Effects of Different Annuloplasty Rings on Anterior Mitral Leaflet Dimensions

Abstract

OBJECTIVE

To assess the effects of annuloplasty rings (ARs) on anterior mitral leaflet (AML) dimensions.

METHODS

Sixteen radiopaque markers were sutured evenly spaced over the surface of the AML in 57 sheep. Size 28 mm Cosgrove (n=11), rigid saddle-shaped AR (RSAR) (n=12), Physio (n=12), IMR-ETlogix (n=10) and GeoForm (n=12) rings were implanted in a releasable fashion. Under acute open-chest conditions, 4-D marker coordinates were measured using biplane videofluoroscopy with the AR inserted and after AR release. Septal-lateral and commissure-commissure dimensions were calculated from opposing marker pairs on the septal-lateral and commissure-commissure aspect of the AML at end-diastole (ED) and end-systole (ES). To assess changes in AML shape, a “planarity index” was assessed by calculating the root mean square values as distances of the 16 AML markers to a best fit AML plane at ES.

RESULTS

At ED, AML septal-lateral and commissure-commissure dimensions did not change with Cosgrove compared to Control, while RSAR, Physio, IMR-ETlogix and GeoForm reduced AML commissure-commissure, but not septal-lateral AML dimensions. At ES, the septal-lateral AML dimension was smaller with IMR-ETlogix and GeoForm, but did not change with Cosgrove, RSAR and Physio. AML shape was unchanged in all five groups.

CONCLUSIONS

With no changes in AML planarity, the 4 complete, rigid rings (RSAR, Physio, IMR-ETlogix, and GeoForm) reduced the AML commissure-commissure dimension at ED. IMR-ETlogix and GeoForm decreased the septal-lateral AML dimension at ES, probably due to inherent disproportionate downsizing. These changes in AML geometry could perturb the stress patterns, which in theory may affect repair durability.

INTRODUCTION

Surgical mitral valve (MV) repair is the gold standard for the treatment of mitral regurgitation (MR)¹ and most commonly includes the implantation of an annuloplasty ring. Annuloplasty rings are available in various shapes with different material properties. The optimal size and material properties of an annuloplasty ring for any individual patient is, however, still subject of a subjective debate: Some surgeons implant complete rigid rings² while others prefer partial flexible bands only³. The choice of the annuloplasty ring may directly affect patient outcome^4,5. Annuloplasty rings are sewn to the mitral annulus (MA), a saddle-shaped and dynamic junctional zone of fibrous and muscular tissue between the left atrium and left ventricle.

Previous studies have shown that annuloplasty ring implantation may abolish the dynamics of the MA and change its dimensions and shape⁶. Such alterations in MA geometry and dynamics have been demonstrated to affect anterior mitral leaflet (AML) curvature⁷ and increase AML strains which, in turn, has been associated with impaired MV repair durability⁸. As a consequence, new annuloplasty rings have been introduced that aim to mimic the physiological MA shape (e.g., St. Jude Medical rigid saddle-shaped annuloplasty ring (RSAR), Medtronic Profile 3D)⁹; however, the most commonly implanted rings are flat (Carpentier-Edwards Classic and Physio). Furthermore, even less physiological shaped annuloplasty rings with a disproportional reduction in the septal-lateral dimension (e.g., Edwards GeoForm and IMR-ETlogix) have been introduced, designed specifically for the treatment of patients with functional/ischemic MR with the rationale to restore a greater amount of leaflet coaptation^10,11.

It is reasonable to speculate that implantation of such differently designed annuloplasty rings may affect not only the geometry and dynamics of the MA heterogeneously, but also the dimensions and shape of the AML.

The recent introduction of three-dimensional (3-D) real-time echocardiography allows more accurate tracking of anatomic borders of the mitral valve complex¹² and, consequently can estimate dimensions of AML and MA during the cardiac cycle. Such changes in AML dimensions and shape may directly translate to leaflet strains and stresses and, thus, could in theory impair long-term repair durability. The effects of annuloplasty rings on mitral leaflet dimensions, however, are currently unknown.

