Biopsychosocial factors should be considered when evaluating central sensitization in axial spondyloarthritis

Abstract

To identify the determinants of central sensitization (CS) in patients with axial spondyloarthritis (axSpA). Central Sensitization Inventory (CSI) was used to determine CS frequency. Disease-related variables including Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Ankylosing Spondylitis Disease Activity Score (ASDAS-CRP/-ESR), Maastricht Ankylosing Spondylitis Enthesitis Score (MASES), Bath Ankylosing Spondylitis Functional Index (BASFI), Ankylosing Spondylitis Quality of Life Questionnaire (ASQoL) and Numeric Rating Scale (NRS)_GLOBAL were assessed. Biopsychosocial variables were evaluated by the Multidimensional Scale of Perceived Social Support (MSPSS), Brief Illness Perception Questionnaire (B-IPQ), Hospital Anxiety and Depression Scale (HADS) and subscales for Anxiety (HADS-A) and Depression (HADS-D), and Jenkins Sleep Evaluation Scale (JSS). To determine the predictors of the development and severity of CS, multiple linear and logistic regression analyses were performed. The frequency of CS was 57.4% in the study population (n = 108). CSI score was correlated with the duration of morning stiffness, BASDAI, ASDAS-CRP, ASDAS-ESR, NRS_GLOBAL, BASFI, MASES, ASOoL, JSS, HADS, and B-IPQ total scores (ρ ranged from 0.510 to 0.853). Multiple regression analysis indicated that BASDAI (OR: 10.44, 95% CI: 2.65–41.09), MASES (OR: 2.47, 95% CI: 1.09–5.56) and HADS-A (OR: 1.62, 95% CI: 1.11–2.37) were independent predictors of the development of CS. Additionally, higher NRS_GLOBAL, JSS, HADS-D, and HADS-A scores appeared to determine the severity of CS. This study confirms that worse disease activity, more enthesal involvement, and anxiety independently predict the development of CS. Additionally, higher patient-perceived disease activity, sleep impairment and poor mental health significantly contribute to the severity of CS.

Introduction

Axial spondyloarthritis (axSpA) is a chronic autoimmune inflammatory rheumatic disease that affects mainly the sacroiliac joints and the spine. It is characterized by inflammatory back pain, spinal stiffness, and enthesis. Various extra-musculoskeletal manifestations and comorbidities frequently accompany [1]. The major challenge of axSpA and other chronic inflammatory rheumatic diseases is the persistence of intense and disabling pain. Although patients with axSpA are adequately managed with specific therapies such as non-steroidal anti-inflammatory drugs (NSAIDs) and biological disease-modifying anti-rheumatic drugs (bDMARDs), many of them still complain of persistent pain and fatigue [2, 3]. This situation cannot be explained only by the inflammation of the articular/periarticular tissues. There may also be neuroinflammation and related pain-processing mechanisms such as central sensitization (CS) [4].

Central sensitization has been defined recently as “the hyperexcitability of the central nervous system”. It exerts a crucial role in the persistence of chronic, widespread pain and other medically unexplained symptoms [5]. It may present with clinically significant pain states such as allodynia and hyperalgesia. On the other hand, CS can also manifest with cognitive and/or emotional signs and symptoms. Over the last decade, several studies have reported the presence of CS in a broad range of musculoskeletal conditions [6–8]. Among these, the prototype is fibromyalgia. Yet, CS can also be observed in patients with autoimmune inflammatory rheumatic diseases. The chronic inflammatory state can pave the way to the sensitization of central neural pathways; thus, contribute to widespread pain, allodynia and hyperalgesia [9]. Limited pain-relieving response to appropriate treatment regimen among patients with autoimmune inflammatory rheumatic diseases is regarded as an indicator of the presence of CS. Research on this subject mostly involved patients with rheumatoid arthritis [2]. Nevertheless, the relation of CS with axSpA has been less studied so far.

Limited evidence regarding the relationship between CS and SpA revealed that CS could interfere with sleep quality, perceived pain, disease activity, and overall quality of life (QoL) [10–12]. Yet, knowledge of the potential contributors of CS in patients with axSpA is scarce [13]. Given the increased frequency of cognitive and mood disorders in patients with CS, the biopsychosocial status appears as an important determinant of centrally-mediated pain [14]. From a biopsychosocial point of view; biological, psychological, and social factors are regarded to be working independently and jointly to affect an individual’s experience through the management of central sensitivity syndromes [15].

