Article Text
Abstract
Objective To compare spinal symptoms and spinal/hip mobility at baseline and 2 years in early axial spondyloarthritis (axSpA) and non-axSpA chronic back pain (BP) patients.
Methods Baseline and 2 years data of the SPondyloarthritis Caught Early cohort were analysed. Outcomes assessed: overall BP, BP at night, morning stiffness (MS) intensity, MS duration, occiput-to-wall distance (OWD), cervical rotation, chest expansion, lateral spinal flexion (LSF), modified Schober test (mSchober), intermalleolar distance (IMD) and Bath Ankylosing Spondylitis Metrology Index (BASMI). Linear or zero-inflated negative binomial regression was used to compare 2 years outcomes between groups (adjusting for baseline value, sex, age and use of non-steroidal anti-inflammatory drugs).
Results There were 294 axSpA and 123 non-axSpA patients (mean symptom duration: 13 months). At baseline, non-axSpA patients had worse symptoms and mobility, except OWD (eg, mean(SD): BP at night 3.6 (2.9) axSpA vs 4.6 (2.7) non-axSpA; OWD 0.5 (1.2) vs 0.1 (0.7)). After 2 years, all symptoms and cervical rotation significantly improved in both groups, but LSF and mSchober only in axSpA. In multivariable analyses, axSpA was associated with larger improvements in BP at night (β (95% CI): −0.85 (−1.47; −0.23)), mSchober (0.26 (0.03; 0.50)), IMD (4.86 (1.93; 7.80)) and BASMI (−0.24 (−0.41; −0.08)), and with lower likelihood of a normal OWD (OR (95% CI): 0.09 (0.01; 0.83)).
Conclusion Over 2 years, all spinal symptoms and some mobility measures improved in both groups, but impairments remained prevalent (particularly in non-axSpA). Nevertheless, axSpA was associated with larger improvements in BP at night, mSchober, IMD and BASMI, but with more OWD impairment.
- Axial Spondyloarthritis
- Low Back Pain
- Patient Reported Outcome Measures
Data availability statement
Data are available on reasonable request. Data are available on reasonable request, requiring permission from collaborating centres.
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/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Spinal symptoms and impaired mobility are features of axial spondyloarthritis (axSpA) and non-axSpA chronic back pain (CBP), but comparisons between groups with early disease are lacking.
WHAT THIS STUDY ADDS
After 2 years, although spinal symptoms and some mobility measures improve in both axSpA and non-axSpA patients, impairments remain prevalent, most notably in non-axSpA CBP.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Our study highlights the high disease burden of both early axSpA and non-axSpA CBP that persists even after 2 years of follow-up.
Introduction
Axial spondyloarthritis (axSpA) is a chronic inflammatory disease that may lead to irreversible structural damage of the sacroiliac joints (SIJ) and spine. The main symptom of the disease is chronic back pain (CBP), frequently in combination with substantial morning stiffness.1 In fact, in one study conducted in patients with long-standing disease, stiffness and pain were the most common concerns.2 Pain ensues mostly due to inflammation and structural spinal damage.3 4 Moreover, in addition to natural ageing and degenerative conditions, inflammation and structural spinal damage (the latter predominantly in more advanced disease stages) contribute to reduced spinal mobility in axSpA.3–6 Therefore, improving spinal symptoms and mobility is important, as they may lead to impaired functioning, disability and, consequently, low health-related quality of life.3 7
CBP is also a highly prevalent symptom in the general population, with up to 80% of individuals reporting CBP at some point in life, and is considered the primary cause of disability worldwide.8–10 Furthermore, reduced spinal and hip mobility have been reported in non-axSpA CBP patients.11
The Assessment of SpondyloArthritis international Society (ASAS) recommends assessing back pain and morning stiffness by asking patients to report the intensity of these symptoms (and duration for morning stiffness) on a Numeric Rating Scale (NRS).12 13 For the assessment of spinal and hip mobility, several measures have been proposed and validated,12 and a composite index, the Bath Ankylosing Spondylitis Metrology Index (BASMI), was developed to monitor mobility over time.14 Percentile curves were also defined for these measures in healthy individuals, thus allowing comparisons with the general population.15
In axSpA, spinal mobility has been mostly studied in patients with long-standing disease, often associated with established radiographic spinal damage.16–20 More recently, data have shown that spinal and hip mobility are also frequently impaired in early axSpA, as well as in non-axSpA CBP.11 21 However, very little information is available about the course over time of spinal symptoms and spinal/hip mobility in these two groups of patients. We hypothesised that, compared with non-axSpA CBP, patients with axSpA would have more spinal symptoms and impaired mobility at baseline. Moreover, as specific and effective treatment is available for axSpA, but it is to some extent lacking for non-axSpA CBP, we would expect less symptoms and impairments in mobility after 2 years in those with axSpA. However, to our knowledge, no data comparing the course of these features in early axSpA,22 as compared with non-axSpA CBP patients referred to rheumatology outpatient clinics, are currently available.
