Diagnostic accuracy of 4D flow MRI comparing 2mm3 and 3mm3 spatial resolution

Abstract Funding Acknowledgements Type of funding sources: Public Institution(s). Main source(s): National Institute for Health Research, UK University Leeds Background Cardiac magnetic resonance (CMR) examinations requiring repeated breath-holds are challenging in younger patients. While 4-dimensional phase-contrast (4D flow) CMR does not require breath-holds, acquisition has been lengthy. Therefore to date spatial resolution influenced mainly by scan length. With accelerated sequences becoming available, higher is clinically feasible. Purpose We therefore evaluated the minimum 4D flow necessary accurate clinical assessment. Methods Ten healthy volunteers (mean age 24.8 years) underwent cardiac on a 3T scanner using Flow prototype sequence at 2x2x2mm3 (4DFlow2) and 3x3x3mm3 (4DFlow3) resolution. Net forward (FF) peak velocity (PV) valve tracking were calculated with commercially available software kinetic energy (KE) left ventricle (LV) was analysed research tool. Bland-Altman analysis used statistical assessment reported as bias ± limits agreement. Results Aortic metrics similar 4DFlow2 (FF 94ml; PV 133cm/s) 4DFlow3 95ml; 130cm/s), both showed good agreement 2D PC MRI 93ml, Bland-Altman:1.6 9.7 2.2 13.5, respectively). Similar results obtained pulmonary 138cm/s; Bland-Altman:4.7 15.1 8.1 18.2, Branch artery (PA) FF main PA (Bland-Altman:1.1 15.9 1.1 10.6, respectively), but 32.5). Global LV KE measured average 12% lower compared 4DFlow2, whereas maximum systolic methods. Conclusions 3mm3 appears be sufficient evaluation aortic valves. Smaller vessels such branch arteries no gold standard assessment, our suggest that some parameters energetic sensitive. Differences SNR might also contribute differing results. Figure. plots