Background/Purpose:

In studies of rheumatoid arthritis using ultrasonography (US), findings of synovial thickening are often reported in semiquantitative scores. For synovial biopsies of small joints, selection by US score has been suggested. However, it is not known how consistently US scores translate into a specific measurable synovium thickness in a cohort of RA patients. The aim of this study was to assess consistency of the scoring and the relationship to measurable synovial thickness.

Methods:

Consecutive CCP+ clinic patients fulfilling the 2010 ACR/EULAR RA criteria with ≥1 tender/swollen joint were enrolled. US examinations of dorsal and ulnar wrist and MCPs 2-5 were obtained. All patients were independently scanned by two experienced rheumatologists. A GE Logiq E9 unit was used, with gray scale (GS) frequencies of 15-18 MHz. Measurements were taken using the machine’s calipers at maximal joint distension in the dorso-palmar plane. Anatomic landmarks included the hyperechoic bony cortex (or deep hyperechoic duplication of the fibrous joint capsule, if present and visible); intervening hypoechoic synovial tissue; and hyperechoic fibrous joint capsule superficially. Still images were scored for GS synovitis (0-3) by both rheumatologists independently. Tenosynovitis within the extensor carpi ulnaris (ECU) tendon sheath was scored as present (1) or absent (0).

Results:

10 sets of images were scored (7 female and 3 male; biologics, n=3; DMARDs alone, n=4; steroids alone, n=1). 45 joints/regions were measured for maximum depth of synovial thickening (MCP = 35, wrist = 8 and ECU = 2).

Excellent inter operator and inter-reader agreement was achieved for semiquantitative scoring with weighted kappas of 0.98 (95% CI 0.93, 1.00, p<0.001, PABAK 0.99)

Average synovial thicknesses for the semi-quantitative scores at each joint region are shown below and were significantly different across the semiquantitative US scores (independent samples Kruskal-Wallis test p<0.001). 26 of 28 joints (92.9%) scoring GS ≥2 had a synovial depth of ≥2mm. All wrist joints (6/6) with ≥2 GS synovitis had ≥2mm depth, while 20 of 22 MCP joints with ≥2 GS synovitis had ≥2mm depth.

Conclusion:

The vast majority of hand and wrist joints graded semiquantitatively at 2 or 3 had at least 2 mm synovial thickening and would be theoretically amenable to biopsy (with a typical 2 mm instrument such as Quick-Core®). Even though experienced rheumatologists can reach excellent inter-reader reliability for semiquantitave US scores, we found significant variation of the actual measured synovial thickness within a given score category. The same score can translate into different mean measured synovial thickness across different joint areas. Greater synovial thickness was measured for wrist joints than MCP joints for a given category. This needs to be taken into account when planning synovial biopsies of small joints.

Funding support provided in part through NIAMS AMP UH2AR067690

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/correlating-semiquantitative-ultrasound-scores-with-measured-synovial-thickness/