The SPondyloarthritis Caught Early (SPACE) cohort recruited patients referred to rheumatologists with a history of CBP of ≥3 months and ≤2 years duration, with onset before 45 years of age. The main goal of the cohort was to make a diagnosis of axSpA (or non-axSpA) early in the disease course. All eligible patients were followed up for 2 years, allowing reliable comparisons between patients with axSpA or non-axSpA CBP over time. Therefore, by using data from the SPACE cohort, we aimed to make comparisons between axSpA and non-axSpA patients in what concerns spinal symptoms and spinal/hip mobility, at baseline and after 2 years.
Methods
SPACE23 is a European observational cohort with recruitment between 2009 and 2016 in the Netherlands, Italy, Sweden and Norway. All adults (≥16 years) referred to rheumatology outpatient clinics (tertiary care) due to CBP of ≥3 months and ≤2 years duration, beginning before the age of 45 years, were included. At recruitment, all patients were subjected to a full clinical (medical history and physical examination, including spinal and hip mobility measures), laboratory (C-reactive protein and HLA-B27) and imaging (radiographs and MRI of the SIJ and spine) diagnostic workup. For classification purposes, images were scored independently by three central readers. Sacroiliitis was defined according to the modified New York criteria for radiographs and ASAS definition for MRI, and a positive finding was only considered in case of agreement between ≥2 readers. Eligible patients were followed up for 2 years and the diagnostic workup was repeated at each visit. Disease activity was assessed by calculating the Axial Spondyloarthritis Disease Activity Score (ASDAS) for those patients with an axSpA diagnosis. Based on clinical judgement, a rheumatologist established a diagnosis of axSpA or non-axSpA CBP with a level of confidence (LoC) ranging from 0 (‘not confident at all’) to 10 (‘very confident’), at each evaluation. A non-axSpA diagnosis meant that CBP could not be attributed to axSpA and an alternative diagnosis was proposed (eg, non-specific CBP, fibromyalgia). Three diagnostic categories were defined based on the 2 years diagnosis, its LoC, (in)consistency over time, and, for non-axSpA categories only, the presence/absence of an alternative diagnosis provided by the rheumatologist: definite axSpA/non-axSpA, most likely axSpA/non-axSpA and possible axSpA/non-axSpA (online supplemental text S1).24 In order to allow reliable comparisons between groups, patients with a diagnosis of possible axSpA/non-axSpA (n=16) were not considered for this analysis. Furthermore, data on ≥1 of the assessed outcomes (described below) had to be available at both baseline and 2 years.
Supplemental material
Spinal symptoms
Spinal symptoms were evaluated by asking patients the intensity of overall back pain (during the entire day), back pain at night and morning stiffness in the previous week on an NRS, ranging from 0 (‘absent’) to 10 (‘unbearable’). The duration of morning stiffness in the previous week was reported by the patient on an NRS ranging from 0 (‘0 hours’) to 10 (‘≥2 hours’). By calculating the mean of the scores recorded for morning stiffness intensity and duration, a composite score was created to assess overall morning stiffness.13 The proportion of patients with back pain (overall and at night) and morning stiffness (intensity and/or duration), arbitrarily defined by all authors as a score >2, was also reported at baseline and 2 years.