Our goal was to assess the effects of five annuloplasty rings with different shapes (physiological/non-physiological) and material properties (flexible/semi-rigid/rigid) on AML and MA dimensions and shape (planarity) in vivo in an experimental ovine model using four-dimensional radiopaque marker tracking. The annuloplasty rings were implanted in a releasable fashion, i.e., each animal served as its own control.

MATERIAL AND METHODS

Annuloplasty ring preparation

To allow annuloplasty ring release in the beating heart, five different rings (the flexible, partial Cosgrove-Edwards band (Edwards Lifesciences, Irvine, CA, USA), St. Jude Medical rigid saddle-shaped annuloplasty ring (RSAR) (St. Jude Medical Inc, St. Paul, MN, USA), semi-rigid Carpentier-Edwards Physio, rigid Edwards IMR-ETlogix and rigid Edwards GeoForm (Edwards Lifesciences, Irvine, CA, USA), (Figure 1B–F) were prepared before the operation in the following manner: The middle part of eight double-armed polyester braided sutures were stitched evenly spaced around the ring through the fabric of the annuloplasty ring from the bottom to the top side using a Spring Eye needle. The resulting loops were “locked” with two polypropylene sutures. Two drawstring sutures were attached to the ring¹³.

Figure 1.

Schematic representation of the radiopaque marker array used in this study (A). Intraoperative photograph of the Cosgrove-Edwards band (B), St. Jude Medical rigid saddle-shaped annuloplasty ring (RSAR) (C), Carpentier-Edwards Physio (D), Edwards IMR-ETlogix (E) and Edwards GeoForm (F) from a surgical view. AC=anterior commissure. PC=posterior commissure. *=mitral annulus saddle horn.

Surgical preparation

Fifty seven adult, Dorsett-hybrid, male sheep (49 ± 4 kg) were premedicated with ketamine (25mg/kg intramuscularly), anesthetized with sodium thiopental (6.8mg/kg intravenously), intubated and mechanically ventilated with inhalational isoflurane (1.0–2.5%). All animals received humane care in compliance with the Principles of Laboratory Animal Care formulated by the National Society of Medical Research and the Guide for Care and Use of Laboratory Animals prepared by the National Academy of Sciences and published by the National Institute of Health (DHEW NIHG publication 85-23, revised 1985). This study was approved by the Stanford Medical Center Laboratory Research Animal Review committee and conducted according to Stanford University policy.

Through a left thoracotomy, 13 radiopaque markers were implanted to silhouette the left ventricle (LV) at the cross-section points of two equally spaced longitudinal and three equatorial meridians as described earlier ¹⁴. Using cardiopulmonary bypass and cardioplegic arrest, 16 radiopaque markers were sewn to the MA and 16 markers were sewn to the AML (single tantalum loops, 0.6 mm ID, 1.1 mm OD, 3.2 mg each). Then, the annuloplasty ring was implanted in a releasable fashion as follows: the eight polyester sutures of the ring were stitched equidistantly in a perpendicular direction from the ventricular to the atrial side through the mitral annulus. The annuloplasty ring was secured to the mitral annulus by tying these sutures. The “locking” polypropylene sutures and the drawstrings were exteriorized, and the left atrium was closed. All annuloplasty rings were “true-sized” by assessing the height and the entire AML area. As all animals had similarly sized leaflets, all received size 28 rings. The sheep while intubated and anesthetized, were immediately transferred with the chest open to the experimental catheterization laboratory.

