Edge-to-Edge Mitral Repair
Tension on the Approximating Suture and Leaflet Deformation During Acute Ischemic Mitral Regurgitation in the Ovine Heart
Background Edge-to-edge approximation of the mitral valve leaflets (Alfieri procedure) is a novel surgical treatment for patients with ischemic mitral regurgitation (IMR). Long-term durability may be limited if abnormal mitral leaflet stresses result from this procedure. The aim of the current study was to measure Alfieri stitch tension (FA) and to explore its geometric determinants in an ovine model of acute IMR as a reflection of the mitral leaflet stresses imposed by the procedure.
Methods and Results Eight sheep were studied immediately after surgical placement of (1) a force transducer interposed between sutures approximating the central leaflet edges and (2) radiopaque markers around the mitral annulus and leaflet edges. Computer-aided analysis of videofluorograms was used to obtained 3D marker coordinates. Simultaneous measurements of FA, septal-lateral annular dimension (LS-L), leaflet edge separation (LSEP), anterior (LAL) and posterior (LPL) leaflet length, and hemodynamic variables were obtained at baseline (CTL) and during acute IMR (circumflex artery occlusion). FA was significantly elevated throughout the cardiac cycle during IMR compared with CTL, with maximum FA in diastole (0.26±0.05 versus 0.46±0.08 N, CTL versus IMR; P<0.05). Multivariable analysis revealed LS-L as the single independent predictor of maximum FA (P<0.001). Positive linear correlations were shown between values of FA and LAL and LPL (dependent variables).
Conclusions These experimental data demonstrate higher FA during IMR and cyclic changes in FA closely paralleling changes in LS-L, eg, being greatest in diastole when the annulus is largest. Increased FA during IMR is probably indicative of successful therapeutic intent, but higher diastolic leaflet stresses resulting from persistent or progressive mitral annular dilatation may adversely affect repair durability. This indirectly implies that concomitant mitral ring annuloplasty should be added to the Alfieri repair.
Mitral valve repair has become the preferred method for correcting ischemic mitral regurgitation (IMR). It provides durable and predictable results in most patients1 and probably superior clinical outcomes compared with mitral valve replacement.2–4 Mitral annuloplasty, with either a prosthetic ring or suture plication, has been applied to correct IMR5–7 with the intent of reducing mitral annular area and improving leaflet apposition. Ring annuloplasty has also been shown to prevent IMR in an ovine model of acute LV ischemia,8 but ring annuloplasty abolishes normal mitral annular dynamics9 and restricts posterior leaflet mobility.10 Ring annuloplasty alone, however, can be unpredictable in some patients with IMR.11
Recently, a simple mitral valve repair method in which the leading edges of the mitral leaflets are approximated by use of a suture (Alfieri procedure or “bow-tie” repair) was introduced by Alfieri and colleagues.12 The technique involves short aortic cross-clamp times, and the location of the suture can be customized on the basis of the location of the regurgitant jet with both central and paracommissural leaflet approximation.13 The procedure has been shown to provide predictable and durable control of MR in patients with a wide spectrum of valvular pathology13,14 and was used successfully in a small series of patients with IMR.15 From theoretical models of the normal mitral valve, it is thought that the tension on the approximating suture between the mitral valve leaflet edges is close to zero.15,16 Long-term durability, however, may be limited in mitral valve diseases with preexisting deformation of the mitral valve apparatus if abnormal leaflet stresses result from the procedure.
In evaluating the Alfieri procedure in an ovine model of acute IMR, we hypothesized that mitral annular dilatation during IMR would significantly increase Alfieri stitch tension (FA) and consequently augment stresses on the mitral valve leaflets. Therefore, the present study was performed to assess FA and deformation of the mitral leaflets as a reflection of the mitral leaflet stresses imposed by the procedure and to explore the geometric determinants of FA in this experimental setting of acute IMR.
