FRI0051 The risk of aseptic arthroplasty loosening in patients with rheumatoid arthritis. (12th June 2018)

• Record Type: Journal Article
• Authors: Böhler, C.; Weimann, P.; Alasti, F.; Smolen, J.S.; Windhager, R.; Aletaha, D.
• Is Part Of: Annals of the rheumatic diseases. Volume 77(2018)Supplement 2
• Page Start: 572
• Page End: 573
• Publication Date: 2018-06-12
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2018-eular.3747
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Total joint arthroplasty (TJA) of the hip (THA) and knee (TKA) are well-established operations for end stage degenerative or inflammatory joint disease, and has excellent outcomes. In rheumatoid arthritis (RA), it is performed in about 25% of the patients. 1 According to registry data septic complications after TJA are more frequent in RA than in osteoarthritis (OA), which is likely linked to the immunomodulatory therapy that RA patients receive. 2 However, aseptic prosthesis loosening (APL) is the most common complication and it remains unclear whether RA per se is also a risk factor for non-infectious complications, e.g. by the presence of higher levels of systemic inflammation. Objectives: To compare the rates of APL between OA and RA patients, and to investigate the influence of disease activity levels on the risk for APL in RA patients. Methods: We identified all patients who underwent THA and TKA between 2002 and 2015 at our academic centre, and linked them with an existing prospective RA database to identify documented RA patients. Age and sex-matched OA patients were used as controls. Our primary endpoint were radiographic signs of APL as previously established. 3, 4 Radiographs were evaluated by two independent observers blinded to the clinical diagnosis. To explore the effects of systemic inflammation, we compared the time integrated level of disease activity by the Simplified Disease Activity Index (SDAI) during the year before an x-ray indicated loosening (for those with loosening) with the respective levels over one year before the last available x-ray (for those without loosening). We used nonparametric tests and the chi-square test to compare rates of loosening between RA and OA patients and to compare AUC SDAI between patients with and without APL. Additionally, we calculated a Cox proportional hazard regression to estimate the risk of loosening with increasing SDAI, stratified for TKA and THA. Results: Data of 49 RA patients and 88 OA patients were available for analysis. The rate of APL was 36.7% in the RA and 13.6% in the OA group (p=0.002). This was explained by a much higher rate of APL in the TKA group (RA: 34.4%; OA: 6.5%; p=0.001), while the rates in the THA group were only numerically different (RA: 41.2%; OA: 30.8%; p=0.528). In the RA group one year time integrated SDAI was significantly higher prior to loosening than in controls without loosening (p=0.043). In the Cox model, SDAI was also significantly related to time to APL with a Hazard ratio of 1.125 (95% CI 1.021–1.241) (p=0.018). Figure 1 depicts cox regression adjusted for AUC SDAI 3, 13 and 26 separately for TKA and THA. Conclusions: RA is not only a risk factor for infectious complications after TJA, but also for APL after THA or TKA. This effect might– at least partly – be explained by systemic and local inflammation in RA patients as depicted by higher levels of disease activity in RA patients with APL. In the context of treatment-to-target of RA, the presence of an arthroplasty should be considered as an indication for even more stringent control of disease activity. References: [1] Kapetanovic MC, et al. Ann Rheum Dis2008. [2] Schrama JC, et al. Arthritis Care Res2010. [3] Higuchi Y, et al. J Arthroplasty2017. [4] Lee BS, et al. Clin Orthop Relat Res2013. Disclosure of Interest: None declared