Data acquisition and analysis

Videofluoroscopic images (60 frames/sec) of all radiopaque markers were acquired using biplane videofluoroscopy (Philips Medical Systems, North America, Pleasanton, CA, USA). First, images were acquired with the ring inserted (Cosgrove-Control, RSAR-Control, Physio-Control, ETlogix-Control, GeoForm-Control). To acquire a data set for a parallel study, acute LV ischemia was then induced by tightening an encircling vessel loop around the circumflex branch of the left coronary artery for 90 sec. Thereafter, the locking sutures were pulled out and the ring was lifted away from the mitral annulus towards the left atrial roof using the drawstrings. After hemodynamic values returned to baseline, a third data acquisition was performed and images were acquired with the ring released (Cosgrove, RSAR, Physio, ETlogix, GeoForm)

Marker coordinates from three consecutive sinus rhythm heart beats from each of the biplane views were then digitized and merged to yield the time-resolved 3-D coordinates of each marker centroid in each frame using semi-automated image processing and digitization software^15,16. ECG and analog left ventricular pressure (LVP) were recorded in real-time on the video images during data acquisition.

For each cardiac cycle, end-systole (ES) was defined as the time frame preceding maximum peak rate fall of the LVP (−dP/dt), and end-diastole (ED) as the time frame containing the peak of the R-wave on the ECG. Systolic dimension changes are described as changes from ED to ES.

Hemodynamics

Instantaneous LV volume was computed from the epicardial LV markers using a space-filling multiple tetrahedral volume method¹⁵. Hemodynamic data were calculated from marker derived instantaneous LV volumes and analog LV pressures.

Anterior mitral leaflet and mitral annular dimensions

The septal-lateral MA dimension was calculated as the distance in 3-D space between markers (#17 and #19, (Figure 1A)), and the commissure-commissure MA dimension as the distance between markers (#18 and #20, (Figure 1A)). The septal-lateral AML dimension was determined as the distance between the AML marker nearest the saddle horn (#1, (Figure 1A)) and the central free edge of the AML (#3, (Figure 1A)). The commissure-commissure AML dimension was calculated as the distance between the leaflet marker opposite to the anterior and posterior commissure (#2 and #4, respectively, Figure 1A).

To determine the relative changes of native septal-lateral and commissure-commissure MA dimensions with ring insertion, the percentage difference in MA dimensions between Control state and ring state was calculated at ED and ES as: 100*((MA dimension in the ring state) – (MA dimension in the Control state))/MA dimension in the ring state.

Anterior mitral leaflet and mitral annulus shape

In order to estimate changes in AML shape, a “planarity index” was assessed as the root mean square (RMS) value of the distances from the 16 leaflet markers to a best fit plane through the same 16 leaflet markers at ES as:

$$x_{\mathit{rms}}=\sqrt{\frac{x_{\#1}^2+x_{\#2}^2\cdots x_{\#16}^2}{16}}$$

where x describes the distance to the plane. Similarly, to determine the planarity index of the MA, we calculated the RMS values of the distances from the 16 MA markers to a best fit plane through the same 16 annular markers at ES.

Statistical analysis

Data are reported as mean ± 1 SD unless otherwise stated. Control data (Cosgrove-Control, RSAR-Control, Physio-Control, ETlogix-Control, GeoForm-Control) and data with the respective ring inserted (Cosgrove, RSAR, Physio, ETlogix, GeoForm) were compared using two-way repeated measures ANOVA with a Holm-Sidak posthoc test. The RMS values of the AML and MA, with and without annuloplasty ring inserted, were compared using two-way repeated measures ANOVA with a Holm-Sidak posthoc test (Sigmastat 3.5, Systat Software, Inc, San Jose, CA, USA). A p-value of less than 0.05 was considered to be statistically significant.

RESULTS

Hemodynamics

Table 1 summarizes the hemodynamics. No significant differences were found between groups with respect to heart rate, LV dP/dt (except for Cosgrove, where dP/dt was slightly higher compared to Control (1547 ± 398 vs. 1366 ± 331 mmHg/sec, p<.05)) and LVEDV (except for GeoForm, where LVEDV was slightly smaller compared to Control (109 ± 13 vs. 112 ± 13 ml, p<.05)).

Table 1.