The surgical procedures used have been previously described.17 Eight adult sheep were used in the study. A diagram of the study protocol is shown in Figure 1. With the chest open but before marker implantation and mitral repair, the proximal circumflex coronary artery was dissected from the surrounding myocardium, and a silicone rubber snare occluder was placed proximal to the first obtuse marginal branch. Doppler echocardiography was used to demonstrate the presence of acute IMR in all animals after a 2-minute occlusion of the proximal left coronary circumflex artery. Subsequently, 8 miniature radiopaque markers were inserted into the subepicardium of the left ventricle along 4 equally spaced longitudinal meridians with 2 levels between the left ventricular (LV) apex and base and 1 marker at the LV apex. After establishment of cardiopulmonary bypass and with the heart arrested, 8 markers were sutured around the circumference of the mitral annulus (1 near each commissure and 3 along the anterior and posterior annuli). Subsequently, the central edges of the anterior and posterior mitral leaflets were approximated with a 5-0 polypropylene suture reinforced with 2 small Teflon-felt pledgets as shown in Figure 2A and 2B. The approximating suture (Alfieri stitch) was placed ≈5 mm from each leaflet edge and secured a miniature force transducer that served as another radiopaque marker (No. 13 in Figure 2B). The force transducer was constructed of a slit copper ring 5 mm in diameter and 0.4 mm thick. On each side of the slit, 2 small holes were incorporated to permit suture fixation. The transducer used a simple semiconductor strain gauge connected in a quarter-bridge electrical circuit. Technical specifications of the transducer have been described previously.18 Four additional radiopaque markers were placed on the posterior and anterior leaflet edges at the center of each of the 2 newly created valve orifices. Micromanometer-tipped catheters (Millar SPC-500) were placed in the LV chamber via the apex and in the left atrial lumen via the atriotomy for monitoring of LV pressure (LVP) and left atrial pressure (LAP).
After completion of marker implantation, the heart was defibrillated, and the animals were weaned from cardiopulmonary bypass and transferred immediately to the experimental animal catheterization laboratory where they were studied intubated with open chests and anesthetized with ketamine (1 to 4 mg · kg−1 · h−1 IV infusion) and diazepam (5-mg IV bolus as needed). Esmolol infusion (20 to 50 μg · kg−1 · min−1 IV) was used to minimize reflex sympathetic responses. Hemodynamic data and simultaneous biplane videofluoroscopy and epicardial color Doppler echocardiography were obtained at steady-state conditions and during vena caval occlusion immediately before and after a 2-minute occlusion of the circumflex artery (proximal to the first obtuse marginal branch) with a silicone rubber snare occluder. The sheep were allowed to stabilize 3 to 5 minutes between data acquisition runs. The extent of MR during acute proximal circumflex occlusion after the edge-to-edge repair was compared with the extent of MR induced by circumflex occlusion before marker implantation and valve repair, thus allowing evaluation of the efficacy of the Alfieri stitch in the setting of acute IMR.
All animals received humane care in compliance with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the Guide for Care and Use of Laboratory Animals prepared by the National Academy of Sciences and published by the National Institutes 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.
Videofluoroscopic images were acquired with the animals in the right lateral decubitus position (with the chest open) with a Philips Optimus 2000 biplane Lateral ARC 2/Poly DIAGNOST C2 system (Phillips Medical Systems, North America) with the image intensifier in the 9-in fluoroscopic mode. Data from the 2 radiographic views were digitized and merged by use of custom-designed software19 to yield the 3D x, y, and z coordinates for each of the radiopaque markers every 16.7 seconds throughout the cardiac cycle. Ascending aortic pressure, LVP, LAP, FA, and ECG voltage signals were digitized and recorded simultaneously during data acquisition.
Hemodynamic and Cardiac Cycle Timing
Two or three consecutive steady-state beats during control and after occlusion of the circumflex artery were averaged and defined for each animal as CTL and IMR, respectively. In each cardiac cycle, end systole was defined as the frame containing peak rate of LVP fall (−dP/dt). End diastole was defined as the videofluoroscopic frame containing the peak of the ECG R wave. Instantaneous LV volume at end systole and end diastole was calculated from epicardial LV markers by use of a space-filling multiple tetrahedral volume method.20 Although myocardial volume is included in this calculation of LV volume, it accurately reflects changes in LV chamber size.
Mitral Annular Dynamics
Mitral annular area throughout the cardiac cycle was computed from 3D coordinates of the 8 markers sutured around the mitral annulus (markers 1 through 8, Figure 2B). The annular centroid was first calculated, and then the annular area was divided into 8 individual “pie slices” whose areas were summed to yield total annular area. The septal-lateral mitral annular diameter (LS-L) was determined as the distance in 3D space between markers placed on the midseptal (anterior) and midlateral (posterior) mitral annuli (1 through 5 in Figure 2B).
