A clinical study of annular geometry and dynamics in patients with ischemic mitral regurgitation: new insights into asymmetrical ring annuloplasty☆

Ischemic mitral regurgitation (IMR) is a severe disease that has a complex pathogenesis and is associated with poor prognosis in patients with coronary artery disease [1]. The clinical definition of ischemic mitral regurgitation implicates that valve insufficiency occurs after myocardial infarction and that the leaflets and subvalvular apparatus are structurally normal [2]. Recent studies showed that ischemic mitral regurgitation is mostly due to a regional annulus enlargement, occurring predominantly at the posterior leaflet [3,4]. It has also been postulated that ischemic mitral regurgitation mainly occurs at the region P3 and new surgical approaches based on asymmetrical annuloplasty rings have been suggested [4]. However, it is commonly acknowledged that ischemic mitral regurgitation has a much more complex pathogenesis, thereby involving interactions of several anatomic structures: leaflets, annulus, chordae tendineae, papillary muscles, and left ventricle. As the geometry and the function of these structures are difficult to obtain in the single patients, the choice of the surgical approach to ischemic mitral regurgitation is still a controversial issue. During the last decade, our group has developed several methods based on three-dimensional echocardiography, which are ideally suitable for assessing the complex geometry of mitral valve in all its components [5,6] and for quantifying the degree of mitral regurgitation [7]. As the surgical approach to ischemic mitral regurgitation is strictly dependent on accurate three-dimensional measurements of cardiac structures and their function, we performed this study in order to analyze changes in annular geometry in patients with mitral regurgitation, and hence to investigate the relationship between ischemic mitral regurgitation and the pattern of asymmetrical annular dilatation. One hundred and ten patients were included in this study. They were divided into three groups: group 1, 30 patients with coronary artery disease without mitral regurgitation; group 2, 38 patients with chronic ischemic mitral regurgitation; group 3, 42 patients with mitral regurgitation due to primary valvular lesions. All patients in group 2 had a history of previous myocardial infarction. The main regional distribution of myocardial infarction was: anterior in 16 (42.1%) patients, posterior in 14 (36.8%), lateral in 3 (7.9%), and indeterminate location in 5 (13.2%) (Table 1). Characteristics of the study groups In group 2, 3 patients (7.8%) were in NYHA class I, 8 (21%) in class II, 16 (42.1%) in class III, 11 (28.9%) in class IV. The etiology of mitral valve regurgitation in group 3 was degenerative in 32 patients (76.2%) endocarditis in 10 (23.8%). In group 3, 16 patients (38.1%) were in NYHA class I, 18 (42.8%) in class II, 6 (14.3%) in class III, 2 (4.7%) in class IV. All patients were examined by two- and three-dimensional transesophageal echocardiography. Echocardiographic data were obtained using a 5.0 MHz multiplane transesophageal echo-probe (Model 21369 A, connected to SONOS 5500 Imaging System, Philips, Andover, MA, USA). The probe was advanced to midesophageal level to visualize a modified transversal four-chamber view that entirely visualizes left ventricle. At each point of measurement, three cardiac cycles were recorded for later evaluation. Multiplane echocardiographic data were processed by the EchoAnalyzer®, a multitask system developed at our institution for three-dimensional reconstruction and measurements [6,7]. The three-dimensional acquisitions were obtained by rotating the tip of the transesophageal transducer, which was steered by a step motor. The acquisition was accomplished by increments of 2° until to obtain 90 heart cycles. The time needed for obtaining the complete multiplanar data set ranged from 51 s to 3 min and 45 s (mean 1 min, 49 s) according to the spatial resolution of the data and the heart rate of the patients. The 3D acquisition was triggered to the ECG and the respiratory cycle. Three-dimensional shapes of left ventricular annuli were obtained by using a semiautomatic technique for segmentation of heart cavities, which was implemented into the EchoAnalyzer® (Fig. 1). Longitudinal and septal-lateral annulus diameters; global diastolic and systolic annular area and its percentual shortening, diastolic and systolic areas of six regions corresponding to Carpentier classification were measured by 3D-echocardiography (Fig. 2). Changes in annulus dimension, area and shortening of the two groups of patients with mitral regurgitation (groups 2 and 3) were calculated as the percent of the differences between the values of group 1 and group 2 or 3, respectively. Three-dimensional reconstructions of left ventricular annulus obtained by three-dimensional echocardiography. Schematics showing