The objectives of the current study were (i) to determine the frequency of CS among patients with axSpA, (ii) to examine the interrelation of CS and patients’ health status (disease activity, function, QoL, etc.), and (iii) to elucidate the potential contributors of CS in axSpA. We hypothesized that psychosocial determinants are the main drivers of CS in patients with axSpA.

Methods

Participants and study design

The study was conducted at the Department of Physical Medicine and Rehabilitation in Cukurova University Faculty of Medicine between July 2022 and December 2022. In this cross-sectional study, patients were recruited by rheumatologists and physiatrists consecutively from an ordinary outpatient clinic during routine evaluation. Individuals diagnosed as either axSpA according to the Assessment of Spondyloarthritis International Society (ASAS) classification criteria for axSpA [16] or ankylosing spondylitis (AS) according to the Modified New York Criteria [17] were included in the current study. Patients were excluded if they (i) were below the age of 18 years, (ii) had other concomitant systemic inflammatory rheumatic diseases or fibromyalgia, (iii) had neurological conditions such as myelopathy and radiculopathy, (iv) were receiving centrally acting drugs (e.g. pregabalin, gabapentin, amitriptyline, duloxetine, opioids), (v) had alcohol/substance abuse, (vi) had any uncontrolled systemic diseases or malignancy, (vii) were pregnant, and/or (viii) had the inability to understand and fill in the questionnaires.

The ethical approval of the study was obtained from the Local Ethics Committee of Cukurova University Faculty of Medicine (Date of approval: July 22, 2022, Number: 124/15). Written informed consent was obtained from the patients before participating in the study. The study was conducted in line with the principles of the Declaration of Helsinki.

Study variables and evaluation methods

Sociodemographic data of patients (age, gender, marital status, level of education, annual income), smoking and alcohol use, body mass index (BMI), symptom duration (years), diagnosis time (years), duration of morning stiffness (minutes), history of coronavirus disease-2019 (COVID-19), history/presence of extra-articular involvement (uveitis, psoriasis, inflammatory bowel disease, dactylitis), presence of peripheral arthritis, and medications (NSAIDs, disease-modifying antirheumatic drugs [DMARDs]) were recorded. Laboratory measures included erythrocyte sedimentation rate (ESR) (mm/h), C-reactive protein (CRP) (mg/L) and human leucocyte antigen-B27 (HLA-B27).

Disease activity was assessed by using the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), the Ankylosing Spondylitis Disease Activity Score with CRP (ASDAS-CRP) and ESR (ASDAS-ESR) [18, 19]. Patient-perceived disease activity was evaluated by a numeric rating scale (NRS)_GLOBAL (0 = least severe, 10 = most severe). Entheseal involvement was evaluated by using the Maastricht Ankylosing Spondylitis Enthesitis Score (MASES). Thirteen entheseal sites were analyzed by MASES and the overall score ranged from 0 to 13 [20]. Bath Ankylosing Spondylitis Functional Index (BASFI) was used to evaluate the functional status of the patients. BASFI score was calculated as the mean of the ten scales and ranged between 0 and 10 [21, 22]. Quality of life was assessed by the Ankylosing Spondylitis Quality of Life Questionnaire (ASQoL). Scores ranged from 0 to 18, where higher scores indicated poor QoL [23, 24].

Central sensitization inventory (CSI) was used to screen for possible central sensitization. The scale consists of two parts. The first part of the scale includes 25 items evaluating the frequency of symptoms associated with central sensitization. This first section is scored between 0 and 100 points and a CSI score ≥ 40 in patients with chronic pain indicates a high probability of central sensitization. In the second part of the questionnaire, the presence of a previous diagnosis of any specific disease associated with central sensitivity syndrome is questioned [25, 26]. Higher CSI scores indicate clinically severe CS [27].