Mitral Leaflet Dynamics
The anterior and posterior mitral leaflet lengths (LAL and LPL, respectively) throughout the cardiac cycle were calculated as the 3D distances from the Alfieri stitch at the leaflet edges to the anterior and posterior leaflet bases, respectively (1–13 and 5–13, Figure 2B). Leaflet separation in the 2 newly created mitral orifices was computed as the distance between the respective leaflet edge markers (9–10 and 11–12, Figure 2). LSEP was defined as the average of the distances between the leaflet edge markers in the 2 orifices.
All data are reported as mean±SEM unless otherwise noted. Hemodynamic and marker-derived data from consecutive steady-state beats from each heart were time aligned at end systole. Marker data were calculated from 20 frames before to 20 frames after end systole, a duration of 650 ms. Differences between variables (FA, LS-L, LSEP, LAL, LPL) during CTL and IMR were compared at end diastole, end systole, peak FA in diastole, and their minimum and maximum by use of a multivariate ANOVA with variables of time, before versus after circumflex artery occlusion, and sheep identification. Statistical significance was inferred if P<0.05.
Consecutive diastolic values of the transvalvular pressure gradient (LAP−LVP), LS-L, LSEP, and CTL/IMR were entered into a general linear model multivariable analysis with FA as the dependent variable. Because of the variability between animals, analysis was performed for each animal, and tests of significance of the regression coefficients (different from zero) were subsequently performed with the use of the ensemble of data from all experiments. Relationships between LS-L, LSEP, and LAP−LVP (independent variables) and FA (dependent variable) with caval occlusion at CTL and during IMR were tested at their maximum in diastole, at time of maximum FA, and at end systole in similar linear regression models. Sheep identification was treated as a model factor to reduce the influence of animal-to-animal variability on the analysis. With the use of linear regression analysis, correlations between values at maximum FA of LAL and LPL (dependent variables) and FA (independent variable) during vena caval occlusion at CTL and IMR were tested for each experiment and all experiments together (see above). Statistical models in StatView 227 were used for the analysis.
The average weight of the animals was 68±5 kg. The total duration of cardiopulmonary bypass was 82±9 minutes, with a mean aortic cross-clamp time of 61±7 minutes. Proper marker position and integrity of the leaflet-approximating stitch were confirmed in all animals at postmortem examination (Figure 2A).
Hemodynamics and MR
Hemodynamic variables after the edge-to-edge repair at CTL and during IMR are shown in Table 1. As expected, myocardial ischemia decreased LVP and dP/dtmax and increased LV volumes and mitral annular area at end systole and end diastole. Figure 3 shows the degree of MR for each animal at CTL and IMR before and after the Alfieri procedure. The location of the regurgitant jet was broadly central and holosystolic before the Alfieri repair, with some animals having additional smaller commissural jets. In most animals, IMR persisted after the repair, and the location of the jet was predominantly central (on both sides of the approximating stitch). As indicated by the thin lines in Figure 3, occlusion of the circumflex artery caused moderate MR in most animals. Heavy lines demonstrate the change in IMR before and after the Alfieri procedure. Except for 1 animal that had grade III MR after the edge-to-edge repair (possibly because of leaflet distortion from the transducer), the Alfieri procedure in general had a small effect on the degree of IMR (grade 1.5±0.3 after the Alfieri procedure versus grade 1.8±0.1 before the procedure, P=0.4). Thus, it is important to note that in the setting of the Alfieri repair, acute occlusion of the proximal circumflex artery still resulted in significant IMR in this ovine preparation. In addition, there was no relationship between Alfieri stitch tension and the degree of MR during occlusion of the circumflex artery or between the change in Alfieri stitch tension and MR from CTL to IMR.
Alfieri Stitch Tension
The time course of FA, LVP, LS-L, and LSEP at CTL and IMR (mean of 8 animals) at steady state and during vena caval occlusion at CTL and during IMR is shown in Figure 4. FA was significantly elevated throughout the cardiac cycle during IMR compared with CTL. Differences in FA between the 2 conditions compared at peak FA in diastole, at minimum FA in early systole, and at end systole were all statistically significant (Table 2), although the increase in tension was most remarkable in late systole with maximum FA increase at end systole. The change in FA from end systole to maximum in diastole was the same during IMR and CTL (0.18±0.02 versus 0.12±0.05 N, P=NS).
