Abstract
Objective
This study (the SpACT study) aimed to investigate the frequency and localisations of different types of spinal new bone formation (NBF) in patients with axial spondyloarthritis (axSpA), psoriatic arthritis (PsA), rheumatoid arthritis (RA) and healthy controls (HCs) using low-dose CT (ldCT).
Methods
Patients with axSpA, PsA or RA, and HC without a history of chronic back pain were included and underwent ldCT of the entire spine. Three readers blinded to all clinical information, including diagnosis, assessed sagittal and coronal images for NBF: (1) marginal syndesmophytes, (2) non-marginal syndesmophytes and (3) osteophytes.
Results
69 participants (33 females, mean age 51.4 years) were included: AxSpA: 30; PsA: 19; RA: 10; HC: 10. Across all groups, the thoracic spine consistently showed the highest number of NBFs, especially for marginal syndesmophytes and osteophytes. Furthermore, on sagittal images, NBF, regardless of type, occurred predominantly at the anterior vertebral corners. Coronal images showed right-sided dominance of NBFs, particularly osteophytes and non-marginal syndesmophytes, whereas marginal syndesmophytes had an almost equal overall distribution. Both sagittal and coronal reconstructions demonstrated high inter-reader reliability (intraclass correlation coefficient >0.9) for almost all groups for detecting any type of NBF.
Conclusion
AxSpA exhibited a distinct NBF pattern characterised by frequent marginal syndesmophytes, particularly in the thoracic spine. In contrast, the most prevalent findings in the other groups were osteophytes, and no consistent NBF pattern was observed to distinguish the groups from each other. Further studies, especially longitudinal assessments using ldCT or equally bone-sensitive imaging methods, are needed to further increase our understanding of NBF patterns.
Keywords: Axial Spondyloarthritis; Arthritis, Psoriatic; Arthritis, Rheumatoid; Spondylitis, Ankylosing
WHAT IS ALREADY KNOWN ON THIS TOPIC
CT is the reference standard for bone visualisation, but its clinical use is limited by radiation exposure.
Low-dose CT (ldCT) allows sensitive detection of syndesmophytes in axial spondyloarthritis (axSpA).
Distinct radiographic patterns of different types of spinal new bone formation (NBF) in patients with various inflammatory arthritides have previously been described, yet systematic corresponding CT data are lacking.
WHAT THIS STUDY ADDS
It provides a systematic ldCT evaluation of spinal NBF across axSpA, psoriatic arthritis, rheumatoid arthritis and healthy controls.
It quantifies and describes the localisation of different types of spinal NBF, offering CT-based insights into bone formation patterns.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
This study demonstrates that ldCT (1.60 mSv, SD 0.43 mSv) delivers detailed multiplanar visualisation of different types of spinal NBF while maintaining low radiation exposure, which may prove relevant for detecting and monitoring disease processes in clinical practice and trials.
Introduction
Axial spondyloarthritis (axSpA) is a chronic rheumatic inflammatory disease that primarily affects the spine and sacroiliac joints. Over time, it can lead to irreversible structural damage significantly impacting function and quality of life.1 Psoriatic arthritis (PsA), a related condition within the spondyloarthritis (SpA) spectrum, involves inflammation in both peripheral and axial joints, as well as entheses. These diverse manifestations contribute to pain, fatigue and functional impairment.2
While axial PsA shares similarities with axSpA, it is also described to exhibit distinct axial joint manifestations.3 However, data on these patterns of axial involvement in PsA, particularly regarding characteristics of spinal structural damage, are limited. In recent years, there has been a growing interest in the type and pattern of axial joint involvement in PsA4 and the appropriate treatment thereof. Consequently, multiple clinical trials in axial PsA are being conducted,5 6 highlighting the need for improved axial structural damage assessment and outcome measures.
In rheumatoid arthritis (RA), the typical manifestations are inflammation of peripheral joints; however, spinal involvement is not uncommon, particularly in the cervical spinal segment.7 The frequency of radiographic findings in the spine of patients with RA varies substantially across studies (0.7–95%),8 making studies investigating alternative methods for structural damage assessment in the spine of patients with RA highly relevant.
Assessing new bone formation (NBF) in the spine—a hallmark of structural damage—and monitoring its progression in patients with inflammatory rheumatic diseases is crucial for long-term management.9 Radiography is the only validated and recommended method for assessing structural damage in the spine,9 but it has limited sensitivity to change and can only detect changes over a minimum of 2 years.10 11 MRI is the preferred imaging method for evaluating spinal inflammation. However, MRI has limitations in detecting NBF, especially in the early stages, because conventional T1-weighted images do not allow differentiation of the NBF (eg, a syndesmophyte) from ligaments (both black) before the NBF becomes so large that it contains fat.12
CT is the gold standard for bone visualisation.13,15 Nonetheless, this method is not routinely used in clinical practice or trials due to the high dose of radiation it involves.16 Recent advancements in noise and radiation reduction have led to the use of low-dose CT (ldCT) in spinal imaging.13 ldCT has been shown to be highly sensitive for detecting syndesmophytes in the spine of patients with axSpA and also to be able to detect progression over 2 years.17
However, data on more detailed assessments using ldCT of the patterns and various types of spinal NBF in different inflammatory rheumatic diseases are lacking.
Traditionally, marginal syndesmophytes, described as slim, vertical ossifications, are considered characteristic of axSpA,18 whereas radiographic studies19 20 have described more bulky, vertically oriented non-marginal bone formations in PsA and reactive arthritis. Additionally, a third pattern of predominantly horizontal NBF, osteophytes, has been associated with degenerative spine disease.20 21 Nonetheless, these distinctions are based primarily on radiography, and there is a lack of scientific data evaluating such NBF patterns using CT, the current gold standard.
This study aims to address the abovementioned knowledge gap by investigating the frequency and localisations of different types of spinal NBF in patients with axSpA, PsA and RA and healthy controls (HCs) using ldCT.
Methods and materials
Design
Patients were recruited from the rheumatological outpatient clinic at Copenhagen University Hospital, Rigshospitalet, Glostrup, Denmark, between July 2023 and August 2024. HCs were recruited through a poster campaign at the same hospital.
