|Home | About | Journals | Submit | Contact Us | Français|
Diastolic mitral valve (MV) opening characteristics during ischemic mitral regurgitation (IMR) are poorly characterized. The diastolic MV opening dynamics was quantified along the entire valvular coaptation line in an ovine model of acute IMR.
Ten radiopaque markers were sutured in pairs on the anterior (A1-E1) and corresponding posterior (A2-E2) leaflet edges from the anterior (A1/A2) to the posterior (E1/E2) commissure in 11 adult sheep. Immediately after surgery, 4-D marker coordinates were obtained before and during occlusion of the proximal left circumflex coronary artery. Distances between marker pairs were calculated throughout the cardiac cycle every 16.7 ms. Leaflet opening was defined as the time after end-systole (ES) when the first derivative of the distance between marker pairs was greater than a threshold value of 3 cm/s. Valve opening velocity was defined as the maximum slope of marker pair tracings.
Hemodynamics were consistent with acute ischemia, as reflected by increased MR grade (0.5 ± 0.3 versus 2.3 ± 0.7, p <0.05), decreased contractility (dP/dtmax: 1,948 ± 598 versus 1,119 ± 293 mmHg/s, p <0.05), and slower left ventricular relaxation rate (dP/dtmin: −1,079 ± 188 versus −538 ± 147 mmHg/s, p <0.05). During ischemia, valve opening occurred earlier (A1/A2: 112 ± 28 versus 83 ± 43 ms, B1/B2: 105 ± 32 versus 68 ± 35 ms, C1/C2: 126 ± 25 versus 74 ± 37 ms, D1/D2: 114 ± 28 versus 71 ± 34 ms, E1/E2: 125 ± 29 versus 105 ± 33 ms; all p <0.05) and was slower (A1/A2: 16.8 ± 9.6 versus 14.2 ± 9.4 cm/s, B1/B2: 40.4 ± 9.9 versus 32.2 ± 10.0 cm/s, C1/C2: 59.0 ± 14.9 versus 50.4 ± 18.1 cm/s, D1/D2: 34.4 ± 10.4 versus 25.5 ± 10.9 cm/s; all p <0.05), except at the posterior edge (E1/E2: 13.3 ± 8.7 versus 10.6 ± 7.2 cm/s). The sequence of regional mitral leaflet separation along the line of coaptation did not change with ischemia.
Acute posterolateral left ventricular ischemia causes earlier leaflet opening, probably due to a MR-related elevation in left-atrial pressure; reduces leaflet opening velocity, potentially reflecting an impaired left ventricular relaxation rate; and does not perturb the homogeneous temporal pattern of regional valve opening along the line of coaptation. Future studies will confirm whether these findings are apparent in patients with chronic IMR, and may help to refine the current strategies used to treat IMR.
Extensive studies have been conducted to investigate systolic mitral valve dynamics during ischemic mitral regurgitation (IMR) (1–3). Recent data have suggested that myocardial ischemia may also restrict mitral valve leaflet opening during ventricular relaxation (4,5), although few studies have described the diastolic mitral valve opening dynamics under physiological conditions (6) or during IMR. A more complete understanding of diastolic mitral valve opening dynamics may be important, as alterations are associated with changes in left ventricular inflow patterns (4) with potential deleterious consequences (7,8), and could affect optimal mitral valve closure during systole (9,10). Furthermore, IMR has been shown to result in asymmetrical ventricular dilatation and regional mitral leaflet tethering (11), and it is reasonable to assume that alterations of diastolic mitral valve dynamics occur predominantly in the region affected by the ischemic insult. A more detailed investigation of diastolic mitral valvular geometry and dynamics along the entire valvular coaptation line may, therefore, provide further insight into the pathophysiology of IMR and help to optimize the current strategies used to treat this disease.
The study aim was to quantify the diastolic mitral valve leaflet opening geometry and dynamics at multiple discrete points along the entire valvular coaptation line, in an ovine model of acute posterolateral myocardial ischemia. It was hypothesized that acute ischemia would regionally alter mitral valve leaflet opening due to localized tethering effects via a displacement of the posterior papillary muscle.
