Background
Axial spondyloarthritis (axSpA) is an inflammatory disease predominantly affecting the axial skeleton. The Assessment of SpondyloArthritis International Society (ASAS) classification criteria divides axSpA into the following two categories [
1,
2]: patients with radiographic axSpA (r-axSpA), who have developed radiographic damage in the sacroiliac joint (SIJ) and fulfil the modified New York criteria (mNYc) [
3], also termed ankylosing spondylitis (AS), and patients with non-radiographic axSpA (nr-axSpA) who did not meet the mNYc. However, there is currently no unified understanding of the nr-axSpA concept. One viewpoint considers that nr-axSpA and AS are homogeneous in different stages of the disease course because they share common characteristics, such as clinical presentation, clinical disease activity, treatment response, and comorbidity burden [
4,
5]. However, there are nr-axSpA cases that have not progressed to AS during long-term follow-up and even evolved to other diseases, supporting the notion that nr-axSpA partially overlaps with pre-AS but is not fully equal to the early phase of AS; additionally, nr-axSpA is more clinically heterogeneous than AS [
6,
7]. Although nr-axSpA is widely distributed globally, its clinical characteristics vary among ethnic and geographic groups[
8]. Previous studies on nr-axSpA have focused on Caucasian populations, so the clinical characteristics of nr-axSpA in Chinese patients are not well known. Therefore, it is necessary to further study the characteristics of Chinese patients with nr-axSpA.
Several important differences in demographic features, patterns of disease onset, and clinical manifestations between adult-onset AS (AoAS) and juvenile-onset AS (JoAS) have been reported [
9‐
11]. For example, JoAS has a higher frequency of peripheral joint involvement and a lower prevalence of HLA-B27 positivity and uveitis than AoAS. As mentioned previously, nr-axSpA and AS belong to different axSpA subgroups. However, the similarities and differences between adult- and juvenile-onset nr-axSpA remain obscure. Thus, this study aimed to summarise the clinical characteristics of Chinese patients with nr-axSpA and compare the differences between adult-onset and juvenile-onset nr-axSpA.
Discussion
In this study, we summarised the demographic and clinical characteristics of nr-axSpA in China and compared the differences between adult-onset and juvenile-onset nr-axSpA for the first time, using a current study on a large sample size of patients with nr-axSpA in China.
In this study, the median age at disease onset of nr-axSpA was 22 years, with a median of 23 years in the adult-onset group, similar to the results reported in the Korean and Indian cohorts (26.4 and 24.8 years, respectively) [
16,
17]. By contrast, the median of this study was lower than that in Western cohorts, in which the mean ages at disease onset were > 30 years [
4,
18], suggesting that the disease may develop at an earlier age in Asian populations than in Western populations [
8].
In our study, the prevalence of HLA-B27 positivity was 72.2% in patients with nr-axSpA. To date, there has been a lack of large-scale epidemiological survey results on the prevalence of HLA-B27 positivity among Chinese patients with nr-axSpA. A previous study reported that more than 90% of AS patients in China were HLA-B27 positive [
19], which was significantly higher than the results of our study, indicating that the rate of HLA-B27 positivity in nr-axSpA was lower than that of AS in China.
Sex has consistently been identified as a major demographic difference in nr-axSpA compared with AS, usually exhibiting a comparable sex ratio [
6]. In this study, the male-to-female ratio in nr-axSpA was 1.26:1, which is lower than that in AS, but higher than that in Western cohorts of nr-axSpA such as DESIR (0.89:1) [
18], SPACE (0.82:1) [
18], and SCQM (0.88:1) [
20]. This may reflect regional differences in the demographic characteristics of nr-axSpA patients, as in other studies of Asian nr-axSpA population, a higher proportion of male was also observed than that in Western nr-axSpA population [
8,
21], suggesting that the proportion of male in nr-axSpA in Asian population may be higher than that in Western population. However, it is important to point out that the male-to-female ratio in nr-axSpA is much lower than that in AS for the same race, no matter in Asian population or European population. In addition, several differences exist in the demographic characteristics between sexes. For example, male patients had a younger age at disease onset and diagnosis and a higher prevalence of HLA-B27 positivity than female patients. A similar finding was reported in a Swiss Clinical Quality Management cohort study [
22]. It is supposed that genotypic differences between sexes may be a reason for these demographic differences.
IBP is a key symptom of axial involvement and is present in most patients with axSpA. The total prevalence of IBP in our study was 83.6%, which is consistent with a large meta-analysis reporting a prevalence ranging from 71.5% to 95.4% [
23]. In addition, nearly 75% of patients with nr-axSpA had peripheral involvement during the disease course. Extra-articular manifestations occurred in 10.4% of patients, and uveitis is considered one of the most common extra-articular manifestations; 4.9% of patients experienced uveitis in our study, much lower than that reported in Indian and Western cohorts (8%–12.4%) [
4,
16,
18]. This finding can be partly explained by the varying prevalence of uveitis among ethnicities and geographic regions. A previous study reported that the prevalence of uveitis in Chinese patients with AS was much lower than that reported in Western, Korean, and Indian studies [
21,
24]. Another explanation might be the shorter disease duration of patients with nr-axSpA in our study, which is supported by the viewpoint that uveitis in axSpA does not always occur in the early stage of the disease and requires time to develop [
25,
26].