Social support was evaluated by the Multidimensional Scale of Perceived Social Support (MSPSS). The questionnaire includes 12 items measuring social support on a 7-point Likert scale. The MSPSS questionnaire is scored between 0 and 84, where higher scores indicate higher levels of perceived social support [28, 29]. The Brief Illness Perception Questionnaire (B-IPQ) was used to evaluate the disease insights of the patients. The questionnaire has 9 items; the first 5 items evaluate cognitive insights, the 6th and 8th items assess emotional aspects, and the 7th item asks about the degree of understanding of the disease. The 9th question asks about the three most important factors that cause the illness [30, 31].

The Hospital Anxiety and Depression Scale (HADS) was used to evaluate depression and anxiety symptoms. The 14-item measure produces two subscales HADS-Depression (HADS-D) and HADS-Anxiety (HADS-A). Scores ≥ 11 indicate the probability of a mood disorder [32, 33]. Sleep quality was evaluated using the Jenkins Sleep Evaluation Scale (JSS). In this 4-item questionnaire, the total score is ranging from 0 to 20, showing more disturbed sleep as it increases [34, 35].

Participants were asked to complete the study questionnaires themselves under the supervision of the researcher during a face-to-face visit in a silent room. The patients were given approximately 30 min to complete the questionnaires. The questionnaires were validated in Turkish.

Statistical analyses

The G*Power^® software (Heinrich-Heine-Universitat Düsseldorf, Düsseldorf, Germany) was used to calculate the sample size. Based on the values obtained from previous studies [11–13] in the literature by accepting the CSI score as the primary assessment tool, the required sample sizes were obtained in numbers ranging from 5 to 42 at 5% error and 90% power. One hundred patients were planned to be included in the study under the assumption that CS would be newly diagnosed and the rate of patients with a CSI score ≥ 40 would be 45%.

Data analyses were performed by the 20.0 version of IBM^® SPSS^® (IBM Corp, Armonk, NY, USA) statistical software. The Shapiro–Wilk test was used to confirm the normality of the continuous variables. The sociodemographic, psychosocial and other clinical variables were evaluated by descriptive statistics. Comparative analysis of continuous variables between patients with and without CS was performed by the Mann–Whitney U test. Pearson’s chi-squared or Fisher's Exact Test was used to compare categorical variables between groups. Results of categorical variables were presented as numbers and percentages, whereas continuous variables were shown as either mean ± standard deviation or median [25% (q1)—75% (q3) quartiles). The potential correlation of the CSI score with clinical variables was tested by Spearman’s correlation analysis. Values were presented as Spearman’s rho (ρ). Variables significant at p < 0.25 level in univariate analysis were included in regression analysis. Dependent variables were the CSI score and the presence of CS (CSI score ≥ 40). Multiple linear regression analysis was applied to determine the predictors of the CSI score. Multiple logistic regression analysis was performed to identify the risk factors for the development of CS in patients with axSpA. In univariate analysis, variables significant at the p < 0.25 level were entered in logistic regression analysis. p values below 0.05 were accepted as “statistically significant”.

Results

Patient characteristics

One hundred forty-three patients with axSpA were evaluated for eligibility for the study. After applying the exclusion criteria, 108 patients with axSpA (mean age 43.8 ± 11.3 years, 61.1% male) were included in this study (Fig. 1). The frequency of patients with CS was 57.4%. Of the patients with CS, 83.8% had clinically severe/extreme CS (CSI score ≥ 50). The socio-demographic and clinical characteristics of the patients were displayed in Tables 1 and 2.

Fig. 1.

Study flow chart. axSpA axial sponyloarthritis, CSI Central Sensitization Inventory

Table 1.

