In six sheep, radiopaque markers were placed on the left ventricle (LV), the mitral annulus, the left atrium (LA), and the central edge of both mitral leaflets to investigate the effects of acute LV ischemia on atrial contraction, mitral annular area (MAA), and mitral regurgitation (MR). Animals were studied with biplane videofluoroscopy and transesophageal echocardiography before and during balloon occlusion of the left anterior descending (LAD), distal circumflex (dLCX), and proximal circumflex (pLCX) coronary arteries. MAA and LA area were calculated from the corresponding markers. LAD occlusion did not alter LA area reduction or presystolic MAA reduction, whereas dLCX occlusion resulted in a mild decrease in the former with no change in the latter. Neither occlusion resulted in MR. pLCX occlusion, however, significantly decreased LA area and presystolic MAA reduction and resulted in increased end-diastolic MAA, delayed valve closure from end diastole, and MR. Decreased atrial contractile function, as observed during acute posterolateral ischemia, is linked to diminished presystolic mitral annular reduction, a larger mitral annular size at end diastole, and MR.
- mitral annulus
- left ventricular ischemia
- mitral regurgitation
the normal mitral annulus is a dynamic structure that undergoes area changes throughout the cardiac cycle, reaching a maximum in diastole and a minimum in systole, thus facilitating both left ventricular (LV) filling and competent valve closure (29). Recent ovine experiments showed that 89% of this area reduction occurs before ventricular systole (8), suggesting an atrial influence on annular dynamics. Presystolic annular area reduction has also been reported in human subjects (19) and has been shown in experimental animals to be dependent on the strength and duration of atrial contraction (27, 29). Furthermore, the absence of properly timed atrial contraction, as during ventricular pacing (8) or atrial fibrillation (22), abolishes presystolic annular area reduction. Whether altered atrial contraction affects valvular competence is unclear, but some experimental evidence suggests that left atrial dimensions affect mitral annular size (23) and that valve closure may be compromised in the absence of appropriate atrial systole (4, 13, 25). Although more recent studies of mitral valve closure and atrial fibrillation have suggested the importance of atrial contraction in competent valve closure (22), older data do not concur (2). Differences could arise from subject heterogeneity, associated ventricular and valvular pathology, and different methods of data acquisition.
Acute posterolateral ischemia in sheep has been shown to result in mitral regurgitation and altered mitral annular dynamics (7), but the effect of ischemia on atrial contraction has not been examined. With the myocardial marker technique, we have chosen to use acute ischemia to investigate its effect on atrial size reduction, annular dynamics, and valvular competence during sequential, isolated acute occlusion of the left anterior descending coronary artery (LAD), distal circumflex coronary artery (dLCX), or proximal left circumflex coronary artery (pLCX) in sheep.
Six castrated adult male sheep were used in the study. As described previously (8), a left thoracotomy was performed, and eight tantalum myocardial markers (inner diameter 0.8 mm, outer diameter 1.3 mm, length 1.5–3.0 mm) were inserted beneath the LV epicardial surface along four equally spaced longitudinal meridians as shown in Fig.1. After establishment of cardiopulmonary bypass and with the heart arrested, eight tantalum markers were sutured around the circumference of the mitral annulus (1 near each commissure and 3 along the anterior and posterior annulus; Fig. 1). Miniature gold markers were sutured at the central edge of each mitral leaflet. Four additional markers were sewn to the endocardial surface of the left atrium at midchamber level, corresponding to the annular septal, lateral, anterior commissure and posterior commissure markers (Fig. 1). After completion of marker placement, the animal was rewarmed, the atriotomy was closed, and the cross-clamp was removed. After resuscitation, the animal was weaned from bypass. A micromanometer pressure transducer (PA4.5-X6; Konigsberg Instruments, Pasadena, CA) was placed in the LV chamber through the apex.
