Background: Up to 40% of ankylosing spondylitis patients report persistently high pain scores of >4 (scale of 0-10) even after responding to long-term TNF-alpha blocking therapy.[1] In other rheumatic diseases, nociplastic pain (due to altered functioning of the nervous system leading to peripheral and central sensitization) is common.[2] In axial spondyloarthritis (axSpA), patient illness and pain perceptions were shown to influence disease outcome.[3] Therefore, we hypothesized that central sensitization and patients’ illness perceptions are associated with persistently high disease activity in axSpA.

Objectives: To investigate to what extent central sensitization, pain catastrophizing and patients’ perceptions play a role in axSpA and to explore associations with disease activity.

Methods: Between April and September 2019, consecutive outpatients from the Groningen Leeuwarden axSpA (GLAS) cohort,[4] an ongoing large prospective cohort, were included in this study. Besides the standardized assessments, patients filled out three additional questionnaires: Central Sensitization Inventory (CSI), Pain Catastrophizing Scale (PCS) and Revised Illness Perception Questionnaire (IPQ-R). Univariable and multivariable linear regression analyses were used to investigate the association of CSI, PCS and each of the eight subscales of the IPQ-R, and disease activity assessments ASDAS-CRP, BASDAI, and CRP. We corrected for the following potential confounders: gender, symptom duration, BMI, educational level, smoking status and HLA-B27 status.

Results: Of 171 included patients, 58% were male, 79% were HLA-B27 positive, median symptom duration was 21 (IQR 10-32), mean ASDAS-CRP 2.1 ± 1.0, mean BASDAI 3.9 ± 2.2 and median CRP 2.9 (IQR 1.2-6.3). Mean CSI score was 37.8 ± 14.1 (scale of 0-100), and 44% of patients scored ≥40 on the CSI.[5] Median PCS score was 15 (IQR 7-22) (scale of 0-52), median IPQ-R illness identity subscore 3 (IQR 2-4) (scale of 0-14) and mean IPQ-R treatment control subscore 18.1 ± 3.4 (scale of 5-25). In univariable regression analysis, CSI and PCS scores and IPQ-R subscores all showed significant associations with ASDAS-CRP, and all except the IPQ-R subscale personal control showed significant associations with BASDAI. Only IPQ-R treatment control was significantly associated with CRP. Central sensitization, two IPQ-R subscales (perceived treatment control and the number of symptoms patients attributed to their axSpA: illness identity) and BMI were independently associated with disease activity assessments BASDAI (R²=0.46) and ASDAS-CRP (R²=0.36) (Figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Conclusion: In this axSpA population with long-term disease, 44% scored above the CSI cutoff point of 40, indicating a high probability of central sensitization. CSI score, illness identity and treatment control were independently associated with disease activity assessments.

References: [1]Arends S et al. Clin Exp Rheumatol 2017;35(1):61-8.

[2]Meeus M et al. Semin Arthritis Rheum 2012;41(4):556-67.

[3]Van Lunteren M et al. Arthritis Care Res (Hoboken) 2018;70(12):1829-39.

[4]Arends S et al. Arthritis Res Ther 2011;13(3):R94.

[5]Neblett R et al. J Pain 2013;14(5):438-45.

Disclosure of Interests: Stan Kieskamp: None declared, Davy Paap: None declared, Marlies Carbo: None declared, Freke Wink Consultant of: Abbvie, Janssen, Reinhard Bos: None declared, Hendrika Bootsma Grant/research support from: Unrestricted grants from Bristol-Myers Squibb and Roche, Consultant of: Consultant for Bristol-Myers Squibb, Roche, Novartis, Medimmune, Union Chimique Belge, Speakers bureau: Speaker for Bristol-Myers Squibb and Novartis., Suzanne Arends Grant/research support from: Grant/research support from Pfizer, Anneke Spoorenberg: None declared