Patients who fulfilled the Assessment of Axial Spondyloarthritis International Society classification criteria for axSpA,22 or the ClASsification criteria for Psoriatic Arthritis for PsA23 or the European Alliance of Associations for Rheumatology/American College of Rheumatology 2010 criteria for RA were included.24 Furthermore, HCs who did not have a history of chronic back pain, defined as no back and/or buttock pain in the last 3 months as well as no previous known severe degenerative spine disease (which has required hospitalisation and/or surgery or chronic use of opioids) were included. Additionally, inclusion criteria for all groups were: age ≥18 years at the time of consent in addition to the ability and willingness to provide written informed consent and compliance with the requirements of the study protocol.
For axSpA, half of the patients were required to be radiographic axSpA according to the modified New York Criteria,25 and half of these patients should have at least one syndesmophyte in either the cervical or lumbar spine on radiography assessed by a central SpA expert at baseline.
Patients with PsA should have negative rheumatoid factor and anti-cyclic citrullinated peptide. Patients with PsA were further subgrouped based on the presence/absence of axial symptoms, defined as Bath Ankylosing Spondylitis Score (BASDAI) ≥2 and BASDAI spine pain ≥2 and/or presence of inflammatory back pain.26 27
Patients with RA were not allowed to have psoriasis or a history of psoriasis.
The following exclusion criteria for this CT study applied to all groups: alcohol and/or substance abuse, and the inability to complete examination protocol due to physical or psychological reasons.
All participants underwent ldCT of the entire spine, along with a clinical rheumatological examination and the collection of patient-reported outcomes.
Image acquisition
ldCT examination of the spine (cervical, thoracic and lumbar spine) was performed in a photon-counting CT system (Siemens Naeotom Alpha; Siemens Healthineers, Erlangen, Germany). CT examination z-axis coverage was from 1 cm above dens through the pelvis, with scanning performed in the craniocaudal direction. CT images were reconstructed in the axial plane at 1 mm thickness with 1 mm increments, and in the sagittal and coronal plane at 3 mm slice thickness with 3 mm increments. Automated tube current modulation was used for all examinations, and kilovoltage peak was set at Sn100 kVp. The scan field of view was 50 cm and the displayed field of view was set for the patient body size to include the entire spine. Quantum iteration level 3 was used, the pitch was 0.5, collimation 144×0.4 and rotation time 0.5 s. The requested CT-dose index was 1.36 mGy. Image quality was ensured using dose-modulation techniques based on patient size. The mean effective radiation dose was 1.60 mSv (SD 0.43 mSv). Images were reconstructed with a Br64 kernel.
Image evaluation
Images were pseudonymised and scored independently by three experienced readers blinded to all other imaging and clinical information, including diagnosis.
Before image evaluation, the different types of NBF were defined by the study group. The following definitions were used; (1) Marginal syndesmophyte: Slim NBF originating from the margin/corner of the vertebra with a vertical (>45 degrees compared with the endplate) direction at origin, (2) Non-marginal syndesmophyte: Non-slim NBF that arises away from the margin/corner of the vertebra, with a vertical (>45 degrees) direction at origin and (3) Osteophyte: NBF originating from the corner of the vertebrae with a horizontal (<45 degrees) orientation at origin (figure 1).
Figure 1. Schematic drawing and low-dose CT image of the three defined types of spinal new bone formation: marginal syndesmophyte (1), non-marginal syndesmophyte (2) and osteophyte (3).
Marginal syndesmophytes, non-marginal syndesmophytes and osteophytes were assessed on sagittal and coronal spinal images at each discovertebral unit (DVU, 23 units in total). At each DVU, the evaluation was performed for the individual vertebral entities. On sagittal images, the anterior corners, central locations and posterior corners were assessed, and on coronal images, the right and left sides. Both bridging and non-bridging lesions were scored at all locations. If a bridging lesion was scored, non-bridging lesions could not be scored, and vice versa.
Before scoring, the readers were calibrated in a 2-hour session, where four ldCT scans of the spine were evaluated.
The readers were one expert rheumatologist with extensive expertise in scoring magnetic resonance images and radiographs of the spine, one expert musculoskeletal radiologist and one experienced musculoskeletal senior radiographer.
Patient involvement
Patients were involved as research partners in the preparation of participant information.
Statistical analyses
Descriptive statistics were used to summarise patient characteristics. For each reader, the sum scores of spinal NBF were calculated across predefined vertebral segments and locations (see ‘image evaluation’). Additionally, concordant sum scores per patient were calculated, in which a lesion was considered present per location if at least two of the three readers independently scored the lesion.
Inter-reader reliability was assessed using a two-way random effects single measure model of intraclass correlation coefficient (ICC) based on absolute agreement for sum scores on patient level.
All the above calculations were performed for each group and each type of NBF, both bridging and non-bridging lesions and these combined. Finally, the calculations were also done for any type of NBF, that is, if one reader at a given location scored a syndesmophyte and another scored a non-marginal syndesmophyte at that same location, that would satisfy as a concordant read, in the ‘any type of NBF’ row.
Analyses were performed using R V.4.4.1.
Results
Patients
69 participants (33 females) were included: 30 with axSpA, 19 with PsA, 10 with RA and 10 HCs. Participant characteristics are summarised in table 1.
Table 1. Participant characteristics.