Socio-demographic and biopyschosocial variables of the study population stratified for CSI score

	All patients n = 108	CSI < 40 (n = 46)	CSI ≥ 40 (n = 62)	p
Age (years),	43.92 ± 11.396	42.72 ± 11.91	44.81 ± 11.02	0.349
Gender
Female	42 (38.9%)	13 (28.3%)	29 (46.8%)	0.051
Male	66 (61.1%)	33 (71.7%)	33 (53.2%)
Marital status- married	78 (72.2%)	30 (65.2%)	48 (77.4%)	0.162
History of COVID-19	46 (42.6%)	21 (45.7%)	25 (40.3%)	0.580
Education level (years)	11 (5–13)	11 (8–14)	8 (5–13)	0.039
Annual income (TL)	6000 (4000–10,000)	7500 (5500–12,000)	5500 (3000–7000)	0.001
BMI (kg/m2)	27.02 (24.36–30.24)	26.7 (23.5–30.1)	27.4 (25–31)	0.113
Current smoking	38 (35.2%)	15 (32.6%)	23 (37.1%)	0.817
Current alcohol use	23 (21.3%)	10 (21.7%)	13(21%)	0.470
MSPSS (0–84)	59 (42–70.8)	70 (55–75)	49 (37–61)	< 0.001
JSS (0–20)	13 (5–17)	5 (3–10)	17 (13–18)	< 0.001
HADS-D score (0-21)	8 (4–12)	4 (2–6)	11 (8–14)	< 0.001
HADS-A score (0-21)	8 (3–12)	4 (1–8)	10 (8–14)	< 0.001
B-IPQ total score (0–80)	45 (34.3–54)	34 (23–42)	51 (45–57)	< 0.001
Consequences (0–10)	6 (4–8)	4 (3–6)	8 (6–9)	< 0.001
Timeline (0–10)	10	10	10	0.043
Personal control (0–10)	5 (3–7)	6 (5–8)	5 (3–6)	0.007
Treatment control (0–10)	8 (5–9)	9 (7–10)	7 (5–9)	0.001
Identity (0–10)	6.5 (5–8)	5 (3–6)	8 (7–9)	< 0.001
Concern (0–10)	7 (3.3- 9)	4.5 (2–7)	8.5 (6–10)	< 0.001
Comprehensibility (0–10)	10 (7–10)	10 (9–10)	9 (6–10)	0.031
Emotional response (0–10)	7 (3–9)	3 (2–6)	8 (7–10)	< 0.001

Values are presented in: n (%), mean ± standard deviation or median (q1–q3)

COVID-19 Coronavirus disease-2019, TL Turkish Lira, BMI body mass index, CSI Central Sensitization Inventory, MSPSS Multidimensional Scale of Perceived Social Support, JSS Jenkins Sleep Evaluation Scale, HADS-D Hospital Anxiety and Depression Scale – Depression, HADS-A Hospital Anxiety and Depression Scale – Anxiety, B-IPQ Brief Illness Perception Questionnaire

Table 2.

Disease-related clinical variables of the study population stratified for CSI score

	All patients (n = 108)	CSI < 40 (n = 46)	CSI ≥ 40 (n = 62)	p
Diagnosis AS	90 (83.3%)	40 (87%)	50 (80.6%)	0.384
Diagnosis time (years)	11 (4–19.5)	11.5 (4–20)	10 (3–18)	0.513
Symptom duration (years)	13 (6–20)	14 (7–21)	13 (5–20)	0.627
HLA-B27 positivity	59 (54.6%)	29 (63%)	30 (48.4%)	0.130
Duration of morning stiffness (min.)	20 (0–120)	5 (0–20)	30 (20–60)	< 0.001
Peripheral arthritisa	34 (31%)	11 (23.9%)	23 (37.1)	0.088
Extra-articular manifestationb	26 (24%)	10 (21.7%)	16 (25.8%)	0.506
Medication
NSAID	24 (22.2%)	14 (30.4%)	10 (16.4%)	< 0.001
csDMARD	7 (6.5%)	3 (6.5%)	4 (6.6%)
bDMARD	58 (53.7%)	29 (63%)	29 (47.5%)
NSAID + csDMARD	18 (16.7%)	0 (0%)	18 (100%)
CRP (mg/L)	5.07 (2.99–9.34)	5 (2.8–7.3)	5.2 (3.3–10.8)	0.216
ESR (mm/h)	17 (10–26.5)	15.5 (10–24)	18.5 (10–29)	0.371
BASDAI (0-10)	4.75 (2.65–6.58)	2.3 (1.7–3.6)	6.5 (5.4–7.5)	< 0.001
ASDAS-CRP	3 (2.13–3.6)	2 (1.7–2.4)	3.5 (3–3.7)	< 0.001
ASDAS-ESR	2.85 (2.1–3.6)	2.1 (1.5–2.4)	3.5 (3–3.8)	< 0.001
NRSGLOBAL (0–10)	6 (3–7)	3 (2–49	7 (6–8)	< 0.001
MASES (0–13)	0 (0–2)	0	2 (0–3)	< 0.001
BASFI (0–10)	3.8 (1.7–6.4)	2.2 (0.9–3.3)	6 (3.7–7.2)	< 0.001
ASQoL (0–18)	10 (4–15)	4 (1–7)	13 (10–17)	< 0.001
ASOoL score > 8	61 (56%)	9 (19.6%)	52 (83.9%)	< 0.001