After 7 ± 1 (mean ± SD) days, each animal was taken to the experimental cardiac catheterization laboratory, sedated with ketamine (1–4 mg · kg−1 · h−1 iv infusion) and diazepam (5 mg iv bolus as needed), intubated, and mechanically ventilated (veterinary anesthesia ventilator 2000; Halowell EMC). Esmolol (20–50 μg · kg−1 · min−1) intravenous infusion was used to minimize reflex sympathetic responses. The coronary anatomy of each animal was identified by contrast angiography (Fig. 2) before vessel occlusion. Regional LV ischemia was induced by coronary artery balloon occlusion (1–4 min), as described previously (7), performed at three locations in sequential fashion: proximal LAD (proximal to first diagonal), dLCX (distal to second obtuse marginal), and pLCX (proximal to first obtuse marginal). This sequence of coronary occlusion was chosen so as to facilitate animal survival through all three ischemic intervention because pLCX occlusion has the highest rate of inducing malignant arrhythmias in sheep. Although some coronary artery branches to the left atrium are present proximal to the takeoff of the first obtuse marginal branch, during pLCX occlusion the most distal portion of the balloon was positioned just proximal to the takeoff of the first obtuse marginal branch, and because the balloon was 2 cm in length, the 2-cm segment of the LCX proximal to the first obtuse marginal was also occluded. Thus most proximal atrial branches were occluded during pLCX occlusion by the proximal portion of the balloon. There was no appreciable difference in the PR interval between the ischemic interventions.
Videofluoroscopic marker data recordings were acquired before and during ischemia at each location, and simultaneous transesophageal color Doppler echocardiography was performed to assess mitral regurgitation. Respiration was arrested during all data acquisitions (15–30 s). After each data acquisition run, a 3- to 5-min period of stabilization was allowed before the next occlusion. Degree of mitral regurgitation was graded by an experienced echocardiographer on the basis of the width and extent of the regurgitant jet into the left atrium as none (0), mild (+1), moderate (+2), moderate-severe (+3), and severe (+4).
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 the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH Publication No. 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. At the end of the experiment the animals were euthanized with an overdose of thiopental sodium and potassium chloride.
Images were acquired with the animal in the right lateral decubitus position with a Philips Optimus 2000 biplane Lateral ARC 2/Poly DIAGNOST C2 system (Philips Medical Systems North America, Pleasanton, CA) with the image intensifier in the 9-in. fluoroscopic mode. Data from two radiographic views were digitized (18) and merged to yield three-dimensional coordinates for each of the radiopaque markers every 16.7 s with custom-designed software (5). Ascending aortic pressure, LV pressure, and ECG voltage signals were digitized and recorded simultaneously during data acquisition.
Hemodynamic and cardiac cycle timing markers.
Two to three consecutive steady-state beats in sinus rhythm immediately before and during acute regional LV ischemia (LAD, dLCX, pLCX) were averaged and defined as “preischemia” and “ischemia” data for each animal, respectively. For each cardiac cycle, end systole was defined as the videofluoroscopic frame containing the point of peak rate of LV pressure fall (−dP/dt), and end diastole was defined as the videofluoroscopic frame containing the peak of the ECG R wave. Instantaneous LV volume was calculated from the epicardial LV markers with the use of a space-filling multiple tetrahedral volume method for each frame, i.e., every 16.7 ms. Although epicardial LV volume calculated in this manner overestimates true chamber volume, the change in LV volume is an accurate measurement of the change in chamber volume (standard error of estimate < 1 ml) (17). Thus stroke volume and change in end-diastolic volume or end-systolic volume are accurately calculated from the change in epicardial LV volume, but the ejection fraction is substantially underestimated.
Left atrial, mitral leaflet, mitral annular, and LV dynamics.
