Correspondence to Dr Fabian Proft; [email protected]
WHAT IS ALREADY KNOWN ON THIS TOPIC
Variants of sacroiliac joint anatomy are common in patients with low back pain and have been shown to influence imaging changes in patients with axial spondyloarthritis (axSpA).
Certain variants have been hypothesised to be radiological mimicker of sacroiliitis, but they may also modulate disease by changing biomechanical conditions within the joint.
WHAT THIS STUDY ADDS
The overall clinical presentation of axSpA is not significantly influenced by the anatomy of the sacroiliac joint, results reveal a similar distribution of clinical features of axSpA between the groups with normal and atypical sacroiliac joints.
The individual subtypes of the sacroiliac joints may modulate the disease mechanically in distinct ways, which requires validation in larger studies.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Atypical sacroiliac joints do not appear to generally determine the clinical presentation of axSpA; their role in difficult-to-treat axSpA remains a future topic of research.
Introduction
Axial spondyloarthritis (axSpA) is a common rheumatological disease typically affecting the axial skeleton with the sacroiliac joints (SIJs) being the most frequently affected anatomical site.1 2 The leading clinical symptom of the disease is chronic low back pain.2 This pain often shows typical characteristics of inflammatory back pain (IBP), which can be defined according to the Berlin criteria by the presence of various clinical symptoms such as: morning stiffness >30 min, improvement with exercise but no improvement with rest, alternating buttock pain, pain at night with awakening in the second half of the night.3 4 Peripheral manifestations often accompany axial components, with the most common being peripheral arthritis, enthesitis and dactylitis, as well as extramusculoskeletal manifestations (EMMs) such as uveitis, psoriasis and inflammatory bowel disease (IBD).5 6 Among these, uveitis is the most frequent EMM.7 The management of axSpA involves identifying the clinical features characteristic of the disease, leading to an expert-based diagnosis, followed by an evaluation of disease activity. Disease activity can be assessed using tools such as the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) or the Axial Spondyloarthritis Disease Activity Score.2 8
The SIJs play a crucial biomechanical role in force transmission from the lower extremities to the torso and stabilisation of the axial skeleton.1 9 Current research on the pathophysiology of axSpA postulates an influence of the mechanical joint conditions on disease development and progression.10 11 Among the six distinguished anatomical variations of the SIJs by Prassopoulos et al the accessory SIJ, the iliosacral complex and the crescent-shaped ilium seem to lead to chondral stress and consequently to an increased inflammatory response, suggesting a mechanical contribution of these latter two variations to disease modulation.12 13 Additionally, an association between anatomical variations of the SIJs and changes on MRI images in axSpA was established: an increase of bone erosion and bone marrow oedema has been demonstrated in such cases, especially in those with accessory SIJs.14–16 In a study by Schett et al it has been displayed that overall mechanical influence is part of the process of enthesitis by triggering or maintaining inflammation at affected sites.17 In patients with an atypical SIJ form, without axSpA, more subchondral sclerosis has been described, laying additional emphasis on the mechanical properties of these joints.18 While these studies seem to illustrate a mechanical role of certain SIJ types in disease modulation, at the same time anatomical variation on imaging has also been discussed to be a radiological mimicker of sacroiliitis: dorsally located accessory joint facets in adults may cause local irritation, which may be hard to distinguish from sacroiliitis. In the developing skeleton, so-called joint facet defects may mimic erosion.19–21 To date, it remains unclear whether patients with axSpA with atypical joint forms constitute a clinically distinct phenotype, in whom mechanical loading plays a more prominent role in the disease process and furthermore, if the described imaging changes may represent a radiological mimicker of sacroiliitis, making false positive diagnoses more likely, as has been hypothesised before.22
In this analysis, we evaluated whether clinical phenotypes differ between patients with axSpA with and without atypical joint forms by comparing the distribution of clinical spondyloarthritis (SpA) features and disease activity (BASDAI) between groups and analysing this for both women and men independently.
Methods
Study-specific sample
This analysis encompasses our study-specific sample, which is aggregated from multiple well-defined clinical cohorts from an Assessment of SpondyloArthritis international Society expert centre. The German Spondyloarthritis Inception Cohort (GESPIC) which contributed with three separate subgroups: GESPIC-ankylosing spondylitis (GESPIC-AS; patients with radiographic axSpA), GESPIC-Crohn (patients with Crohn’s disease, with or without axSpA) and GESPIC-uveitis (patients with non-infectious acute anterior uveitis, with or without axSpA).23–25 The Identification of Optimal Referral Strategy for Early Diagnosis of axial spondyloarthritis (OptiRef) cohort included patients with chronic back pain lasting for at least 3 months, with symptom onset before the completion of 45 years of age.26 27 The SIJ MRI and CT cohort involved patients presenting with chronic low back pain and suspected axSpA.28–30 Finally, the virtual non-contrast susceptibility-weighted imaging cohort included patients referred for MRI due to suspected axSpA, based on symptoms as IBP or confirmed axSpA diagnoses.31
684 patients had sufficient data to be assessed across these cohorts, with 379 receiving a clinical diagnosis of axSpA and 305 being excluded after axSpA was ruled out.14 Patients with axSpA and atypical joints were matched by age and sex, with those patients with axSpA with normal joints. The presence of a duplicate file for the same patient led to the exclusion of 18 patients.
MRI assessment and final patient inclusion
The MRI evaluation of SIJs was conducted on the entire sample of 684 patients. A semi-quantitative scoring system was used to classify joints as either normal or atypical. Atypical joints were defined by the presence of specific features, including accessory joints, iliosacral complexes and crescent-shaped iliac bones, due to their distinctive biomechanical properties compared with normal joints.14 18 Examples of these variants are illustrated in figure 1.
