The influence of biological DMARDs on aseptic arthroplasty loosening: a retrospective cohort study

Abstract

Abstract Objective To investigate whether biological DMARDs affect the risk of aseptic loosening after total hip/knee arthroplasty (THA/TKA) in patients with RA. Methods We retrospectively identified all patients suffering from RA who underwent THA/TKA at our academic centre between 2002 and 2015 and linked them with an existing prospective observational RA database at our institution. The risk of aseptic loosening was estimated using radiological signs of component loosening (RCL). A time-dependent Cox regression analysis was used to compare the risk of implant loosening between patients treated with traditional DMARDS and biological DMARDs, or alternately both over time. Results A total of 155 consecutive total joint arthroplasties (TJAs) (103 TKA vs 52 THA) was retrospectively included in the study. Mean age at implantation was 59 ± 13 years. Mean follow-up time was 69 ± 43 months. Overall, 48 (31%) TJAs showed signs of RCL, with 28 (27.2%) RCLs occurring after TKA compared with 20 after THA (38.5%). A significant difference regarding the incidence of RCL between the traditional DMARDs group (39 cases of RCL, 35%) and the biological DMARDs group (nine cases of RCL, 21%) (P = 0.026) was observed using the log-rank test. This was also true when using a time-dependent Cox regression with therapy as well as arthroplasty location (hip vs knee) as variables (P = 0.0447). Conclusion Biological DMARDs may reduce the incidence of aseptic loosening after TJA in patients with RA compared with traditional DMARDs. This effect seems to be more pronounced after TKA than THA.