Background

Mitral valve (MV) repair using annuloplasty rings is the preferred method of treatment for MV regurgitation, but the impact of annuloplasty ring placement on left ventricular intraventricular flow has not been studied.

Methods

Annuloplasty rings of varying sizes were placed in 5 healthy sheep (intercommissural ring sizes were 24, 26, 28, 30, and 32 mm), and three-dimensional phase contrast magnetic resonance imaging (4D flow MRI) was performed before and 1 week after ring placement.

Results

Normal diastolic flow consisted of diastolic intraventricular vortices that naturally unwound during systole. Postsurgical intraventricular flow was highly disturbed in all sheep, and the disturbance was greatest for undersized rings. Ring size was highly correlated with the diastolic inflow angle (Pearson's r = -0.62, p < 0.1, 95% confidence interval: -0.92 to 0.14). There was a mean angle increase of mean diastolic inflow angle increase of 12.3 degrees (< 30 mm, p < 0.01, 95% confidence interval: 4.8 to 19.6) for rings less than 30 mm. There was an inverse relationship between peak velocity and annuloplasty ring area (Pearson's r = -0.80, p < 0.05, 95% confidence interval: -0.96 to -0.2). Transmitral pressure gradients increased significantly from baseline 0.73 ± 0.18 mm Hg to after annuloplasty 2.31 ± 1.04 mm Hg (p < 0.05).

Conclusions

Mitral valve annuloplasty ring placement disturbs normal left ventricular intraventricular flow patterns, and the degree of disturbance is closely associated with annuloplasty ring size.

Background

Rapid determination of the left ventricular (LV) pressure-volume (PV) relationship as loading conditions are varied is the gold standard for assessment of LV function. Cine magnetic resonance imaging (MRI) does not have sufficient spatiotemporal resolution to assess beat-to-beat changes of the LV PV relationship required to measure the LV end-systolic elastance (EES) or preload-recruitable stroke work (PRSW). Our aim was to investigate real-time MRI and semiautomated LV measurement of LV volume to measure PV relations in large animals under normal and inotropically stressed physiologic conditions.

Methods

We determined that PV relationships could be accurately measured using an image exposure time Tex less than 100 ms and frame rate Tfr less than 50 ms at elevated heart rates (∼140 beats per minute) using a golden angle radial MRI k-space trajectory and active contour segmentation.

Results

With an optimized exposure time (Tex=95 ms and frame rate Tfr=2.8 ms), we found that there was no significant difference between cine and real-time MRI at rest in end-diastolic volume, end-systolic volume, ejection fraction, stroke volume, or cardiac output (n=5, p<0.05) at either normal or elevated heart rates. We found EES increased from 1.9±0.7 to 3.1±0.3 mm Hg/mL and PRSW increased from 6.2±1.2 to 9.1±0.9 mm Hg during continuous intravenous dobutamine infusion (n=5, p<0.05).

Conclusions

Real-time MRI can assess LV volumes, EES, and PRSW at baseline and elevated inotropic states.