Enlarge this image.Figure 1. Normal and atypical sacroiliac joint forms as seen on T1-weighted MRI-images: (A) normal sacroiliac joint, (B) iliosacral complex, (C) accessory joint facet, (D) crescent-joint shape.
Patients were included in the analysis based on the presence of these atypical joint forms, which have been associated with imaging changes in prior studies.18 Figure 2 depicts the inclusion process and detailed assessment steps.
Enlarge this image.Figure 2. Flow chart illustrating patient inclusion and exclusion. AS, ankylosing spondylitis; axSpA, axial spondyloarthritis; GESPIC, German Spondyloarthritis Inception Cohort; LBP, low back pain; OptiRef, Identification of Optimal Referral Strategy for Early Diagnosis of axial spondyloarthritis; SIMACT, sacroiliac joints MRI and CT; VNC-SWI, virtual non-contrast susceptibility-weighted imaging.
Clinical features
In addition to age and sex, which were adjusted for during matching, the following clinical factors were considered as potential confounders: body mass index (BMI), symptom duration, human leucocyte antigen (HLA)-B27 positivity, elevated C-reactive protein (CRP) values, positive family history of SpA and treatment with biological disease-modifying anti-rheumatic drugs (bDMARDs). Elevated CRP was defined as any value equal to or above 5 mg/L. For BASDAI (n=9, 5.5%), BMI (n=3, 1.8%) and symptom duration (n=13, 8%), mean imputation was applied. All other variables showed complete data sets.
Outcome variables
We evaluated the association between atypical joint shapes and other clinical characteristics of axSpA, namely characterisation of the low back pain as IBP by the treating rheumatologist, documented current or ever diagnosed extra-axial manifestations of the disease such as peripheral SpA manifestations (peripheral arthritis, enthesitis, dactylitis) and EMMs (uveitis, psoriasis and IBD). Furthermore, we analysed whether there was an association between the joint shapes and disease activity as measured by BASDAI.
Propensity score matching
We a priori assumed sex and age to be the most important predictors of heterogeneity regarding the clinical characteristics of axSpA in our patient cohort. Sex is known to be related to the incidence of axSpA and its phenotype, and associated with the presence or absence of an SIJ variation; we have regarded it in our study as the principal influencer of the clinical picture in the overall cohort.32 33 Additionally, we matched for age, as axSpA typically begins before the age of 40 and progresses over the course of the patient’s lifetime.2 Matching patients with and without atypical joints was performed using propensity score matching with an intended ratio of 1:2; matching variables were age and sex, and matching tolerance was set to 0.01. While this approach aimed to address key confounders, we acknowledge that additional variables were considered in subsequent analyses, as detailed in the Results and Discussion sections. Group differences in the matched cohort were also evaluated to explore other factors potentially influencing outcomes.
Statistical analysis
Group differences were assessed using χ² tests for categorical variables and t-tests or one-way analysis of variance (ANOVA) for continuous variables. We applied linear and logistic regression models for the outcome analysis, adjusting for all factors with a p value<0.15. This threshold was chosen based on simulations demonstrating its adequacy for including confounding variables that may not reach significance in univariate analyses.34 Furthermore, using both propensity score matching and regression adjustment can help improve the precision of effect estimates and reduce residual confounding, providing a more robust analysis by combining the strengths of both methods.35 Additionally, a sex-specific sensitivity analysis was conducted separately for men and women. Finally, a sensitivity analysis was performed in the bDMARD-naïve group and according to SIJ type. The analysis according to joint type was carried out descriptively, given the small sample size of the subtypes. To adjust for multiple testing, we employed the Bonferroni method, prespecifying 4 subgroup analyses (all patients, women, men, joint type) for 8 primary outcomes, totalling 32 tests (8×4). Since ANOVA tests the null hypothesis in one step, we considered the multigroup joint type comparison as one test. Thus, we set the Bonferroni-adjusted significance threshold at p<0.002 (0.05÷32, rounded). The sensitivity analyses were purely confirmatory and not aimed at new insights, so no extra corrections were made. Nonetheless, we consistently applied the stricter p<0.002 threshold for all tests and provided raw p values for transparency. All statistical analyses were performed using SPSS V.29.0.1.1 (IBM, New York, New York, USA) and R V.4.3.3 (The R Project for Statistical Computing).
Results
Study cohort
A total of 165 patients were included in this analysis. Patients with an atypical SIJ had a mean age of 38.7 years (SD 12.9), while patients with a normal SIJ had a mean age of 36.1 years (SD 11.2)—per study design, there was no significant difference between the two groups (p=0.175). Similarly, sex distribution was nearly identical in both groups: in the group of atypical joints 34 (55.7%) were women, in the normal joint group 52 (50.0%) were women (p=0.582). Table 1 summarises the baseline characteristics of the groups with normal and atypical SIJs in the matched cohort. No statistically significant differences in distribution or mean were found for any clinical parameters under investigation. HLA-B27 positivity was documented in 88 (84.6%) of patients with normal SIJs and in 50 (82.0%) in those in the atypical SIJ cohort (p=0.821). No significant difference was found regarding the prevalence of elevated CRP (40.4% vs 47.5%, p=0.463). A positive family history of SpA was reported in 28 (26.9%) of patients with a normal SIJ and in 16 (26.2%) of those with an anatomical variation (p>0.999). Most included patients were bDMARD-naïve (n=147, 89%). There were no significant differences in BMI (24.4 (4.7) vs 25.8 (4.4), p=0.073), however, the p value was below the predefined threshold for inclusion in the regression analysis and therefore it was included in the outcome analysis.
