Introduction Prospective studies of individuals at increased risk of developing rheumatoid arthritis (RA) will further improve the understanding of disease development. Ultrasound emerges clinically useful in detecting subtle inflammatory changes in rheumatic diseases.

Objectives To investigate ultrasound (US) detected changes as markers for future arthritis development.

Methods Patients presenting with musculoskeletal complaints and a positive Anti-Citrullinated Protein Antibody (ACPA) test at primary care, were referred to Karolinska rheumatology clinic for further rheumatic joint disease assessments. Those lacking arthritis by clinical and US examination (defined as synovial hypertrophy with Doppler activity) were recruited into the Risk-RA clinical research program, and followed-up by our multidisciplinary rheumatology team. A total of 64 patients with complete US records were included between years 2015→2016. Our patient demographics were 84% (n 54/64) females, mean age 49 (range 22–82) years, median relative ACPA (times cut-off) 26 (range 1–174) titer, median visual analogue scale (VAS) pain 30, median VAS patient global 28 and median C-reactive protein was 1 (0–20) at inclusion. Hand (Wrists, MCP’s, PIP’s, DIP’s) and feet joints were US-evaluated for synovial hypertrophy, hyperemia and bone erosions. The presence of wrist (compartments 1–6) and finger (flexor and extensor) tenosynovitis were assessed. Data from inclusion→follow-up visits until September 2017 were analysed. SPSS software version 25 was used (Univariate, Chi-square, T-test and Mann Whitney U-test) for comparisons.

Results At inclusion, none of the 64 patients had any signs of active joint inflammation. However, ultrasound changes for tenosynovitis were seen in 7 out of 64 patients, 3 of who also presented with mild hypertrophy without Doppler activity and one patient with mild Doppler hyperemia (without hypertrophy), and none with bone erosions at inclusion. Among all tendons evaluated bilaterally, tenosynovitis of the Extensor Carpi Ulnaris (ECU) wrist tendons (4 of 7 patients) and the 2nd finger flexor-tendons (3 of 7 patients) were most commonly affected.

Of the 57 patients without US-tendon changes, one had mild Doppler hyperemia (without hypertrophy). Patients with US-tendon changes were 86% (n 6/7) females, had mean age 56 years, median VAS pain 42, median VAS global health 20, mean 2.7 mg/L CRP, median relative ACPA titer 70 in comparison to patients without US-tendon changes, 75% females, mean age of 48 years, median VAS pain 24, median VAS global health 28, mean 2.7 mg/L CRP and median ACPA titer of 23. The numerical difference in pain and relative ACPA titer were non-significant (p>0.05).

After follow up for mean 18 months (range 1–18), 7 out of 7 (100%) with US-tendon changes at inclusion and 18 out of 57 (32%) without US-tendon changes developed arthritis. Patients with US-tendon changes compared to those without tenosynovitis at inclusion, developed arthritis within 12 and 11 mean months follow-up, respectively.

Conclusions Our study shows that tenosynovitis is a specific marker for arthritis development in ACPA-positive patients with musculoskeletal symptoms. The role of inflammatory spreading from tendons (synovial sheath) to synovial tissue within joints need to be further investigated.

Disclosure of interest None declared