Eleven adult castrated male sheep (mean body weight 73 ± 8 kg) were premedicated with ketamine (25 mg/kg, intramuscularly), anesthetized with sodium thiopental (6.8 mg/kg, intravenously), intubated, and mechanically ventilated with inhalational isoflurane (1.0–2.5%). Miniature radiopaque tantalum markers (n = 13) were then implanted surgically into the subepicardium to silhouette the left ventricular chamber along four equally spaced longitudinal meridians, as described previously (1). In order to gain detailed insight into the mitral valvular leaflet opening dynamics and geometry during acute IMR, 10 markers were sutured as pairs along the anterior and posterior mitral leaflet free-edges. Five radiopaque markers (A1-E1) were placed on the anterior mitral leaflet edge (AML), and five markers (A2-E2) on the edge of the posterior mitral leaflet (PML) from the anterior commissure (A1/A2) to the posterior commissure (E1/E2; Fig. 1). Markers were also placed on second-order chordae insertion points of the atrial side of the AML, and at the tips of the anterior and posterior papillary muscle. In addition, eight markers were sewn equidistantly around the mitral annulus.
Postoperatively, the animals were transferred to the catheterization laboratory and studied while intubated, open-chest, and anesthetized with inhalational isoflurane (1.5–2.0%). All animals were studied in normal sinus rhythm, with ventilation arrested at end-expiration during data acquisition. In order to create acute posterolateral left ventricular ischemia, the left circumflex coronary artery was occluded proximal to the first obtuse marginal artery by cinching an encircling suture until the occurrence of moderate or more MR (2.6 ± 0.7 rain). Before and during ischemia, data were acquired at 60 Hz (every 16.7 ms) using biplane videofluoroscopy. Using transesophageal Doppler echocardiography, the MR was graded semi-quantitatively by an echocardiography expert, based on the regurgitant jet extent and width as none (grade 0), trace (+0.5), mild (+1), moderate (+2), moderate-severe (+3), or severe (+4), using a three-chamber view before and during the induction of acute posterolateral myocardial ischemia. Two-dimensional images from each of the two X-radiography views were digitized and merged to yield 3-D coordinates every 16.7 ms using custom software (12,13). The analog left ventricular pressure and electrocardiographic voltage were digitized and recorded simultaneously with the marker images. Other data derived from this study have been reported previously (14).
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 [NIH] Publication 85 to 23, revised 1985). This study was approved by the University’s Medical Center Laboratory Research Animal Review Committee, and conducted according to University policy.
Data from three consecutive hemodynamically stable beats during control and acute ischemic conditions were averaged and analyzed. End-systole was defined as the videofluoroscopic frame immediately preceding the peak negative rate of left ventricular pressure change (−dP/dtmax), and early diastole was defined as within two frames (34 ms) after end-systole (ES; representing the early phase of isovolumetric relaxation). The instantaneous left ventricular volume was computed from the epicardial left ventricular markers, using a space-filling multiple tetrahedral volume method (12). The hemodynamic data were calculated from instantaneous left ventricular volumes and analog left ventricular pressures.
To quantify mitral valve opening, the onset of marker pair separation and the maximum marker pair separation velocity were calculated, as shown schematically Figure 2A and B, respectively. First, the distances between marker pairs were calculated throughout the cardiac cycle. The onset of marker pair separation was defined as the time after ES when the first derivative of the distance between marker pairs was greater than a threshold value of 3 cm/s (Fig. 2A). The maximum of this first derivative was used to determine the maximum marker pair separation velocity (Fig. 2B). To illustrate whether potential differences in marker pair separation were due to changes in separation of the anterior or the posterior leaflet, or both, the positions of the marker pairs were plotted in three-dimensional space, using a right-handed Cartesian coordinate system. At each sample time, the origin of a right-handed internal coordinate system was defined as the centroid of the annular markers (#15–22, Fig. 1). The positive y-axis was defined from the origin to the midpoint of the papillary muscle marker tips. A plane containing the y-axis and the lateral annular marker (#18) was determined, and the positive x-axis was defined normal to the y-axis and pointing in the direction of marker #18. The positive z-axis was defined as normal to the x- and y-axes pointing towards the posterior commissure. Marker positions were plotted at ES and 102 ms after ES (as a late time point during iso-volumetric relaxation, where the leaflets are still closed under physiological conditions), under baseline conditions, and during acute left ventricular ischemia.