Table 1Clinical characteristics in patients with normal and atypical sacroiliac joints
| Normal joint (n=104) | Atypical joint (n=61) | P value | |
| Age, years (mean (SD)) | 36.1 (11.2) | 38.7 (12.9) | 0.175 |
| Female patients (n (%)) | 52 (50.0) | 34 (55.7) | 0.582 |
| BMI (mean (SD)) | 24.4 (4.7) | 25.8 (4.4) | 0.073 |
| Symptom duration, months (mean (SD)) | 120.0 (107.4) | 116.5 (98.3) | 0.838 |
| HLA-B27 positivity (n (%)) | 88 (84.6) | 50 (82.0) | 0.821 |
| Elevated CRP (n (%)) | 42 (40.4) | 29 (47.5) | 0.463 |
| Positive family history of SpA (n (%)) | 28 (26.9) | 16 (26.2) | >0.999 |
| bDMARD therapy (n (%)) | 12 (11.5) | 6 (9.8) | 0.936 |
| Cohort (%) | 0.609 | ||
| GESPIC-AS cohort (%) | 34 (32.7) | 22 (36.1) | |
| GESPIC-Crohn cohort (%) | 5 (4.8) | 0 (0.0) | |
| GESPIC-uveitis cohort (%) | 22 (21.2) | 13 (21.3) | |
| OptiRef cohort (%) | 15 (14.4) | 11 (18.0) | |
| SIMACT cohort (%) | 19 (18.3) | 11 (18.0) | |
| VNC-SWI cohort (%) | 9 (8.7) | 4 (6.6) |
Elevated CRP was defined as any value equal or above 5 mg/L.
AS, ankylosing spondylitis; bDMARD, biological disease-modifying anti-rheumatic drug; BMI, body mass index; CRP, C-reactive protein; GESPIC, German Spondyloarthritis Inception Cohort; HLA-B27, human leucocyte antigen B27; OptiRef, Identification of Optimal Referral Strategy for Early Diagnosis of axial spondyloarthritis; SIMACT, sacroiliac joint MRI and CT; SpA, spondyloarthritis; VNC-SWI, virtual non-contrast susceptibility-weighted imaging.
Outcomes and adjusted results
We observed no differences in disease activity, expressed by mean BASDAI, between patients with normal and atypical joints (adjusted ß=−0.118 (95% CI -0.713, 0.476), p=0.696). IBP revealed a similar distribution between the groups (adjusted OR 0.614 (95% CI 0.274, 1.377), p=0.236). None of the EMMs (uveitis, psoriasis, IBD) or peripheral manifestations (peripheral arthritis, enthesitis, dactylitis) were more prevalent in patients with normal SIJs than in those with atypical joint shapes—details on unadjusted and adjusted results are displayed in table 2.
Table 2Regression analysis of clinical characteristics in patients with normal and atypical sacroiliac joints
| Unadjusted estimate (95% CI) | Unadjusted p value | Adjusted estimate (95% CI) | Adjusted p value | |
| BASDAI | −0.021 (−0.617, 0.575) | 0.945 | −0.118 (−0.713, 0.476) | 0.696 |
| Inflammatory back pain | OR 0.61 (0.274, 1.358) | 0.226 | OR 0.614 (0.274, 1.377) | 0.236 |
| Uveitis ever | OR 1.915 (0.843, 4.35) | 0.121 | OR 1.777 (0.774, 4.076) | 0.175 |
| Psoriasis ever | OR 1.157 (0.445, 3.011) | 0.765 | OR 1.112 (0.424, 2.919) | 0.829 |
| IBD ever | OR 0.972 (0.273, 3.467) | 0.966 | OR 1.173 (0.32, 4.306) | 0.810 |
| Peripheral arthritis current | OR 0.675 (0.276, 1.652) | 0.390 | OR 0.622 (0.251, 1.54) | 0.304 |
| Enthesitis current | OR 1.215 (0.575, 2.571) | 0.610 | OR 1.244 (0.583, 2.656) | 0.572 |
| Dactylitis current | OR 1.293 (0.28, 5.981) | 0.742 | OR 1.215 (0.26, 5.675) | 0.804 |
Estimates for BASDAI are calculated using linear regression and presented as β (95% CI), reflecting the mean difference in outcome per unit change. Estimates for the other outcomes are calculated using logistic regression and are presented as OR (95% CI), indicating the OR of the outcome occurring. Adjusted p values were adjusted for BMI.
BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; bDMARD, biological disease-modifying anti-rheumatic drugs; BMI, body mass index; IBD, inflammatory bowel disease.
Sex-specific analysis: outcomes and adjusted results
Table 3 and table 4 display the distribution of clinical characteristics in our subgroup analysis for women and men, respectively. Individual female cohorts show a similar distribution of most clinical parameters under investigation, as, for example, for HLA-B27 positivity (76.9% vs 82.4%, p=0.738) and rates of a positive family history for SpA (25.0% vs 32.4%, p=0.619). Age revealed to be different in female patients with an atypical SIJ with a mean age of 40.7 years (SD 12.8), while those with a normal SIJ had a mean age of 35.0 years (SD 10.9) (p=0.031). Additionally, differences in BMI of the female cohorts were noted: 23.6 (SD 5.5) versus 25.9 (SD 5.3) (p=0.063), again the p value was below the predefined threshold for inclusion in the regression analysis. Consequently, age and BMI were adjusted for in a female-sex-specific regression analysis, as illustrated in table 5. No statistically significant results were noted in this outcome evaluation. Uveitis was the only outcome variable close to reaching statistical significance, observed more often in female patients with an atypical joint than with a normal joint (adjusted OR=5.174 (95% CI 1.464, 18.285), p=0.011). In the male cohort distribution of clinical parameters was similar, no statistically significant differences were detected. In the male cohort, no values were adjusted since no variables surpassed the established threshold. Further details on sex-specific differences in the male cohort as shown in table 6.