| Characteristics | All participants | Axial spondyloarthritis | Psoriatic arthritis | Rheumatoid arthritis | Healthy control |
|---|---|---|---|---|---|
| n (%) | 69 (100) | 30 (43.5) | 19 (27.5) | 10 (14.5) | 10 (14.5) |
| Sex, female n (%) | 33 (48) | 11 (37) | 11 (58) | 6 (60) | 5 (50) |
| Age (years), mean (±SD) | 51.4 (14.0) | 50.4 (14.0) | 49.7 (14.2) | 61.5 (9.5) | 47.6 (14.8) |
| Symptom duration (years), mean (±SD) | 19.2 (12.9) | 21.9 (15.0) | 16.7 (10.3) | 15.8 (9.5) | NA |
| Time since diagnosis (years), mean (±SD) | 9.5 (9.5) | 9.2 (11.2) | 8.2 (6.9) | 12.8 (8.0) | NA |
| C-reactive protein (mg/mL), mean (±SD) | 5.6 (9.8) | 8.9 (12.8) | 2.6 (2.3) | 4.7 (10.7) | 2.3 (1.6) |
| HLA-B27 positive, n (%) | 31 (46) | 23 (77) | 5 (26) | 1 (11) | 2 (20) |
| Anti-CCP positive, n (%) | 11 (16) | 1 (3) | 0 (0) | 10 (100) | 0 (0) |
| Rheumatoid factor positive, n (%) | 9 (13) | 0 (0) | 0 (0) | 9 (90) | 0 (0) |
| Inflammatory back pain, n (%) | 41 (59) | 24 (80) | 14 (74) | 2 (20) | 0 (0) |
| BASDAI (VAS 0–100), mean (±SD) | 33.9 (21.5) | 38.0 (20.9) | 39.5 (14.6) | 40.7 (21.4) | 3.9 (4.8) |
| BASFI, mean (±SD) | 23.7 (21.5) | 30.8 (20.2) | 24.7 (18.9) | 23.6 (26.6) | 0.8 (1.6) |
| ASDAS, mean (±SD) | 2.1 (1.0) | 2.5 (1.0) | 2.1 (0.6) | 2.1 (0.9) | 0.9 (0.3) |
anti-CCP, anti-citrullinated protein; ASDAS, Ankylosing Spondylitis Disease Activity Score; BASDAI, The Bath Ankylosing Spondylitis Disease Activity Index; BASFI, The Bath Ankylosing Spondylitis Functional Index; HLA-B27, human leucocyte antigen B27; NA, not applicable; VAS, Visual Analogue Scale.
Marginal syndesmophytes
Marginal syndesmophytes were most frequent in the axSpA group (table 2, figure 2), with a total mean of 13.1 lesions scored per patient on sagittal images (figure 3), predominantly located at anterior corners (mean 7.3) and in the thoracic spine (mean 7.8) (figure 4). Coronal images (figure 3) confirmed the thoracic spine as the most affected spinal segment (mean 7.7), with an overall equal distribution of lesions between the right and left sides of the vertebra (table 2).
Table 2. Different types of NBF scored (mean per patient) in the spine of patients with axial spondyloarthritis, psoriatic arthritis, rheumatoid arthritis and healthy controls (concordant reads, all readers, combined bridging and non-bridging new bone formation).
| All participants (n=69) |
Axial spondyloarthritis (n=30) |
Psoriatic arthritis (n=19) |
Rheumatoid arthritis (n=10) |
Healthy control (n=10) |
|
|---|---|---|---|---|---|
| Sagittal plane—stratified by vertebral location | |||||
|
Total Ant/Cent/Post |
Total Ant/Cent/Post |
Total Ant/Cent/Post |
Total Ant/Cent/Post |
Total Ant/Cent/Post |
|
| Marginal (1) syndesmophytes |
6.6 3.8/0.6/2.1 |
13.1 7.3/1.1/4.6 |
1.9 1.4/0.5/0.1 |
2.0 1.6/0.0/0.4 |
0.5 0.4/0.0/0.1 |
| Non-marginal (2) syndesmophytes |
1.1 1.1/NA/0.0 |
1.7 1.7/NA/0.0 |
0.2 0.2/NA/0.0 |
0.2 0.2/NA/0.0 |
1.8 1.8/NA/0.0 |
| Osteophytes (3) |
10.8 9.8/NA/1.0 |
8.5 8.0/NA/0.5 |
11.5 10.1/NA/1.5 |
15.8 14.3/NA/1.5 |
11.0 10.0/NA/1.0 |
| (1) or (2) |
8.4 5.6/0.6/2.2 |
15.7 9.7/1.1/4.9 |
2.4 1.8/0.5/0.1 |
3.6 3.1/0.0/0.5 |
2.9 2.8/0.0/0.1 |
| Any type of NBF |
20.0 16/0.6/3.4 |
25.2 18.5/1.1/5.6 |
14.4 12.2/0.5/1.7 |
20.7 18.6/0.0/2.1 |
14.2 13.1/0.0/1.1 |
| Sagittal plane—stratified by spinal segment | |||||
|
Total Cerv/Thor/Lumb |
Total Cerv/Thor/Lumb |
Total Cerv/Thor/Lumb |
Total Cerv/Thor/Lumb |
Total Cerv/Thor/Lumb |
|
| Marginal (1) syndesmophytes |
6.6 1.3/4.0/1.3 |
13.1 2.6/7.8/2.6 |
1.9 0.2/1.7/0.1 |
2.0 0.5/0.7/0.8 |
0.5 0.1/0.3/0.1 |
| Non-marginal (2) syndesmophytes |
1.1 0.1/0.6/0.3 |
1.7 0.2/0.9/0.7 |
0.2 0.1/0.2/0.0 |
0.2 0.1/0.1/0.0 |
1.8 0.3/1.1/0.4 |
| Osteophytes (3) |
10.8 2.0/6.6/2.1 |
8.5 1.7/5.3/1.5 |
11.5 2.7/6.2/2.6 |
15.8 2.7/10.4/2.7 |
11.0 1.1/7.3/2.6 |
| (1) or (2) |
8.4 1.6/4.9/1.9 |
15.7 3.1/9.1/3.6 |
2.4 0.2/2.1/0.2 |
3.6 0.8/1.3/1.5 |
2.9 0.5/1.7/0.7 |
| Any type of NBF |
20.0 3.9/11.8/4.2 |
25.2 5.1/14.6/5.4 |
14.4 3.0/8.7/2.7 |
20.7 4.1/12.2/4.4 |
14.2 1.7/9.2/3.3 |
| Coronal plane—stratified by vertebral location | |||||
|
Total Right/Left |
Total Right/Left |
Total Right/Left |
Total Right/Left |
Total Right/Left |
|
| Marginal (1) syndesmophytes |
6.2 3.2/3.0 |
12.9 6.5/6.3 |
0.8 0.4/0.4 |
2.2 1.3/0.9 |
0.4 0.4/0.0 |
| Non-marginal (2) syndesmophytes |
1.5 0.9/0.6 |
2.2 1.2/1.0 |
0.3 0.3/0.1 |
1.7 1.1/0.6 |
1.8 1.2/0.6 |
| Osteophytes (3) |
6.9 4.4/2.5 |
5.4 3.3/2.1 |
7.9 5.1/2.9 |
9.5 6.1/3.4 |
7.1 4.8/2.3 |
| (1) or (2) |
8.7 4.6/4.1 |
16.2 8.3/7.9 |
1.8 1.2/0.6 |
4.7 2.7/2.0 |
3.0 1.9/1.1 |
| Any type of NBF |
16.5 9.5/7 |
22.8 12.3/10.5 |
10.3 6.5/3.8 |
15.5 9.4/6.1 |
10.4 6.8/3.6 |
| Coronal plane—stratified by spinal segment | |||||
|
Total Cerv/Thor/Lumb |
Total Cerv/Thor/Lumb |
