Objective The aim of this work was to investigate the association between disease activity measured by the Ankylosing Spondylitis Disease Activity Score (ASDAS) and radiographic spinal progression in patients with early axial spondyloarthritis (axSpA).
Methods Altogether, 178 patients with definite axSpA (100 with ankylosing spondylitis and 78 with non-radiographic axSpA) were included. Spinal radiographs (baseline and year 2) were assessed according to the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) and for the presence of syndesmophytes. Clinical and lab data were collected at baseline and every 6 months thereafter. Time-averaged (over 2 years) values of the C-reactive protein based ASDAS were calculated.
Results There was a clear positive association between disease activity according to ASDAS and radiographic spinal progression. In the logistic regression analysis, mSASSS progression by ≥2 points over 2 years was significantly associated with the time-averaged ASDAS: unadjusted OR=1.64 (95% CI 1.03 to 2.62), adjusted (for presence of syndesmophytes at baseline, smoking status and intake of non-steroidal anti-inflammatory drugs) OR=1.80 (95% CI 1.04 to 3.13). Syndesmophyte formation/progression demonstrated an even stronger association with the time-averaged ASDAS: unadjusted OR=2.62 (95% CI 1.46 to 4.68), adjusted OR=2.45 (95% CI 1.26 to 4.77).
Conclusions Persisting high disease activity according to the ASDAS is associated with accelerated radiographic spinal progression in early axSpA.
Statistics from Altmetric.com
Previously, we demonstrated that elevated acute phase reactants (C reactive protein, CRP, and erythrocyte sedimentation rate, ESR) were predictive of radiographic spinal progression over 2 years in patients with axial spondyloarthritis (axSpA) included in the GErman SPondyloarthritis Inception Cohort (GESPIC).1 There was, however, no clear association between radiographic spinal progression and clinical disease activity as assessed by the Bath Ankylosing Spondylitis Disease Activity Score (BASDAI). Recently, an association between disease activity measured by the Ankylosing Spondylitis Disease Activity Score (ASDAS)2 ,3—a score combining four patient-reported outcomes and one objective parameter, either CRP or ESR—and radiographic spinal progression was demonstrated for patients with advanced ankylosing spondylitis (AS) in the Outcome in AS International Study (OASIS) cohort.4 Of note, the models with ASDAS as disease activity measure fitted the data from the OASIS cohort better than the models with BASDAI, CRP or BASDAI plus CRP.4 These findings have not been confirmed so far, and there are no published studies addressing this issue in early axSpA. In the present work, we investigated the association between disease activity measured by the ASDAS and radiographic spinal progression over 2 years in patients with early axSpA.
Patients and clinical assessment
A detailed description of the GESPIC cohort has been given elsewhere.5 Briefly, patients with axSpA were included if they had either definite AS with symptom duration of up to 10 years or non-radiographic axSpA (nr-axSpA) with symptom duration of up to 5 years. Clinical and laboratory data were collected at baseline and every 6 months thereafter and radiographic data (cervical spine lateral view, lumbar spine lateral and anteroposterior views) were collected at baseline and after 2 years. Patient's global assessment of disease activity (PG) was included in the study protocol after study initiation, and, therefore, it was collected starting from month 6 on. ASDAS values (CRP based) were calculated for up to four follow-up visits between month 6 and month 24 with the recently proposed correction for low-level (<2 mg/L) CRP.6 Time-averaged values of the ASDAS, BASDAI and CRP over 2 years were also calculated.
Treatment of GESPIC patients has been done according to the local rheumatologist, without any limitations.5 Considering that GESPIC was initiated before anti-tumour necrosis factor α (TNF-α) treatment of AS was approved, only a minority of the included patients did receive anti-TNF-α treatment at baseline and during the first 2 years of follow-up. For the quantification of non-steroidal anti-inflammatory drugs (NSAIDs) intake, the NSAID intake score7 developed by the Assessment of SpondyloArthritis International Society (ASAS) was used.
Spinal radiographs were centrally collected, digitised if necessary, anonymised and subsequently scored independently by two trained readers (DP and HH) in a concealed and randomly selected order according to the modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS).8 In addition to the mSASSS, we assessed syndesmophytes on both lateral views and lumbar anteroposterior view. Meaningful radiographic spinal progression was defined as: (1) worsening of the mSASSS value by ≥2 units after 2 years or (2) development of at least one new syndesmophyte or progression of existing syndesmophytes (formation of a bridging syndesmophyte from two single syndesmophytes) after 2 years. Syndesmophytes were considered to be present only if both readers scored them as such (mSASSS ≥2 for an individual vertebral corner).
