Sensitivity and specificity of ultrasonography and low-field magnetic resonance imaging for diagnosing arthritis.

Objectives: To evaluate the value of grey-scale ultrasonography (US) including power Doppler ultrasonography (PDUS) and low-field magnetic resonance imaging (MRI) for the diagnosis of arthritis in a diagnostic phase III study.

Methods: Fifty consecutive patients with suspected arthritis were included in the study. Following a standardised protocol, US of the carpus and the metacarpophalangeal (MCP) joints of the dominant hand was performed. Subsequently, low-field MRI was done using standard sequences, with contrast agent (Gadolinium DTPA) administered to 29 patients.

Results: In 32 out of 50 patients a clinical diagnosis of arthritis was established. In grey-scale ultrasonography including PDUS, sensitivity and specificity were determined as 0.94 and 0.5, respectively, for synovitis (effusion and hypertrophy), 0.72 and 0.94, respectively, for Doppler signals, and 0.38 and 1.0, respectively, for bone erosions. In low-field MRI, sensitivity and specificity values were 0.77 and 0.75, respectively, for synovitis (when using contrast agent), 0.48 and 0.78, respectively, for bone marrow oedema, and 0.58 and 0.83, respectively, for bone erosion.

Conclusions: Both grey-scale ultrasonography including PDUS and low-field MRI are suitable imaging methods for diagnosing arthritis at an early stage. However, PDUS displays a higher specificity and almost the same sensitivity as compared to contrast-enhanced MRI, while being a much simpler and less costly procedure.