Hemodynamic data

	Heart rate (min−1)	dP/dtmax (mmHg/sec)	LVEDV (ml)	LVPmax (mmHg)
Cosgrove-Control	96±12	1366±331	116±15	97±8
Cosgrove	97±12	1547±398*	117±15	98±6
RSAR-Control	89±17	1283±409	119±21	100±9
RSAR	89±15	1226±297	119±19	98±5
Physio-Control	92±12	1309±333	123±22	95±8
Physio	92±11	1348±338	121±24	95±6
ETlogix-Control	82±6	1165±387	123±20	94±5
ETlogix	80±9	1201±394	121±19	96±4
GeoForm-Control	92±10	1313±315	112±13	96±8
GeoForm	93±10	1388±410	109±13	97±7

Values are mean ± SD.

dP/dt _max: maximum positive rate of change of left ventricular pressure;

LVEDV: left ventricular end-diastolic volume

LVP_max: maximum left ventricular pressure.

^*p<.05

Anterior mitral leaflet and mitral annular septal-lateral dimensions

Figure 2A–E shows the septal-lateral dimensions of the AML and the MA of the five ring groups with the ring inserted and after ring release at ED and ES. While in the Control state the septal-lateral dimension of the AML did not change from ED to ES in all five groups, the septal-lateral dimensions of the MA decreased significantly from ED to ES in all ring groups. Although annuloplasty ring implantation decreased the MA septal-lateral dimensions at both ED and ES compared to Control, only IMR-ETlogix and GeoForm reduced the septal-lateral AML dimension at ES compared to the Control state. The RSAR ring increased the AML septal-lateral dimension at ED. The percentage reduction in the septal-lateral MA dimension compared to Control were similar for Cosgrove, RSAR and Physio at ED (−14 ± 7%, −17 ± 8%, and −21 ± 9%) and ES (−5 ± 7%, −10 ± 7%, and −12 ± 7%),while the GeoForm decreased the MA septal-lateral dimensions to a greater extent (ED: −45 ± 8%, ES: −33 ± 9%), followed by the IMR-ETlogix (ED: −33 ± 6%, ES: −24 ± 8%). Whereas, annuloplasty ring implantation abolished the decrease in the MA septal-lateral dimension from ED to ES for all groups, ring insertion did not affect the systolic changes in the septal-lateral dimension of the AML for the Control state nor with Cosgrove, RSAR, IMR-ETlogix and PHYSIO rings implanted. With the GeoForm implanted the change of AML septal-lateral dimension from ED to ES was significantly greater compared to Control.

Figure 2.

Septal-lateral dimension of the anterior mitral leaflet and the mitral annulus of Cosgrove-Control and Cosgrove (A), RSAR-Control and RSAR (B), Physio-Control and Physio (C), ETlogix-Control and ETlogix (D), GeoForm-Control and GeoForm (E) with ring inserted (“brick” pattern and black bars) and after ring release (small dots and white bars) at end-diastole (small dots and “brick” pattern bars) and end-systole (white and black bars). AML, anterior mitral leaflet; MA, mitral annulus; ED, end-diastole; ES, end-systole; CTRL, Control. Data are mean ± 1 SD. * p<.05

Anterior mitral leaflet and mitral annular commissure-commissure dimensions

Figure 3A–E shows the commissure-commissure dimensions of the AML and the MA in the five groups with the ring inserted and after ring release at ED and ES. In the Control state the commissure-commissure dimensions of both AML and MA decreased from ED to ES in all groups. Implantation of RSAR and GeoForm resulted in a significant decrease of the commissure-commissure dimension of the AML at ED, but not at ES compared to Control. The Physio and IMR-ETlogix rings significantly decreased the commissure-commissure dimensions both at ED and ES. The Cosgrove ring did not change the commissure-commissure dimension of the AML, neither at ED nor at ES. The commissure-commissure dimension of the MA decreased significantly with ring implantation in all five groups compared to Control. The percentage reductions in the commissure-commissure dimension of the MA were similar for Cosgrove, RSAR and GeoForm at ED (−7 ± 6 %, −12 ± 10 %, −8± 5 %) and ES (−6 ± 6 %, −8 ± 10 %, −7 ± 5%) and slightly greater for Physio (ED: −20 ± 7%, ES: −18 ± 5%) and IMR-ETlogix (ED: −22 ± 5%, ES: −20 ± 5%). Although annuloplasty rings abolished the decrease in the MA commissure-commissure dimension from ED to ES in all groups, the commissure-commissure dimensions of the AML still became smaller from ED to ES with the ring inserted in Cosgrove, Physio, IMR-ETlogix and GeoForm, although not significantly so with RSAR.

