026 Dimensional flow cardiovascular magnetic resonance: two-centre, 1.5t, phantom and in-vivo validation study

Abstract

Background Validation of four-dimensional (4D) flow CMR accelerated acquisition methods is needed to make them more robust for clinical applications.1 Our aim was to compare three widely-used acceleration methods in 4D flow CMR: 4D segmented fast-gradient-echo (4D- turbo-field-echo, 4D-TFE), 4D non-segmented gradient-echo with echo-planar imaging (4D- EPI) and 4D-k-t Broad-use Linear Acquisition Speed-up Technique accelerated TFE (4D-k-t BLAST). Methods CMR was performed in two institutions on identical 1.5T systems. Acceleration methods were compared in static/pulsatile phantoms (Figure 1) and 25 volunteers. In volunteers, the CMR protocol included: cines, 2D phase contrast (PC) at the aortic valve (AV) and mitral valve (MV) and three whole-heart free-breathing (no respiratory motion correction) 4D flow CMR pulse sequences. Field-of-view, slices, phases (30), voxel size and VENC were the same for each subject. In volunteers, net acquisition time for each 4D flow sequence was recorded, as well as a visual grading of image quality on a four-point scale: 0, no artefacts to 3, non-evaluable. Results For the pulsatile phantom experiments, the mean error against the reference flow by time beaker measurements for 4D-TFE was 4.9%±1.3%, for 4D-EPI 7.6%±1.3% and for 4D-k-t BLAST 4.4%±1.9%. In vivo, acquisition time was shortest for 4D-EPI at 7 min59s±2 min30s. 4D- EPI and 4D-k-t BLAST had minimal artefacts, while for 4D-TFE, 40% of AV and MV assessments were non-evaluable because of phase dispersion artefacts. Peak velocity assessment using 4D-EPI demonstrated best correlation to 2D PC (AV: r=0.78, p Conclusion Of the three 4D flow CMR methods tested, 4D-EPI demonstrated the least susceptibility to artefacts, good image quality, modest agreement with the current reference standard for peak intra-cardiac velocities and the highest consistency of intra-cardiac flow quantifications. Competing interests The authors declare that they have no competing interests. Acknowledgement We thank Gavin Bainbridge, Caroline Richmond, Margaret Saysell and Petra Bijsterveld for their invaluable assistance in recruiting and collecting data for this study. Funding Sources This work was supported by the British Heart Foundation [FS/10/62/28409 to S.P.] and Dutch ZonMw [Project Number: 104003001 to J.W]. References . Dyverfeldt P, Bissell M, Barker AJ, Bolger AF, et al. 4D flow cardiovascular magnetic resonance consensus statement. J Cardiovasc Magn Reson2015;17:72.