Background Syndesmophyte formation in ankylosing spondylitis (AS) is still insufficiently understood. Previous studies have suggested an association between MRI vertebral edge inflammation (VEI) and vertebral edge fat deposition (VEFD) and the subsequent development of radiographic syndesmophytes (SYN).
Objectives To investigate the relationship between MRI inflammation and fat deposition at a VE and the development of a SYN at the same VE.
Methods A random 80% sample of the ASSERT database was used for this analysis. ASSERT was a 24-week (wk) RCT comparing infliximab monotherapy and placebo in patients with active AS, with an open extension until 102wks with all patients on infliximab. Spinal MRIs (T1-weighted and STIR images) at baseline, 24wks and 102wks were assessed by 2 readers independently who scored each VE for the presence or absence of VEI and VEFD. Spinal X-rays (baseline and 102wks) were scored by 2 different readers independently using the mSASSS. Data were analysed at the VE level in the 24 VEs assessed by the mSASSS (anterior VEs of the cervical and lumbar spine), using a multi-level approach to adjust for within-patient correlation and MRI reader (GEE for binomial outcomes). Particular emphasis was on a sequence analysis, which aimed at the specific sequence VEI→VEFD→SYN. The following variables were considered potential covariates and adjusted for when appropriate: treatment, gender, presence of SYN/ankylosis at baseline (patient level), HLA-B27, baseline- and time-averaged BASDAI/ASDAS/CRP, age, BMI and disease duration. Readers were unaware of the patient's identity, their treatment, the scores of the other imaging modality, and the true time-order of the images (fully unbiased scores).
Results After excluding VEs with SYN or ankylosis at baseline and non-evaluable VEs, data from 3079 to 3363 vertebral edges, belonging to 182 patients, were analysed. The presence of VEI (adjusted (ad)OR=1.98-1.93) as well as the presence of VEFD (adOR=1.59-2.32) at any time point (TP) was significantly associated with the development of a new SYN. The strength of the association increased by combining VEI and VEFD, both when analysing VEs with VEI and VEFD present across any of the three TP (simultaneously or not) (adOR=2.29-2.73) and also when analysing only newly developed VEFD preceded by inflammation in the same VE at a previous TP (adOR=2.45-3.01) (sequence analysis, VEI→VEFD→SYN). The complete absence of both VEI and VEFD across the three TP was protective for the development of a new SYN (aOR=0.45-0.62) (Table). However, a large proportion of SYN developed in VEs without MRI inflammation or fat degeneration at any of the three TP (42-66%, depending on the combination of MRI- and X-ray reader).
Conclusions Both VEI and VEFD contribute to the development of a new SYN in AS, especially if VEI precedes VEFD. But this typical sequence only partially explains the development of new SYN in AS.
Acknowledgements We thank Janssen Biotech Inc for providing the images and data necessary to perform these analyses. Janssen Biotech Inc had no influence on data analyses, interpretation of results or writing of the abstract.
Disclosure of Interest None declared
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.