To identify mitral valve closure objectively, the distance between leaflet edge markers throughout the cardiac cycle for each animal was represented by a continuous function generated by spline-fitting computer software (S-Plus; Mathsoft, Cambridge, MA). Valve closure was defined as the time after end diastole at which the first time derivative of the associated spline function decreased below 0.002 (a value reflecting the beginning of a plateau near the minimum of the leaflet edge distance). Mitral annular area and atrial cross-sectional area were determined by summation of triangular slices from the centroid of the annulus to the eight annular markers and from the centroid of the atrium to the four atrial markers, respectively. Left atrial cross-sectional area change and mitral annular area reduction were calculated as the percent change between maximum area in late diastole [time window of 10 frames (167 ms) before end diastole] and minimum area in early systole. “Presystolic” mitral annular area contraction was defined as the percentage of total area reduction occurring between late diastolic maximum and end diastole. Left atrial and annular septal-lateral and commissure-commissure dimensions were calculated as a distance in three-dimensional space between the markers tagging these locations (Fig. 1) on the atrium and annulus, respectively. LV septal-lateral and commissure-commissure dimensions were calculated as the distance between the respective epicardial markers at the equatorial level (Fig. 1).
All data are reported as means ± SD. Hemodynamic and marker-derived data from two to three consecutive steady-state beats were time aligned at end diastole, and data from these beats were averaged for each animal. The data were analyzed over 20 frames before and 20 frames after end diastole, thereby allowing evaluation of the studied variables over the entire cardiac cycle. Data were compared with Student's t-test for paired observations with a significance level set at P < 0.05.
The mean weight of the animals used in the study was 71 ± 13 kg. Cardiopulmonary bypass time was 81 ± 6 min, and aortic cross-clamp time was 61 ± 3 min. Proper myocardial marker position was confirmed in all animals on postmortem examination of excised hearts. Hemodynamic conditions before and after each artery occlusion are summarized in Table 1. During all ischemic interventions, stroke volume, LV dP/dt, and peak LV pressure decreased, whereas end-systolic volume increased significantly, with similar change in these parameters relative to preischemic controls seen across ischemic interventions. Thus a comparable insult to LV systolic function was induced by LAD, dLCX, and pLCX occlusion. During balloon occlusion of the coronary arteries, the aim of the occlusion was to reduce ventricular pressure by ∼30%. The range of occlusion times varied between 1 and 4 min, with dLCX at the higher end of the spectrum and pLCX at the lower end. It is likely that the longer occlusion times in the dLCX group were responsible for equivalent ischemic insults with distal and proximal LCX ischemia.
Mitral Annular Dynamics
Group mean left atrial area, mitral annular area, and leaflet separation throughout the cardiac cycle before and after each ischemic intervention are shown in Fig.3. Data summarizing atrial and annular dynamics and valvular competence are presented in Table2. Mitral regurgitation during pLCX ischemia was broadly central and holosystolic. Late diastolic left atrial area reduction did not change with LAD ischemia, decreased somewhat with dLCX ischemia, and was reduced substantially during pLCX ischemia (Table 2 and Fig. 3). This significant compromise in presystolic atrial area reduction was associated with decreased presystolic mitral annular area reduction and an 18.9 ± 9.9% increase in end-diastolic mitral annular area during pLCX ischemia. With dLCX occlusion, annular area at end diastole was 7.3 ± 4.5% larger than in the preischemic control, whereas during LAD ischemia mitral annular area increased by 4.4 ± 8.7%, which was not statistically significant. Left atrial area at end diastole increased in similar fashion to annular area with a 21.8 ± 10.6% (P = 0.005), 8.5 ± 5.3% (P = 0.02), and 3.1 ± 7.8% [P = not significant (NS)] increase observed during pLCX, dLCX, and LAD occlusion, respectively. The distance between the anterior leaflet edge marker and posterior leaflet edge marker in early systole [assessed 33.4 ms after end diastole to study “leaflet loitering” (7)] was significantly increased during pLCX occlusion (0.52 ± 0.22 vs. 1.24 ± 0.44 cm before and during occlusion; P = 0.03) but not with LAD (0.84 ± 0.30 vs. 0.91 ± 0.31 cm; P = 0.8) or dLCX (0.52 ± 0.22 vs. 0.73 ± 0.39 cm; P= 0.2) ischemia (Fig. 3). The early systolic distance between anterior and posterior leaflets seen in pLCX contributed to delayed valve closure (Table 2) and was accompanied by significant mitral regurgitation. These perturbations were not observed in either LAD or dLCX ischemia.