Total Cerv/Thor/Lumb |
Total Cerv/Thor/Lumb |
Total Cerv/Thor/Lumb |
|
| Marginal (1) syndesmophytes |
6.2 0.8/3.7/1.7 |
12.9 1.7/7.7/3.4 |
0.8 0.0/0.7/0.1 |
2.2 0.0/0.9/1.3 |
0.4 0/0.4/0.0 |
| Non-marginal (2) syndesmophytes |
1.5 0.0/0.8/0.8 |
2.2 0.0/1.0/1.2 |
0.3 0.0/0.3/0.1 |
1.7 0.0/1.0/0.7 |
1.8 0.0/0.9/0.9 |
| Osteophytes (3) |
6.9 0.3/4.4/2.2 |
5.4 0.1/3.9/1.4 |
7.9 0.3/5.1/2.6 |
9.5 0.4/5.7/3.4 |
7.1 0.7/3.3/3.1 |
| (1) or (2) |
8.7 0.9/5.0/2.7 |
16.2 2.1/9.2/4.8 |
1.8 0.0/1.5/0.4 |
4.7 0.0/2.5/2.2 |
3.0 0.0/1.8/1.2 |
| Any type of NBF |
16.5 1.3/10.0/5.3 |
22.8 2.4/13.9/6.6 |
10.3 0.3/6.9/3.2 |
15.5 0.6/8.8/6.1 |
10.4 0.7/5.2/4.5 |
NA: not applicable, the central vertebral location, is only scored for marginal syndesmophytes.
ant, anterior vertebral corner; cent, central vertebral location; cerv, cervical spinal segment; lumb, lumbar spinal segment; NBF, new bone formation; post, posterior vertebral corner; thor, thoracic spinal segment.
Figure 2. Different types of NBF (mean per patient, sagittal images) in the spine of patients with axial spondyloarthritis, psoriatic arthritis, rheumatoid arthritis and healthy controls. NBF, new bone formation.
Figure 3. Different types of NBF on sagittal (a, d, e, f) and coronal spine images (b, c), in patients with axial spondyloarthritis (a–c), psoriatic arthritis (d), rheumatoid arthritis (e) and a healthy control (f). (a) Short thin arrow: slim, vertically oriented (>45°) NBF originating from the vertebral corner, scored as a marginal syndesmophyte. Long thin arrow: bulky NBF arising away from the vertebral corner, with vertical origin, scored as a non-marginal syndesmophyte. (b–c) Thick arrows show the same lesion in two coronal slices. Differences in shape and orientation (vertical in b, more horizontal and bulkier in c) illustrate the difficulty of NBF classification and emphasise the need for clear hierarchical scoring rules. (d) Dotted arrow: a bulky/non-slim NBF originating above the vertebral margin with a vertical direction at origin, scored as a non-marginal syndesmophyte. (e) Asterisk (*): NBF arising from the vertebral corner with a horizontal origin (<45°), scored as an osteophyte. (f) Dashed arrow: Non-marginal syndesmophyte. Dots: Osteophytes. NBF, new bone formation.
Figure 4. Occurrence of different types of new bone formation lesions (ie, osteophytes, non-marginal syndesmophytes and marginal syndesmophytes and any of these lesions), scored at each discovertebral unit, sagittal plane (see online supplemental table 5 for the exact numbers).
In PsA, marginal syndesmophytes were less frequent, with a total mean of 1.9 lesions scored per patient, again mainly at the anterior corner (mean 1.4) and almost exclusively in the thoracic spine (mean 1.7). As for the axSpA group, coronal images showed a balanced right-left distribution of lesions in the PsA group, again primarily in the thoracic spine.
In RA, a total mean of 2.0 marginal syndesmophytes was scored per patient in the sagittal plane, most often at anterior corners (mean 1.6), but with a more even distribution across spinal segments than the axSpA and PsA groups (table 2 and figure 2). On coronal images, lesions were more commonly scored on the right side of the vertebra, with the thoracic and lumbar spinal segments most affected.
In HCs, marginal syndesmophytes were rare, with a total mean of 0.5 lesions scored per individual on sagittal images, almost entirely at anterior corners (mean 0.4) and on coronal images only on the right side of the vertebra.
Non-marginal syndesmophytes
Non-marginal syndesmophytes were most frequently scored in the axSpA group, with a mean of 1.7 per patient on sagittal images, almost exclusively at the anterior corners and with the thoracic spine most involved (mean 0.9). In contrast, very few non-marginal syndesmophytes were scored in the PsA and RA groups. In PsA, a mean of 0.2 lesions per patient was scored on sagittal images, with similarly low scores on coronal images (mean 0.3), although individual reader scores (online supplemental tables 1–3) varied. Patients with RA also had a mean of 0.2 non-marginal syndesmophytes in the sagittal plane, exclusively anterior; however, coronal images revealed more lesions, particularly in the thoracic (mean 1.0) and lumbar spine (mean 0.7), with right-sided dominance (mean 1.1). HCs had somewhat higher scores of non-marginal syndesmophytes than the PsA and RA groups, with a mean of 1.8 lesions per individual on sagittal images, primarily in the thoracic spine (mean 1.1) and merely at anterior corners. Coronal images confirmed a similar mean score of 1.8 per individual, equally distributed between the thoracic and lumbar spine, with most lesions scored on the right side of the vertebra (mean 1.2).