Values are presented in: n (%) or median (q₁–q₃)

CSI Central Sensitization Inventory, AS ankylosing spondylitis, HLA-B27 Human Leucocyte Antigen-B27, NSAID non-steroidal anti-inflammatory drugs, csDMARD conventional synthetic disease-modifying antirheumatic drug, bDMARD biological disease-modifying antirheumatic drug, CRP C-reactive protein, ESR erythrocyte sedimentation rate, BASDAI Bath Ankylosing Spondylitis Disease Activity Index, ASDAS-CRP Ankylosing Spondylitis Disease Activity Score with C-reactive protein, ASDAS-ESR Ankylosing Spondylitis Disease Activity Score with erythrocyte sedimentation rate, NRS numeric rating scale, MASES Maastricht Ankylosing Spondylitis Enthesitis Score, BASFI Bath Ankylosing Spondylitis Functional Index, ASQoL Ankylosing Spondylitis Quality of Life Questionnaire

^aDefined as swollen joint count ≥ 1

^bHistory of inflammatory bowel disease, uveitis, psoriasis

Comparison of clinical and biopsychosocial variables between patients with and without CS

Demographic, clinical, and biopsychosocial characteristics of patients with and without CS were compared. No significant difference was found between the groups in terms of age, gender, BMI, marital status, history of COVID-19, smoking, alcohol use, type of SpA, disease duration, duration of symptoms, HLA-B27 positivity, extra-articular manifestations, current peripheral arthritis, CRP and ESR level (p > 0.05 for all) (Tables 1 and 2). However, disease-related variables (NRS_GLOBAL, BASDAI, ASDAS-CRP, ASDAS-ESR, MASES, BASFI, and ASOoL) were significantly worse in patients with CS (p < 0.001 for all) (Table 2). Biopsychosocial variables including JSS, HADS, and B-IPQ scores were also higher in patients with CS (p < 0.001 for all). Additionally, MSPSS score was lower in this group (p < 0.001) (Table 1).

Correlation of CSI score with socio-demographic, biopsychosocial and disease-related variables

The results of the correlation analysis were given in Table 3. A significant positive moderate-strong correlation (Spearman’s rho ranged from 0.510 to 0.853) was found between CSI score and duration of morning stiffness, BASDAI, ASDAS-CRP, ASDAS-ESR, NRS_GLOBAL, BASFI, ASOoL, MASES, JSS, HADS, and B-IPQ total score (ρ ranged 0.510–0.853). CSI score showed a negative correlation with MSPSS, annual income, and education level (ρ = – 0.511, – 0.308 and – 0.228, respectively). CSI score showed no correlation with age, BMI, diagnosis time, symptom duration, CRP, and ESR.

Table 3.

Correlation of CSI score with socio-demographic, disease-related and biopsychosocial variables in patients with ax-SpA

	CSI score
	ρ	p
Age	0.144	0.138
Education level (years)	– 0.228	0.018
Annual income (TL)	– 0.308	0.001
BMI (kg/m2)	0.131	0.175
Diagnosis time (years)	– 0.007	0.944
Symptom duration (years)	0.021	0.832
Duration of morning stiffness (min.)	0.622	< 0.001
CRP (mg/L)	0.050	0.609
ESR (mm/h)	0.146	0.131
BASDAI	0.853	< 0.001
ASDAS-CRP	0.751	< 0.001
ASDAS-ESR	0.748	< 0.001
NRSGLOBAL	0.799	< 0.001
MASES	0.510	< 0.001
BASFI	0.672	< 0.001
ASOoL	0.717	< 0.001
MSPSS	– 0.511	< 0.001
JSS	0.838	< 0.001
HADS-D	0.650	< 0.001
HADS-A	0.586	< 0.001
B-IPQ total	0.692	< 0.001
Consequences	0.669	< 0.001
Timeline	0.232	0.016
Personal control	– 0.281	0.003
Treatment control	– 0.270	0.005
Identity	0.728	< 0.001
Concern	0.503	< 0.001
Comprehensibility	– 0.206	0.032
Emotional response	0.626	< 0.001