the measured segments, areas, and diameters of mitral valve annulus. Left ventricular geometry, and hence global left ventricular remodeling, was assessed by the ‘sphericity index’ as the ratio of longitudinal to short axis diameters of left ventricle. This index represents an indirect parameter of the degree of global leaflet tethering because the more spherical the left ventricle becomes, the greater the displacement of papillary muscles, according to Hung et al. [8]. The assessment of regional left ventricular remodeling was accomplished by measuring the anterior and posterior tethering of papillary muscles. Lateral and posterior displacements of anterior and posterior papillary muscles were quantified as the distances from two anatomic landmarks defined by the septal insertions and by a mid septal perpendicular line at early systole according to Yiu et al. [9]. The mitral valve deformation was evaluated by measuring the tenting area, i.e. the area enclosed between mitral leaflets and the line of annular plane and the coaptation depth, i.e. the distance between leaflet coaptation and mitral annular plane at early systole [9]. The degree of mitral regurgitation was assessed by three-dimensional color Doppler as the volume of regurgitant jets [7], which were calculated from the three-dimensional color Doppler data sets by counting the voxels containing high velocity and turbulence components. The origin, direction, and three-dimensional spatial distribution of mitral regurgitant jets were analyzed in the two groups of patients with mitral regurgitation and classified according to the main jet direction: posterior, anterior, central, or indefinable. The technique that allows the three-dimensional reconstruction of color Doppler data and the measurement of regurgitant jet volumes for quantitative assessment of mitral regurgitation has been developed at our institution and has been previously described [10]. All results are presented as mean ± standard deviation, and p-values <0.01 were considered statistically significant. Linear regression analysis was used to describe the correlations between different annular dimensions and the degree of mitral regurgitation. Differences between groups with and without impaired cardiac function were assessed by Student's t-test for unpaired data. Patients with significant ischemic mitral regurgitation (group 2) showed significantly larger longitudinal (52.7 ± 3.9 mm vs 41.8 ± 2.9 mm; p < 0.01) and antero-lateral (31.8 ± 3.5 mm vs 26.7 ± 2.8 mm; p < 0.01) annular diameters (Table 2) than the patients with mitral regurgitation due to primary valvular lesions (group 3). Diastolic (997.8 ± 64.9 mm2 vs 700.7 ± 46.8 mm2; p < 0.01) and systolic (894.9 ± 57.3 mm2 vs 547.3 ± 35.0 mm2; p < 0.01) annular areas (Table 2) were larger in the group with ischemic mitral regurgitation than in patients with mitral regurgitation due to primary valvular lesions. Moreover, annular area change, an index of the sphincteric function of mitral annulus, was significantly lower in the group with ischemic mitral regurgitation than in the group with primary valvular lesions (10.3 ± 1.1% vs 21.9 ± 1.6%; p < 0.01). Annular dimensions, left ventricular geometry, valve deformation, and regurgitation parameters The analysis of regional annular areas of six sectors according to the segmental classification of Carpentier (Fig. 2) showed that these sectors were homogeneously larger in the patients with ischemic mitral regurgitation than in the patients with primary valvular lesions. The area corresponding to the P3 region did not show larger areas than the other sectors. The percent of changes in the above mentioned parameters, compared to the patients with coronary artery disease without mitral regurgitation (group 1), are reported in Table 2. The sphericity index was significantly higher in the patients with ischemic mitral regurgitation compared to the controls and to the patients with primary valvular lesions (1.13 ± 0.44 vs 1.97 ± 0.51; p < 0.01). The lateral and posterior displacement of anterior and posterior papillary muscles showed significantly higher values in the group 2 than in controls and than in primary valve lesions (Table 2). No specific patterns of symmetric or asymmetric tethering were clearly identified in group 2 [11]. Tenting area and coaptation depth were measured only in groups 1 and 2, because mitral closure patterns of patients in group 3 (mainly valve prolapse or endocarditis) were not comparable to an approximately regular triangle as suggested by Yiu et al. [9]. The values of tenting area and the coaptation depths were significantly larger in patients with ischemic mitral regurgitation than controls. The degree of mitral regurgitation, as assessed by the volumes of regurgitant jets, was higher in the group with primary valvular lesions than in the patients with ischemic mitral regurgitation (32.6 ± 13.4 cm3 vs 23.1 ± 11.1 cm3; p < 0.01). The analysis of