Spinal and hip mobility
Five spinal mobility measures (SMM: occiput-to-wall distance (OWD), cervical rotation performed in sitting position, chest expansion, lateral spinal flexion (LSF), modified Schober test (mSchober)) and one reflecting hip mobility (intermalleolar distance (IMD)) were collected according to ASAS recommendations.12 Each measure was performed twice and the best result was recorded. Results were presented in cm with one decimal place, except for cervical rotation and IMD that were recorded in degrees or cm, respectively, and rounded to the nearest integer. The composite index BASMI14 was calculated using the BASMI linear function,25 ranging from 0 to 10 (higher scores imply worse mobility). For patients recruited in Italy and Sweden, tragus-to-wall distance (TWD) was measured instead of OWD. For these patients, OWD was derived from TWD by subtracting 11.1 cm, in accordance with the formula derived from healthy individuals.15 21 Negative values were set to zero. Impairments for each spinal and hip mobility measure were defined as being below the 2.5th percentile curves derived from healthy individuals (>0 cm for OWD and >97.5th percentile for BASMI).15 This strict cut-off was applied in order to reliably catch patients with definitely abnormal mobility and is also in line with the WHO definition of abnormal growth based on the children’s growth curves.26 Nevertheless, the proportion of axSpA and non-axSpA patients with an impaired measure using the less restricted definition of being below the 5th or above the 95th percentile curve (as appropriate) was also calculated.
Statistical analysis
Categorical variables are presented as frequencies and proportions, and continuous variables as means and respective SD. Baseline patient and disease characteristics, as well as treatment at baseline and 2 years, were reported for both axSpA and non-axSpA patients and comparisons between groups were performed using a χ2 test for dichotomous variables and an unpaired t-test for continuous variables. For each group (axSpA and non-axSpA), baseline and 2 years outcomes for spinal symptoms and spinal/hip mobility were compared using a paired t-test (Wilcoxon signed rank test for OWD). Additionally, except for OWD, 2 years outcomes were compared between groups (axSpA vs non-axSpA) by conducting linear regression models. The dependent variable was the 2-year value of each spinal symptom and mobility measure assessed and the diagnosis the independent variable, and adjustments were made for the baseline value of the respective outcome (equivalent to modelling a 2-year change in the outcome), sex, age and use of non-steroidal anti-inflammatory drugs (NSAIDs) during the entire 2-year follow-up (‘ever use’ vs ‘never use’). Sex was considered as a potential confounder as the subjective report of spinal symptoms and the performance of mobility measures may differ between males and females. As a consequence of degenerative conditions or other age-related comorbidities, age may also influence these outcomes, even though, in SPACE, patients were overall young. Furthermore, naturally, treatment with NSAIDs may improve pain and stiffness and, therefore, differences may be expected between those ever or never taking these drugs over time. For OWD, due to highly skewed data with an excess of zeros, a zero-inflated negative binomial (ZINB) model was performed instead, considering the same dependent and independent variables and confounders. ZINB provides two models (and respective regression coefficients), assuming that zeros derive from two processes: a logit model for the ‘certain zero’ cases (positive coefficients imply a higher likelihood of scoring zero) and a negative binomial model for the ‘not certain zero’ cases (higher coefficients indicate larger impairments).27 28 Coefficients were exponentiated and results are reported in OR and incidence rate ratio (IRR) for the logit and negative binomial models, respectively. Statistical significance was defined as a p<0.05. Statistical analysis and graphs were made using STATA software V.16 (Statacorp).
Results
Study population
In the SPACE cohort, 434 patients with an established axSpA and non-axSpA diagnosis at the 2 years follow-up had undergone baseline and 2 years visits (online supplemental figure S1). Of these, 417 patients (294 axSpA and 123 non-axSpA) had available data at both time points on ≥1 of the assessed outcomes (overall back pain, back pain at night, morning stiffness intensity, morning stiffness duration, OWD, cervical rotation, chest expansion, LSF, mSchober, IMD or BASMI). Although age and symptom duration were similar between groups, axSpA (vs non-axSpA) patients were more frequently male (52% vs 24%) and had more SpA features (mean (SD): 5 (2) vs 3 (1)), such as sacroiliitis on imaging (positive MRI: 39% vs 3.3%; positive mNY: 5.1% vs 0%) and elevated C-reactive protein (35% vs 23%) (table 1).
