THU0250 Impact of gut involvement in early spondyloarthritis. the desir cohort

Abstract

Background: Inflammatory bowel disease (IBD) is a well-known extra articular feature of spondyloarthritis (SpA), with increasing evidence of a pathophysiological relationship. Objective(s): The aims of this study were to evaluate in the DESIR cohort a) the prevalence of IBD at baseline and M60, and b) the incidence of IBD over the first 5 years of follow-up, c) factors (epidemiological, clinical, imaging, biological data available in the database) associated with IBD, a on the prevalent cases at baseline and at year 5 and also on the incident cases. Method(s): DESIR is a prospective observational cohort of patients with recent onset (<3 years) inflammatory back pain, suggestive of axial SpA. All available factors in the database were compared between patients with and without past or present IBD at baseline and at M60, and with and without incident cases of IBD over the 5 years of follow-up, by uni and multivariate analysis. Result(s): At baseline, 706 patients were analysed, 35 had a past history or a concomitant IBD: prevalence 4.94% [CI 95%: 3.3-6.5]. IBD was significantly associated (univariate) with family history of IBD, DMARD use, steroid use, history of uveitis, elevated ESR and negatively associated with psoriasis, HLA-B27 and NSAID score. In multivariate analysis IBD was positively and independently associated with history of uveitis; OR 3.62 [1.95-6.74], levels of DKK-1: OR 1.03 [1.02-1.05] and TNF serum levels: OR 1.17 [1.08-1.26]. IBD was not associated with phenotypic presentation (peripheral arthritis, enthesitis, dactylitis, uveitis) or baseline serum levels of other cytokines (IL-6, IL-17 A, IL-17 F, IL-23, IL-23). At M60, 480 patients were analysed, 58 with IBD: prevalence 12.1% [9.17-14.99]. In univariate analysis on prevalent cases, IBD was associated with lower NSAID score, worse activity and function indexes (ASDAS-CRP, BASFI, SF-36, HAQ, ASQoL), use of DMARD anti TNF, more sick leave. In multivariate analysis, IBD was associated with fulfilment of modified New York criteria: OR 4.85 [2.23- 10.57], sick leave: OR 1.01 [1.005-1.014], BASDAI: OR 1.10 [1.05-1.16], and with smoking: OR 2.79 [1.53-5.07]. No association with MRI scores, enthesitis, psoriasis, BMD. After a 5 year follow-up period, 23 new incident cases of IBD were recorded, giving an estimated occurrence rate of 0.95/100 [0.57-1.35] patient-years in this population. Incidence of IBD was independently associated (multivariate) with: HLA B27: OR 0.36 [0.22-0.59], fulfilment of modified New York criteria at M0: OR 3.35 [1.85-6.08], family history of IBD: OR 3.31 [1.62-6.77]. Conclusion(s): In early SpA, IBD occurs with an incidence of around 1/100 patient-years, and is associated with poor outcome at 5 years, family history of IBD, absence of HLA-B27, fulfilment of modified New York criteria.