three-dimensional spatial distribution of mitral regurgitation showed that the patients with ischemic mitral regurgitation had mainly centrally directed jets than patients with primary valvular lesions (Fig. 3). Three-dimensional reconstruction of central regurgitation jet and mitral valve annulus. This figure shows the spatial relationship between the spreading of regurgitant jet and annular shape. Patients with impaired left ventricular function due to coronary artery disease and ischemic mitral valve regurgitation show poor long-term survival when compared to coronary artery disease without mitral regurgitation. Both the presence and the degree of ischemic mitral regurgitation influence the patients’ prognosis after myocardial infarction by increasing the probability of pulmonary edema, cardiogenic shock, and death [12]. The choice of the optimal therapy for ischemic mitral regurgitation is still one of the most controversial issues of cardiac surgery. This is, at least in part, due to the fact that ischemic mitral regurgitation has a composite pathogenesis involving the interactions of different anatomic structures as leaflets, annulus, chordae tendineae, papillary muscles, and left ventricle [2]. Accordingly, different surgical treatments including mitral valve replacement [13] or repair [14,15] revascularization [16] or left ventricular volume reduction [17] have been proposed. Two major clinical studies, which compare different surgical approaches, have shown that annuloplasty has better clinical outcome than mitral valve replacement [18,19]. However, in spite of the fact that these studies incontestably demonstrated the superiority of annuloplasty to mitral valve replacement, their predictive value is reduced by the lack of an accurate double-blind cross-randomization of the patients to the different surgical procedures. Recently, experimental studies in sheep demonstrated that myocardial ischemia might cause an asymmetrical distortion of mitral valve annulus, which likely accounts for the pathogenesis of ischemic mitral regurgitation [3,4,20]. Such ovine models of ischemic mitral regurgitation have been rashly transferred to postinfarction dilated cardiomyopathy of humans, so that today they represent the rationale for treating ischemic mitral regurgitation with asymmetrical annulopasty procedures. According to these concepts, it has also been proposed that the underlying mechanism of ischemic mitral regurgitation is based on a regional annulus distortion, occurring predominantly at the P3 region of posterior mitral leaflet, and accordingly, asymmetrical annuloplasty rings have been already manufactured and implanted in patients. The specific geometrical characteristics of these asymmetrical annuloplasty rings include a reduced anteroposterior distance to increase leaflet coaptation, a reduced P2–P3 curvature to compensate for the tethered P3 segment, and a dipped P3 region to accommodate its downward displacement [21]. Although these theoretical models of changes in the annular geometry and function are very suggestive, there is poor evidence that they might really reflect the actual individual pathomechanism of every single patient with ischemic mitral regurgitation. Today we observe the growth of quite a number of different annulopasty rings, whose design seems to be based more on subjective judgment than on objective measurements. Often these approaches pretend to provide a definitive solution for surgical treatment of ischemic mitral regurgitation. In our opinion, the underlying pathomechanism of ischemic mitral regurgitation is different in each patient and needs an individualized surgical procedure, which is based on the actual specific dysfunction. The aim of the present study was to analyze the changes in three-dimensional geometry of mitral valve annulus in patients with ischemic mitral regurgitation compared to patients with primary valvular lesions. According to the results, the underlying pathogenetic mechanism of ischemic mitral regurgitation seems to be associated to a global, uniform dilatation of mitral valve annulus occurring at each and every one of the different regions. An increased occurrence of an annular distortion localized at the P3 region could not be observed in our series of patients, thus suggesting that perhaps a more physiologically shaped annuloplasty ring that homogeneously reduces mitral annulus at each of the valve regions might improve the functional outcome, as demonstrated in a recent study, where a restrictive annuloplasty with the ‘physio-ring’ was performed with excellent clinical results [22]. In addition, a continuous geometrical remodeling of left ventricle, which is also thought to be responsible for the recurrent mitral regurgitation occurring after ring annuloplasty [8], is still recognized as a major pathogenetic determinant of ischemic mitral regurgitation. Since coronary artery disease is, as per definition, a disease that regionally