Supplemental material
Spinal symptoms
Data on spinal symptoms are presented in table 2. The mean (SD) intensity of overall back pain was 4.0 (2.4) in axSpA and 5.5 (2.2) in non-axSpA patients at baseline and it significantly decreased to 2.6 (2.3) and 3.9 (2.6) at 2 years, respectively. For both groups and at both time points, back pain was most frequently localised to the buttocks (58% axSpA vs 39% non-axSpA, at baseline) or lumbar spine (21% axSpA vs 35% non-axSpA, at baseline) (online supplemental figure S2). The mean intensity of back pain at night was also higher in non-axSpA patients (3.6 (2.9) axSpA vs 4.6 (2.7) non-axSpA, at baseline) and significantly improved at 2 years in both groups. Similar to what was observed for back pain, the mean intensity and duration of morning stiffness significantly decreased after 2 years (table 2). Despite the observed improvements in all spinal symptoms, 33%–49% of axSpA and 48%–67% of non-axSpA patients had a score >2 at 2 years. Furthermore, at 2 years, 27% (n=79) of axSpA patients in the cohort had an ASDAS ≥2.1 points (vs 42% (n=52) of non-axSpA patients). Out of these 79 axSpA patients with high or very high disease activity, 13 were not taking any pharmacological treatment, 44 were under NSAIDs monotherapy, 3 were using both NSAIDs and conventional synthetic disease-modifying antirheumatic drugs (csDMARDs), and 19 patients were under biological DMARDs (bDMARDs) either with or without NSAIDs or csDMARDs.
Supplemental material
In adjusted multivariable analysis (figure 1), after 2 years, axSpA was significantly associated with 0.85 points larger improvement in back pain at night (β (95% CI) −0.85 (−1.47; −0.23)), compared with non-axSpA. No significant effect of an axSpA diagnosis was found for overall back pain, morning stiffness intensity or duration, although there was a strong trend for all variables for a higher improvement in axSpA, nearly reaching statistical significance. In all models, higher baseline values were a predictor of worse symptoms at 2 years. Additionally, female sex was associated with significantly worse back pain (overall and at night), but only a trend for morning stiffness. The use of NSAIDs was associated with a trend for worse symptoms at 2 years, reaching statistical significance for morning stiffness intensity.
Spinal and hip mobility
At baseline, the following mean (SD) values of SMM were observed for axSpA and non-axSpA (table 3): OWD 0.5 (1.2) vs 0.1 (0.7), cervical rotation 74.8 (12.0) vs 70.7 (12.0), chest expansion 5.9 (1.9) vs 5.3 (1.9), LSF 17.9 (4.8) vs 16.4 (4.3) and mSchober 4.9 (1.2) vs 4.7 (1.1). At 2 years, significant improvements were seen in both groups for cervical rotation (mean improvement: 2.4 (11.9) axSpA vs 2.8 (12.1) non-axSpA) and only in axSpA for LSF and mSchober. Impairments in ≥1 SMM, defined as being <2.5th percentile curve (>0 cm for OWD), were present in about half of the patients in both groups at baseline (51% axSpA vs 53% non-axSpA) and persisted in >40% of patients after 2 years. Applying a less strict definition of impairment of <5th percentile curve, 59% of axSpA and 64% of non-axSpA patients had ≥1 impaired SMM at baseline and >50% of patients in both groups had persistent impairment at 2 years (online supplemental file 3). Only OWD was more frequently impaired in axSpA patients at both baseline (23% axSpA vs 5.1% non-axSpA) and 2 years (22% axSpA vs 3.4% non-axSpA).