ρ Spearman’s rho, CSI Central Sensitization Inventory, TL Turkish Lira, BMI Body Mass Index, CRP C-reactive protein, ESR erythrocyte sedimentation rate, BASDAI Bath Ankylosing Spondylitis Disease Activity Index, ASDAS-CRP Ankylosing Spondylitis Disease Activity Score with C-reactive protein, ASDAS-ESR Ankylosing Spondylitis Disease Activity Score with erythrocyte sedimentation rate, NRS numeric rating scale, MASES Maastricht Ankylosing Spondylitis Enthesitis Score, BASFI Bath Ankylosing Spondylitis Functional Index, ASQoL Ankylosing Spondylitis Quality of Life Questionnaire, MSPSS Multidimensional Scale of Perceived Social Support, JSS Jenkins Sleep Evaluation Scale, HADS-D Hospital Anxiety and Depression Scale – Depression, HADS-A Hospital Anxiety and Depression Scale – Anxiety, B-IPQ Brief Illness Perception Questionnaire

The results of multiple regression analysis

After univariate correlation analysis, multiple linear regression analysis was performed to determine the predictors of the CSI score. NRS_GLOBAL, BASDAI, sleep quality (JSS), HADS-D and HADS-A were the clinical variables affecting the CSI score most (Table 4).

Table 4.

Multiple linear regression analysis to identify the determinants of CSI score

	β	p
NRSGLOBAL	1.78	0.011
BASDAI	2.81	0.001
JSS	1.08	< 0.001
HADS-D	0.47	0.035
HADS-A	0.40	0.040

Adjusted R² = 0.844

CSI Central Sensitization Inventory, NRS numeric rating scale, BASDAI Bath Ankylosing Spondylitis Disease Activity Index, JSS Jenkins Sleep Evaluation Scale, HADS-D Hospital Anxiety and Depression Scale-Depression, HADS-A Hospital Anxiety and Depression Scale-Anxiety

Multiple logistic regression analysis was performed to identify the predictors of the development of CS. BASDAI (OR: 10.44, 95% CI: 2.65–41.09), MASES (OR: 2.47, 95% CI: 1.09–5.56), and HADS-A (OR: 1.62, 95% CI: 1.11–2.37) were determined as the independent risk factors for the development of CS in patients with axSpA (Table 5).

Table 5.

Multiple logistic regression analysis to identify the risk factors for the development of CS

	Odds ratio	95% CI for odds ratio	p
BASDAI	10.44	2.65–41.09	0.001
MASES	2.47	1.09–5.56	0.030
HADS-A	1.62	1.11–2.37	0.013

BASDAI Bath Ankylosing Spondylitis Disease Activity Index, MASES Maastricht Ankylosing Spondylitis Enthesitis Score, HADS-A Hospital Anxiety and Depression Scale – Anxiety

Further multiple linear regression among patients with CS (CSI score ≥ 40) revealed that higher NRS_GLOBAL, JSS, HADS-D, and HADS-A scores and diagnosis of non-radiographic axSpA appeared as the determinants of CS severity (Table 6).

Table 6.

