Undersizing mitral annuloplasty alters left ventricular mechanics in a swine model of ischemic mitral regurgitation

Abstract

Objective: Undersizing mitral annuloplasty (UMA) is a frequently used surgical repair technique to correct ischemic mitral regurgitation in patients with heart failure. In this study, we sought to test the hypothesis that downsizing the mitral annulus can adversely affect the shape and mechanics of the left ventricle inhibiting its functional recovery.

Methods: Eighteen farm swine that underwent an inferolateral myocardial infarction and developed ischemic mitral regurgitation of >2+ severity after 2 months were assigned as follows: 9 swine received an undersized mitral annuloplasty, 6 received papillary muscle approximation (PMA), and 3 animals did not receive any other intervention. Animals lived another 3 months and cardiac magnetic resonance imaging was performed before termination to assess ventricle mechanics and function.

Results: Ejection fraction was comparable between the 2 repair groups before surgery, but was significantly lower in UMA at 38.89% ± 7.91% versus 50.83% ± 9.04% in the PMA group (P = .0397). Animals receiving UMA had lower regional peak fractional shortening and reduced systolic and diastolic radial velocities compared with PMA and in some regions were lower than sham. Animals that underwent UMA had higher circumferential strain than sham, but lower than PMA. UMA animals have lower longitudinal strain compared to sham group and lower LV torsion than PMA.

Conclusions: Undersizing the mitral annulus with an annuloplasty ring can restore valvular competence, but unphysiologically impair ventricle mechanics. Mitral valve repair strategies should focus not only on restoring valve competence, but preserving ventricle mechanics.

Keywords: annuloplasty; functional mitral regurgitation; heart failure; ischemic mitral regurgitation; mitral valve repair; mitral valve replacement; secondary mitral regurgitation.

Figure 1:

(A) Schematic depicting the effect of undersizing mitral annuloplasty on the native mitral valve. The annuloplasty ring draws the native annulus to the size and shape of the ring, thus drawing the native leaflets into the mitral orifice for coaptation; (B) Long axis of a heart after undersizing mitral annuloplasty depicting the deformed inferolateral wall of the left ventricle; (C) Long axis of a heart without annuloplasty but a sub-annular papillary muscle approximating repair that does not deform the inferolateral wall.

Figure 2:

Temporal changes in the average fractional wall thickening through the cardiac cycle beginning with end diastole, depicted by sub-sectors at the basal, equatorial and apical levels among UMA, PMA and sham groups. In general, PMA and Sham group show higher systolic fractional thickening than UMA group. *: p < 0.05 versus sham, +: p < 0.05 versus PMA.

Figure 3:

Temporal changes of radial velocity through the cardiac cycle beginning with end diastole, and different graphs compare these traces at different regions of basal, equatorial and apical levels among UMA, PMA and Sham groups. Generally, it is evident that the radial velocity in the UMA group remains low consistently in all the regions, whereas the PMA and sham group shows early rapid relaxation and then gradual plateauing. *: p < 0.05 versus sham, +: p < 0.05 versus PMA.

Figure 4:

Temporal traces of circumferential strain and radial strain through the cardiac cycle beginning with end diastole, and different graphs compare these traces at different regions of basal, equatorial and apical levels of UMA, PMA and sham groups. *: p < 0.05 versus sham, +: p < 0.05 versus PMA.

Figure 5:

Temporal traces of longitudinal strain through the cardiac cycle beginning with end diastole, and different graphs compare these traces at different regions of basal, equatorial and apical levels of UMA, PMA and sham groups. *: p < 0.05 versus sham, +: p < 0.05 versus PMA.

Figure 6:

(A, B) Temporal traces of basal and apical rotation of UMA, PMA and sham groups respectively; (C, D) Temporal traces of net LV twist (the net differences between apical and basal rotations) and LV torsion (LV net twist normalized by LV long-axis length); (E, F) Peak basal and apical rotations of UMA, PMA and sham groups.

Figure 7:

Graphical abstract: (Top row) Schematic representation of experimental design of this study; (Bottom row) Key findings from this study that demonstrate the deleterious effect of undersizing mitral annuloplasty on left ventricular mechanics, compared to papillary muscle approximation alone.