Correlates of Alfieri Stitch Tension
The time course of FA at CTL and IMR was closely related to cyclic changes in LS-L. Maximum FA was found in diastole coincident with maximum LS-L and before maximum LSEP. FA decreased during middiastole before the closing motion of the mitral leaflets. In most hearts, minimum FA occurred in early systole at the same time as minimum LS-L. In 4 animals, the FA curves in diastole were bimodal with second peaks at end diastole, which explains the observed end-diastolic “humps” in the mean curves. Graphically, the FA curves closely paralleled changes in LS-L but did not appear to be related to changes in LSEP or the transvalvular pressure gradient. LS-L was also significantly higher during IMR compared with CTL, whereas LSEP was virtually unchanged before and after ischemia. FA and LS-L decreased during vena caval occlusions at CTL and during IMR, but LSEP remained almost unchanged (Figure 4).
In multivariate linear regression analysis, values of FA in diastole (versus corresponding values of LS-L, LSEP, and LAP−LVP) were significantly correlated to LS-L (the only statistically significant factor was regression coefficient, β=0.014 to 0.086 N/mm; R2=0.54 to 0.97; P<0.05 in all animals). Multivariate linear regression analysis of relationships between maximum LS-L, LSEP, and LAP−LVP (independent variables) and maximum FA (dependent variable) during steady state and vena caval occlusions at CTL and IMR showed that maximum LS-L was the single statistically significant predictor of FA in all hearts (P<0.001).
Values at maximum FA and at end systole of LS-L, LSEP, and LAP−LVP (independent variables) at CTL and during IMR (steady state and caval occlusions) were also entered into the model with corresponding values of FA as the dependent variable. In both analyses, FA was positively correlated with LS-L and LSEP; however, in multivariable tests for each experiment, statistical significance fell out to either of the covariates, which indicates interaction between the factors.
Relation Between Alfieri Stitch Tension and Mitral Leaflet Deformation
Figure 5 illustrates the relationship between FA and LAL and LPL (mean of 8 experiments) from the time of minimum to maximum FA at CTL and during IMR. Positive statistically significant correlations were demonstrated between maximum FA and LAL and LPL (dependent variables) at CTL and IMR (steady state and caval occlusion) in each experiment (LAL versus FA: β=3.0 to 39.4 mm/N, R2=0.57 to 0.97, P<0.05 in all hearts; LPL versus FA: β=1.2 to 17.8 mm/N, R2=0.51 to 0.79, P<0.05 in all hearts) and for all animals together (P<0.001).
Mitral valve reconstruction has become the procedure of choice for patients with most forms of mitral regurgitation. The Alfieri edge-to-edge repair as an adjunctive technique to correct IMR may be a valuable addition to the surgeon’s armamentarium for selected patients who otherwise would require more complicated repair techniques or replacement of a morphologically normal valve. Several advantages make the Alfieri repair appealing: As an adjunctive technique for repair of IMR, the Alfieri stitch can reduce the risks associated with mitral valve replacement in this high-risk population. The Alfieri technique involves short aortic cross-clamp times and can be customized to the location of the regurgitant jet. Our previous ovine studies attributed acute IMR to annular dilatation8 and leaflet tethering,21 and most likely, both mechanisms contributed to the genesis of IMR in the current experiment. Although annuloplasty rings prevent acute ovine IMR, the mechanisms responsible for the clinical efficacy of annuloplasty rings in chronic IMR remain incompletely understood. An Alfieri stitch prevents leaflet separation and creates apposition of the mitral leaflets at that site; it can be a valuable “bail-out” procedure when ring annuloplasty fails to correct IMR satisfactorily. The edge-to-edge repair is intended to counteract leaflet tethering and incomplete mitral leaflet coaptation, which is considered to be a mechanism of IMR in patients.22 In the present study, acute occlusion of the circumflex artery still resulted in significant IMR even after the Alfieri repair, and there was no relationship between the magnitude of Alfieri stitch tension and the success of the repair. It is important to note, however, that the biomechanical approach of this study was not designed to evaluate the efficacy of the procedure in terms of valvular regurgitation (see Study Limitations).