The mean IMD remained stable between baseline and 2 years in both groups (table 3) but was more frequently impaired in non-axSpA patients (4.6% axSpA vs 23% non-axSpA, at baseline). BASMI could be calculated in 219 axSpA and 105 non-axSpA patients. The mean score at baseline was 1.8 (0.9) in axSpA and 2.3 (0.8) in non-axSpA and it significantly decreased to 1.6 (0.7) and 2.1 (0.9), respectively, at 2 years.
In adjusted multivariable analysis (figure 2), for OWD, axSpA patients had a significantly lower likelihood of belonging to the ‘certain zero’ group after 2 years (OR (95% CI) 0.09 (0.01; 0.83)), meaning they had higher odds of having an impaired OWD. However, no differences between groups were observed in the degree of OWD impairment (IRR (95% CI) 0.90 (0.17; 4.71), data not shown). Furthermore, compared with non-axSpA, axSpA was associated with larger improvements in mSchober (β (95% CI) 0.26 (0.03; 0.50)), IMD (β (95% CI) 4.86 (1.93; 7.80)) and BASMI (β (95% CI) −0.24 (−0.41; −0.08)) at 2 years. No significant differences were observed for cervical rotation, chest expansion or LSF. Similarly to spinal symptoms, worse baseline values were associated with larger impairments in all mobility measures also at 2 years, and female sex with a worse performance in chest expansion and LSF. The use of NSAIDs was associated with worse LSF and BASMI at 2 years.
Discussion
This study revealed that, in general, spinal symptoms and impairments in spinal and hip mobility were prevalent in this cohort of patients with early axSpA and, most notably, non-axSpA CBP patients. After 2 years, although spinal symptoms significantly improved in both groups, this was not the case for most SMMs, and considerable impairments in both symptoms and mobility persisted in a substantial proportion of patients. Nevertheless, axSpA was associated with larger improvements in back pain at night, mSchober, IMD and BASMI, but also with more OWD impairment at 2 years.
Being common features of axSpA,1 the observed high prevalence and mean intensity of back pain and morning stiffness at baseline in axSpA patients was expected and in accordance with prior studies.29 30 As demonstrated by previous evidence, treatment of axSpA generally leads to an improvement in spinal symptoms over time.31 In our study, back pain and morning stiffness did improve after 2 years, however, they still persisted in a substantial proportion of patients (>30% with a score >2 points). Despite the persistency of symptoms, only 19% of axSpA patients were using bDMARDs at 2 years, which may suggest a necessity for more intensive treatment earlier in the disease course (at 2 years 27% of axSpA patients in the cohort had an ASDAS ≥2.1 points).32
Overall, spinal symptoms were more severe in non-axSpA (vs axSpA) patients, with mean scores of back pain intensity comparable to those reported in a previous meta-analysis of studies performed in patients with non-specific CBP that revealed a mean pain score of 51 (out of 100) points on a Visual Analogue Scale.33 Remarkably, back pain at night and morning stiffness, frequently recognised as features of the inflammatory nature of CBP in axSpA, were also more common and severe in non-axSpA patients.34 Despite being prevalent in the general population, the clinical course of non-specific CBP is difficult to predict.9 33 In our study, we observed significant improvements in all spinal symptoms after 2 years at the group level. As no specific treatment is currently available for non-specific CBP (the most common diagnosis in non-axSpA CBP patients in SPACE24 35), spontaneous recovery of symptoms and/or regression to the mean may be more reasonable explanations for this improvement. Illness perception and coping strategies, such as acceptance of the chronicity of the disease or adaptation of daily activities, may also have led patients to report lower pain scores over time. Furthermore, these patients were under regular clinical follow-up, which probably allowed patients to have better access to pharmacological and non-pharmacological treatments, such as physiotherapy and counselling. Other frequently reported diagnoses in the non-axSpA CBP group were myalgia and fibromyalgia24 and treatment of these conditions may likewise partially explain the improvements observed in our cohort. Nevertheless, symptoms still persisted in most patients after 2 years, which is in line with a previous systematic literature review (SLR) reporting that 65% of non-specific CBP patients continue to experience back pain 1 year after its onset.36
Comparisons between groups revealed there was a trend for larger improvements in all spinal symptoms in patients with axSpA (vs non-axSpA) after 2 years, but statistical significance was reached only for back pain at night. With initiation of treatment for axSpA aiming not only at reducing symptoms but also at controlling inflammation, this could partially explain the lower scores seen in axSpA for back pain at night at 2 years. These findings, however, may question the validity of inflammatory symptoms, such as nocturnal back pain and morning stiffness, in distinguishing patients with and without axSpA in rheumatology clinics. Indeed, a recent SLR and meta-analysis of spondyloarthritis features reported a lower specificity for inflammatory back pain than previously reported.37 Also noteworthy, the regression models showed a trend for worse symptoms in patients taking NSAIDs, possibly reflecting the greater need for this therapy in those with more severe symptoms.