Osteophytes
Osteophytes were the most abundant type of NBF in all participants (table 2) but varied in frequency and distribution across the groups (figure 2). In axSpA, osteophytes were mainly found anteriorly on sagittal images (mean 8.0) and concentrated in the thoracic spine, with a right-sided dominance on coronal images. Patients with PsA exhibited slightly more osteophytes (mean 11.5), again primarily at the anterior corners (mean 10.1) and in the thoracic spine (mean 6.2) in the sagittal plane, and with most lesions scored on the right side of the vertebra in the coronal plane (mean 5.1).
The RA group demonstrated the highest frequency of osteophytes with 15.8 lesions per patient on sagittal images, almost exclusively at the anterior corners (mean 14.3) and most pronounced in the thoracic spine (10.4), followed by an equal distribution in the cervical and lumbar spine (mean 2.7). Coronal images confirmed the RA groups as the one with the most osteophytes (mean 9.5), more frequently scored on the right side of the vertebra (mean 6.1), spanning the cervical (mean 0.4), thoracic (mean 5.7) and lumbar (mean 3.4) spine. HCs also exhibited osteophytes, with numbers comparable to PsA (mean 11.0), nearly all located anteriorly and in the thoracic spine on sagittal images, though more evenly distributed between the lumbar (mean 3.1) and the thoracic (3.3) spine on coronal images, but with fewer overall lesions (mean 7.1).
Any type of new bone formation
When considering any type of NBF together, the axSpA group demonstrated the highest overall burden. On sagittal images, a total mean of 25.2 lesions was scored per patient, of which 18.5 were at the anterior corners and 14.6 in the thoracic spine (figure 4). The mean score per patient for a marginal or a non-marginal syndesmophyte (table 2 and figure 2 ‘(1) or (2)’) was 15.7, with most lesions located at the anterior corners (mean 9.7), followed by posterior corners (mean 4.9). Coronal images showed a mean of 22.8 lesions per patient, with a distribution of 12.3 on the right side of the vertebra and 10.5 on the left side, again primarily in the thoracic spine (mean 13.9), followed by the lumbar spine (mean 6.6).
In the PsA group, the burden was lower, with a mean of 14.4 lesions scored per patient on sagittal images and 10.3 on coronal images, most located in the thoracic spine. Patients with RA had the next highest score of any type of NBF, with a mean of 20.7 lesions on sagittal images and 15.5 on coronal images. The HC group had scores of NBF similar to the PsA group (mean 14.2 on sagittal images and 10.4 on coronal images) (table 2 and figure 2).
Inter-reader agreement
For marginal syndesmophytes on sagittal images, ICCs were 0.90 overall, 0.89 in axSpA, 0.78 in PsA, 0.52 in RA and 0.27 in HC (table 3). Agreement was highest in the thoracic spine and at the anterior vertebral corners, except in PsA, where the posterior corners showed the highest agreement (online supplemental table 4). In the coronal plane, ICCs for detecting marginal syndesmophytes were 0.81/0.79/0.69/0.61/0.18 (all participants/axSpA/PsA/RA/HC).
Table 3. Inter-reader reliability (ICC) for all readers stratified by spinal segment.
| ICC, All readers |
All participants (n=69) |
Axial spondyloarthritis (n=30) |
Psoriatic arthritis (n=19) |
Rheumatoid arthritis (n=10) |
Healthy controls (n=10) |
|---|---|---|---|---|---|
| Sagittal plane |
Total Cervical/Thoracic/Lumbar |
||||
| Marginal syndesmophytes (1) |
0.90 0.82/0.88/0.88 |
0.89 0.81/0.89/0.89 |
0.78 0.30/0.75/0.39 |
0.52 0.72/0.33/0.23 |
0.27 0.12/0.27/0.24 |
| Non-marginal syndesmophytes (2) |
0.19 0.10/0.17/0.22 |
0.16 0.02/0.18/0.21 |
0.06 0.30/0.02/0.19 |
0.12 0.46/0.03/–0.10 |
0.42 0.78/0.31/0.22 |
| Osteophytes (3) |
0.80 0.72/0.78/0.80 |
0.73 0.63/0.72/0.65 |
0.91 0.82/0.85/0.93 |
0.65 0.68/0.72/0.65 |
0.80 0.77/0.80/0.77 |
| (1) or (2) |
0.93 0.82/0.92/0.92 |
0.93 0.81/0.92/0.93 |
0.92 0.43/0.91/0.72 |
0.62 0.77/0.52/0.24 |
0.65 0.72/0.57/0.63 |
| Any type of NBF |
0.94 0.81/0.93/0.93 |
0.93 0.79/0.92/0.94 |
0.98 0.78/0.97/0.93 |
0.80 0.92/0.83/0.39 |
0.93 0.81/0.91/0.91 |
| Coronal plane |
Total Cervical/Thoracic/Lumbar |
||||
| Marginal syndesmophytes (1) |
0.81 0.55/0.82/0.87 |
0.79 0.52/0.80/0.87 |
0.69 - /0.69/0.29 |
0.61 0/0.68/0.54 |
0.18 -/0.25/–0.00 |
| Non-marginal syndesmophytes (2) |
0.12 0.03/0.10/0.17 |
0.08 0.02/0.07/0.13 |
0.09 - /0.00/0.36 |
0.39 0/0.17/0.48 |
0.38 -/0.39/0.58 |
| Osteophytes (3) |
0.77 0.44/0.78/0.77 |
0.69 0.16/0.73/0.62 |
0.81 0.20/0.78/0.89 |
0.89 0.31/0.88/0.70 |
0.86 0.98/0.84/0.70 |
| (1) or (2) |
0.92 0.66/0.92/0.91 |
0.92 0.64/0.92/0.91 |
0.81 - /0.83/0.62 |
0.85 0.00/0.83/0.79 |
0.66 -/0.59/0.62 |
| Any type of NBF |
0.93 0.64/0.94/0.90 |
0.93 0.62/0.93/0.91 |
0.95 0.20/0.96/0.90 |
0.90 0.27/0.93/0.72 |
0.92 0.98/0.91/0.91 |
(-): not calculated as the denominator was 0.
ICC, intraclass correlation coefficient; NBF, new bone formation.