affects different myocardial regions, according to the distribution of coronary lesions, the concept of an asymmetrical annuloplasty ring still remains very suggestive and provides a useful background for selecting the surgical options. However, individual information on structure and function of the different components of mitral valve should be obtained by the current preoperative diagnostics in the single patients. Today resonance imaging may be considered the gold standard technique for non-invasive measurements of intracardiac structures and of left ventricular volumes [23,24], nevertheless its availability is limited by the induction of strong magnetic fields that prevent its application in the operating room. In contrast, transesophageal color Doppler echocardiography, along with three-dimensional reconstruction systems, are now routinely available in the operating room and may provide the necessary quantitative assessment needed for choosing the different surgical approaches to ischemic mitral regurgitation [6,7,10,17]. Downsizing the valvuloplasty ring is another well-established technique for surgical repair of ischemic mitral regurgitation. In a recent study, Bax et al. [22] showed that a group of 51 patients who underwent CABG and restrictive annuloplasty with a downsized mitral valve ring had good results and significant reverse left ventricular remodeling after a two-year follow-up [22]. Recent studies, based on elegant computational models, by Maisano et al. [25] demonstrated that modifying the shape of the annular prosthesis might provide the key for treating patients with functional mitral regurgitation. Different degrees of leaflet tethering were applied to the different models in order to simulate special patterns of ventricular remodeling. The results of this experimental study open the new interesting concept that a specifically designed annuloplasty ring may improve the functional outcome by directly acting on the tethering mechanisms. A direct comparison of different surgical techniques for treating ischemic mitral regurgitation was beyond the design of this study. This is a clinical investigation designed specifically to examine the geometry and dynamics of mitral valve annulus as a determinant of ischemic mitral regurgitation. One of the major drawbacks is the exact clinical definition of the patients’ study cohort according to the different pathogenetic mechanisms, which essentially is the main limitation that affects every clinical investigation dealing with ischemic mitral regurgitation. The clinical characterization of the study groups critically influences the results and the comparability between different clinical trials, performed in different institutions, is not practicable. In our study, the significantly lower values of annular area changes, which can be considered as an index of the sphincteric function of mitral annulus, in the group with ischemic mitral regurgitation compared to the group with primary valvular lesions is once more consistent with the concept that an important factor of ischemic mitral regurgitation is the global impairment of left ventricular function and the global dilatation of left ventricle. The analysis of global and regional left ventricular geometry, the tethering indexes, and the regurgitant jet characteristics showed that the patients with ischemic mitral regurgitation had poor ventricular function, global enlargement of left ventricle, and predominantly central jets. For that reason the results of this study should be cautiously interpreted, because this small series of patients might not reproduce the wide range of different pathomechanisms and the complex clinical presentation of ischemic mitral regurgitation. Further clinical investigations with more accurate 3D diagnostic means are mandatory for improving the practical impact of these findings. In summary, the findings of this study suggest that the use of asymmetrical annuloplasty rings for ischemic mitral regurgitation should be cautiously applied to selected patients after individual assessment of geometrical changes of overall mitral valve apparatus, including left ventricle, papillary muscles, leaflets, and annulus. An accurate individual three-dimensional assessment and quantification of the mitral valve in all its anatomical and functional components is likely to provide more accurate information for choosing an individualized surgical approach that is based on the exact knowledge of the underlying pathomechanism. The ideal surgical repair for ischemic mitral regurgitation should be individually tailored after the assessment of the three-dimensional geometry and function of each single component of the mitral valve complex. This work was supported by the Deutsche Forschungsgemeinschaft – SFB 414: Information Technology in Medicine – Computer and Sensor Supported Surgery. Project H1: Clinical Application and Development of 3D-Procedures for the Diagnosis and Therapy.