Objective The purpose of the study is to evaluate the role of ultrasonography in the detection of musculoskeletal disease in subclinical form, and the possible association with different disease features.

Methods Sixty-four consecutive SLE patients followed at our Lupus Clinic were enrolled. Patients were divided into 3 groups according to the type of musculoskeletal involvement: GROUP 1: 22 patients with no previous or current musculoskeletal involvement; GROUP 2: 21 patients with arthralgias without objective signs of synovitis; GROUP 3: 21 patients with objectively detectable musculoskeletal involvement. All patients underwent osteoarticular ultrasound exploring 26 sites bilaterally: MCP II-V, PIP II-V, wrist, elbow, shoulder, knee, Achilles tendon. Images were grayscale (GS) and Power Doppler (PD) evaluated according to the OMERACT EULAR Score System definitions. The presence of ultrasound-evident arthritis was defined for each patient if there was at least 1 joint with GS score≥1 and PD≥1.

Results In 7 patients in GROUP 1+2 (16.3%), subclinical arthritis could be detected by ultrasound, particularly in 3 (6.9%) in Group 1 and in 4 (9.3%) in GROUP 2. In GROUP 1+2, a statistically significant difference in mean age at the time of ultrasound (50.92 ± 10.56 and 40.43 ± 10.40, respectively, p = 0.019) and mean age at diagnosis (38.50 ± 10.18 and 25.95 ± 9.71, p = 0.004) could be detected between patients with and without subclinical arthritis. Higher age at diagnosis (OR = 1.518, p = 0.049) and at the time of ultrasound (OR = 1.336, p = 0.018) were confirmed to be predictive factors for subclinical arthritis. The presence of subclinical arthritis was associated with normal C3 values (p = 0.015).

Conclusions The sensitivity of ultrasonography in detecting subclinical joint inflammation in patients with SLE is confirmed. The results should prompt consideration of the presence of arthritis even in patients without complement consumption. The correlation between age and subclinical arthritis suggests on the one hand that synovial inflammation may develop even in more advanced stages of disease, and on the other hand makes it imperative to exclude the possible bias of the presence of osteoarthritic pathology.