Table 3Clinical characteristics in women with normal and atypical sacroiliac joints
| Normal joint (n=52) | Atypical joint (n=34) | P value | |
| Age, years (mean (SD)) | 35.0 (10.9) | 40.7 (12.8) | 0.031 |
| BMI (mean (SD)) | 23.6 (5.5) | 25.9 (5.3) | 0.063 |
| Symptom duration, months (mean (SD)) | 115.7 (110.9) | 130.8 (108.1) | 0.534 |
| HLA-B27 positivity (n (%)) | 40 (76.9) | 28 (82.4) | 0.738 |
| Elevated CRP (n (%)) | 18 (34.6) | 16 (47.1) | 0.353 |
| Positive family history of SpA (n (%)) | 13 (25.0) | 11 (32.4) | 0.619 |
| bDMARD therapy (n (%)) | 3 (5.8) | 4 (11.8) | 0.555 |
| Cohort (%) | 0.812 | ||
| GESPIC-AS cohort (%) | 18 (34.6) | 13 (38.2) | |
| GESPIC-Crohn cohort (%) | 2 (3.8) | 0 (0.0) | |
| GESPIC-uveitis cohort (%) | 12 (23.1) | 7 (20.6) | |
| OptiRef cohort (%) | 6 (11.5) | 6 (17.6) | |
| SIMACT cohort (%) | 10 (19.2) | 5 (14.7) | |
| VNC-SWI cohort (%) | 4 (7.7) | 3 (8.8) |
Elevated CRP was defined as any value equal or above 5 mg/L.
AS, ankylosing spondylitis; bDMARD, biological disease-modifying anti-rheumatic drug; BMI, body mass index; CRP, C-reactive protein; GESPIC, German Spondyloarthritis Inception Cohort; HLA-B27, human leucocyte antigen B27; OptiRef, Identification of Optimal Referral Strategy for Early Diagnosis of axial spondyloarthritis ; SIMACT, sacroiliac joint MRI and CT; SpA, spondyloarthritis; VNC-SWI, virtual non-contrast susceptibility-weighted imaging.
Table 4Clinical characteristics in men with normal and atypical sacroiliac joints
| Normal joint (n=52) | Atypical joint (n=27) | P value | |
| Age, years (mean (SD)) | 37.2 (11.5) | 36.2 (13.0) | 0.745 |
| BMI (mean (SD)) | 25.2 (3.7) | 25.6 (3.1) | 0.623 |
| Symptom duration, months (mean (SD)) | 124.2 (104.7) | 98.5 (82.9) | 0.272 |
| HLA-B27 positivity (n (%)) | 48 (92.3) | 22 (81.5) | 0.288 |
| Elevated CRP (n (%)) | 24 (46.2) | 13 (48.1) | >0.999 |
| Positive family history of SpA (n (%)) | 15 (28.8) | 5 (18.5) | 0.466 |
| bDMARD therapy (n (%)) | 9 (17.3) | 2 (7.4) | 0.388 |
| Cohort (%) | 0.741 | ||
| GESPIC-AS cohort (%) | 16 (30.8) | 9 (33.3) | |
| GESPIC-Crohn cohort (%) | 3 (5.8) | 0 (0.0) | |
| GESPIC-uveitis cohort (%) | 10 (19.2) | 6 (22.2) | |
| OptiRef cohort (%) | 9 (17.3) | 5 (18.5) | |
| SIMACT cohort (%) | 9 (17.3) | 6 (22.2) | |
| VNC-SWI cohort (%) | 5 (9.6) | 1 (3.7) |
Elevated CRP was defined as any value equal or above 5 mg/L.
AS, ankylosing spondylitis; bDMARD, biological disease-modifying anti-rheumatic drug; BMI, body mass index; CRP, C-reactive protein; HLA-B27, human leucocyte antigen B27; OptiRef, Identification of Optimal Referral Strategy for Early Diagnosis of axial spondyloarthritis ; SIAMCT, sacroiliac joint MRI and CT; SpA, spondyloarthritis; VNC-SWI, virtual non-contrast susceptibility-weighted imaging.
Table 5Regression analysis of clinical characteristics in women with normal and atypical sacroiliac joints
| Unadjusted estimate (95% CI) | Unadjusted P value | Adjusted estimate (95% CI) | Adjusted p value | |
| BASDAI | −0.099 (−0.839, 0.641) | 0.793 | −0.224 (–0.972, 0.524) | 0.559 |
| Inflammatory back pain | OR 0.757 (0.211, 2.706) | 0.668 | OR 0.747 (0.203, 2.749) | 0.660 |
| Uveitis ever | OR 5.739 (1.648, 19.985) | 0.006 | OR 5.174 (1.464, 18.285) | 0.011 |
| Psoriasis ever | OR 2.816 (0.626, 12.662) | 0.177 | OR 2.577 (0.562, 11.82) | 0.223 |
| IBD ever | OR 1.022 (0.266, 3.928) | 0.974 | OR 1.297 (0.32, 5.258) | 0.716 |
| Peripheral arthritis current | OR 1.378 (0.42, 4.52) | 0.597 | OR 1.213 (0.361, 4.082) | 0.755 |
| Enthesitis current | OR 1.25 (0.475, 3.293) | 0.652 | OR 1.282 (0.476, 3.455) | 0.623 |
| Dactylitis current | OR 3.187 (0.278, 36.597) | 0.352 | OR 2.87 (0.244, 33.739) | 0.402 |
Estimates for BASDAI are calculated using linear regression and presented as β (95% CI), reflecting the mean difference in outcome per unit change. Estimates for the other outcomes are calculated using logistic regression and are presented as OR (95% CI), indicating the OR of the outcome occurring. Adjusted p-values were adjusted for BMI.
BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; bDMARD, biological disease-modifying anti-rheumatic drugs; BMI, body mass index; IBD, inflammatory bowel disease.
Table 6Regression analysis of clinical characteristics in men with normal and atypical sacroiliac joints
| Unadjusted estimate (95% CI) | Unadjusted p value | |
| BASDAI | −0.019 (–0.955, 0.916) | 0.968 |
| Inflammatory back pain | OR 0.476 (0.166, 1.369) | 0.169 |
| Uveitis ever | OR 0.525 (0.132, 2.096) | 0.362 |
| Psoriasis ever | OR 0.597 (0.147, 2.419) | 0.470 |
| IBD ever | OR 0 (0,>1000) | 0.997 |
| Peripheral arthritis current | OR 0.267 (0.055, 1.292) | 0.101 |
| Enthesitis current | OR 1.086 (0.324, 3.634) | 0.894 |
| Dactylitis current | OR 0.628 (0.062, 6.346) | 0.694 |
Estimates for BASDAI are calculated using linear regression and presented as β (95% CI), reflecting the mean difference in outcome per unit change. Estimates for the other outcomes are calculated using logistic regression and are presented as OR (95% CI), indicating the OR of the outcome occurring.
BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; bDMARD, biological disease-modifying anti-rheumatic drugs; BMI, body mass index; IBD, inflammatory bowel disease.
Sensitivity analysis in a subset of bDMARD-naïve patients
A sensitivity analysis was conducted in the subset of bDMARD-naïve patients (n=147, 89%). Clinical characteristics were analysed both in the overall cohort and stratified by sex, as displayed in online supplemental tables 1.2, 2.2 and 3.2. Detailed results are provided in online supplemental tables 1.3, 2.3 and 3.3. The outcomes of these analyses were consistent with our initial observations, showing no statistically significant differences in the evaluated features.
Outcomes according to subtype of atypical sacroiliac joint
Finally, we assessed the differences in frequency of the documented clinical parameters according to the specific joint shapes in a descriptive manner: accessory SIJ (n=4), crescent-shaped ilium (n=34) and iliosacral complex (n=23). IBP was found to be significantly different among the three analysed subtypes (88.2% in the crescent-shaped ilium group vs 73.9% in the iliosacral complex group vs 0.0% in the accessory joint group, p<0.001). Apart from this finding, no other statistically significant distinctions were observed between the subgroups. Details on this subanalysis are displayed in online supplemental table 4.
Discussion
To the best of our knowledge, this is the first analysis investigating potential differences in clinical presentation between patients with axSpA with and without atypical SIJ forms.
Our analysis did not yield tangible differences between both groups when compared individually. Reassuringly, the results of our analysis, especially the presence of disease activity according to laboratory findings by CRP or clinically according to the BASDAI score in patients with atypical joint forms, question the hypothesis, that the imaging findings in atypical joints lead to false positive diagnoses.36 These results accentuate the possibility of mechanical and inflammatory pathways concomitantly modulating disease in axSpA, as assumed in previous studies.10 11 Similarly, all other known SpA features were equally prevalent in patients with atypical SIJs, indicating that no distinct phenotype could be identified between the two cohorts based on anatomical sacroiliac differences. In the performed subanalysis according to joint shape classification of back pain as inflammatory showed to be statistically significantly different among the subgroups, most frequently documented in patients with a crescent-shaped ilium and lacking any prevalence in the group with an accessory joint. An individual analysis in a larger cohort should be performed to confirm this result, as it is important to note, that our accessory SIJ group consisted of only four patients, which remains a viable alternate justification for the low prevalence of this clinical characteristic in the respective subgroup.
Results from sex-disaggregated analyses were comparatively less revealing, showing no statistically significant differences in the evaluated clinical parameters. In the female cohort, uveitis revealed an increased frequency in those with atypical SIJs with a rather low p value, however without reaching statistical significance after correction for multiple testing with the Bonferroni method. The possible result of women with atypical SIJs being more frequently affected by uveitis than women with normal joints would be interesting to be validated in larger studies, as the exploratory nature of our analysis does not allow a conclusion to be drawn in this case. A possible explanation for the frequency of this EMM in this specific cohort is the fact that a diagnosis of acute anterior uveitis represents one of the inclusion criteria for the GESPIC-uveitis-subcohort, possibly contributing to the increased frequency of this EMM in our overall study population.
