AB0264 THE EXTENT OF JOINT DESTRUCTION IN THE KNEE JOINT AT THE INITIATION OF BDMARDS TREATMENT IS A PREDICTIVE FACTOR OF SUBSEQUENT TOTAL KNEE ARTHROPLASTY AND PROGRESSION OF JOINT DESTRUCTION IN PATIENTS WITH RHEUMATOID ARTHRITIS DURING LONG-TERM BDMARDS TREATMENT

Abstract

Background The joint destruction inhibitory effect of biological disease-modifying anti-rheumatic drugs (bDMARDs) in patients with rheumatoid arthritis (RA) has been reported in a lot of studies. Most of them are evaluated in small joints such as hands and feet, there are few reports on the joint destruction inhibitory effect in large joints such as knees which are frequently affected in disease course of RA1). Although bDMARDs is very effective drugs, we experience RA patients in whom total knee arthroplasties (TKA) are performed after the initiation of bDMARDs treatment2). Objectives To investigate the rate of TKA during bDMARDs treatment and to explore associated factors of subsequent TKA after the initiation of bDMARDs treatment in RA patients. Methods 356 knee joints of 184 RA patients who initiated bDMARDs treatment in our institute from March 2004 to November 2013 were included in this retrospective study. They were retrospectively studied over the course of a minimum of 5 years of follow up. Knees in which TKA were already performed before the initiation of bDMARDs treatment were excluded in this study. Patients’ characteristics and rates of subsequent TKA over time were investigated. Next, factors associated with subsequent TKA were explored using univariate and multivariate analysis. The X-ray change (Larsen grade3), ARASHI score4)) at the last observation after 5 years or more was also examined. ARASHI score was developed in Japan to assess destruction of large joints such as knees or hips. Results Baseline patients’ characteristics were below. Mean age was 57.1 years old. 152 female and 32 male. MTX was concomitant in 85.7% of patients. Prednisolone was concomitant in 53.7% of patients, First bDMARDs were infliximab in 67 cases, etanercept in 67 cases, adalimumab in 35 cases, tocilizumab in 6 cases, abatacept in 9 cases, golimumab in 0 cases and certolizumab in 0 cases. Agents were changed according to clinical necessity. The median follow-up period was 7.8 years. TKA were performed in 20 knees (5.95%) in total. Baseline extent of joint destruction of knees which was evaluated using Larsen grade and ARASHI score, joint destruction, age, disease duration, steinbrocker class, concomitant corticosteroids, concomitant methotrexate, DAS28-CRP, SDAI, MMP-3, mHAQ, swelling of the knees,and tenderness of the knees were associated with subsequent TKA in univariates analysis. Multivariate analysis revealed that the extent of baseline joint Destruction and mHAQ were associated with subsequent TKA (Larsen grade: Odds ratio; 6.17, 95%, confidence interval; 2.93-12.98, ARASHI score: Odds ratio; 3.50, 95% confidence interval; 2.18-5.62, mHAQ : Odds ratio; 3.56, 95% confidence interval; 1.06-11.93). Cut-off value was calculated as 2 in Larsen grade, 3 in ARASHI score and 1.5 in mHAQ using receiver operating characteristic analysis. There were significant differences in the rates of subsequent TKA if knees were divided according to baseline Larsen grade, ARASHI score and mHAQ (Fig1). Baseline ARASHI status score in progressed knees which was worsened after over 5 years were increased compared with that in knees which was stable. There was no significant difference in baseline Larsen grade between progressed knees and non-progressed knees. Conclusion It was suggested that baseline joint destruction and mHAQ at the initiation of bDMARDs in RA patients was predictive factors of subsequent TKA and progression of joint destruction of knees during long-term bDMARDs treatment. Early intervention with effective bDMARDs is necessary to prevent destruction of knee joints and subsequent TKA in RA patients whose knee joints are affected.