AB1471-HPR The reliability and validity of a novel MRI-based tool for the evaluation of forefoot bursae in patients with rheumatoid arthritis: The “FFB-score”

Background We recently demonstrated forefoot bursae (FFB) are a clinically important feature in patients with rheumatoid arthritis (RA) 1 . However, there is a need for an observer-independent, reliable and valid method of characterising FFB. Magnetic Resonance Imaging (MRI) allows improved visualisation and characterisation of FFB in multiple imaging planes. Objectives To determine the reliability and validity of a novel MRI-based scoring tool for the evaluation of FFB in patients with RA. Methods A collaborative process of tool design was completed by a team of rheumatologists, radiologists, and podiatrists from centres within the UK & Germany. In an iterative process of tool design, items to be included, grading criteria, overall utility and MRI sequences were determined. The FFB-Score assesses 9 distinct forefoot regions and contains 5 items; lesion presence, shape, enhancement and T1 & T2 characteristics. The final tool was evaluated on 42 consecutive patients with RA, mean(SD) age 62.2(±12) years, disease duration 15.3(±10.3) years, & DAS-28 3.1(±1.4), who were recruited from a UK rheumatology clinic. Images were obtained using a 1.5T whole body scanner and 4-channel flex extremity coil. The MRI protocol included coronal T1 & STIR, coronal & sagittal T1 post-gadolinium, and long axis 3D SPACE sequences. The intra and inter-reader reliability were evaluated using Percentage Exact/Close Agreement (PEA/PCA). Content validity was evaluated using Lawshes’ Content Validity Ratio. Discriminant validity, with regards to differentiation between high/low clinical markers of disease activity (DAS-28), MRI-determined disease activity (erosion, bone marrow oedema, synovitis) or foot-related disability (Foot Impact Scale), was evaluated using receiver operator characteristic curves and area under the curve analysis. Results The intra-reader reliability was determined as moderate for intermetatarsal soft tissue lesion shape (PEA:40% PCA:50%) and T1/T2 characteristics (PEA:60% PCA:90%, PEA:30% PCA:80% respectively) and good for all other items (PEA:50-90% PCA:100%). Inter-reader reliability was determined as good for all items (PEA:50-90% PCA:100%). Content validity was determined as good (Lawshes’ CVR:0.625). The FFB-score total lesion enhancement grading has good discriminant validity when differentiating between patients with high/low MRI-determined disease activity in the forefoot (erosion:p=0.011, synovitis:p=0.004, oedema:p=0.018). The FFB-score has good discriminant validity for high/low reported foot impairment and activity limitation (p=0.006, p=0.033 respectively). Conclusions The FFB-Score is a viable tool for the detection and evaluation of FFB in patients with RA, which demonstrates satisfactory intra/inter-reader agreement and validation. Further validation, assessment of responsiveness and refinement of the FFB-score is needed in order to maximize the potential utility in clinical trials and epidemiological studies. References Bowen CJ et al. The clinical importance of ultrasound detectable forefoot bursae in rheumatoid arthritis. Rheumatol. (Ox.) 2010;59:1:191-2 Disclosure of Interest None Declared