Haemodynamic outcome at four-dimensional flow magnetic resonance imaging following valve-sparing aortic root replacement with tricuspid and bicuspid valve morphology

OBJECTIVE: To provide a more complete characterization of aortic blood flow in patients following valve-sparing aortic root replacement (VSARR) compared with presurgical cohorts matched by tricuspid and bicuspid valve morphology, age and presurgical aorta size. METHODS: Four-dimensional (4D) flow magnetic resonance imaging (MRI) was performed to analyse three-dimensional (3D) blood flow in the thoracic aorta of n = 13 patients after VSARR with reimplantation of native tricuspid aortic valve (TAV, n = 6) and bicuspid aortic valve (BAV, n = 7). Results were compared with presurgical age and aortic size-matched control cohorts with TAV (n = 10) and BAV (n = 10). Pre- and post-surgical aortic flow was evaluated using time-resolved 3D pathlines using a blinded grading system (0-2, 0 = small, 1 = moderate and 2 = prominent) analysing ascending aortic (AAo) helical flow. Systolic flow profile uniformity in the aortic root, proximal and mid-AAo was evaluated using a four-quadrant model. Further analysis in nine analysis planes distributed along the thoracic aorta quantified peak systolic velocity, retrograde fraction and peak systolic flow acceleration. RESULTS: Pronounced AAo helical flow in presurgical control subjects (both BAV and TAV: helix grading = 1.8 ± 0.4) was significantly reduced (0.2 ± 0.4, P < 0.001) in cohorts after VSARR independent of aortic valve morphology. Presurgical AAo flow was highly eccentric for BAV patients but more uniform for TAV. VSARR resulted in less eccentric flow profiles. Systolic peak velocities were significantly (P < 0.05) increased in post-root repair BAV patients throughout the aorta (six of nine analysis planes) and to a lesser extent in TAV patients (three of nine analysis planes). BAV reimplantation resulted in significantly increased peak velocities in the proximal AAo compared with root repair with TAV (2.3 ± 0.6 vs 1.6 ± 0.4 m/s, P = 0.017). Post-surgical patients showed a non-significant trend towards higher systolic flow acceleration as a surrogate measure of reduced aortic compliance. CONCLUSIONS: VSARR restored a cohesive flow pattern independent of native valve morphology but resulted in increased peak velocities throughout the aorta. 4D flow MRI methods can assess the clinical implications of altered aortic flow dynamics in patients undergoing VSARR.