Although analysed patients differed in inclusion criteria and cohort-specific methodologies, results regarding BMI and bDMARD therapy display no statistically relevant differences between the matched cohorts. Most included patients were bDMARD-naïve at the time of the investigation, enabling a homogenous comparison of groups. Despite matching patients based solely on age and sex, we are aware that other confounders, such as BMI or therapy, influence the clinical picture of axSpA.37 38 In order to question and discuss the impact cohort discrepancies might have on our regressively analysed findings, we included adjusted results according to BMI for women. Additionally, we performed a sensitivity analysis on a smaller subset of bDMARD-naïve patients as online supplemental data. As for the results presented initially in the overall cohort, no statistically significant differences in the distribution of clinical characteristics among bDMARD-naïve groups with atypical and normal SIJs were observed. When analysing groups according to specific joint variants, no statistically significant differences were stated. However, the rather small sample size becomes apparent, as only four patients have an accessory SIJ. Differences in the classification of back pain as inflammatory are also statistically significant in this group, likely emphasised by the limited data source.
This analysis has several limitations to its generalisability. First, despite drawing from different patient cohorts, our study design with age-matched and sex-matched controls limits the available data sets substantially, thus reducing the statistical power. This was also the reason why matching was performed for the overall presence of atypical joint form rather than for each shape separately, although we concede that the impact of each of the variants in question may differ. A variant-specific analysis in a larger cohort would be valuable to further validate our findings, especially given the low frequency of certain variants, such as in the subgroup with accessory SIJ types, which included only four patients. Additionally, our approach of using matching followed by multivariate regression may have resulted in a more cautious analysis, as the exclusion of unmatched individuals during the matching process, coupled with regression on the reduced sample, likely impacted statistical power. Future propensity score analyses could be more inclusive, as the incorporation of cofounders before matching could further enhance the homogeneity in the compared cohorts. Selection bias may arise from the inclusion and exclusion criteria, however, matching of patients by age and sex and inclusion of eligible patients with complete data attempted to minimise this effect. While this approach ensures comparability, it may limit the generalisability of findings to broader populations. Furthermore, certain clinical information had to be gathered from electronic patient records, with the known limitations of this source of information, such as the lack of standardised/systematic data documentation. For instance, the history of childbirth in women, which is associated with mechanical joint disease (eg, osteitis condensans ilii), was unavailable but would be valuable for future analyses.39 Similarly, information on physical activity, a cornerstone of axSpA therapy that improves quality of life, spinal mobility and reduces disease activity, was not systematically documented.11 Including these variables in future studies would enhance understanding of their impact on clinical outcomes. Tied in with this fact, it is important to acknowledge that, although data collection in these cohorts was conducted prospectively and in a standardised manner, the current analysis represents a post hoc evaluation of pooled data from different cohorts. This distinction highlights the inherent challenges in interpreting the clinical picture, as retrospective analyses may not always fully capture the clinical reality as initially recorded. We applied mean imputation for variables with minor proportions of missing data (BASDAI, BMI and symptom duration). Although we acknowledge that mean imputation has limitations, as it may underestimate variability and could potentially bias results under certain conditions, the overall low proportion of missing data across all variables minimises the risk of significant bias. Finally, longitudinal data, for example, on patients with axSpA with limited response to therapy, would further illuminate the connection between clinical symptoms and anatomical joint form variations.
In summary, our findings suggest that clinical phenotypes do not significantly differ between patients with axSpA with normal and atypical SIJs and rather weaken the role of these variants in radiological disease mimicking. We believe additional research on the topic is crucial to further evaluate the need for diagnosing such anatomical variations and clarify their individual influence on symptomatic joint disease.
Data availability statement
Data are available upon reasonable request. Sharing of detailed data by the corresponding author is possible upon justified request.
Ethics statements
Patient consent for publication
Not applicable.
Ethics approval
The ethics review board of the Charité – Universitätsmedizin Berlin approved the included studies respectively before study initiation (EA4/161/15, EA4/170/16, EA1/0886/16, EA1/073/10, NCT01277419). Participants gave informed consent to participate in the study before taking part.
X @ProftDr
Contributors Conception and design: CDA, FP, KZ, ML. Acquisition of data: CDA, JR, VRR, DP, FP. Analysis and interpretation of data: CDA, KZ, FP. Drafting the article: CDA, KZ, FP, ML, STU. Critically revising the article: all authors. Preparation of graphical content: STU, CDA, KZ, ML. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: KZ, FP, TD, DP. Statistical analysis: KZ, CDA, ML. Administrative/technical/material support: KZ, JG. Study supervision: STU, FP, KZ. Study guarantor: FP.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None of the authors report funding by organisations that may gain or lose financially through the publication of this manuscript. Outside the submitted work, KZ reports funding (research grant) from the Assessment of Spondyloarthritis international Society (ASAS) during the conduct of this study. STU is a participant in the BIH-Charité Junior Digital Clinician Scientist Program funded by the Charité–Universitätsmedizin Berlin and the Berlin Institute of Health. TD reports funding (research grant) from the Berlin Institute of Health and from Canon Medical Systems; he also reports lecture fees from Canon Medical, UCB, Biocad and Novartis pharma, outside the submitted work and is part of the advisory board of Eli Lilly. VRR reports personal fees from AbbVie and the Falk Foundation. JR is a participant in the BIH-Charité Clinician Scientist Programme funded by the Charité–Universitätsmedizin Berlin and the Berlin Institute of Health, and reports personal fees from Janssen and support for attending meetings/travel from AbbVie, Novartis and Janssen outside the submitted work. DP reports grants and personal fees from AbbVie, Eli Lilly, MSD, Novartis, Pfizer and personal fees from Bristol-Myers Squibb, Roche, UCB, Biocad, GlaxoSmithKline, Gilead, Janssen, Samsung Bioepis and UCB outside the submitted work. FP reports grants and personal fees from Novartis, Eli Lilly and UCB and personal fees from AbbVie, AMGEN, BMS, Celgene, Janssen, Hexal, Medscape, MSD, Pfizer and Roche outside the presented work. For all other authors, none were reported.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
1 Wong M, Sinkler MA, Kiel J. Anatomy, abdomen and pelvis, sacroiliac joint, disclosure: margaret sinkler declares no relevant financial relationships with ineligible companies. disclosure: john kiel declares no relevant financial relationships with ineligible companies. StatPearls. Treasure Island (FL) ineligible companies; 2023.