Multiple linear regression analysis to identify the predictors of the severity of CS

	β	p
Constant	37.85	< 0.001
AS	– 5.47	0.013
NRS GLOBAL	2.64	< 0.001
JSS	0.58	0.019
HADS-D	0.48	0.025
HADS-A	0.46	0.017

Adjusted R² = 0.627

AS ankylosing spondylitis, NRS numeric rating scale, MSPSS Multidimensional Scale of Perceived Social Support, JSS Jenkins Sleep Evaluation Scale, HADS-D Hospital Anxiety and Depression scale-Depression, HADS-A Hospital Anxiety and Depression Scale-Anxiety

Discussion

In this cross-sectional study among patients with axSpA, the frequency of CS was 57.4%. In previous studies, the frequency of CS ranged from 45 to 60% in SpA [7, 11–13]. The present study also revealed that patients with axSpA and CS had lower socioeconomic status, lower perceived social support, worse sleep quality and mood (depression and anxiety), as well as higher negative illness representation of disease compared to patients without CS. Additionally, disease-related parameters (disease activity measurements, functional status and QoL) of patients with CS were more impaired compared to patients without CS. We also found a moderate-strong correlation between the CSI score and all of the clinical variables mentioned above. The findings of the current study support the data on axSpA in the literature. Karlıbel et al. [11] found a moderate-strong correlation between CSI score and pain, MASES, disease activity, QoL and sleep quality. In a study by Yücel et al. [13], clinical variables such as pain, duration of morning stiffness, BASDAI, MASES, disease-specific QoL, fatigue, sleep quality, disability, and depression were correlated with CSI score. In line with these findings, Kieskamp et al. [10] found that CS and illness perceptions were independently and significantly associated with disease activity (BASDAI and ASDAS-CRP). Kieskamp et al. [12] in their recently published subsequent study, determined that CS was strongly related to lower disease-specific and patient-reported QoL. Additionally, consistent with the literature [10, 11, 13], we found no significant associations between CSI score and objective inflammatory parameters including ESR and CRP. This finding indicates that patients with axSpA who experienced CS have a worse clinical status although the underlying inflammation is adequately managed. It points out that there is much more than inflammation among the pathophysiological mechanisms involved in the development of CS. In this regard, it would lead to several questions in terms of the development and severity of CS: Why do some patients with axSpA develop CS while others do not? Why does the clinical severity of CS differ between patients with axSpA? Further results of the current study helped us answer the above-mentioned questions. These results will be discussed in detail.

To identify the risk factors in the development of CS among the study sample, we performed a multiple logistic regression analysis. The logistic regression analysis revealed that the risk of developing CS is markedly increased in patients with higher disease activity (BASDAI), enthesal involvement, and anxiety score. BASDAI score appeared as a major determinant of CS. Each 1-unit increase in BASDAI score increased the risk of CS by 10.44 times. In addition, each 1-unit increase in MASES score and HADS-A score increased the risk of CS by 2.47 and 1.62 times, respectively. Increased MASES score indicates a multi-site state of inflammation and pain. In this regard, it would be an expected finding that increased enthesal involvement relates to increased CS risk. BASDAI, which was found as a more important predictor of CS, also evaluates enthesitis along with morning stiffness/inflammation, spinal pain, fatigue, and pain/swelling in peripheral joints. Yet, a very high-risk ratio of BASDAI (RR: 10.44) can also be related to other factors such as the limitations of BASDAI itself. BASDAI is based solely upon patient-reported data. The measure does not include objective parameters such as CRP and ESR. On the other hand, ASDAS is a composite measure that includes CRP or ESR [36]. Both ASDAS-ESR and -CRP did not appear to predict the risk of CS in axSpA. This finding also indicates that inflammation is not the only factor in the development of CS in axSpA patients. This result is also important in terms of the management of patients. In patients with high BASDAI score, it is of value to evaluate potential CS before assigning the patient as “non-responsive” to ongoing treatment, as the presence of CS may lead to over-exaggerated BASDAI scores and thereby modify the bDMARD treatment plan in axSpA [36]. If a patient is detected to have CS, proper management of the condition would be of value. Non-pharmacological strategies such as exercise and electrotherapy are used in CS syndromes [37]. Besides, pharmacological regimens (e.g. duloxetine, pregabalin) should be tailored when necessary. In their review, Nijs et al. underlined that using pharmaceutical treatments for CS alone did not produce satisfying outcomes. CS entails intricate psycho-neuro-immunological connections. Therefore, clinicians should create a personalized multimodal treatment strategy that includes pain neuroscience education, cognitively targeted exercise therapy, sleep management, and/or stress management [38].

