Background/Purpose:  Ultrasound has been shown to detect subclinical synovitis in patients who are in clinical remission, in rheumatoid arthritis (RA). The value of power Doppler ultrasound (PDUS) to predict flares in patients with RA in remission has not been fully studied. To determine whether PDUS assessment of synovitis predicts short term flares in patients with RA in clinical remission.

Methods:  Consecutive RA patients in clinical remission (DAS28 < 2.6) were included. US examinations were performed at baseline by the same rheumatologist, blinded to clinical data, using an Esaote MyLab 70 machine (6-18 MHz broad band multifrequency linear transducer). A total of 20 joints of both hands were assessed: wrists, first to fifth MCPs and second to fifth PIPs. PDUS signal was evaluated on a semi-quantitative scale from 0 to 3. PD synovitis was defined as the presence of intraarticular USPD signal ≥ 1, and was treated as a dichotomous variable. On the same day a complete clinical assessment was performed by another rheumatologist. Patients were followed-up and regularly assessed every two/three months. Flare was defined as the requirement of a change in disease modifying antirheumatic drugs (DMARDs) (increasing dose, adding or changing DMARDs or biologics therapy) by the treating rheumatologist (blinded to baseline PDUS findings); or an increase in DAS28 >1.2 or a DAS28 >3.2 on follow-up. Baseline variables and the presence of PDUS, were compared among patients with and without flares in univariate analysis. Multivariable analysis using a Cox proportional hazards model, with flare as the outcome variable, and PD signal, demographic characteristics, and baseline disease activity as independent variables were also calculated.

Results: 80 patients fulfilling DAS28 remission criteria were included. Baseline patients’ characteristics are shown in the table. Among the 80 patients, 20 (25%) showed at least one joint with positive PDUS signal. Mean number of joints with PD signal was 1.75 (SD: 1.16, range 1-5). In 35 patients a treatment reduction was initiated by the treating rheumatologist, after inclusion in the study (blinded to baseline PDUS findings). Among the 80 patients on remission, 36 (45%) experienced a flare within follow up (median follow up (IQR): 15.4 (9.4-27.3) months). Flare occurred a median of 9.4 (IQR: 4.9-15.7) months after inclusion in the study (US date). In univariate analysis neither a positive PDUS signal, nor the reduction of treatment, nor the use of DMARDs in contrast with biologics, were associated with flares (table). In the multivariate Cox proportional hazards model, none of the variables were associated with an increased risk of flare.

Conclusion:  Among RA patients in clinical remission synovial inflammation by PDUS was seen in 25%, but it was not associated with disease flare in the short term. None of the variables studied were associated with increased risk of flare. Table. Patients’ characteristics.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/utility-of-power-doppler-ultrasound-detected-synovitis-for-the-prediction-of-short-term-flare-in-rheumatoid-arthritis-patients-in-clinical-remission/