Figure 3.

Commissure-commissure dimension of the anterior mitral leaflet and the mitral annulus of Cosgrove-Control and Cosgrove (A), RSAR-Control and RSAR (B), Physio-Control and Physio (C), ETlogix-Control and ETlogix (D), GeoForm-Control and GeoForm (F) with ring inserted (“brick” pattern and black bars) and after ring release (small dots and white bars) at end-diastole (small dots and “brick” pattern bars) and end-systole (white and black bars). AML, anterior mitral leaflet; MA, mitral annulus; ED, end-diastole; ES, end-systole; CTRL, Control. Data are mean ± 1 SD. * p<.05

Anterior mitral leaflet and mitral annular shape

Figure 4A shows the RMS values derived from the AML markers at ES for all five ring groups without and with the ring implanted. RMS values of the AML markers were identical between the Control state and with the ring implanted in all five groups; Figure 4B shows the RMS values derived from the MA markers at ES. While implantation of the Cosgrove band and RSAR did not change RMS values derived from MA markers compared to Control, these values became significantly smaller with the Physio, IMR-ETlogix and GeoForm.

Figure 4.

Root mean square (RMS) values derived from the anterior mitral leaflet (AML) (A) and mitral annular (MA) (B) markers for all five ring groups with the ring inserted and after ring release at end-systole. CTRL, Control. Data are mean ± 1 SD. * p<.05.

DISCUSSION

In this study we measured, for the first time, the effects of different annuloplasty rings on AML and MA dimensions and planarity in an in vivo experimental ovine model. In summary we found that: 1.) The MA predominantly shortens in the septal-lateral dimension during systole in the Control state, while the AML septal-lateral dimension does not change from ED to ES. 2.) All rings decreased both MA septal-lateral and commissure-commissure dimensions at ED and ES compared to Control, whereas Physio, RSAR, IMR-ETlogix and GeoForm only reduced AML commissure-commissure dimensions at ED. Cosgrove did not affect AML dimensions. 3.) Ring insertion abolished the changes in MA dimensions (both, septal-lateral and commissure-commissure) during systole although ring implantation did not affect the decrease in the commissure-commissure dimension of the AML from ED to ES. 4.) AML septal-lateral dimension does not change with physiologically shaped rings at ES, but is smaller with IMR-ETlogix and GeoForm. 5) Cosgrove and RSAR did not change annular planarity, while Physio, IMR-ETlogix and GeoForm resulted in a more planar annulus. In contrast to the ring effects on MA planarity, none of the rings affected the planarity of the AML.

The AML decreases in the commissure-commissure dimension during systole independent of MA contraction

In accordance with previous marker studies we found that the MA predominantly decreases septal-lateral dimension from ED to ES¹⁷. In contrast, the AML decreases in the commissure-commissure dimension, but not in the septal-lateral dimension. We found that annuloplasty ring implantation abolishes the systolic decrease in the septal-lateral and commissure-commissure dimension of the MA; however, as in the Control state, the AML dimensions still decreased from ED to ES in the commissure-commissure dimension in all five ring groups. These findings therefore demonstrate that the systolic changes in AML dimensions are independent of MA contraction.