Genetic predisposition, biopsychosocial, and cognitive behavioral factors including anxiety, depression, decreased perceived social support, kinesiophobia, pain catastrophizing and maladaptive illness perception are related to CS [14]. These bidirectional associations have been studied in several populations suffering from chronic pain, especially in patients with chronic low back pain [39], shoulder pain [40] and knee osteoarthritis [41]. Bilika et al. [40] found that patients with chronic shoulder pain and CS presented significantly higher scores in kinesiophobia, pain catastrophizing, anxiety, depression, and illness perceptions than patients without CS. Additionally, they reported that catastrophizing, depression, and functionality are predictive factors of the extent of CS symptoms in patients with chronic shoulder pain. Clark et al. [39] reported trait anxiety and the extremely defensive high anxious personality type as predictors of CS in patients with chronic low back pain. In a study by Campbell et al. [41], low sleep efficiency and high catastrophizing scores were associated with increased levels of CS in patients with knee osteoarthritis. Although the presence of biopsychosocial variables has been reported to interfere with treatment response [42], disease activity [10], medication adherence [43], mood and health-related QoL [44, 45], there is limited evidence regarding the potential relation of biopsychosocial variables and CS in patients with axSpA. In the current study, biopsychological variables appeared as the main drivers of CS in axSpA. Patients with higher perceived disease activity, poor sleep quality and mental health (depression and anxiety) were observed to have an increased risk of severe CS. Psychological distress is well-known to be correlated to pain sensitization. The current study confirmed previously reported association between psychological factors and CS. Severe stress, as well as, sleep disturbances appeared as glial activators which may lead to neuroinflammation and subsequent sensitization of the central nervous system [46] Sleep disturbances in patients with rheumatic diseases can cause exacerbated inflammation, fatigue, psychological distress, and impaired QoL [47]. Even in healthy subjects, one-night total sleep deprivation was found to develop anxiety, mechanical pain sensitivity and hyperalgesia [48]. A recent cross-sectional study found that sleep impairment and psychological status were important confounders of the severity of CS in patients with familial Mediterranean fever [49].

There are a number of limitations and strengths to be discussed. Our results are not easily generalizable to all patients with axSpA. AxSpA covers a broad spectrum of different inflammatory conditions such as inflammatory bowel disease-related SpA, psoriasis-related SpA, and reactive arthritis-related SpA. The majority of our study population comprised patients with AS. Another limitation is related to the tools used in the study. Many selected tools are subjective and can create a bias risk. An important strength of our study is its comprehensiveness in evaluating the potential contributors (including biopsychosocial factors) of CS in patients with axSpA.

Conclusions

In this study, CS was detected in more than half of patients with axSpA. In line with these data, the presence of CS should be investigated especially in patients with axSpA experiencing chronic pain. High disease activity (BASDAI), more enthesal involvement, and anxiety were detected to be risk factors for the development of CS. The severity of CS in patients with axSpA may increase in the presence of predictive factors including higher patient-perceived disease activity and impaired biopsychological variables (sleep disturbance and poor mental health). If predictive factors of development and clinical severity of CS can be recognized, at-risk patients can be targeted with multifaceted and individualized approaches in the treatment to diminish the risk of CS and associated burden.

Author contributions

All co-authors take full responsibility for the integrity and accuracy of all aspects of the research. All authors approve the submitted version of the manuscript. Authorship contribution: conception and design of study; AS, ICB, IT, SSZA, IU. Acquisition of data; AS, ICB, SSZA. Analysis and interpretation of data; AS, ICB, IU. Drafting the article or revising it critically for important intellectual content: AS, ICB, IT, SSZA, IU. Final approval of the version to be submitted: AS, ICB, IT, SSZA, IU.

Funding

No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this article.

Data availability

The datasets gathered during the preparation of this manuscript are available from the corresponding author upon reasonable request.

Declarations

Conflict of interest

The authors have declared no conflicts of interest regarding the publication of this manuscript.

Ethical approval and consent to participate

The study protocol was approved by the Local Ethics Committee of Cukurova University Faculty of Medicine (Date of approval: July 22, 2022, Number: 124/15) and the study was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Written informed consent was obtained from each study participant.