Background/Purpose:

Microwave Radiometry (MR) is an easy-to perform, rapid, non-invasive method detecting in-depth tissue temperature that may be useful for joint inflammation assessment in RA (1,2). We tested the hypothesis that MR can diagnose and grade (teno)synovitis in small and large joints in RA; thus, may be used for the development of an MR-derived combined joint-temperature score to measure RA disease activity.

Methods:

Eighty-two RA patients and 23 age- and sex-matched healthy individuals, underwent MR, clinical and laboratory assessments and joint ultrasound, as described (2); 21 patients were re-examined 2 months after treatment initiation. The temperature of each joint was expressed by Δt (difference between MR-derived temperature of pre-defined joint points resulting in lower Δt values in warmer joints). A thermo-score was created by summing the Δt of 7 small joints according to the US7 ultrasound score (wrist, 2^nd-3^rd MCP, 2^nd-3^rd PIP, 2^nd and 5^th MTP), as well as of elbow, knee and lower leg of the clinically dominant upper/lower extremity. This thermo-score was reproducible among healthy individuals (Intraclass Correlation Coefficient 0.714).

Results:

At the joint level, MR performed better than clinical examination in the knees to predict ultrasound confirmed joint effusion [area under the curve (AUC) 0.815 vs. 0.688, respectively), or power Doppler (AUC 0.901 vs. 0.791, respectively), whereas a Δt≤0.2 had 80% sensitivity and 82% specificity for power Doppler. Also, 85% of knee joints with subclinical inflammation could be diagnosed by MR. In all other joints MR performed similarly to clinical examination and its diagnostic performance was moderate. At individual patient level, the thermo-score strongly correlated to DAS28 score, tender and swollen joint counts, patient’s and physician’s visual analogue scale, CRP and ESR levels (all p<0.001), as well as to the US7 ultrasound scores (p<0.026). Moreover, the thermo-score could discriminate patients in high/moderate disease activity (mean±SD=4.6±3.0), low disease activity/remission (8.0±2.9) (p<0.001), or healthy subjects (8.8±1.9) (p<0.001). A thermo-score≤7.4 could distinguish highly/moderately active patients from healthy controls with a sensitivity and specificity of 81% and 83%, respectively (AUC 0.866), whereas a cut-off of 8.1 could distinguish highly/moderately active from patients in low activity/remission with both a sensitivity and specificity of 71% (AUC 0.802). Finally, individual changes from baseline to follow-up mirrored the corresponding DAS28 changes in 17/21 patients (p=0.001).

Conclusion:

Increased in-depth joint temperature detected by MR performed well for the detection of knee joint inflammation in RA. The MR-derived score of 10 selected joints correlated to clinical, laboratory and ultrasound measures of RA disease activity, could discriminate disease activity stages, as well as mirrored treatment-induced disease activity changes. Multicenter studies are needed to confirm these novel findings.

References

1. Zampeli E, et al. PLoS ONE 2013;8(5):e64606

2. Pentazos G, et al. J Clin Rheumatol 2018. E-pub ahead of print.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/microwave-radiometry-as-a-novel-additional-method-for-rheumatoid-arthritis-disease-activity-assessment-a-prospective-single-center-study/