For non-marginal syndesmophytes, the overall ICC in the sagittal plane was 0.19, ranging from 0.42 in HC to 0.06 in PsA (table 3). In axSpA, the lumbar spine demonstrated the highest agreement, whereas in PsA, RA and HC, agreement was highest in the cervical spine. In the coronal plane, the overall ICC was 0.12 for non-marginal syndesmophytes.
For osteophytes on sagittal images, the overall ICC was 0.80, with the highest values in the thoracic spine (axSpA, RA and HC) and lumbar spine (PsA). On coronal images, the overall ICC was 0.77, with the highest agreement in RA and the lowest in axSpA (table 3).
For any type of NBF together, sagittal ICCs were 0.94 overall, 0.93 in axSpA, 0.98 in PsA, 0.80 in RA and 0.93 in HC. On coronal images, the overall ICC was 0.93.
Discussion
This study provides new insights into the patterns and distribution of different types of spinal NBF across three major inflammatory diseases—axSpA, PsA and RA. In contrast to most previous studies investigating types of spinal NBF,21 this study is favoured by the use of ldCT, rather than radiography. Our findings address a knowledge gap and highlight the potential of ldCT and similarly bone-sensitive imaging methods as feasible and sensitive tools for detecting subtle bone changes—changes that are less apparent on radiographs and MRI.
Prior studies using CT to investigate syndesmophytes in the spine of patients with axSpA have reported that this type of NBF was most frequent in the thoracic spine17 28 29 and at the anterior location of the vertebra.17 Another study by Tan et al,30 using lumbar spine CT scans, examined the spatial distribution of syndesmophytes along the vertebral rim and reported that these most often occurred at the posterior-lateral rim. These studies all described distribution patterns of syndesmophytes; however, the separation of these into subtypes (ie, marginal and non-marginal syndesmophytes) and the presence of other types of NBF (ie, osteophytes) and their distribution in different planes and vertebral locations throughout the entire spine were not addressed. Furthermore, to our knowledge, such descriptions have not been reported from CT studies of the spines of patients with PsA and RA.
In this study, we found that, consistent with the existing literature on the distribution of syndesmophytes, the thoracic spine exhibited the highest number of NBFs, particularly for marginal syndesmophytes and osteophytes. Moreover, NBF—regardless of type—occurred predominantly at the anterior vertebral corners on sagittal images. The coronal images showed right-sided dominance of NBFs, particularly for osteophytes and non-marginal syndesmophytes. In contrast, marginal syndesmophytes had an almost equal overall distribution between the right and left sides of the vertebra in axSpA and PsA.
Both sagittal and coronal reconstructions provided complementary insights, with comparable numbers of NBF scored on coronal and sagittal images, except for osteophytes, which were more common on sagittal images, and with excellent inter-reader reliability (ICC >0.9) in both planes for almost all diagnostic groups for detecting any type of NBF. These findings highlight the strength of ldCT for detailed multiplanar visualisation of spinal NBF while maintaining low radiation exposure (1.60 mSv (SD 0.43 mSv)), which may especially prove relevant for future studies investigating the progression over time in spinal structural damage.
The high prevalence of marginal syndesmophytes in patients with axSpA, particularly in the thoracic spine and at the anterior vertebral corners, supports the established understanding that inflammatory lesions in axSpA, often located in the vertebral corners, evolve into new bone through tissue metaplasia, typically as a slender ossification in a symmetric and vertically oriented fashion.31 32
Patients with PsA and RA also exhibited marginal and non-marginal syndesmophytes, although the frequency and distribution of these lesions were markedly different. Patients with PsA demonstrated a lower burden of syndesmophytes and greater inter-reader variability, possibly reflecting both the heterogeneous nature of axial involvement in PsA and the current lack of consensus definitions on PsA related types of NBF.33 Furthermore, it is possible that the type of NBF seen in axial PsA is not as pronounced or classic as in axSpA. A previous study by de Hooge et al concluded that the likelihood of syndesmophyte formation in PsA is low, with a higher probability of developing a syndesmophyte when axial involvement was documented radiographically rather than clinically.34 In the present study, only clinical symptoms of axial disease were required, which most likely have contributed to the lower prevalence of syndesmophytes observed in patients with PsA.
The PsA group showed relatively higher prevalence of osteophytes compared with syndesmophytes, suggesting either a morphological overlap and/or potential misinterpretation due to an almost similar appearance. Moreover, inter-reader agreement in the PsA group improved when marginal and non-marginal syndesmophytes were assessed collectively rather than separately, further suggesting that clearer definitions and more comprehensive reader training could enhance consistency in detecting the individual structural changes in axial PsA.
In RA, osteophytes were the most common finding, but a subset of patients also exhibited marginal syndesmophytes (mean of 2 per patient). However, the relatively lower inter-reader agreement for marginal syndesmophytes in RA (ICC 0.52 sagittal, 0.61 coronal), together with low frequency, suggests some challenges in distinguishing the different types of NBF also in this group. Notably, most of the scored NBF in the RA group were observed in the thoracic spine. This is somewhat unexpected, as cervical spine involvement is more typically described in RA and has been well-documented as the most common site of spinal pathology in these patients.35 This probably suggests that in the examined patients with RA, degenerative disc disease was the main cause of NBF, rather than RA-related findings.
As expected, HCs displayed minimal but still some marginal syndesmophytes, yet had comparable numbers of non-marginal syndesmophytes and osteophytes to the other groups, especially in the thoracic and lumbar spine. The findings in HCs were most likely all degenerative, supporting the need for clear, standardised definitions to avoid potential overinterpretation in clinical and research settings.
Consistent with this, a previous study using radiographs and MRI to assess degenerative spinal lesions in young patients found that the prevalence of these lesions was similar in patients with back pain without axSpA and in those with axSpA.36
Figure 3 illustrates the range of NBF types found within single patients and exemplifies the challenges in classifying lesions, such as distinguishing, for example, an osteophyte from a non-marginal syndesmophyte. In this study, no hierarchical rules were applied when lesion characteristics did meet some but not all criteria for a single type of NBF. Implementation of hierarchical decision-making frameworks (ie, the NBF’s origin—marginal vs non-marginal—should be given greater weight than its shape—bulky vs slim), along with clearer reader guidelines, to enhance scoring consistency and interpretability, may be recommended for future research.
