AB1127 ULTRASOUND INTER-READER RELIABILITY OF INFLAMMATORY FINDINGS IN PATIENTS WITH POLYARTHRITIS

Background Ultrasonography is an imaging technique that allows rheumatologists to visualise structural and inflammatory changes within a joint. Objectives The objective of this study was to assess the inter-reader reliability of interpretation of inflammatory and destructive changes in a wide range of joints in patients with polyarthritis. Methods This study was divided in two parts: 1) consensus process and 2) reliability exercise. For the first part, a written questionnaire was sent by email to 6 sonographers from 3 portuguese hospitals with the highest level of competence (EULAR competency assessment level 2). The questionnaire included 17 questions divided in two groups: 1) elementary components in B-mode and Doppler assessment (effusion, synovial hypertrophy (SH), power Doppler (PD), erosions and synovitis definition) and 2) approach at the joint level (the definition of which plan and recess will be assessed in each joint). The participants were asked to rate their level of agreement/disagreement for each statement using a 1-5 Likert scale (1=strongly disagree to 5-strongly agree). For the reliability exercise, video clips of US examinations of 40 joints (wrist, metacarpophalangeal (MCP) from 1 to 5, proximal interphalangeal (PIP) from 1 to 5, knee, tibiotarsal (TT) and metatarsophalangeal (MTP) joints from 1 to 5, elbow and shoulder) from each of 15 patients were collected (showing a multiplanar bilateral ultrasound approach). Each joint in each video was scored by individual ultrasonographers for the presence/absence of elementary components: effusion (Yes/No), SH (No/Grade 1 to 3), PD (No/Grade 1 to 3) and erosions (Yes/No). Inter-reader agreement analysis was assessed through Fleiss’ kappa coefficient and classified according to Landis and Koch[8]: κ values Results Thirty seven joints of the 600 joints were excluded due to dislocation of the joint or presence of objects (rings/catheters) and the videos of a total of 563 joints were analysed by the 6 ultrasound experts. Inter-reader agreement was superior for TT joints and inferior for wrist; the identification of the erosions had the better agreement in the elementary components (Table 1). Conclusion The reliability of interpretation of inflammatory and destructive changes using video clips was in general good to excellent and it was better for erosions and tibiotarsal joint (regarding elementary component and anatomical region, respectively). Disclosure of Interests None declared