Reduced spinal mobility is a major concern in longstanding axSpA. In one study addressing established axSpA, 79% of patients had ≥1 impaired SMM, mainly LSF.19 In a previous study performed in SPACE with a smaller number of patients, Fongen et al also documented impairments in spinal and hip mobility in early disease in >50% of patients at baseline, with LSF remaining as the most frequently affected measure.11 The same results were found in our larger sample of axSpA patients. Contrary to long-standing disease, structural damage is far less prominent in early axSpA and reduced spinal mobility in earlier disease stages may be mostly due to pain and inflammation.5
Corroborating the findings by Fongen et al, reduced mobility was also prevalent in non-axSpA. Additionally, non-axSpA patients showed a somewhat worse performance for most SMMs, compared with those with axSpA. One possible explanation is the higher severity of spinal symptoms observed in non-axSpA patients, which may lead to more restricted movements, in line with the already demonstrated relationship between high-intensity back pain and reduced mobility in non-specific CBP.38 39 After 2 years, only cervical rotation significantly improved.
In general, Fongen et al observed that impairments in spinal and hip mobility are prevalent in all CBP patients at baseline, thus having no diagnostic value to distinguish those with axSpA from those with non-axSpA. As only baseline data were assessed, a certain diagnostic uncertainty could be expected in this population with short symptom duration, as patients initially classified as non-specific CBP could turn out to have axSpA later in the disease course (or vice versa). To overcome this, in our study, patients were classified as having axSpA or non-axSpA based on the diagnosis at 2 years with a high LoC, strengthening the certainty surrounding the diagnosis. Furthermore, a larger sample population was included. Despite this, we achieved similar findings at baseline, confirming the limited diagnostic value of spinal/hip mobility measures. Additionally, after 2 years, only few SMMs significantly improved in either group. This may reflect a real lack of improvement in mobility over time or be the consequence of persistency of symptoms. Regardless of the limited change in spinal mobility over 2 years in both axSpA and non-axSpA patients, axSpA was associated with larger improvements in mSchober, IMD and, consequently, BASMI, possibly partially reflecting the somewhat higher improvements in spinal symptoms observed in these patients. However, axSpA patients had also higher odds of having an impaired OWD at 2 years, although the difference between groups in the absolute OWD may not be clinically relevant. Furthermore, NSAID use was associated with worse LSF and BASMI after 2 years, which may be a consequence of more severe spinal symptoms in patients taking NSAIDs.
To our knowledge, this is the first study comparing spinal symptoms and spinal/hip mobility in axSpA and non-axSpA CBP patients over 2 years. In SPACE, every patient was subjected to the same inclusion criteria and, for this analysis, we excluded uncertain diagnoses (n=16), which allowed for more reliable comparisons between the axSpA and non-axSpA groups. A limitation of our study is the difference in demographic characteristics between groups, which may have influenced the results. In our cohort, the non-axSpA group had a higher proportion of females and female sex was associated with worse back pain, chest expansion and LSF at 2 years. Previous studies have also demonstrated higher prevalence of CBP, higher pain levels in general and lower performance of some SMMs in women compared with men.15 40–45 However, we corrected the analyses for the effect of sex. Also noteworthy is the lack of established definitions of clinically significant improvements and impairments. For non-specific CBP, an improvement of ≥2.5 points on an 11-point NRS was considered by experts as being clinically significant for back pain intensity,46 47 but, to our knowledge, no such definition has been proposed for the other spinal symptoms we have analysed. For axSpA, improvement over time is defined by a decrease in ASDAS ≥1.1 points,48 taking into account a multitude of factors and not only pain intensity.1 49 No consensual definition of improvement is currently available for back pain or morning stiffness. In what concerns spinal mobility, the poor reliability of SMMs over time justified their removal from the ASAS core set of domains for the evaluation of axSpA patients. This needs to be taken into consideration when interpreting changes over time in spinal/hip mobility.13 21 50 Finally, although all patients were treated as per regular clinical practice, data about the performance of regular exercise or other complementary treatments were not consistently collected which may have also influenced our results to some extent.
