OBJECTIVES

To analyse the performance of ultrasonography (US) to detect bone erosion progression at the patient level and at the joint level by the US score for erosions (USSe) in early-stage and late-stage rheumatoid arthritis (RA) over a 2-year follow-up.

METHODS

Clinical and demographic information was recorded at baseline, and hands and feet RX were scored according to the Sharp erosion score. USSe was performed at baseline and over 2 years of follow-up on six bilateral joints (MCP2, 3, 5; MTP2, 3, 5). Inter-examiner reproducibility was performed on 14 patients, and the smallest detectable change (SDC) was calculated. US progression was defined as a change in USSe > SDC.

RESULTS

71 patients were included: 22 (31.0 %) early RA, and 49 (69.0 %) late RA. The intra-class correlation coefficient values of the USSe for intra- and inter-examiner studies were 0.96 (CI95: 0.93-0.98), and 0.92 (CI95: 0.75-0.97), respectively. On US, erosions prevailed at baseline in MTP5 joints followed by MCP2 and MCP5 joints. With an SDC calculated at 2.3, 28 patients (39.4 %) were classified as progressors, 30 (42.3 %) were stable, and 13 (18.3 %) were regressors during the follow-up. At the joint level, erosion progression was significant on the MCP2 and MTP5 joints in early RA (p < 0.01) and on the MCP5 and MTP5 joints for all RA (p < 0.05).

CONCLUSIONS

US is a highly reproducible method that is able to detect erosion progression at the patient level for both early and late RA and at the joint level (MCP2 and MTP5) for only early RA.