To quantify the opening motion of the AML edge, an angle (α) in 3-D space between the annular mid-septal-lateral diameter (line between #18 and #22) and marker #C1 (Figs. 1 and and3A)3A) at ES and maximum diastolic opening was calculated. To gain insight into the AML-motion at the strut chordal (SC) insertion points (representing the AML-belly), the angles β and γ were calculated at ES and maximum diastolic opening; β was calculated as the angle between the anterior septal-lateral diameter (line between midpoints of markers #22 and #15 and markers #18 and #17) and the anterior strut chordal marker (SCant) (Figs. 1 and and3B).3B). Similarly, angle γ was calculated as the angle between the posterior septal-lateral diameter (chord between midpoints of markers #22 and #21 and markers #18 and #19) and the posterior strut chordal marker (SCpost; Figs 1 and and3C3C).
To describe the opening motion of the middle scallop of the PML, λ was calculated as the angle formed by the mid-septal-lateral diameter (line between #18 to #22) and marker C2 (Figs. 1 and and3D)3D) at ES and maximum diastolic opening.
The excursion angles Δα, Δβ, Δγ and Δλ, were calculated as differences between the end-systolic (αES, βES, γES and λES) and maximum diastolic (αmax, βmax, γmax and λmax) values of the respective leaflet angles.
The septal-lateral mitral annular diameter was calculated as the distance between markers placed on the mid-septal (#22) and mid-lateral (#18) annulus, while the commissure-commissure annular dimension was calculated as the distance between the commissural markers #16 and #20.
The change in septal-lateral leaflet length has previously been used to describe leaflet extension during ischemia (15). For the AML, septal-lateral length was measured as the distance between the mid-septal marker #22 and marker C1; similarly, septal-lateral length for the PML was assessed as distance between mid-lateral marker #18 and marker C2.
All results were reported as mean ± SD, unless otherwise stated. The geometric variables measured before and during ischemia were compared using a two-tailed Student’s t-test for paired observations. For comparisons between different valve segments and/or time points, a one-way ANOVA with a Tukey post-hoc test was performed (Sigmastat 2.03; SPSS, Chicago, IL, USA). A p-value <0.05 was considered to be statistically significant.
The hemodynamic variables, measured before and during ischemia, are listed in Table I. The hemodynamics were consistent with acute ischemia, as reflected by an increased MR grade and left ventricular end-diastolic pressure (LVEDP), a decreased maximum left ventricular pressure (LVP), and dP/dt absolute values and a slower left ventricular relaxation rate (as reflected by a less-negative minimum dP/dt). The heart rate remained unchanged during ischemia. In the majority of the experiments, both, under baseline conditions and with acute ischemia, mitral opening was a fusion wave with no discrete E or A waves discernable using pulsed-wave Doppler echocardiography, most probably due to the fairly high heart rate observed (baseline 97 ± 14 bpm; ischemia 90 ± 15 bpm). The transmitral flow was highest during the rapid left ventricular filling phase in all animals, both at baseline and during ischemia.
The results for marker pair separation onset time, maximum velocity, and maximum distance are shown in Figure 4A. The onset of marker pair separation was not statistically different between the individual regions (A–E) under baseline, nor ischemic conditions. The mitral valve opened significantly earlier along all coaptation regions (A–E) during ischemia. At the time point of valve opening, the LVP was significantly higher with ischemia (17 ± 6 versus 32 ± 6 mmHg; p = 0.0001), suggesting a higher left atrial pressure with ischemia.
The maximum marker pair separation velocities are shown in Figure 4B. With ischemia, the maximum velocity of the mitral valve opening motion was significantly slower for all regions, except for the posterior region, which was smaller but this did not attain statistical significance (E1/E2: 13.3 ± 8.66 versus 10.6 ± 7.19 cm/ms; p = NS). Despite significant annular dilatation in both commissure-commissure and septal-lateral dimensions (Table III), the maximum marker pair distances along the middle portion of the mitral valve did not increase during ischemia (Fig. 4C).
The plotted positions of the mitral annular saddle horn (#22) and mid-lateral (#18) marker, as well as the anterior and posterior mitral leaflet edge markers pairs at ES (circles) and 102 ms after ES (triangles) under baseline conditions (closed symbols) and during acute ischemia (open symbols) in the X-Y and X-Z plane, respectively, are shown in Figures 5 and and6.6. These data show that the marker positions are almost unchanged under baseline conditions (closed symbols) for both time points, whereas the marker distances are increased with ischemia for all marker pairs at 102 ms after ES. In all coaptation regions, the wider separation was due to increased marker displacement of both the anterior and posterior leaflet markers; this suggested that an earlier opening consisted of earlier opening of both the anterior and posterior mitral leaflet in all coaptation regions.
