Δημήτριος Τούσουλης Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Εμμανουήλ Βαβουρανάκης Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Κωνσταντίνα Αγγέλη Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Ελευθέριος Τσιάμης Αναπληρωτής Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Ιωάννης Παρασκευαίδης Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Ιγνάτιος Οικονομίδης Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Κωνσταντίνος Τσιούφης Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Background: A comprehensive non-invasive evaluation of prosthetic mitral valve function can be challenging. We described the development and functional testing of an MRI compatible flow loop system for comparison of multimodality imaging techniques to assess prosthetic mitral valve function. We presented a novel method to assess BMV effective orifice area (EOA) based on phase contrast (PC) cardiovascular magnetic resonance (CMR) and 4D Flow MRI derived data. We compared the performance of these new methods to Doppler and in vitro reference standards.
Methods: Three prosthetic mitral valves with normal function, a 31 mm Medtronic Mosaic valve, a 31 mm Hancock II valve and a 31 mm Open Pivot mechanical prosthesis were placed into a mock ventricle and imaged using Doppler echocardiography, 3D color Doppler echocardiography and cardiovascular magnetic resonance, at 70 beats/min with DFVs of 70, 90 and 110 ml/beat. The diastolic orifice area was assessed by 2D spectral Doppler method (Doppler-EOA), 3D color Doppler method (3D-CDOA), cine MRI planimetry (ΑOA) and phase contrast MRI planimetry (PC-MROA) for each valve at each flow rate, followed by a head to head comparison of echocardiography and MRI-based techniques. In addition, four sizes of normal BMVs (27, 29, 31, 33 mm) and 4 stenotic BMVs (27 and 29 mm, with mild or severe leaflet obstruction) were evaluated using the CMR-compatible flow loop. Valves were evaluated with PC-CMR and Doppler methods under the same flow conditions of 70, 90 and 110 ml/beat, at 70 beats/min (N = 24). PC-EOA was calculated as PC-CMR flow volume divided by the PC-time velocity integral (TVI). Finally, 4 stented porcine BMVs (27, 29, 31, 33 mm) underwent 4D Flow imaging. 4D Flow-EOA was calculated by dividing the diastolic forward volume with the calculated 4D Flow-TVI. PC-ΕΟΑ and 4D Flow MRI-EOA were compared with the Doppler-EOA reference standards.
Results: PC-DFV demonstrated an excellent correlation (r = 0.99, p < 0.0001) when compared with the flow transducer standard across all valve types and flow conditions. A two way ANOVA analysis showed that 3D-CDOA, ΑOA and PC-MROA were not significantly different from each other (p = NS), however the Doppler derived EOA was significantly smaller than the cine MRI, PC-CMR or the 3D color Doppler area measures (p < 0.001). PC-CMR measurements of the diastolic peak velocity and TVI correlated strongly with Doppler values (r = 0.99, p < 0.001 and r = 0.99, p < 0.001, respectively). Across all conditions tested, the Doppler and PC-CMR measurement of EOA (1.4 ± 0.5 vs 1.5 ± 0.7 cm2, respectively) correlated highly (r = 0.99, p < 0.001), with a minimum bias of 0.13 cm2, and narrow limits of agreement (-0.2 to 0.5 cm2). 4D Flow-EOA revealed a strong correlation when compared with Doppler derived EOA (r = 0.985, p < 0.001) and the mean difference was -0.1 ± 0.05 cm2 between the two methods. Calculated 4D Flow-TVI by our method correlated well with Doppler derived TVI (r = 0.99, p < 0.001 and mean difference 3.2 ± 1.97 cm).
Conclusion: With the development of the flow loop we explored and defined the conditions in which a specific imaging modality is limited or best suited to the non-invasive assessment of prosthetic MV function. We described a novel method to assess bioprosthetic mitral valve function based on PC measures of transvalvular flow volume and velocity integration. PC-CMR methods can be used to accurately measure EOA for both normal and stenotic BMV’s and may provide an important new parameter of prosthetic valve function when Doppler methods are unobtainable or unreliable. Our 4D Flow data showed high agreement with Doppler-EOA calculations, demonstrating a potential for 4D Flow MRI in bioprosthetic valve evaluation.