Background: Remission in rheumatoid arthritis (RA) is now achievable in a significant proportion of patients using a combination of a treat to target strategy and biologic therapy. A number of clinical assessment tools exist for assessing remission. Several reports have shown that ultrasound (US) may have a role in better characterising this group of patients suggesting that subclinical synovitis increases the risk of erosive disease and flares in patients in DAS28 remission.^1–3 These observations hint at a stratum of patients where remission is incomplete and may require a more tailored approach to therapy.

Objectives: In an early RA cohort we examine the relationship between US imaging and histological synovitis in the context of clinical remission and examine the predictive value of US to confirm synovial inflammation.

Methods: A prospective, observational study of 122 DMARD naïve, early RA patients classified according to the 1987 ACR criteria, with a maximum disease duration of 12 months (MRC PEAC study www.peacmrc.mds.qmul.ac.uk). One hundred and three synovial biopsies were analysed at baseline and 75 at 6 months, with 85 paired US 12 joint scores (US 12: 10 metacarpophalangeal joints (MCPJ) and 2 wrists). US images were analysed using a using a semi-quantitative score for synovial thickness (ST) and power Doppler (PD).⁴ Synovial inflammation was assessed using the Krenn synovitis scoring system. Fisher exact statistical test and Spearman’s rank was used to determine the association and correlations.

Results: Demographics of this cohort are listed in table 1. There was a good correlation between US ST and PD scores with the Krenn histology score at the level of the single biopsied joint (ST: r=0.47, p<0.001, PD: r=0.5, p<0.001). An association continued to be demonstrated when extending the US data set to 12 joints (ST r=0.27 p<0.01, PD r=0.28 p<0.01).

Twenty-two patients with paired histology and ultrasound data were in DAS28 remission at 6 months and were eligible for analysis. A significant association between low PD (≤3/36) signal (but not ST) and low Krenn score (≤4) was demonstrated (Fisher exact test p<0.03) with a predictive value of 90%. This reduced to 80% in patients not in DAS remission at 6 months (n=42). Lastly, a clear relationship was noted between patients with US PD score recorded after 6 months of DMARD therapy (n=62, flares n=19). No subsequent flares were recorded during the course of the follow up period of 6 months with a low US PD score (p<0.002, negative predictive value 76%) and a high US PD score had a 86% positive predictive value for disease flare within this time course.

Conclusions: This study demonstrates that there is considerable validity in the use of US to assess disease activity, which reflects histological synovitis in patients in low disease activity states and remission. Ultrasound imaging may demonstrate a distinct clinical and histological remission cohort of patients and may be a useful predictive tool in terms of predicting subsequent clinical disease activity.

References:

1. Haschka, et al. ARD 2016.
2. Naredo, et al. Rheum 2015.
3. Brown, et al. Arthritis Rheum 2006.
4. D’Agostiono, et al. RMD open 2017.

Disclosure of Interest: None declared

DOI: 10.1136/annrheumdis-2018-eular.1264