Background Synovitis is an important biomarker in arthritis and is visualised well using contrast enhanced magnetic resonance imaging (MRI). In osteoarthritis (OA) synovitis has been shown to correlate with histological change, pain and structural progression . Currently, MRI combined with gadolinium-based contrast agents is the gold standard for assessing synovitis but it has significant potential risks. Imaging of synovitis without contrast agent has been explored before  but with little success. Multi-parametric imaging has been used in MRI to increase the accuracy of detection of pathology within the prostate gland .
Objectives This study investigated if multi-parametric can be used to image synovitis associated with knee OA without the need for contrast agent.
Methods 28 patients (10 male, 18 female; average age: 57.2 years) with a diagnosis of osteoarthritis of the knee were imaged on a 3T Siemens® Verio scanner using two small flex receiver coils. 3D Gradient echo sequences were used to obtain parametric images based on Magnetisation Transfer (MT) and T1 relaxation time. MT Ratio (MTR) images and T1 maps were calculated and analysed using OsiriX® software. Additionally, fat images obtained using a water/fat separation technique (VIBE Dixon) were superimposed onto MTR images. The Gold standard contast-enhanced images were obtained using T1 VIBE with Water Excitation [WE] and 0.1mmol/kg Dotarem® contrast agent. 50 regions of interest representing both normal structures of the knee and synovitis were identified on the contrast-enhanced images and copied to corresponding points on T1 maps and MTR images and the mean values calculated.
Results A significant difference in mean values between the synovial tissue and other structures of the knee was identified on T1 maps and MTR images. The use of fat images superimposed onto the MTR images increased the visual conspicuity of synovitis.
By assessing and recording the differences in image signal intensity on quantifiable images, the distribution and volume of synovitis is better defined than when derived from one sequence.
Conclusions MT has previously shown the ability to demonstrate synovitis  and multi-parametric imaging has been utilised to assess prostate pathology on MRI scans prior to biopsy . In the knee the conspicuity of synovitis may be increased by using two or more different sequences in combination. The addition of fat images from VIBE Dixon scanning provides a visual demarcation of fat tissue allowing for better delineation of synovial tissue. The MTR and T1 images also provide a quantitative measure that may be useful in assessing the response to treatment in a longitudinal study.
Multi-parametric imaging of synovitis appears to offer a patient-friendly, low risk method for visualising synovitis in the knee of patients with osteoarthritis.
Loeuille D, et al. Arthritis and rheumatism. 2005 Nov;52(11):3492-501.
Pelletier JP et al Osteoarthritis and cartilage/OARS, Osteoarthritis Research Society. 2008;16 Suppl 3:S8-13.
Delongchamps NB et al. Prostate Cancer and Prostatic Diseases (2011) 14, 232–237.
Burnett C et al.Poster 2793. ISMRM2011.
Acknowledgements The work is supported by funding from The National Institute of Health Research.
Disclosure of Interest None declared