For the analysis of the association between disease activity parameters (ASDAS, BASDAI and CRP) and radiographic spinal progression over 2 years, we calculated time-averaged values of the mentioned parameters which were available at up to five time-points during 2 years. The Kruskal–Wallis test, the Mann–Whitney U test and the χ2 test were applied for the comparisons of the subgroups with different disease activity defined according to ASDAS thresholds.3 The Mann–Whitney U test and the Fischer's exact test were applied for the comparison of the radiographic spinal progression rates between AS and nr-axSpA. A logistic regression was used for the analysis of the association between disease activity parameters and radiographic spinal progression. ORs with 95% CIs were calculated.
The study protocol was approved by the ethical committee of the coordinating centre (Charité Universitätsmedizin Berlin, Berlin, Germany) and by all local ethical committees of the participating centres. Written informed consent was provided by all included patients.
A total of 178 patients with definite axSpA (100 with AS and 78 with nr-axSpA) with available radiographic data of the spine at baseline and after 2 years, as well as available clinical data allowing calculation of the CRP-based ASDAS (BASDAI, PG and CRP) were included in the current analysis. Baseline characteristics of the patients included in this analysis (table 1) were similar to those previously reported.1 ,5
A total of 27 (15.2%) patients progressed by ≥2 mSASSS points, and 20 (11.2%) patients developed new syndesmophytes or had progression of existing syndesmophytes after 2 years of follow-up.
There was a clear increase in radiographic spinal progression (figure 1) with increasing disease activity according to the time-averaged ASDAS, although the statistical significance for intergroup differences was reached for syndesmophyte formation/progression only. Patients with very high disease activity as defined by ASDAS had a remarkably high radiographic spinal progression rate according to any definition in comparison to subgroups with lower disease activity.
In the logistic regression analysis, mSASSS progression by ≥2 points over 2 years was significantly associated with the time-averaged ASDAS: OR=1.64 (95% CI 1.03 to 2.62) in the univariable analysis and OR=1.80 (95% CI 1.04 to 3.13) after adjusting for other factors with known impact on radiographic spinal progression (presence of syndesmophytes at baseline, smoking status and NSAID intake score) (table 2).
Syndesmophyte formation/progression demonstrated an even stronger association with the time-averaged ASDAS: OR=2.62 (95% CI 1.46 to 4.68) in the univariable model and OR=2.45 (95% CI 1.26 to 4.77) after adjusting for baseline syndesmophytes, smoking status and NSAID intake score (table 2).
At baseline, four patients were on anti-TNF-α treatment; exclusion of these patients from the analysis did not result in a substantial change of the effect estimates for the time-averaged ASDAS in the adjusted models: OR for mSASSS progression by ≥2 points over 2 years=1.91 (95% CI 1.08 to 3.36); OR for syndesmophyte formation/progression=2.37 (95% CI 1.21 to 4.64). During 2 years of follow-up, 17 patients started anti-TNF-α therapy, however only 12 patients in total received anti-TNF-α therapy for at least 6 months. Inclusion of anti-TNF-α therapy for at least 6 months as a covariate in the adjusted model with model with time-averaged ASDAS resulted in an OR=1.91 (95% CI 1.09 to 3.33) for mSASSS progression by ≥2 points and an OR=2.79 (95% CI 1.39 to 5.58) for syndesmophyte formation/progression.
There was a trend for a positive association between single components of the ASDAS (BASDAI, CRP and PG) and radiographic spinal progression as defined by the mSASSS worsening by two points. Such an association between syndesmophyte formation/progression and CRP as well as between syndesmophyte formation/progression and PG was statistically significant. However, the goodness of fit of the adjusted model (as assessed by the R2 value) with ASDAS was always better as compared with the models with BASDAI, CRP or PG alone and was very close to that of the model with all three single parameters put together (table 2).
Since BASDAI provided a signal for an association with radiographic spinal progression, we calculated crude and adjusted ORs for single components of the index (see online supplementary table S1). Interestingly, the significant association with radiographic spinal progression could be found for the duration of morning stiffness and morning stiffness severity only.
Crude and adjusted odds ratios for the association between the time-averaged total BASDAI score, single BASDAI items and radiographic spinal progression over two years in patients with early axial spondyloarthritis.
Radiographic spinal progression after 2 years (as well as the structural damage extent at baseline) was generally more prominent in the AS subgroup as compared with nr-axSpA subgroup: mSASSS change over 2 years—0.90±2.91 vs 0.56±1.74, respectively, p=0.77; mSASSS progression by ≥2 points—in 20% and 9% of the patients, respectively, p=0.057; syndesmophyte formation/progression—in 17% and 4% of the patients, respectively, p=0.007. However, the logistic regression analysis showed similar effect estimates for the association of radiographic spinal progression with the time-averaged ASDAS in AS and nr-axSpA: adjusted ORs for mSASSS worsening by ≥2 points were 1.81 (95% CI 0.95 to 3.47) and 1.86 (95% CI 0.61 to 5.72) for the AS and nr-axSpA subgroups, respectively; adjusted ORs for syndesmophyte formation/progression were 2.39 (95% CI 1.14 to 4.99) and 3.17 (95% CI 0.47 to 21.18) for the AS and nr-axSpA subgroups, respectively.