Additional limitations must be acknowledged. First, defining and distinguishing between marginal, non-marginal syndesmophytes and osteophytes remains challenging, despite a structured scoring approach. This ambiguity may have contributed to lower inter-reader agreement for certain lesion types and groups (eg, non-marginal syndesmophytes in PsA and RA). A recent systematic literature review by Gazel et al21 found heterogeneity and variations in the radiographic definitions of NBF within SpA, particularly regarding the shape, location and growth patterns of syndesmophyte subtypes. These variations underline the need for concise and formal NBF definitions in SpA and related conditions. Additionally, the potential presence of diffuse idiopathic skeletal hyperostosis (DISH) was not evaluated, nor was it described as a distinct ossification pattern within the framework of this study.
Second, while concordant scoring enhances reliability, it may underestimate the true lesion burden by excluding discordant findings. Future improvements to the assessment system—particularly the incorporation of hierarchical rules—could help mitigate these issues. It is also possible that a higher number of patients, particularly in the PsA group, could have provided a clearer picture of the pattern of lesions and localisation in patients with PsA.
Lastly, while this study focused on imaging-defined bone formation, correlating these structural findings with clinical and biochemical measures of disease activity or function would provide a more comprehensive understanding of their clinical significance. Furthermore, considering the clinical context is crucial in routine practice, where radiological interpretations are improved by the availability of clinical information.
In conclusion, this study demonstrates that ldCT enables detailed assessment of spinal NBF across inflammatory arthritis conditions. Among the groups assessed—axSpA, PsA, RA and HC—axSpA exhibited a distinct pattern characterised by frequent marginal syndesmophytes, particularly in the thoracic spine. In contrast, the most prevalent findings in the other groups were osteophytes, and no consistent or distinguishing NBF pattern was observed in patients with PsA relative to the other groups. The high prevalence of osteophytes suggests that the primary cause of the observed NBF in the examined participants with PsA and RA was degenerative rather than inflammatory.
Additional studies, especially longitudinal assessments using ldCT or equally bone-sensitive imaging methods, are needed to further increase our understanding of these NBF patterns and to explore their utility in differential diagnostics and disease monitoring.
Supplementary material
Acknowledgements
The authors would like to thank the Danish Rheumatism Association for the salary of STW and for financial support of the study.
Footnotes
Funding: The Danish Rheumatism Association.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants and was approved by the Danish Scientific Health Ethics Committee, H-23016294. Participants gave informed consent to participate in the study before taking part.
Data availability statement
Data are available upon reasonable request.
References
- 1.Sieper J, Poddubnyy D. Axial spondyloarthritis. Lancet. 2017;390:73–84. doi: 10.1016/S0140-6736(16)31591-4. [DOI] [PubMed] [Google Scholar]
- 2.Gladman DD, Antoni C, Mease P, et al. Psoriatic arthritis: epidemiology, clinical features, course, and outcome. Ann Rheum Dis. 2005;64 Suppl 2:ii14–7. doi: 10.1136/ard.2004.032482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Feld J, Chandran V, Haroon N, et al. Axial disease in psoriatic arthritis and ankylosing spondylitis: a critical comparison. Nat Rev Rheumatol. 2018;14:363–71. doi: 10.1038/s41584-018-0006-8. [DOI] [PubMed] [Google Scholar]
- 4.Maksymowych WP, Østergaard M. What Constitutes a Positive MRI for Clinical Trial Recruitment of Psoriatic Arthritis Patients With Axial Involvement? J Rheumatol. 2022;49:72–4. doi: 10.3899/jrheum.211340. [DOI] [PubMed] [Google Scholar]
- 5.Poddubnyy D, Baraliakos X, Van den Bosch F, et al. Axial Involvement in Psoriatic Arthritis cohort (AXIS): the protocol of a joint project of the Assessment of SpondyloArthritis international Society (ASAS) and the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) Ther Adv Musculoskelet Dis. 2021;13:1759720X211057975. doi: 10.1177/1759720X211057975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gladman DD, Mease PJ, Bird P, et al. Efficacy and safety of guselkumab in biologic-naïve patients with active axial psoriatic arthritis: study protocol for STAR, a phase 4, randomized, double-blinded, placebo-controlled trial. Trials. 2022;23:743. doi: 10.1186/s13063-022-06589-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.de Souza MC, de Ávila Fernandes E, Jones A, et al. Assessment of cervical pain and function in patients with rheumatoid arthritis. Clin Rheumatol. 2011;30:831–6. doi: 10.1007/s10067-011-1687-4. [DOI] [PubMed] [Google Scholar]
- 8.Drosos AA, Pelechas E, Voulgari PV. Radiological Findings of the Cervical Spine in Rheumatoid Arthritis: What a Rheumatologist Should Know. Curr Rheumatol Rep. 2020;22:19. doi: 10.1007/s11926-020-00894-8. [DOI] [PubMed] [Google Scholar]
- 9.Ramiro S, Nikiphorou E, Sepriano A, et al. ASAS-EULAR recommendations for the management of axial spondyloarthritis: 2022 update. Ann Rheum Dis. 2023;82:19–34. doi: 10.1136/ard-2022-223296. [DOI] [PubMed] [Google Scholar]
- 10.van der Heijde D, Landewé R. Selection of a method for scoring radiographs for ankylosing spondylitis clinical trials, by the Assessment in Ankylosing Spondylitis Working Group and OMERACT. J Rheumatol. 2005;32:2048–9. [PubMed] [Google Scholar]
- 11.Spoorenberg A, de Vlam K, van der Linden S, et al. Radiological scoring methods in ankylosing spondylitis. Reliability and change over 1 and 2 years. J Rheumatol. 2004;31:125–32. [PubMed] [Google Scholar]
