Background/Purpose: To further decrease the burden of RA we need to identify patients early, preferably in absence of clinical apparent synovitis. Recent studies in US suggest that earlier detection is possible. We aim to identify arthralgia patients developing clinical apparent inflammatory arthritis (IA) within a year using US to detect subclinical synovitis at first consultation.

Methods: We followed arthralgia patients with at least two painful joints of hands, feet or shoulders without clinical apparent synovitis over one year in a multi-centre cohort study. Their symptoms needed to be present <1 year and were not explained by other conditions (e.g. fibromyalgia). We collected data at baseline, 6 months and 12 months follow-up including physical examination, laboratory variables (ESR, CRP, auto-antibodies), diagnosis and medication. At baseline we also examined 26 joints (bilateral MCP2-5, PIP2-5, wrist, MTP2-5) by US and images were scored semi-quantitatively on greyscale (GS; 0-3) and power Doppler (PD; 0-3). US synovitis was defined as GS grade 2 or 3 and/or presence of PD. One-year incident IA was defined as clinical soft tissue swelling. Univariate logistic regression was used to analyse the association between demographic characteristics, clinical characteristics, and US findings and the incidence of IA for all patients. This was then further explored selecting the strongest variables (p<0.157) in a multivariate logistic regression. Missing values in independent variables were imputed using STATA (multiple imputation by chained equations (m=5)).

Results: In total, 196 patients were included of whom 160 completed the 12 months follow-up. At baseline 71 (37%) arthralgia patients had US synovitis and in 29 (15%) patients PD signal was detected. Incident IA was present in 37 (19%) patients (Table 1: baseline characteristics) of whom 22 patients initiated DMARD treatment during the one-year follow-up. Strongest associations in the univariate analysis were found for RF (OR 2.4), morning stiffness >30 minutes (OR 2.9), ACPA (OR 3.8), US synovitis (OR 2.1) and PD signal (OR 4.8). In the multivariate logistic regression the presence of morning stiffness >30 minutes (OR 3.7: 95%CI 1.4-9.5), ACPA (OR 4.0: 95%CI 1.5-10.6) and PD signal (OR 4.3: 95%CI 1.7-10.9) were associated with incident IA (Table 2).

Conclusion: One-year incident IA was present in 19% of the early arthralgia patients of which 53% showed US synovitis at baseline. The presence of PD signal, morning stiffness >30 minutes and ACPA were significantly associated with the development of IA after one year.