Background Medication withdrawal for RA patients undergoing total hip (THA) and knee (THA) arthroplasty balances risk of infection with risk of flare, yet little is known about post-op flare.

Objectives To describe rates, characteristics and molecular profile of flare within 6 weeks of THA/TKA.

Methods Baseline (BL) sociodemographic, RA characteristics and tissue/blood samples for molecular profile, BL and 6 week DAS-28, and weekly flare & function questions (MDHAQ, HOOS ADL) were collected in RA patients undergoing THA/TKA. Patient-MD concordant flarers were identified using patient report of flare (“yes” to “Are you in a flare?”), flare intensity/duration combined with MD agreement by review of records. Medication management was standard of care: biologic DMARDs were stopped pre-op; steroids and MTX were continued. Measures of immune activation include Rho associated kinases (ROCKs) pathways, which act via phosphorylation of ezrin-radixin-moesin (pERM) protein substrates.

Results Included are results of the first 24 of 50 RA subjects; 20 THA and 4 TKA. 18/24 (75%) met RA 2010 classification criteria. Patients were (median (IQR)) 54 (18) yrs old with 14.5 (23) yrs of disease; 74% were female, 84% white, 88% college educated; 65% ACPA/RF positive. Median (IQR) time to flare was reported at 3.0 (3.0) wks in 14/24 (63%) of patients (flare intensity of 5 (4), duration >4 days in 79%; 1 already flared by BL. More flarers had stopped biologic DMARDs (67 vs 22%). All but 2 stopping biologics flared. More flarers were on steroids (6 vs 1), and MTX (11 vs 5). At BL, DAS-28 was not different (3.8 (1.5) vs 3.9 (1.3), p=.44) but flarers had higher ESR 14 (41) vs 2.5 (7), CRP 1.1 (4.3) vs 0.68 (0.9), and HAQ 4.7 (2) vs 2.0 (2.4). HOOS ADL was worse 34.6 (28.8) vs 44.2 (37.2). Trend for baseline lymphocyte associated pERM was higher in a subset of 5 flarers.

• Download figure
• Open in new tab
• Download powerpoint

Conclusions Flares are frequent and severe in RA patients undergoing THA/TKA, more after discontinuing biologics. Work to identify clinical and novel molecular signatures to predict flare are needed to optimize perioperative management. Activation of the ROCK pathway may signal and play a role in flare.

Disclosure of Interest S. Goodman: None declared, R. Friedlander: None declared, C. Figgie: None declared, A. Pernis Grant/research support from: Kadmon Pharmaceuticals, R. Khianey Maharaj: None declared, C. Rozo: None declared, E. DiCarlo: None declared, M. Figgie: None declared, V. Bykerk: None declared