In summary, high levels of spinal symptoms and restricted spinal and hip mobility were highly prevalent in both axSpA and non-axSpA CBP patients. After 2 years, spinal symptoms and some mobility measures significantly improved in both groups, but substantial symptoms and impairments in mobility persisted in both axSpA and non-axSpA patients. These findings are of importance as they reflect the high disease burden that persists in both diseases even after 2 years. For axSpA, although the presence of other unknown conditions responsible for the persistent impairments cannot be ruled out, this may reflect an undertreatment of patients, suggesting the need to improve implementation of early treatment. From the rheumatologist’s point of view, a better understanding of the clinical course of non-specific CBP is also very important, as these patients constitute a significant proportion of referred individuals to rheumatology outpatient clinics. Altogether this may contribute to improve the diagnostic and treatment approach of CBP patients, with the ultimate goal of improving the quality of life of those living with these high-burden diseases.
Data availability statement
Data are available on reasonable request. Data are available on reasonable request, requiring permission from collaborating centres.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and the SPACE study protocol was approved by the Medical Ethical Committee of the Leiden University Medical Center (Medisch-Ethische Toetsingscommissie Leiden Den Haag Delft (METC LDD)) as well as the appropriate local ethical committees of the collaborating centres. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors would like to acknowledge the Assessment of SpondyloArthritis international Society (ASAS) for having supported the research fellowship of Ana Bento da Silva.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors ABdS, FAvG, DvdH and SR were involved in the design and implementation of the study, as well as in the core interpretation of the results. FAvG, MvL, MvdS, CF, SE and RR were responsible for data collection. Statistical analysis and interpretation of the results were performed by ABdS, FAvG, DvdH, SR, MvL and MLM. ABdS was responsible for writing the first version of the manuscript. All coauthors were involved in further data interpretation, revising the manuscript and providing final approval to the version to be published. ABdS and FAvG are the guarantors of this study.
Funding ABdS was supported by an ASAS (Assessment of SpondyloArthritis International Society) Fellowship.
Competing interests SR has received research grants and/or consultancy fees from AbbVie, Eli Lilly, Galapagos, Janssen, MSD, Novartis, Pfizer, UCB and Sanofi. MvdS has acknowledged speakers fees from Janssen, Novartis, UCB and Beneke, consulting fees from Janssen, Novartis, Abbvie and UCB, and research support from Janssen, Novartis and UCB. SE has received consulting fees from Amgen, Janssen, Novartis, AbbVie and UCB. RR has received consulting and/or speaking fees from Advisory Boards from Abbvie, Novartis, Janssen, Lilly, MSD, Pfizer and UCB. DvdH has acknowledge consulting fees from AbbVie, Argenx, BMS, Galapagos, Glaxo-Smith-Kline, Janssen, Lilly, Novartis, Pfizer, Takeda and UCB; associate editor Annals Rheumatic Diseases, editorial board member Journal of Rheumatology and RMD Open, Advisor Assessment Axial Spondyloarthritis international Society and director of Imaging Rheumatology bv. FAvG has received grants/research supports from Stichting ASAS, Novartis and UCB, and received honoraria or consultation fees from Novartis, BMS, AbbVie, MSD, Janssen, Lily and UCB. The remaining authors have declared no conflicts of interest.
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.