Some cardiac surgeons have been understandably skeptical of the durability of a single-suture repair in an area of apparently high stress in which high systolic pressure gradients, rapid motion of the mitral leaflets in diastole, and alterations of the mitral valve attachments may potentially contribute to increased stresses on the thin valvular leaflet tissue. In this ovine model of acute IMR, LS-L was the major predictor of Alfieri stitch tension. Accordingly, the time course of FA, with maximum in diastole, was closely related to cyclic changes in LS-L. Increased LS-L resulting from annular dilatation during acute ischemia resulted in higher FA compared with CTL, with maximum FA on the order of 16 to 70 g force. In comparison, the maximum systolic tension of a single chorda tendinea is in the range of 80 to 90 g force. Maximum FA occurred before maximum LSEP, but the observed correlation between FA and LSEP at maximum FA in diastole suggests that the opening motion of the mitral leaflets may exert additional stresses on the Alfieri stitch. Mitral annular dynamics during CTL and IMR in this short-term open-chest study are similar to those observed in our previous conscious, closed-chest ovine experiments17 (7 to 11 days after annular marker implantation) and other published data.23
In 10 very ill patients undergoing edge-to-edge mitral valve repair resulting from IMR, Umana et al15 anticipated that the mitral leaflets are naturally pushed together and that therefore stresses on the approximating stitch would not be excessive in systole when the suture serves mainly to limit the excursion of the leaflets into the left atrium during high LVP in systole. Consequently, they postulated that there would likely be less tension on the leaflet-approximating suture during systole. From a mathematical model of the stresses on the closed mitral valve derived by Arts et al,16 it was proposed that a decrease in annular diameter caused by insertion of a mitral annuloplasty ring reduces the leaflet surface area exposed to the systolic ventricular pressure and results in decreased tension on the edges of the leaflets and therefore on the suture. In accordance with this theory, the present study demonstrated lower Alfieri stitch tension with no relation to LVP in systole. Greater Alfieri stitch tension was seen in diastole, which was directly related to mitral annular dimensions at CTL and during IMR, indicating that the highest stresses on the mitral leaflets after the Alfieri procedure appear in diastole, and this may, at least in theory, affect the durability of the procedure. It has also been speculated that major stresses on the Alfieri suture may occur secondary to the AV pressure gradient pulling the leaflets apart during valve opening in diastole.15 In the present study, however, no relationship between the diastolic transvalvular pressure gradient and Alfieri stitch tension was observed, but there appeared to be a minor contribution of mitral leaflet separation to the Alfieri stitch tension.
The data in this experiment demonstrate that higher FA during IMR was associated with increased strain on the mitral leaflets, as if the Alfieri repair results in an anchoring effect that limits further dilatation of the mitral annulus. Conversely, cyclic changes of mitral annular area and septal-lateral mitral annular diameter after the Alfieri procedure at CTL and IMR were comparable to our previous observations in sheep without valve repair.17 It must be emphasized that the present experiment consists solely of acute IMR in sheep hearts; accurate clinical extrapolation of these results is difficult. Nonetheless, these data suggest that Alfieri stitch tension in conditions in which annular dilatation is part of the pathophysiology, such as in progression of chronic IMR, may become excessively high and in turn have a negative effect on the durability of the procedure. If this is the case, it argues for performing adjunctive procedures on the mitral annulus to relieve these stresses, such as a mitral annuloplasty ring. Indeed, McCarthy et al24 reported on 4 patients undergoing partial ventriculectomy and the Alfieri repair alone who, despite minimal initial mitral regurgitation, developed substantial mitral insufficiency within several months, probably because of progressive annular dilatation. In current clinical practice, a concomitant ring annuloplasty is recommended by those who use the Alfieri technique extensively.12–14
In conclusion, the present study demonstrated higher Alfieri stitch tension during IMR and cyclic changes in FA, with maximum in diastole, which closely paralleled changes in mitral annular septal-lateral dimension. Increased stitch tension during IMR is probably indicative of successful therapeutic intent, but FA certainly is not zero. Hypothetically, any increase in Alfieri stitch tension caused by residual or progressive annular dilatation may adversely affect repair durability
The major limitations of this study are those inherent in the acute animal model used. The study was based on experiments using acute myocardial ischemia and therefore did not take into account long-term changes, such as LV remodeling and resultant compensatory changes in the mitral subvalvular apparatus. Furthermore, although the extent of myocardial ischemia after circumflex artery occlusion has been shown to be very reproducible,25 the effect on mitral valve competency is not simple because it involves many components of the mitral valvular-ventricular complex.23,26 Therefore, these data must be interpreted in the setting of experimental acute IMR and cannot be extrapolated directly to chronic IMR in human subjects. The animals used in this study were anesthetized and studied in open-chest conditions, and no annuloplasty rings were used, all of which limit extrapolation of these experimental results to humans. Anatomical differences between ovine and human mitral leaflets23 and annulus27 must also be considered.