2 Sieper J, Poddubnyy D. Axial spondyloarthritis. Lancet 2017; 390: 73–84. doi:10.1016/S0140-6736(16)31591-4
3 Sieper J, van der Heijde D, Landewé R, et al. New criteria for inflammatory back pain in patients with chronic back pain: a real patient exercise by experts from the Assessment of SpondyloArthritis international Society (ASAS). Ann Rheum Dis 2009; 68: 784–8. doi:10.1136/ard.2008.101501
4 Rudwaleit M, Metter A, Listing J, et al. Inflammatory back pain in ankylosing spondylitis: a reassessment of the clinical history for application as classification and diagnostic criteria. Arthritis Rheum 2006; 54: 569–78. doi:10.1002/art.21619
5 Rudwaleit M, van der Heijde D, Landewé R, et al. The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part II): validation and final selection. Ann Rheum Dis 2009; 68: 777–83. doi:10.1136/ard.2009.108233
6 de Winter JJ, van Mens LJ, van der Heijde D, et al. Prevalence of peripheral and extra-articular disease in ankylosing spondylitis versus non-radiographic axial spondyloarthritis: a meta-analysis. Arthritis Res Ther 2016; 18: 196. doi:10.1186/s13075-016-1093-z
7 Zeboulon N, Dougados M, Gossec L. Prevalence and characteristics of uveitis in the spondyloarthropathies: a systematic literature review. Ann Rheum Dis 2008; 67: 955–9. doi:10.1136/ard.2007.075754
8 van der Heijde D, Lie E, Kvien TK, et al. ASDAS, a highly discriminatory ASAS-endorsed disease activity score in patients with ankylosing spondylitis. Ann Rheum Dis 2009; 68: 1811–8. doi:10.1136/ard.2008.100826
9 Vleeming A, Schuenke MD, Masi AT, et al. The sacroiliac joint: an overview of its anatomy, function and potential clinical implications. J Anat 2012; 221: 537–67. doi:10.1111/j.1469-7580.2012.01564.x
10 McGonagle D, Thomas RC, Schett G. Spondyloarthritis: may the force be with you? Ann Rheum Dis 2014; 73: 321–3. doi:10.1136/annrheumdis-2013-203924
11 Perrotta FM, Lories R, Lubrano E. To move or not to move: the paradoxical effect of physical exercise in axial spondyloarthritis. RMD Open 2021; 7: e001480. doi:10.1136/rmdopen-2020-001480
12 Prassopoulos PK, Faflia CP, Voloudaki AE, et al. Sacroiliac joints: anatomical variants on CT. J Comput Assist Tomogr 1999; 23: 323–7. doi:10.1097/00004728-199903000-00029
13 Ziegeler K, Kreutzinger V, Proft F, et al. Joint anatomy in axial spondyloarthritis: strong associations between sacroiliac joint form variation and symptomatic disease. Rheumatology (Oxford) 2021; 61: 388–93. doi:10.1093/rheumatology/keab318
14 Ziegeler K, Ulas ST, Poddubnyy D, et al. Anatomical variation of the sacroiliac joint carries an increased risk for erosion and bone marrow oedema in axial spondyloarthritis. Rheumatology (Oxford) 2023; 62: 1117–23. doi:10.1093/rheumatology/keac282
15 Vereecke E, Jans L, Herregods N, et al. Association of anatomical variants of the sacroiliac joint with bone marrow edema in patients with axial spondyloarthritis. Skeletal Radiol 2024; 53: 507–14. doi:10.1007/s00256-023-04435-z
16 El Rafei M, Badr S, Lefebvre G, et al. Sacroiliac joints: anatomical variations on MR images. Eur Radiol 2018; 28: 5328–37. doi:10.1007/s00330-018-5540-x
17 Schett G, Lories RJ, D’Agostino M-A, et al. Enthesitis: from pathophysiology to treatment. Nat Rev Rheumatol 2017; 13: 731–41. doi:10.1038/nrrheum.2017.188
18 Ziegeler K, Kreutzinger V, Diekhoff T, et al. Impact of age, sex, and joint form on degenerative lesions of the sacroiliac joints on CT in the normal population. Sci Rep 2021; 11: 5903. doi:10.1038/s41598-021-85303-5
19 Badr S, Khizindar H, Boulil Y, et al. Anatomical Variants of the Sacroiliac Joint. Semin Musculoskelet Radiol 2023; 27: 221–5. doi:10.1055/s-0043-1761954
20 Song R, Lee S, Lee SH. Progressive sacroiliitis due to accessory sacroiliac joint mimicking ankylosing spondylitis: A case report. Medicine (Baltimore) 2019; 98: e15324. doi:10.1097/MD.0000000000015324
21 Zejden A, Jurik AG. Anatomy of the sacroiliac joints in children and adolescents by computed tomography. Pediatr Rheumatol Online J 2017; 15: 82. doi:10.1186/s12969-017-0210-0
22 Ziegeler K, Hermann KGA, Diekhoff T. Anatomical Joint Form Variation in Sacroiliac Joint Disease: Current Concepts and New Perspectives. Curr Rheumatol Rep 2021; 23: 60. doi:10.1007/s11926-021-01033-7
23 Rudwaleit M, Haibel H, Baraliakos X, et al. The early disease stage in axial spondylarthritis: results from the German Spondyloarthritis Inception Cohort. Arthritis Rheum 2009; 60: 717–27. doi:10.1002/art.24483