The anterior and posterior mitral leaflet excursion data are summarized in Table II. The leaflet angle α was significantly smaller at both ES and at αmax with ischemia, resulting in an unchanged excursion angle Δα. All other leaflet angles (β, γ, and λ) did not change at ES during ischemia, but had significantly smaller maximum values (βmax, γmax, and λmax), resulting in smaller excursion angles (Δβ, Δγ, and Δλ).
Both, the anterior and posterior septal-lateral leaflet length increased significantly with ischemia during early diastole (1.99 ± 0.23 versus 2.25 ± 0.34 cm, p <0.001; and 1.32 ± 0.41 versus 1.51 ± 0.63 cm, p <0.05, respectively; Table III), which suggested that the leaflet extension was due to leaflet flattening.
The tethering lengths of the anterior and posterior PMs at early diastole are listed in Table III. The tethering length of the anterior PM remained unchanged with ischemia relative to baseline. The tethering length of the posterior PM significantly increased, suggesting early diastolic tethering via the posterior PM during ischemia.
In this ovine study, mitral leaflet opening geometry and dynamics was examined during the diastolic filling phase of the left ventricle, in different regions along the coaptation, before and during acute posterolateral IMR. Relative to the baseline, acute IMR caused an earlier opening of all valve regions and reduced the mitral leaflet opening velocity, but did not perturb the homogeneous temporal pattern of regional valve opening across the entire line of coaptation, despite regional diastolic tethering. The excursion of the PML-edge and AML-belly was also restricted, but not of the AML-edge.
The opening of both anterior and posterior mitral leaflets was significantly earlier with ischemia. Two mechanisms likely account for this phenomenon, namely:
The slower mitral valve opening velocity found during ischemia was associated with a less negative dP/dtmin (reflecting a slower left ventricular relaxation rate), decreased dP/dtmax, and elevated left ventricular end-diastolic pressures (reflecting impaired myocardial contractility). Left ventricular systolic dysfunction with an impaired left ventricular relaxation rate and higher diastolic left ventricular pressures has been linked to decreased mitral valvular opening velocity (20,21). The slower valve opening motion found in the present study most likely reflected these ischemia-related hemodynamic changes. It should be noted that the increased leaflet tension does not facilitate a more rapid leaflet opening, because the chordae exert little traction on the leaflet subsequent to the start of valve opening.
In the present study, under baseline conditions all regions of the mitral valve opened simultaneously. This was in contrast to results obtained with the isolated swine heart study by Saito et al. (17), who reported a separation of mitral valvular leaflets beginning at both sides near the edges of the mid-portion of the mitral valve, followed by the leaflet edges near the commissure and, lastly, the tips of the mitral valvular mid-portion. Two reasons may explain the discrepancy between these findings. First, the study of Saito et al. was conducted under in vitro conditions, and the mechanism of valve opening might have differed from that in the present in vivo experiments. Second, Saito et al. described the duration from the initial separation of the sides near the central tips to separation of central tips as being very short, approximately 4–16 ms. The present temporal resolution of 16.7 ms may therefore not have been sufficient to detect this difference.
Surprisingly, the proposed hypothesis could not be confirmed, as no difference was found in the regional sequence of mitral valve opening during ischemia, despite regional diastolic tethering. This suggests that traction via the subvalvular apparatus has a minor impact on opening velocity during early leaflet separation. This has already been suggested by Dent et al. (19), who transected both papillary muscles in a dog model of left ventricular dysfunction and found that the excursion of the mitral leaflets was similar to that in animals with a preserved submitral apparatus. Dent and colleagues concluded that the submitral apparatus had a minor effect on alterations in mitral valvular motion, and hypothesized that the mitral valvular opening motion was mainly a passive motion driven by the blood flow. This hypothesis might explain the present finding of a preserved homogeneous temporal pattern of regional valve opening across the entire line of coaptation, despite regional diastolic tethering.
