Pos0947 subclinical synovitis in anti-ccp positive individuals at-risk of ra: how likely is it to resolve?

Background Subclinical synovitis is a relatively common ultrasound (US) finding in individuals at-risk of rheumatoid arthritis (RA) (33%) [1]. It is associated with the development of inflammatory arthritis (IA); which often results in physicians prescribing disease modifying anti-rheumatic drugs (DMARDs) in the absence of clinical inflammation. Prediction of resolution of these changes may prevent overtreatment in a number of these patients. Objectives: 1) To assess reversibility of subclinical synovitis in individuals at risk of RA, within 12 months’ time. 2) To investigate factors associated with resolution of subclinical synovitis. Methods A single centre, prospective, observational study recruited anti-cyclic citrullinated peptide (anti-CCP) antibody positive at-risk individuals with a new musculoskeletal complaint, but no clinical synovitis. US scans of wrists, metacarpophalangeal joints, proximal interphalangeal joints and metatarsophalangeal joints were performed at baseline. Those with subclinical synovitis [grey scale (GS) ≥1 and power doppler (PD) ≥1] in at least one joint were then selected. The following variables were collected: gender, age, smoking exposure, anti-CCP titre, the presence of rheumatoid factor, antinuclear antibodies (ANA) and shared epitope, levels of CRP (mg/dL) and ESR (mm/h), minutes of early morning stiffness (EMS), number of tender joints, number of painful joints reported by the patient, number of joints with subclinical synovitis and number of bone erosions. Another US scan was performed after 12 months. If participants had been diagnosed with an IA by then, they were considered to have persistent synovitis and the US scan was not repeated. Univariate analysis was used to identify relevant factors for resolution of subclinical synovitis. Multivariable binary logistic regression and ROC curves were performed to identify the best prediction model. Results A total of 91 individuals presented with subclinical synovitis on US. This resolved in 47.3% (43/91) of cases, whereas US changes persisted in 52.7% (48/91) of participants [out of which 62.5% (30/48) developed an IA within 12 months]. Univariate analysis showed that a low anti-CCP titre [<3x upper limit of normality (ULN)], a negative rheumatoid factor, no smoking exposure, age ≤55, ESR≤13, absence of bone erosions and <2 joints with subclinical synovitis at baseline were associated with resolution of subclinical synovitis within 12 months (p<0.05). ROC curves were performed to identify the best cut-offs for continuous variables. Statistically significant factors were included in a multivariate binary logistic regression analysis, which was also adjusted for confounders (age, smoking status and gender). The ROC curve (Figure 1) of the final multivariate prediction model showed AUC=0.819 (p<0.001), with anti-CCP low titre having an odds ratio (OR) of 3.43 for resolution of subclinical synovitis [95%CI (1.07-11.05); p=0.039] and ESR≤13 having OR=3.41 [95%CI (1.22-9.56) p=0.019]. Conclusion Subclinical synovitis is a common finding in anti-CCP positive at-risk individuals, which has shown association with progression to IA. However, by 12 months’ time, these US changes disappear in approximately half of the patients, with ESR≤13 and a low anti-CCP titre being the most significant predictors of resolution. Considering the risks associated with immunosuppressants, these data may prevent unnecessary treatment in these patients. Reference [1] Rakieh C. et al. Predicting the development of clinical arthritis in anti-CCP positive individuals with non-specific musculoskeletal symptoms: a prospective observational cohort study. Ann Rheum Dis. 2015 Sep;74(9):1659-66. Figure 1. ROC curve of the multivariate predictive model. Acknowledgements: NIL. Disclosure of Interests None Declared.