Background Musculoskeletal ultrasound (MSUS) is a useful tool for the diagnosis of rheumatoid arthritis (RA), and can be used to help guide treatment escalation in patients with active disease. At the opposite end of the spectrum however, the relationship between clinical and ultrasound measures of remission remains poorly understood.

Objectives To define the relationships between clinical and ultrasound parameters in the setting of RA low disease activity and remission.

Methods Patients with a clinical diagnosis of established RA were recruited as part of an ongoing study of RA clinical remission. Patients who were taking biological therapies or who had received corticosteroids within the past 3 months were excluded. At the baseline visit, a 7-joint ultrasound scan (US-7 scan) was performed blinded to the clinical disease activity assessment. All scans were performed by the same trained operator (KB) using the same ultrasound machine. Greyscale (GS) and power Doppler (PD) synovitis was scored on a 4-point semi-quantitative scale (0–3); erosions were scored as either present (1) or absent (0). MSUS images were then re-scored in a blinded fashion by KB and BT, with intra- and inter-observer reliability assessed by Cohen's Kappa statistic with linear weighting. Multivariate ordinal logistic regression was used to analyse the significance of relationships between total GS, PD and erosion scores with age, sex, disease duration, seropositivity for rheumatoid factor (RhF) and/or anti-citrullinated peptide antibody (ACPA), smoking history, and clinical remission as defined by the 2011 ACR/EULAR Boolean criteria.

Results 29 patients were assessed in this preliminary cross-sectional analysis (19 [66%] female, 22 [76%] RhF and 20 [69%] ACPA positive) of median (IQR) age 67 (53–72) years and 6 (4–13) years following RA diagnosis. Methotrexate, sulfasalazine and hydroxychloroquine were prescribed to 25 (86%), 8 (28%) and 5 (17%) patients respectively. The median (IQR) DAS28-CRP score was 1.18 (0.97–1.81) and 16 (55%) patients satisfied ACR/EULAR remission. Median (range) total GS, PD and erosion MSUS scores were 5 (1–10), 0 (0–2) and 0 (0–8) respectively. Intra- and inter-observer MSUS score agreement ranged from good to excellent; intra/inter κ: GS 0.72/0.43, PD 0.92/0.83, erosions 0.70/0.65. In multivariate analyses, disease duration (p<0.001) and male sex (p=0.038) were independent predictors of a higher total erosion score only. Patients who were seropositive for either RhF and/or ACPA tended to have higher GS and PD scores though this was not statistically significant (p=0.053 and p=0.063 respectively).

Conclusions Joint erosions were more frequently seen in male patients with a longer duration of disease, the latter in keeping with increased cumulative inflammatory joint damage. Substantial levels of synovitis, particularly GS, were visualised by MSUS despite low clinical measures of disease activity. Whether such ultrasound findings provide added value in predicting subsequent disease course following the withdrawal of DMARD therapy is an important focus of our ongoing prospective study.

1. Backhaus et al. (2009) Arthritis Rheum; 61: 1194–201.

Disclosure of Interest None declared