Intravoxel incoherent motion magnetic resonance imaging of the knee joint in children with juvenile idiopathic arthritis.

Abstract

MRI of synovitis relies on use of a gadolinium-based contrast agent. Diffusion-weighted MRI (DWI) visualises thickened synovium but is of limited use in the presence of joint effusion. To investigate the feasibility and diagnostic accuracy of diffusion-weighted MRI with intravoxel incoherent motion (IVIM) for diagnosing synovitis in the knee joint of children with juvenile idiopathic arthritis. Twelve consecutive children with confirmed or suspected juvenile idiopathic arthritis (10 girls, median age 11years) underwent MRI with contrast-enhanced T1-weighted imaging and DWI at 1.5T. Read-out segmented multi-shot DWI was acquired at b values of 0s/mm2, 200s/mm2, 400s/mm2 and 800s/mm2. We calculated the IVIM parameters perfusion fraction (f) and tissue diffusion coefficient (D). Diffusion-weighted images at b=800s/mm2, f parameter maps and post-contrast T1-weighted images were retrospectively assessed by two independent readers for synovitis using the Juvenile Arthritis MRI Scoring system. Seven (58%) children showed synovial hypertrophy on contrast-enhanced imaging. Diagnostic ratings for synovitis on DWI and on f maps were fully consistent with contrast-enhanced imaging, the diagnostic reference. Two children had equivocal low-confidence assessments on DWI. Median f was 6.7±2.0% for synovitis, 2.1±1.2% for effusion, 5.0±1.0% for muscle and 10.6±5.7% for popliteal lymph nodes. Diagnostic confidence was higher based on f maps in three (25%) children and lower in one child (8%), as compared to DWI. DWI with IVIM reliably visualises synovitis of the knee joint. Perfusion fraction maps differentiate thickened synovium from joint effusion and hence increase diagnostic confidence.