High Rate of Discordance Between Clinical and Ultrasound Assessment in RA

In real-life clinical practices, there is a high rate of disagreement between the clinical and ultrasound evaluation of disease activity in patients with rheumatoid arthritis (RA). According to study results published in Joint Bone Spine, factors related to both assessment modes were associated with significant discordance, leading to overestimation and underestimation of disease activity.

The investigators of this real-life study sought to examine the proportion of disagreements between clinical disease activity scores (DAS28) and ultrasound scores in assessing patients with RA and to identify factors associated with discrepancy in assessment.

The study included 1091 patients from the Swiss RA registry who had ≥1 concomitant DAS28 and ultrasound score recorded between March 2009 and January 2017. According to previously established cutoffs for the DAS28 and ultrasound scores, investigators categorized disease activity as remission, low to moderate, and high disease activity. They defined remission as DAS28 <2.6; high disease activity defined as DAS28 >5.1; and low-to-moderate disease activity defined by values between remission and high disease activity.

Researchers performed a longitudinal analysis on patients who had ≥2 assessments. The investigators examined factors potentially associated with disagreements in disease activity, including demographic characteristics (age, sex, care setting, smoking status, body mass index) and disease-related factors (disease duration, types of treatments, treatment duration, swollen joint counts, radiographic erosions, concomitant fibromyalgia or depression).

Of 2369 assessments performed on 1091 participants, 1196 (50.4%) assessments were considered discordant; disagreement varied from 47% to 62% among the 3 disease activity categories. Compared with the DAS28 scores, ultrasound scores overestimated or underestimated disease activity in 23.5% of cases; compared with ultrasound scores, DAS28 overestimated or underestimated disease activity in 26.8% of cases. In total, the overestimation of remission was around 50% whereas the underestimation of high disease activity was around 60%. When investigators used DAS28 as the reference, factors associated with significant disagreement included older age, longer disease duration, and a higher swollen joint count. When they used ultrasound scores as the reference, factors associated with discrepancy included individual clinical parameters of the DAS28 (but not age or disease duration), and the only factor related to ultrasound scores was the presence of tenosynovitis. In longitudinal analysis, the proportion of disagreements did not differ.

Limitations to the study included the use of multiple operators (however, in most cases, the same operator performed both clinical and ultrasound evaluations), insufficient data on potentially confounding factors including osteoarthritis status, and low sensitivity of the Doppler mode on some ultrasound machines.

In measuring the disease activity of patients with RA in real-life clinical practice, a high rate of disagreement was found between clinical and ultrasound assessments. This discrepancy remained high during follow-up when ultrasound assessors were aware of the clinical findings, perhaps indicating that these modes of assessment measure different aspects of disease activity.

Reference          

Zufferey P, Courvoisier DS, Nissen MJ, et al. Discordances between clinical and ultrasound measurements of disease activity among RA patients followed in real life [published online September 23, 2019]. Joint Bone Spine. doi:10.1016/j.jbspin.2019.09.010