POS0993 RADIOLOGICAL CERVICAL INVOLVMENT IN ANKYLOSING SPONDYLITIS

BackgroundCervical involvement in ankylosing spondylitis (AS) is particularly disabling and has been poorly studied yet. Actual radiologic scoring design to assess disease progression does not provide a comprehensive picture of the cervical involvement in AS. [1] Association with clinical features such as psoriasis are discussed.ObjectivesWe aimed to assess radiographic features of cervical involvement in AS and clinico-biological parameters associated with this specific radiographic location.MethodsCross-sectional study based on radiographic analysis of a subgroup of patients included in the French BambooSpine cohort, originally designed to study the genetic risk factors of AS structural severity. The analysis was based on patients included in Parisian hospitals to have access to images of the entire spine on the Picture Archiving and Communication System (PACS) shared by all University Hospitals in Paris. A double reading of the images was performed by a rheumatologist/radiologist pair until consensus was reached, to identify syndesmophytes, zygapophyseal joint (ZJ) involvement and posterior ligament structures (PLS) at each cervical level, syndesmophytes in the thoracic region, and ZJ and PLS involvement in the lumbar region.ResultsOf the 113 assessed patients, 101 were men, mean age 53 years (+/- 11 years) with 27 years (+/- 13 years) of disease duration (Table 1). Of those whose HLA B27 status was known, 85% (70/82) were carriers. Among 86 patients with radiological cervical involvement, 83% had syndesmophytes, 86% had ZJ involvement and 24% had PLS involvement. In the cervical region, 13 patients (15%) had zygapophyseal fusion without syndesmophytes. 26/113 patients were completely free of any cervical involvement while 30/113 had a maximal cervical involvement (anterior and posterior). In univariate analyses, HLA-B27 was significantly associated with ZJ involvement at the cervical and lumbar level (p=0.016). Cervical involvement of any type was not associated with psoriasis. Low educational level (not beyond secondary school) was significantly associated with syndesmophytic involvement (p=0.035). There was a non-significant trend for an association between arthritis and ZJ involvement.Table 1.Clinical featuresMen (n,%)101/11389,4%Age at diagnosis (mean, SD)31,211,68Duration of evolution (mean, SD)27,013,0Smoking ever (n,%)65/11258,0 %Chronic Back Pain (for > 3 months) (n,%)109/11396,5%Chest pain45/11040,9%Buttock pain78/10971,6%Arthritis36/11132,4%Enthesitis38/11233,9%Dactylitis6/1115,4%HLA B2770/8285,4%CRP87/10384,5%X ray sacro-iliitis112/112100%Personnal history -Uveitis41/11336,3% -Inflammatory bowel disease13/11311,5% -Reactive arthritis2/1121,8% -Psoriasis19/11117,1%ConclusionCervical involvement was very frequent (76% of this severe AS population) but not associated with psoriasis, as usually thought. Zygapophyseal involvement was present in 86 % of cases and exclusive in 15 % of cases. This latter radiological form of the disease, i.e. without syndesmophytes, is usually more difficult to diagnose, and should be systematically assessed among AS patients with cervical pain and/or patients with reduced cervical mobility.References[1]Wanders AJB, Landewé RBM, Spoorenberg A, et al. What is the most appropriate radiologic scoring method for ankylosing spondylitis? Arthritis & Rheumatism 2004Disclosure of InterestsNone declared