prevalence in the same locations and patient groups as in A and B. Boxes in A and B represent the prevalence and relative sumscore, whereas boxes in C and D represent the BME depth prevalence in each location

BME in the ligamentous joint compartments was significantly more frequent in the axSpA group where a BME sumscore > 0 occurred in 60%/40% (sacral/iliac) compared to 15%/8% (sacral/iliac, p both < 0.005) in MBP patients.

At least one BME depth score occurred in 60% of axSpA patients. This was significantly more frequent than in the MBP group being 20% (p < 0.005). Figure 1C, D presents the BME depth prevalence in the 14 locations divided in axSpA (C) and MBP (D) patients. The prevalence of BME depth was significantly higher in the axSpA group except at the upper anterior sacrum.

Five patients had a BME intensity score, all occurring in axSpA patients.

The occurrence of a least one FMD lesion was more frequent in axSpA patients compared to the MBP group, occurring in 76% and 47% (p = 0.016), respectively. Figure 2A, B represents the prevalence of FMD in the six cartilaginous locations. AxSpA patients had widespread FMD with significantly higher prevalence in all joint portions compared to MBP patients. MBP patients had the highest FMD prevalence in the middle and upper sacral and middle and lower iliac areas.

Figure 2.

A, B Prevalence of FMD in the six cartilaginous joint portions in axSpA (A) and MBP (B) patients divided in iliac the (left) and sacral (right) sides. C, D Prevalence of FMD depth in the same locations and patient groups as in A and B. Boxes represent the prevalence in each location

The presence of at least one FMD depth score was higher in the axSpA group (52%) compared to the MBP group (19%, p < 0.005). Figure 2C, D represents the prevalence of FMD depth in the same locations as Figure 2A, B. The axSpA group had significantly higher FMD depth prevalence in all locations, except at the upper ilium.

The prevalence of sclerosis and erosions in the six cartilaginous locations are presented in Figure 3A–D. Subchondral sclerosis was rather similar in prevalence and distribution in axSpA (Figure 3A) and MBP patients (Figure 3B) with an occurrence of sclerosis in at least one location of 40% and 42%, respectively. The occurrence of erosions in at least one location was more frequent in axSpA patients compared to MBP patients being present in 80% and 29% (p < 0.005), respectively. The axSpA group had a significantly higher occurrence of erosions in all joint portions (Figure 3C, D) and demonstrated a widespread distribution with the highest prevalence in the three iliac joint portions, whereas the MBP group had the highest prevalence (27%) in the middle iliac joint portion.

Figure 3.

A, B Prevalence of sclerosis in the six cartilaginous joint portions in axSpA (A) and MBP (B) patients divided in the iliac (left) and sacral (right) sides. C, D Prevalence of erosions in the same locations and patient groups as in A and B. Boxes represent the prevalence in each location

Atypical SIJ Morphologies and Lumbar Spine Changes

Table 3 presents the prevalence of four atypical SIJ morphologies, transitional vertebra and disc degenerative changes. The bipartite iliac bony plate was significantly more prevalent in the MBP group, whereas there were no significant between-group differences regarding the remaining variations. Vertebral corner changes (BME and/or FMD) occurred significantly more frequently in axSpA, but there were no between-group differences regarding disc degenerative changes.

Associations Between MRI Patterns and axSpA Diagnosis, Anatomical Variations, Gender, Childbirths, age and BMI

BME in the ligamentous compartment (OR 7.0, p < 0.005), BME depth (OR 5.9, p < 0.005), FMD depth (OR 4.7, p < 0.005), bilateral FMD (OR 4.8, p < 0.005) and erosions (OR 5.9, p < 0.005) were independently associated with axSpA (Table 4). BME in the ligamentous compartment was also a risk factor for having an iliosacral complex (OR 2.8, p = 0.045). BME depth (OR 3.7, p = 0.007) and unilateral FMD (OR 6.4, p < 0.005) were risk factors for having dysmorphic cartilaginous joint facets. Unilateral FMD and bilateral sclerosis were associated with female gender (OR male 0.1, p = 0.011 and 0.2, p = 0.041, respectively) and childbirths (OR 3.8, p = 0.023 and 4.8 p = 0.012, respectively). Furthermore, unilateral sclerosis (OR male 0.1, p = 0.011) and unilateral erosions (OR male 0.2, p = 0.041) were also associated with female gender. None of the MRI variables was associated with bipartite iliac bony plate, lumbar disc extrusion, increasing age or BMI.

processing....