24 Poddubnyy D, Haibel H, Listing J, et al. Baseline radiographic damage, elevated acute-phase reactant levels, and cigarette smoking status predict spinal radiographic progression in early axial spondylarthritis. Arthritis Rheum 2012; 64: 1388–98. doi:10.1002/art.33465
25 Poddubnyy D, Rudwaleit M, Haibel H, et al. Rates and predictors of radiographic sacroiliitis progression over 2 years in patients with axial spondyloarthritis. Ann Rheum Dis 2011; 70: 1369–74. doi:10.1136/ard.2010.145995
26 Poddubnyy D, Proft F, Spiller L, et al. Diagnosing axial spondyloarthritis: estimation of the disease probability in patients with a priori different likelihoods of the diagnosis. Rheumatology (Oxford) 2021; 60: 5098–104. doi:10.1093/rheumatology/keab227
27 Proft F, Spiller L, Redeker I, et al. Comparison of an online self-referral tool with a physician-based referral strategy for early recognition of patients with a high probability of axial spa. Semin Arthritis Rheum 2020; 50: 1015–21. doi:10.1016/j.semarthrit.2020.07.018
28 Diekhoff T, Greese J, Sieper J, et al. Improved detection of erosions in the sacroiliac joints on MRI with volumetric interpolated breath-hold examination (VIBE): results from the SIMACT study. Ann Rheum Dis 2018; 77: 1585–9. doi:10.1136/annrheumdis-2018-213393
29 Diekhoff T, Hermann K-GA, Greese J, et al. Comparison of MRI with radiography for detecting structural lesions of the sacroiliac joint using CT as standard of reference: results from the SIMACT study. Ann Rheum Dis 2017; 76: 1502–8. doi:10.1136/annrheumdis-2016-210640
30 Greese J, Diekhoff T, Sieper J, et al. Detection of Sacroiliitis by Short-tau Inversion Recovery and T2-weighted Turbo Spin Echo Sequences: Results from the SIMACT Study. J Rheumatol 2019; 46: 376–83. doi:10.3899/jrheum.171425
31 Deppe D, Hermann K-G, Proft F, et al. CT-like images of the sacroiliac joint generated from MRI using susceptibility-weighted imaging (SWI) in patients with axial spondyloarthritis. RMD Open 2021; 7: e001656. doi:10.1136/rmdopen-2021-001656
32 Kiil RM, Jurik AG, Zejden A. Anatomical variation at the sacroiliac joints in young adults: estimated prevalence by CT and concomitant diagnostics by MRI. Skeletal Radiol 2022; 51: 595–605. doi:10.1007/s00256-021-03843-3
33 Rusman T, van Bentum RE, van der Horst-Bruinsma IE. Sex and gender differences in axial spondyloarthritis: myths and truths. Rheumatology (Oxford) 2020; 59: iv38–46. doi:10.1093/rheumatology/keaa543
34 Bursac Z, Gauss CH, Williams DK, et al. Purposeful selection of variables in logistic regression. Source Code Biol Med 2008; 3: 17. doi:10.1186/1751-0473-3-17
35 Rubin DB, Thomas N. Combining Propensity Score Matching with Additional Adjustments for Prognostic Covariates. J Am Stat Assoc 2000; 95: 573–85. doi:10.1080/01621459.2000.10474233
36 Ulas ST, Proft F, Diekhoff T, et al. Sex-specific diagnostic efficacy of MRI in axial spondyloarthritis: challenging the “One Size Fits All” notion. RMD Open 2023; 9: e003252. doi:10.1136/rmdopen-2023-003252
37 Liew JW, Huang IJ, Louden DN, et al. Association of body mass index on disease activity in axial spondyloarthritis: systematic review and meta-analysis. RMD Open 2020; 6: e001225. doi:10.1136/rmdopen-2020-001225
38 Baraliakos X, Kiltz U, Kononenko I, et al. Treatment overview of axial spondyloarthritis in 2023. Best Pract Res Clin Rheumatol 2023; 37: 101858. doi:10.1016/j.berh.2023.101858
39 Jurik AG, Linauskas A, Kiil RM. Diagnostic features of osteitis condensans ilii by MRI-a systematic literature review. Skeletal Radiol 2025; 54: 423–30. doi:10.1007/s00256-024-04773-6
© 2025 Author(s) (or their employer(s)) 2025. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group. http://creativecommons.org/licenses/by-nc/4.0/ This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ . Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Ziegeler, Katharina 2
; Ulas, Sevtap Tugce 1
; Diekhoff, Torsten 1
; Greese, Juliane 1
; Lindholz, Maximilian 1
; Rademacher, Judith 3
; Valeria Rios Rodriguez 4
; Poddubnyy, Denis 4
; Proft, Fabian 4
1 Department of Radiology, Charité University Hospital Berlin, Berlin, Germany
2 Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, California, USA
3 Department of Gastroenterology, Infectiology and Rheumatology, Charité - Universitätsmedizin Berlin, Berlin, Berlin, Germany; Berlin Institute of Health at Charité, Berlin, Berlin, Germany
4 Department of Gastroenterology, Infectiology and Rheumatology, Charité - Universitätsmedizin Berlin, Berlin, Berlin, Germany