Our data showed that most patients had high disease activity (ASDAS > 2.1) at baseline; however, there were differences between sexes. The proportions of male patients with ASDAS-CRP > 2.1 and elevated CRP levels and ESRs were higher than those of female patients. Additionally, the median ASDAS-CRP level was higher in male patients than female patients. The conclusion that acute-phase reactants and ASDAS are important positive predictors of radiographic sacroiliitis progression in axSpA has been reported in a previous cohort study [
27]. Whether male patients are more likely to experience progression of structural damage than female patients should be further investigated through long-term cohort studies with larger sample sizes.
In the realm of treatment, the importance of exercise is on par with pharmacological interventions. However, our study revealed that only a small percentage of patients with nr-axSpA received exercise instruction. Research indicates that when applying attentional focus strategies, an External Focus of Attention (EFA) seems to be more effective than an Internal Focus of Attention (IFA) in affecting the movement execution in patients with musculoskeletal disorders [
28]. Although there is limited research comparing the effectiveness of EFA and IFA guidance in axSpA patients, there is a need for further exploration to determine the appropriate rehabilitative strategies and motor learning methodologies for this population.
Several important differences in the demographic features, patterns of disease onset, and clinical manifestations have been reported between JoAS and AoAS. In this study, we also compared adult-onset and juvenile-onset nr-axSpA. Several significant differences were identified between the two groups regarding the demographic characteristics (Table
2). First, our study revealed that there were significantly more male patients in the juvenile-onset group than in the adult-onset group. Male predominance has also been reported in previous studies on juvenile spondyloarthritis (JSpA), enthesitis-related JIA (ERA), and other subtypes of juvenile idiopathic arthritis (JIA), findings which align with our study. Previous studies have reported that male patients were associated with worse functional outcomes and poorer prognosis and thus seek health care more frequently than female patients, which also contributes to the phenomenon (male bias) [
29,
30]. Second, the rate of family history of SpA was slightly higher in the juvenile-onset group than in the adult-onset group; however, no significant difference in HLA-B27 status was identified between the two groups, indicating that genetic factors other than HLA-B27 may play a role in disease pathogenesis. In addition, our data demonstrated that the disease duration was longer in juvenile-onset nr-axSpA than in adult-onset nr-axSpA, indicating that a longer diagnostic delay may occur more frequently in the juvenile-onset group. Currently, no standard classification system exists for juvenile-onset axSpA. Adult classification criteria such as Amor and the European Spondyloarthropathy Study Group have been more often applied to these patients in previous studies [
31,
32]. In paediatric rheumatology, these patients may be categorised as JIA, as defined by the International League of Associations for Rheumatology classification criteria [
33]. The application of this classification system in clinical practice aids in identifying patients with similar clinical and prognostic implications. However, sensitive imaging tools, such as MRI, which have contributed to a better assessment of patients with early disease stages of axSpA are not included in these classification systems. Our previous report confirmed that MRI can predict the progression of juvenile-onset nr-axSpA to JoAS [
12]. However, refining definitions of “positive” MRI in juvenile-onset nr-axSpA is required in future studies [
34].
Several similarities were observed in the clinical manifestations between the two groups. For example, IBP, one of the key clinical manifestations, occurred in the majority of patients in both groups. In addition, morning stiffness, nocturnal pain, buttock pain, groin pain, and extra-articular manifestation were equally frequent between the two groups. However, the axial/peripheral pattern of disease at onset differed significantly; the juvenile-onset group experienced a more frequent onset of peripheral joint involvement, whereas axial involvement was not as frequent as that in the adult-onset group (Table
2). The disease-onset pattern of juvenile-onset nr-axSpA may present as a mode of “peripheral predominant”. A similar conclusion was also demonstrated in an earlier cross-sectional study comparing JoAS and AoAS [
35‐
37]. Therefore, typical axial skeleton involvement should not be expected as the first clinical presentation for patients with juvenile-onset nr-axSpA. Furthermore, this clinical feature of “peripheral predominant” continued in the disease course, especially knee and hip involvement (Fig.
1), which is thought to be associated with a poor functional outcome. This finding has important therapeutic implications, as juvenile-onset nr-axSpA patients may require closer monitoring and more aggressive treatment for hip disease. In addition, enthesitis is more common in juvenile-onset nr-axSpA. Enthesitis is a factor for which the progression to JoAS was suggested in our previous study [
12]. Furthermore, enthesitis was found to be a risk factor for sacroiliitis in an Italian cohort study on ERA [
38].
The ASDAS was used as a clinical tool for measuring disease activity among the patients in our study. Although the level of disease activity was highly comparable between the two groups, juvenile-onset nr-axSpA had more severe disease effects in terms of peripheral pain/swelling measured by a numerical rating scale (Table
2). Moreover, the ASDAS did not include the measure domain of enthesitis. Hence, the ASDAS may fail to assess accurately the disease activity of juvenile-onset patients. Future research might require the exploration of better assessment tools for measuring disease activity in juvenile-onset nr-axSpA while increasing the weight of key stakeholders, such as peripheral arthritis and enthesitis.
Our study had several strengths, particularly the inclusion of a large number of patients from a single centre, thereby reducing the possible effects of population heterogeneity. However, this study also had some limitations. First, as this study used retrospectively collected data from the COCAS cohort, some patient information was lacking (including ASDAS score and SIJ-MRI). Second, although juvenile-onset nr-axSpA was used to define patients who experienced symptom onset at age < 16 years, all patients enrolled in the current study were aged ≥ 16 years at their first visit to the adult Department of Rheumatology. Thus, these patients were not fully representative of the juvenile-onset nr-axSpA population, especially those still in childhood. Third, no comparison between juvenile-onset and adult-onset groups related to functional outcomes was performed in this study, and further studies including a functional assessment and prognosis analysis of nr-axSpA are warranted.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.