AB1222 Comparison of quantitative mri fat-fraction measurement in sij joint on different scanner platforms. (12th June 2018)

• Record Type: Journal Article
• Authors: Bainbridge, A.; Bray, T.; Jones, J.; Tansley, S.; Fulstow, N.; Sengupta, R.; Hall-Craggs, M.
• Is Part Of: Annals of the rheumatic diseases. Volume 77(2018)Supplement 2
• Page Start: 1709
• Page End: 1709
• Publication Date: 2018-06-12
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2018-eular.5411
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Quantitative MRI Proton Density Fat Fraction (PDFF) measurement can objectively identify areas of active inflammation and structural damage in spondyloarthritis by identifying oedema and fat metaplasia in bone marrow. 1 PDFF is typically implemented as a 3D multi-echo Gradient-echo protocol and all of the major MRI equipment manufacturers have PDFF imaging products available that combine a complex multi-peak fat-spectrum model with simultaneous determination and correction for T2* signal decay. However, if the appropriate specialist imaging option has not been purchased, as is commonly the case, then the user may have to rely on simpler base-level protocols for fat-fraction measurement. An important step to implementing quantitative fat-fraction measurement as a clinical tool is to demonstrate that measurements are robust to both scanner platform and measurement technique. Objectives: We aimed to compare quantitative FF measurements obtained using 3 scanners supplied by different manufacturers, and using a range of techniques. Methods: The sacroiliac joints (SIJ) of 10 healthy volunteers (mean age 42 years, range 23–63) were scanned on 3 scanners: a) 3T Philips Ingenia, using MDixon Quant (MDQ) – a specialist PDFF package; b) 1.5T GE Optima MR450W, using IDEAL and LAVA – both base-level options; and c) 3T Siemens Skyra, using DIXON – a base-level option. We also scanned a fat-water phantom constructed using tubes of emulsions of peanut oil and 3% agar. MDQ uses data from 6 different echo-times; IDEAL from 3; and LAVA and DIXON from 2. The excitation flip angle was set to 3 degrees for all acquisitions to minimise the T1 weighting. Each technique yields Fat-only (F) and Water-only (W) images. FF images were calculated using the formula: FF=F/(W+F). Regions of interest (RoIs) were drawn in the SIJ as shown in the figure 1. MDQ PDFF was used as the 'gold-standard' measurement for comparison to IDEAL, LAVA and DIXON. Results: The figure shows the results of the RoI analysis. For the phantom measurements, both LAVA and IDEAL correlated well with MDQ, but Siemens DIXON deviated away from the expected linear relationship. The in vivo data showed good linear relationships with MDQ for all 3 base-level techniques for FF in the range 0.4–0.8, although all 3 slightly underestimated FF compared to MDQ. Conclusions: A clinically useful quantitative image-analysis tool should be applicable to results obtained from several platforms. FF measurement in the bone marrow of the SIJ is robust to both technique and platform. The slight underestimation of FF values on base-level FF measurements is probably due to the influence of T2* weighting. However, these variations are smaller than the FF changes observed in areas of oedema or fat metaplasia. 1 FF measurement is a quantitative and widely applicable method for the diagnosis and assessment of spondyloarthritis. Reference: [1] Bray TJP, Bainbridge A, Punwani S, Ioannou Y, Hall-Craggs MA. Simultaneous Quantification of Bone Edema/Adiposity and Structure in Inflamed Bone Using Chemical Shift-Encoded MRI in Spondyloarthritis. Magn Reson Med. 2017; Acknowledgements: Funding: BANNAR; TB:ARUK grant 21369;MHC, UCLH BRC; Disclosure of Interest: None declared