The diastolic maximum marker pair distances along the medial portion of the mitral valve (Fig. 4C) did not change with ischemia, despite significant annular septal-lateral dilatation. Two mechanisms may account for this finding. First, although diastolic leaflet tethering does not seem to play a major role during initial leaflet separation (see section ‘Valve opening’), the chordae tendineae might tether during ventricular filling and limit maximum valve separation. Second, an impaired left ventricular relaxation has been shown to limit the extent of mitral leaflet opening (19). Both, diastolic valve tethering effects and an impaired left ventricular relaxation, may therefore prevent maximum leaflet separation during ischemia.
In the present study, no significant change was found in the excursion angle Δα (representing excursion of the AML edge) with ischemia. This finding contrasted with that of Otsuji et al. (4), who reported a significantly decreased AML-excursion in patients with incomplete mitral leaflet coaptation and left ventricular dysfunction. However, since in that study the AML-excursion angle Δα represented the excursion of the AML-belly, such a finding may be consistent with the present finding of an ischemia-related decrease in the excursion angles Δγ and Δλ. The finding of an unchanged excursion angle of the AML-edge and decreased excursion angles of AML-belly and PML-edge may either be a secondary effect reflecting alterations in left ventricular inflow patterns that differently affect the AML-edge and PML-edge, or it may be a primary result of chordal tethering.
When using echocardiography, the anterior mitral leaflet excursion has been described as: (i) the maximum distance from E-point to zero (20,22) using M-mode; (ii) the E-point septal separation (19,23) using M-mode; and (iii) the leaflet belly excursion by assessing the angle over which the base of the leaflets moved from systole to its fully open diastolic position (measured as the excursion of a tangent line through the base of the AML in the four-chamber view) (4). If these three different methods had been applied to the present study, the assessment of AML-excursion might have produced different results. By using method (i), no change would have been found in diastolic AML-excursion (here represented by the excursion angle Δα; see Table II). With method (ii), a decreased diastolic AML-excursion (most closely represented by the decrease found in αmax; see Table II) would have been observed. Finally, the use of method (iii) would have resulted in finding a decreased diastolic AML-excursion (represented by the decrease found in excursion angles Δγ and Δλ; see Table II). These different interpretations imply that caution should be exercised when mitral leaflet excursion is assessed, due to the variety of available measures. The evaluation of additional parameters might be needed to account for changes in different anatomical leaflet sites. With an improvement of temporal resolution, real-time three-dimensional echocardiography might represent a useful clinical tool in the evaluation of the temporal sequence and regional characteristics of mitral leaflet opening.
The primary limitation was that the immediate effects of acute ischemic MR in open-chest sheep who previously had normal hearts were investigated; hence, caution should be exercised when extrapolating these findings to the clinical scenario of chronic IMR. A second limitation was that the left atrial pressures were not directly measured., and consequently the transmitral gradients could not be quantified. Third, the left ventricular pressures were not identical between groups, and the left ventricular volumes were not measured. As a result, the effects of differences in loading conditions between groups could not be accounted for. Finally, as all animals developed significant MR and severe left ventricular systolic dysfunction it was not possible to distinguish whether the effects observed were related to MR, to ischemia, or both.
In conclusion, acute posterolateral left ventricular ischemia: (i) causes earlier leaflet opening, most probably due to MR-related elevation in left atrial pressures; (ii) reduces leaflet opening velocity, reflecting impaired left ventricular relaxation; (iii) does not perturb the homogeneous temporal pattern of regional valve opening across the entire valve coaptation, despite diastolic tethering via the posterior PM, suggesting that traction via the subvalvular apparatus plays a minor role during early leaflet separation; and (iv) restricts excursion of the AML-belly and the PML-edge, but does not change excursion of the AML-edge. These observations should be considered for the diagnostic evaluation of mitral valve geometry and dynamics during acute myocardial ischemia. Future studies will confirm whether these findings are apparent in patients with chronic IMR, and may help to refine the present strategies used to treat IMR.
The authors gratefully acknowledge the expert technical assistance of Carol W. Mead BA. The studies were supported by Grants HL-29589 and HL-67025 from the National Heart, Lung and Blood Institute. Dr. Bothe was supported by the Deutsche Herzstiftung, Frankfurt, Germany, Dr. Carlhäll by the Swedish Heart and Lung Foundation and the County Council of Östergötland, Sweden, and Dr. Ennis by NHLBI Pathway to Independence K99-R00 HL-087614
Presented at the Fourth Biennial Meeting of the Society for Heart Valve Disease, 15th–18th June 2007, New York, USA