- 12.Østergaard M, Møller JM, Pedersen SJ, et al. Promising Imaging Methods for Assessment of Structural Progression in Axial Spondyloarthritis. J Rheumatol. 2025;52:1322–3. doi: 10.3899/jrheum.2025-0100. [DOI] [PubMed] [Google Scholar]
- 13.Diekhoff T, Hermann KGA, Lambert RG. Future of Low-Dose Computed Tomography and Dual-Energy Computed Tomography in Axial Spondyloarthritis. Curr Rheumatol Rep. 2022;24:198–205. doi: 10.1007/s11926-022-01075-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lambert RGW, Hermann KGA, Diekhoff T. Low-dose computed tomography for axial spondyloarthritis: update on use and limitations. Curr Opin Rheumatol. 2021;33:326–32. doi: 10.1097/BOR.0000000000000803. [DOI] [PubMed] [Google Scholar]
- 15.Stal R, Ramiro S, Baraliakos X, et al. Good construct validity of the CT Syndesmophyte Score (CTSS) in patients with radiographic axial spondyloarthritis. RMD Open. 2023;9:e002959. doi: 10.1136/rmdopen-2022-002959. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sieper J, Rudwaleit M, Baraliakos X, et al. The Assessment of SpondyloArthritis international Society (ASAS) handbook: a guide to assess spondyloarthritis. Ann Rheum Dis. 2009;68 Suppl 2:ii1–44. doi: 10.1136/ard.2008.104018. [DOI] [PubMed] [Google Scholar]
- 17.de Koning A, de Bruin F, van den Berg R, et al. Low-dose CT detects more progression of bone formation in comparison to conventional radiography in patients with ankylosing spondylitis: results from the SIAS cohort. Ann Rheum Dis. 2018;77:293–9. doi: 10.1136/annrheumdis-2017-211989. [DOI] [PubMed] [Google Scholar]
- 18.Baraliakos X, Listing J, Rudwaleit M, et al. Progression of radiographic damage in patients with ankylosing spondylitis: defining the central role of syndesmophytes. Ann Rheum Dis. 2007;66:910–5. doi: 10.1136/ard.2006.066415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Resnick D. Radiology of seronegative spondyloarthropathies. Clin Orthop Relat Res. 1979;1979:38–45. [PubMed] [Google Scholar]
- 20.Jones MD, Pais MJ, Omiya B. Bony overgrowths and abnormal calcifications about the spine. Radiol Clin North Am. 1988;26:1213–34. [PubMed] [Google Scholar]
- 21.Gazel U, Ayan G, Hryciw N, et al. Disease-specific definitions of new bone formation on spine radiographs: a systematic literature review. Rheumatol Adv Pract. 2024;8:rkae061. doi: 10.1093/rap/rkae061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.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. [DOI] [PubMed] [Google Scholar]
- 23.Taylor W, Gladman D, Helliwell P, et al. Classification criteria for psoriatic arthritis: development of new criteria from a large international study. Arthritis Rheum. 2006;54:2665–73. doi: 10.1002/art.21972. [DOI] [PubMed] [Google Scholar]
- 24.Aletaha D, Neogi T, Silman AJ, et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Ann Rheum Dis. 2010;69:1580–8. doi: 10.1136/ard.2010.138461. [DOI] [PubMed] [Google Scholar]
- 25.van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum. 1984;27:361–8. doi: 10.1002/art.1780270401. [DOI] [PubMed] [Google Scholar]
- 26.Garrett S, Jenkinson T, Kennedy LG, et al. A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol. 1994;21:2286–91. [PubMed] [Google Scholar]
- 27.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. [DOI] [PubMed] [Google Scholar]
- 28.Tan S, Yao L, Ward MM. Thoracic Syndesmophytes Commonly Occur in the Absence of Lumbar Syndesmophytes in Ankylosing Spondylitis: A Computed Tomography Study. J Rheumatol. 2017;44:1828–32. doi: 10.3899/jrheum.170340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.de Bruin F, de Koning A, van den Berg R, et al. Development of the CT Syndesmophyte Score (CTSS) in patients with ankylosing spondylitis: data from the SIAS cohort. Ann Rheum Dis. 2018;77:371–7. doi: 10.1136/annrheumdis-2017-212553. [DOI] [PubMed] [Google Scholar]
- 30.Tan S, Dasgupta A, Yao J, et al. Spatial distribution of syndesmophytes along the vertebral rim in ankylosing spondylitis: preferential involvement of the posterolateral rim. Ann Rheum Dis. 2016;75:1951–7. doi: 10.1136/annrheumdis-2015-208802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Chiowchanwisawakit P, Lambert RGW, Conner-Spady B, et al. Focal fat lesions at vertebral corners on magnetic resonance imaging predict the development of new syndesmophytes in ankylosing spondylitis. Arthritis Rheum. 2011;63:2215–25. doi: 10.1002/art.30393. [DOI] [PubMed] [Google Scholar]
- 32.Maksymowych WP, Chiowchanwisawakit P, Clare T, et al. Inflammatory lesions of the spine on magnetic resonance imaging predict the development of new syndesmophytes in ankylosing spondylitis: evidence of a relationship between inflammation and new bone formation. Arthritis Rheum. 2009;60:93–102. doi: 10.1002/art.24132. [DOI] [PubMed] [Google Scholar]
- 33.Østergaard M, Møller-Bisgaard S, Maksymowych WP. Spinal Lesions in Axial Psoriatic Disease: How Should They Be Identified and Quantified by Magnetic Resonance Imaging? J Rheumatol. 2024;51:25–30. doi: 10.3899/jrheum.2024-0291. [DOI] [PubMed] [Google Scholar]
- 34.de Hooge M, Ischenko A, Steinfeld S, et al. Specific descriptions of axial involvement are associated with radiographic damage development after 2 years in psoriatic arthritis patients. Ann Rheum Dis. 2024;83:194–8. doi: 10.1136/ard-2023-224501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Jurik AG. Imaging the spine in arthritis-a pictorial review. Insights Imaging. 2011;2:177–91. doi: 10.1007/s13244-010-0061-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.de Bruin F, ter Horst S, Bloem HL, et al. Prevalence of degenerative changes of the spine on magnetic resonance images and radiographs in patients aged 16-45 years with chronic back pain of short duration in the Spondyloarthritis Caught Early (SPACE) cohort. Rheumatology (Oxford) 2016;55:56–65. doi: 10.1093/rheumatology/kev283. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon reasonable request.