LAUSR

Background The ASAS axial SpondyloArthritis (axSpA) criteria encompass radiographic (r-axSpA) and non-radiographic (nr-axSpA) disease to enable classification early, while the modified New York criteria (mNY) for Ankylosing Spondylitis (AS) require radiographic sacroiliitis. Studies of agreement between clinical diagnoses of axSpA and classification criteria are sparse, especially for nr-axSpA. Objectives To study the concordance between clinical axSpA diagnoses and classification criteria fulfilment (mNY and ASAS axSpA) in a population-based cohort of established axSpA, and to compare demographics and outcomes between radiographic axSpA and nr-axSpA. Methods Patients with clinical diagnoses (ICD-10) of AS (M45.9) or undifferentiated spondyloarthritis (USpA; M46.0, M46.1, M46.8, M46.9), followed at Skåne University Hospital, and living in a defined area of southern Sweden were assessed in a cross-sectional study. USpA patients had to report back pain ≥3 months before age 45, by telephone screening, to be eligible. To enable classification, included patients underwent clinical assessments, a classification questionnaire, blood testing (including HLA-B27), and scoring of plain X-rays and MRI scans of the sacroiliac joints by an experienced radiologist. Results Out of 233 patients with clinical AS or axial USpA, 198 (85%) fulfilled either mNY or ASAS axSpA classification criteria, while 35 (15%) met neither criteria set. Among 118 patients with a clinical AS diagnosis, 103 fulfilled ASAS axSpA criteria and 84 mNY criteria (positive predictive value [PPV] of clinical AS diagnosis for fulfilling mNY criteria was 71% and for ASAS r-axSpA 69%). For 115 patients with a clinical USpA diagnosis, 89 fulfilled ASAS axSpA criteria, while a higher number was classified as AS (n=48) than nr-axSpA (n=36; PPV of clinical USpA diagnosis for fulfilling nr-axSpA criteria was 31%) (Figure). Comparing characteristics between patients classified as radiographic axSpA (AS [mNY] and/or ASAS r-axSpA) vs. ASAS nr-axSpA, few differences were observed; the former were older, more often men, had longer disease duration and worse spinal mobility (Table).Table Clinical characteristics and outcome measures Radiographic axSpAa nr-axSpA Radiographic SpA vs. nr-axSpA No axSpA n=141 n=57 p-value* n=35 Age, years 54 (14) 45 (12) ≤0.001 49 (12) Male sex, n (%) 88 (62) 18 (32) ≤0.001 10 (29) Symptom duration, years 29 (14) 19 (11) ≤0.001 20 (13)† HLA-B27 positive, n (%) 115 (84) 48 (84) 0.83 4 (11)†§ ASDAS-CRP 1.9 (1.0) 1.9 (0.9) 0.94 2.0 (1.0) BASDAI 3.1 (2.3) 3.2 (2.1) 0.77 3.9 (2.4) BASFI 2.3 (2.4) 2.1 (2.1) 0.51 2.8 (2.3) BASMI 3.4 (1.8) 2.3 (1.1) ≤0.001 2.7 (1.0) † Mean (SD) unless indicated. Missing data 0-14% *Statistical comparisons by Student’s t-test or Chi-Square test aASAS r-axSpA and mNY together. †statistically significant difference (p<0.05) radiographic axSpA vs. no axSpA an d §nr-axSpA vs. no axSpA. Conclusion The overall concordance between clinical diagnoses and fulfilment of axSpA classification criteria was good, with >4/5 meeting any criteria. For disease subtypes, however, the agreement was substantially weaker, and a large group of patients with USpA in this established cohort fulfilled the mNY criteria for AS. The results indicate that in studies aiming to compare radiographic and non-radiographic axSpA, classification according to defined classification criteria is important. Acknowledgement JKW and EM contributed equally Disclosure of Interests Elisabet Lindqvist: None declared, Tor Olofsson: None declared, Anna Jöud: None declared, Mats Geijer Consultant for: Consultant for AbbVie, Novartis, and Pfizer., Johan K Wallman Consultant for: Consultant for AbbVie, Celgene, Eli Lilly, Novartis, and UCB Pharma. Elisabeth Mogard: None declared