In the present study, we could demonstrate that disease activity is associated with radiographic spinal progression in early axSpA, therefore confirming earlier data obtained from patients with advanced AS.4 The time-averaged ASDAS was significantly associated with both mSASSS worsening by ≥2 points and syndesmophyte formation/progression over 2 years indicating that persistently high disease activity should be considered as an important risk factor for radiographic spinal progression at any stage of early axial SpA. Importantly, the model with ASDAS fitted data better than the model with CRP, indicating that radiographic spinal progression is not driven by CRP alone. This could be confirmed by a positive trend for association with radiographic spinal progression observed for BASDAI (morning stiffness severity and duration as components of BASDAI demonstrated the strongest association) and a significant association between PG and syndesmophytes formation/progression.
The association between the time-averaged ASDAS and radiographic spinal progression over 2 years was similar for nr-axSpA and AS that justified pulling both subgroups of axial SpA together. The smaller number of progressors in the nr-axSpA group explains less precise estimates in this subgroup.
The fact that the disease activity as measured by the ASDAS is the factor strongly associated with radiographic spinal progression in axSpA favours early anti-inflammatory treatment in patients with early and active disease in the hope that decreasing the disease activity would also result in slowing down the radiographic progression. This fits the concept of the ‘window-of-opportunity’ in axSpA,9 ,10 and data from two recent observational studies, in which early initiation of treatment with TNF blockers in axSpA was associated with reduced radiographic spinal progression,11 ,12 support this concept. However, this still has to be proven in interventional trials using, for instance, treat-to-target/tight control strategies aimed at remission achievement in axSpA.
In the analysis, we used time-averaged values of ASDAS, CRP, BASDAI and PG as measures of disease activity that do not count for individual variation in the course of the disease and treatment effects; this should be considered as a limitation of our analysis. At the same time, time-averaged values provided the best possible estimation of the inflammation burden over 2 years of follow-up. A substantial number of patients dropped out during a 2-year follow-up (210 out of initially reported 462 patients5 had radiographic follow-up; further 32 patients were excluded because of lack of information required for the calculation of ASDAS). Although we did not find a major difference between patients who dropped out and patients who remained in the study,1 we cannot exclude the fact that the dropouts were not completely at random.
In conclusion, our study has confirmed the association between high disease activity and progression of structural damage in the spine in patients with axSpA. This association seems to be driven by CRP and by other parameters of disease activity such as patient-reported outcome parameters. It should be clarified in future trials whether reduction of disease activity according to ASDAS by intervention would be associated with reduction of radiographic spinal progression in axSpA.
We thank Professor M Leirisalo-Repo, Finland, Professor D van der Heijde, The Netherlands, and Professor M Dougados, France, for scientific advice on the design of the cohort. We are grateful to Beate Buss and Petra Tietz for monitoring the cohort; to Johanna Callhoff, Anja Weiss and Martina Niewerth for the data management support; to Joachim Listing for statistical advice; to Janis Vahldiek and Georg Heine for the X-ray images handling and for the development of the image scoring interface; and to all patients who voluntarily participated in this cohort. We further like to thank the following rheumatologists for inclusion and follow-up of their patients: J Brandt, H Brandt, G-R Burmester, H Deister, E Edelmann, J Emmerich, M Enderlein, E Feist, A Gauliard, E Gromnica-Ihle, F Heldmann, S Hermann, U von Hinüber, Ü Hübner, K Karberg, C Kedor, E Märker-Hermann, H Nüßlein, R Pelle-Lohfink, D Pick, G Reichmuth, E Riehers, M Rihl, R Schmidt, S Schnarr, U Schneider, I-H Song, I Spiller, U Syrbe, V Walz, S Wassenberg, H M Wisseler, H Zeidler and S Zinke.
Handling editor Tore K Kvien
Contributors All authors were involved in data analysis and/or interpretation, drafting the article or revising it critically for important intellectual content. DP had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Funding GESPIC has been financially supported by the German Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung—BMBF). As funding by BMBF was reduced according to schedule in 2005 and stopped in 2007, complementary financial support has been obtained also from Abbott/Abbvie, Amgen, Centocor, Schering-Plough and Wyeth. Since 2010, GESPIC has been supported by Abbvie; additional support has been obtained from ANCYLOSS (grant number FKZ 01EC1002D), ArthroMark (grants numbers FKZ 01EC1009A and FKZ 01EC1401A) and METARTHROS (grant number FKZ 01EC1407A) projects funded by BMBF.
Competing interests None declared.
Ethics approval Ethics committee of the Charité Universitätsmedizin Berlin.
Provenance and peer review Not commissioned; externally peer reviewed.
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.