All annuloplasty rings reduced MA septal-lateral and commissure-commissure dimensions and abolish their dynamics

Interestingly, despite “true sizing” and ring implantation in healthy, non-dilated mitral annuli, implantation of any annuloplasty ring resulted in a decrease in septal-lateral and commissure-commissure mitral annular dimensions. As expected, the GeoForm reduced the MA septal-lateral dimension to the greatest extent (ED: −45± 8%, ES: −33± 9%) followed by the IMR-ETlogix (ED: −33± 6%, ES: −24± 8%); however, surprisingly, the fractional reduction in septal-lateral MA dimension compared to Control was similar for Cosgrove, RSAR and Physio at ED (−14± 7 %, −17± 8 %, and −21± 9%, respectively) and ES (−5± 7 %, −10± 7%, and −12± 6%, respectively). Whereas Cosgrove and RSAR reduced the native mitral annular septal-lateral diameter, these rings provided a greater MA septal-lateral diameter compared to the Physio at ED (+7% and +4%, respectively) and at ES (+7% and +2%, respectively). While the manufacturer claims a reduction in the septal-lateral diameter of 41% with the GeoForm compared to the Physio ring, the in vivo difference in the amount of MA septal-lateral reduction of the GeoForm compared to the Physio was less (approximately 24% at ED and 21% at ES) in this experiment. In a recent study performed in this laboratory we found that the fractional septal-lateral reduction data provided by the manufacturer relates to the inner diameter of the ring without the fabric sewing cloth for Physio and GeoForm. Therefore, it is reasonable to assume that the outer diameter with cloth determines the postoperative dimensions of the mitral annulus, not the inner diameter without cloth. This is consistent with our in vitro measurement of annuloplasty rings where we found that the septal-lateral reduction of the GeoForm compared to the Physio was 24 ± 2 % if the outer septal-lateral dimension with cloth was measured.

Furthermore, because MA shortening is not expected to be affected by the flexible Cosgrove band compared to the Control state, we speculate that the loss of MA contraction may be due to the reduction of the MA diameters and not due to the rigidity of the annuloplasty ring. We hypothesize that implantation of larger bands might have preserved MA dynamics as demonstrated by Sharony et al.¹⁸.

Only complete, rigid annuloplasty ring implantation reduced the commissure-commissure dimension of the AML

Although all annuloplasty rings reduced MA dimensions as described above, only the rigid complete RSAR, Physio, IMR-ETlogix and GeoForm rings (but not the partial flexible Cosgrove band) decreased the AML commissure-commissure. We are, however, currently not able to say whether the changes observed are due to the rigidity of the rings or their being complete rings. These data indicate that if the effects of rigid complete annuloplasty rings on AML dimensions are to be assessed using echocardiography, primary attention should be paid to the AML commissure-commissure dimension since this dimension is predominately affected by the rigid ring implantation. Further, the fact that changes in AML commissure-commissure dimensions predominately occur at ED, points out the importance of assessing changes in the commissure-commissure dimension of the AML during diastole. Because implantation of the flexible Cosgrove band reduced MA dimensions but did not affect AML dimensions and dynamics suggests that Cosgrove least perturbed the physiological configuration of the AML among the five rings tested. Compared to Control, IMR-ETlogix and GeoForm resulted in reduction of the AML septal-lateral dimensions at ES, which may reflect the disproportional reduction of the septal-lateral dimension associated with these rings.

The MA became more planar with Physio, IMR-ETlogix and GeoForm, but no annuloplasty ring affected AML planarity

Compared to Control, the Physio, IMR-ETlogix and GeoForm resulted in a more planar annulus at ES, while the Cosgrove and the RSAR did not alter MA planarity at ES. Despite the different designs and material properties of these annuloplasty rings and their effects on MA shape, none of these rings changed the planarity of the AML at ES. We therefore concluded that the MA shape does not directly translate to the ES shape of the AML. It may, however, be speculated that ring implantation changes the planarity of the posterior mitral leaflet as mentioned by Lancellotti et al.¹⁹.

