AB0983 Contribution of Knee Involvement Evaluated by Doppler Ultrasound and Synovial Fluid Analysis in Rheumatoid Arthritis Disease Activity Assessment

Objectives The objective of this prospective, cross-sectional study was to determinate the contribution of concomitant non-inflammatory knee involvement evaluated by Doppler ultrasound (DUS) in rheumatoid arthritis (RA) disease activity assessment and when indicated, by synovial fluid (SF) analysis in a Rheumatology Clinic of a University Hospital Methods We evaluated 106 patients with RA who consecutively attended. Patients with a history of other musculoskeletal diseases were excluded. RA activity was evaluated by the Disease Activity Score (DAS28). Tenderness and swelling at the knees were recorded separate. Clinical assessment was performed by the same rheumatologist before US assessment. A B-mode (BM) and power Doppler (PD) US assessment was performed by the same rheumatologist, blinded to clinical and laboratory results. For each knee we evaluated and scored synovitis components in BM and PD [synovial effusion (SE), synovial hypertrophy (SH), PD signal] according to Outcome Measurement in Rheumatology (OMERACT) definition and scoring system (0-3). After DUS evaluation the knees were classified in two groups: 1) knee involvement probably due to RA [i.e. SH grade 2-3, with or without PD signal, or SH grade 1 with PD signal (1-3), independently of SE]; 2) knee involvement probably not due to RA: [i.e. SH grade 0 or SH grade 1 without PD signal, independently of SE]. If SE grade 2-3 was detected an arthrocentesis was performed. The SF analysis included a macroscopic and microscopic examination (i.e. cellular count, crystals) and bacterial culture. After SF evaluation knees were re-classified in three groups: 1) knee involvement probably due to RA: inflammatory SF (≥500 cells) and crystals absence; 2) knee involvement probably not due to RA: mechanical SF, with or without crystals; 3) uncertain: inflammatory SF (≥500 cells) with crystals. This last group was included in the group of knee involvement probably due to RA for the final analysis Results We evaluated 212 knees. Pain and/or swelling were present in 174 (72%) knees. Table 1 shows the results of DUS evaluation in clinically affected knees. After DUS evaluation 48 (27.59%) knees were classified in probably due to RA and 126 (72.41%) in not-probably due to RA. DAS 28 was re-calculated excluding the knees classified as probably not due to RA from swollen and pain joint counts. In addition, after SF analysis the knees were re-classified and DAS28 was re-calculated. All patients in remission according to initial DAS28 were also in remission according to both re-calculated DAS 28. The number of patients in remission significantly increased after DUS and DUS+SF assessments (23%, 31.7% and 33.7% respectly; p=0.004 and p=0.001 respectly) Conclusions Knee involvement was common in our RA patients. Knee involvement not due to RA might overestimate RA activity measured by DAS28. DUS and SF analysis may contribute to an accurate diagnosis of knee involvement in RA patients, which should be relevant for better therapeutic management Disclosure of Interest M. Montoro: None declared, I. Janta: None declared, R. Irace: None declared, M. Medina: None declared, B. Serrano: None declared, C. Mata: None declared, L. Martinez: None declared, J. Martinez: None declared, M. Hinojosa: None declared, N. Bello: None declared, J. Ovalles: None declared, J. Nieto: None declared, L. Valor: None declared, F. Lopez: None declared, I. Monteagudo Consultant for: Abbvie, Roche Farma, Bristol-Myers Squibb, Pfizer, UCB, General Electric Healthcare, MSD and Esaote, C. Gonzalez: None declared, E. Naredo Grant/research support: Grant/research support: UCB and MSD Consultant for: Abbvie, Roche Farma, Bristol-Myers Squibb, Pfizer, UCB, General Electric Healthcare, and Esaote, L. Carreno: None declared DOI 10.1136/annrheumdis-2014-eular.3895