Because one of the goals of this study was to investigate leaflet and annular dynamics after the edge-to-edge repair, ring annuloplasty was purposely avoided. In addition, the approximating suture was placed centrally in the mitral orifice in all hearts even though prerepair Doppler echocardiography during ischemia revealed IMR with small regurgitant jets directed toward either of the commissures in some animals, findings that might have prompted a repair not located in the center of the valve. It should also be mentioned that performing the edge-to-edge repair and then inducing acute IMR is distinctly different from the clinical situation in which a broad leak may be present all along the leaflet coaptation zone. Furthermore, the presence of a force transducer might have compromised leaflet apposition adjacent to the approximation point and thereby worsened IMR. The lack of other control animals, ie, sheep without any repair (for comparison with normal annular dynamics) or a group with an Alfieri repair in combination with ring annuloplasty, presents further limitations. The mitral annular dynamics observed in the present study are concordant with our previous investigations of ovine mitral annular dynamics.17
This work was supported by grant HL-29589 from the National Heart, Lung and Blood Institute. Dr Nielsen is supported by grants from the Danish Heart Foundation. Drs Timek and Lai are Carl and Leah McConnell Cardiovascular Research Fellows. Dr Timek is a recipient of the Thoracic Surgery Foundation Research Fellowship Award and was also supported by NHLBI INRSA HL-09569. Dr Lai received fellowship support from the American Heart Association, Western States Affiliate. We appreciate the technical assistance provided by Mary K. Zasio, BA, Carol W. Mead, BA, and Maggie Brophy, AS.
Kay GL, Kay JH, Zubiate P, et al. Mitral valve repair for mitral regurgitation secondary to coronary artery disease. Circulation. 1986; 74 (suppl I): I-88–I-98.
Hickey MS, Smith LR, Muhlbaier LH, et al. Current prognosis of ischemic mitral regurgitation: implications for future management. Circulation. 1988; 78 (suppl I): I-51–I-59.
Enriquez-Sarano M, Schaff HV, Orszulak TA, et al. Valve repair improves the outcome of surgery for mitral regurgitation: a multivariate analysis. Circulation. 1995; 91: 1022–1028.
Czer LS, Maurer G, Trento A, et al. Comparative efficiency of ring and suture annuloplasty for ischemic mitral regurgitation. Circulation. 1992; 86 (suppl II): II-46–II-52.
Cohn LH, Rizzo RJ, Adams DH, et al. The effect of pathophysiology on the surgical treatment of ischemic mitral regurgitation: operative and late risks of repair versus replacement. Eur J Cardiothorac Surg. 1995; 9: 621–626.
Fucci C, Sandrelli L, Pardini A, et al. Improved results with mitral valve repair using new surgical techniques. Eur J Cardiothorac Surg. 1995; 9: 621–626.
Maisano F, Torracca L, Oppizzi M, et al. The edge-to-edge technique: a simplified method to correct mitral insufficiency. Eur J Cardiothorac Surg. 1998; 13: 240–245.
Maisano F, Schreuder JJ, Oppizzi M, et al. The double-orifice technique as a standardized approach to treat mitral regurgitation due to severe myxomatous disease: surgical technique. Eur J Cardiothorac Surg. 2000; 17: 201–205.
Daughters GT, Sanders WJ, Miller DC, et al. A comparison of two analytical systems for three-dimensional reconstruction from biplane videoradiograms. Proc Comp Cardiol (IEEE). 1988; 15: 79–82.
Godley RW, Wann LS, Rogers EW, et al. Incomplete mitral leaflet closure in patients with papillary muscle dysfunction. Circulation. 1981; 63: 565–571.
Gorman JH, III, Jackson BM, Gorman RC, et al. Papillary muscle discoordination rather than increased annular area facilitates mitral regurgitation after acute posterior myocardial infarction. Circulation. 1997; 96 (suppl II): II-124–II-127.
Komeda M, Glasson JR, Bolger AF, et al. Geometric determinants of ischemic mitral regurgitation. Circulation. 1997; 96 (suppl II): II-128–II-133.
Walmsley R. Anatomy of human mitral valve in adult cadaver and comparative anatomy of the valve. Br Heart J. 1978; 40: 351–366.