Clinical inferences

Our findings may have several clinical inferences. If echocardiography is used to assess the effects of annuloplasty rings on commissure-commissure AML dimensions, attention should be paid on measuring these dimensions at ED, since rigid complete annuloplasty rings reduce AML commissure-commissure dimensions compared to Control primarily at ED. If ring effects on septal-lateral leaflet dimensions are assessed, this dimension should be measured at ED and ES, because implantation of a ring with disproportionate septal-lateral downsizing may reduce this dimension at ES and, furthermore, amplify the changes in AML septal-lateral dimension from ED to ES (e.g., GeoForm). Since all parameters measured can be assessed clinically, a comparison between the effects of downsized, physiologically shaped rings and disease-specific rings with a disproportionate septal-lateral reduction on anterior leaflet geometry could give further insight into the optimal ring design for patients with functional/ischemic MR. Hypothetically, combining normally sized rings with a subvalvular approach may help to minimize the effects of ring downsizing on leaflet geometry. Furthermore, the amount of MA septal-lateral downsizing provided by the manufacturer for the Geoform ring refers to the Physio. Whereas in our study Cosgrove and RSAR reduced the MA septal-lateral dimension compared to the Control state at ED and ES (see Results), these rings provide a greater MA septal-lateral diameter compared to the Physio at ED (+7% vs. Physio and +4% vs. Physio, respectively) and at ES (+7% vs. Physio and +2% vs. Physio, respectively). The IMR-ETlogix reduces the MA septal-lateral dimension compared to the Physio by −12% at ED and −12% at ES whereas the GeoForm reduces the MA septal-lateral dimensions compared to the Physio by −24% at ED and by −21% at ES. This degree of downsizing is, however, smaller than the one claimed by the manufacturer (−41% vs. Physio). Consequently, these data may help the surgeon to anticipate the extent of septal-lateral reduction of the MA after an annuloplasty ring implantation more accurately. And finally, these data indicate that a partial, flexible band may be preferred if the surgeon aims to least perturb the dimensions, dynamics and shape of the anterior mitral leaflet.

Limitations

Several limitations should be addressed: First, annuloplasty rings are implanted in patients with mitral regurgitation to prevent postoperative MA dilatation and maintain valve competency. In our experimental setup, however, the rings were implanted in healthy ovine hearts with non-dilated mitral annuli and normal left ventricles; caution, therefore, is necessary translating these findings to patients. Second, the Control data were acquired after the ring was released in the catheterization laboratory. One may speculate that annuloplasty ring implantation could have affected the native function of the MA. Our findings of MA septal-lateral and commissure-commissure dimension values and changes from ED to ES for the Control state, however, correlate well with previous data from control sheep operated on in our laboratory⁶. Third, the brief episode of myocardial ischemia (90s) induced for a parallel study between the Control state and the ring release could have affected the results. And lastly, our study focuses on acute effects of annuloplasty ring implantation on AML dimensions. The question whether ring implantation may have long-term effects on mitral leaflet dimensions remains to be investigated.

In conclusion, the four rigid complete annuloplasty rings (RSAR, Physio, IMR-ETlogix and GeoForm) reduce the AML commissure-commissure, but not the septal-lateral leaflet dimension at end-diastole. Only IMR-ETlogix and GeoForm decrease the AML septal-lateral dimension at ES which may be a result of the disproportionate annular downsizing induced by these rings. While anterior mitral leaflet planarity did not change with any of the rings tested, mitral annular planarity was unperturbed by the Cosgrove and the RSAR, but became more planar with Physio, IMR-ETlogix and GeoForm. Since alterations in mitral annular and anterior mitral leaflet dimensions could alter the stress patterns within the leaflets, this could in theory affect repair durability. Echocardiographic assessment of such dimensions may play an important role in the pre-operative technical planning and the postoperative evaluation of mitral valve repair.

Abbreviations and acronyms

AC

anterior commissure

AML

anterior mitral leaflet

ED

end-diastole

ES

end-systole

MA

mitral annulus

MV

mitral valve

PC

posterior commissure

RMS

root mean square