AB1546 GRADING SYNOVITIS IN WRIST AND HAND USING THE MRI DERIVED APPARENT DIFFUSION COEFFICIENT FAILS TO SHOW RELIABILITY AND REPRODUCIBILITY

Background Diffusion-weighted (DWI) magnetic resonance imaging (MRI) of the hand and wrist has been suggested as an outcome measure of synovitis as an alternative to gadolinium enhanced MRI in patients with rheumatoid arthritis (RA) [1-2]. The apparent diffusion coefficient (ADC), which is a derived parameter from the DWI, may be used to grade synovitis, similar to the OMERACT Rheumatoid Arthritis MRI scoring system (RAMRIS) for synovitis [1]. In a small study criterion validity has been measured but no correlation to contrast-enhanced MRI was revealed [2].Objectives To test the discriminative validity of ADC in the synovium (“synovitis ADC”) in a prospective cohort of patients with RA and psoriatic arthritis (PsA) and healthy controls (HC).Methods The right hand and wrist of all participants were imaged in a 3T MRI system with a dedicated 8 channels coil, applying a 2 mm thick coronal turbo spin echo DWI sequence with an in-plane resolution of 1.5x1.7mm. An ADC map was calculated on basis of two b-values (0;800). Assessment of ADC maps was performed in 7 regions of interest according to the same 7 areas which are assessed by the OMERACT RAMRIS (radioulnar-, radiocarpal-, and intercarpal-carpometacarpophalangeal joints and 2nd-5th metacarpophalangeal joints). Repeatability and intra-observer reproducibility were assessed using intraclass correlation coefficients (ICCs) and Bland-Altman plots.Results Thirty-eight participants were imaged twice within a week (median 7 days, range 3-14 days). Participant characteristics are provided in Table 1. The repeatability was moderate for RA (ICC=0.68; 95%CI: 0.45-0.83), poor for PsA (ICC=0.10; 95%CI: -0.09-0.29) and poor for HC (ICC=0.28; 95%CI: 0.07-0.47). Intra-reader reproducibility was poor for all three groups: RA (ICC=0.37; 95%CI: 0.12-0.58), PsA (ICC=0.22 95%CI: 0.02-0.40) and HC (ICC=0.34; 95%CI: 0.13-0.52).Bland-Altman plots revealed large absolute differences between 1st and 2nd MRI in all three groups of participants (Figure 1).View this table:Table 1. Participant characteristics. RA, rheumatoid arthritis; PsA, psoriatic arthritis; HC, healthy control; VAS visual analogue scale. Values are median (range), if not otherwise indicated.Figure 1. Figure: Bland-Altman plots of the repeatability for the three groups. Left, healthy controls; middle, psoriatic arthritis and right, rheumatoid arthritis. Full line: mean difference. Dotted lines: 95% levels of agreement. The Y-axis indicates the relative difference of apparent diffusion coefficient (ADC) between first and second scanning. The X-axis indicates the mean ADC [x10-3mm/s2] of the first and second MRI.Conclusion ADC, determined from DWI-MRI, is not a reliable outcome measure for grading synovitis in the hand and wrist of patients with inflammatory joint diseases.References [1]Li et al, Magn Reson Imaging. 2014;32:350-3[2]Tanaka et al, Rheumatol Int. 2019;39:2111-2118Acknowledgements: NIL.Disclosure of Interests Jakob Møllenbach Møller: None declared, Anna E F Hadsbjerg Grant/research support from: Novartis, Mette Bjoerndal Axelsen: None declared, René Panduro Poggenborg: None declared, Daniel Glinatsi Speakers bureau: Eli Lilly, Consultant of: Eli Lilly, Abbvie, Iris Eshed: None declared, Mikkel Østergaard Speakers bureau: Abbvie, BMS, Boehringer-Ingelheim, Celgene, Eli-Lilly, Galapagos, Gilead, Hospira, Janssen, MEDAC, Merck, Novartis, Novo, Orion, Pfizer, Regeneron, Roche, Sandoz, Sanofi and UCB, Grant/research support from: Abbvie, BMS, Merck, Novartis and UCB.View this table:Table 1.