Left atrial enlargement at end diastole during ischemia was primarily in the septal-lateral dimension as shown in Table3 and Fig.4. The atrial septal-lateral dimension increased by 3.2 ± 5.3% (P = NS), 6.5 ± 5.0% (P = 0.02), and 16.0 ± 8.0% (P = 0.004), whereas the commissure-commissure (anterior-posterior) dimension increased by 0.1 ± 2.5% (P = NS), 1.7 ± 1.5% (P = 0.03), and 5.1 ± 3.5% (P = 0.01) during LAD, dLCX, and pLCX occlusion, respectively. Mitral annular dilatation was also greatest in the septal-lateral dimension (Table 3), and annular area increase during ischemia resulted primarily from dilatation of the septal-lateral annular diameter, orthogonal to the line of leaflet coaptation, throughout the cardiac cycle as shown in Fig. 4. LV septal-lateral and commissure-commissure (i.e., anterior-posterior) dimensions at end diastole were unchanged with LAD and dLCX ischemia (Table 3, Fig. 4). These dimensions increased only modestly with pLCX, and their increases were of similar magnitude (Table 3). Systolic LV contraction was most effectively reduced by pLCX. Annular septal-lateral diameter change during the cardiac cycle under all conditions closely reflected the dynamic changes in mitral annular area (Figs. 3 and 4). With ischemia, the annular septal-lateral diameter increase was much more substantial during pLCX occlusion than with LAD or dLCX occlusion, particularly at end diastole (increase of 17.5 ± 8.8%, 6.9 ± 4.0%, and 5.5 ± 6.5% for pLCX, dLCX, and LAD occlusion, respectively;P = 0.01 by ANOVA) and during early systole. Thus dysfunctional atrial area reduction observed during pLCX ischemia was accompanied by sizable increase in the atrial and annular septal-lateral dimension at end diastole near the time of delayed valve closure.
Acute posterolateral ischemia has been shown to alter mitral annular dynamics and result in mitral regurgitation in sheep (7, 9), but its effect on atrial contraction has not been investigated. The current study revealed that ischemia produced by pLCX occlusion resulted in increased end-diastolic left atrial area and septal-lateral dimension and decreased atrial contraction accompanied by reduced presystolic mitral annular area reduction, a larger end-diastolic mitral annular septal-lateral size, and mitral regurgitation. Although LAD and dLCX ischemia apparently induced an insult to LV systolic function equal to that of pLCX occlusion, these ischemic interventions did not significantly alter presystolic mitral annular dynamics.
The differential effects of the three ischemic interventions on late diastolic atrial area reduction most likely resulted from the location of the occlusion relative to the atrial blood supply. Angiography revealed that atrial coronary vessels arise from the LCX in these hearts, particularly its proximal segment (Fig. 2), findings consistent with previous anatomic descriptions of ovine (15), goat (14), and human (11) atrial blood supply. It is therefore not surprising that LAD occlusion did not influence atrial contraction. dLCX occlusion only mildly decreased atrial contraction, most likely by occluding only the few atrial vessels that arise from the LCX beyond the second obtuse marginal branch and supply the posterior left atrium. On the other hand, pLCX ischemia would be expected to produce most severe alterations of atrial contraction as indeed observed in the current study.
Alterations in the timing of atrial contraction in sheep have been shown to affect mitral annular dynamics, suggesting an “atriogenic” influence on annular physiology (8, 27). An anatomic basis for this relationship exists because atrial myocardial fibers have been demonstrated to insert into the mitral annulus, especially in the more dynamic lateral region of the annulus (1). Furthermore, the role of atrial systole in the mitral valve has been suggested from pioneering experiments almost four decades ago (21), although its importance has been debated (31). The current experimental series supports such an atriogenic influence because increased atrial size and diminished atrial contraction were accompanied by increased end-diastolic mitral annular size and decreased presystolic mitral annular area reduction.
Increased end-diastolic mitral annular area during acute posterolateral ischemia has been associated with delayed early systolic leaflet coaptation or “loitering” and mitral regurgitation in previous ovine studies (7). Reduction of annular area and septal-lateral diameter with either a rigid or a flexible complete ring abolished both leaflet loitering and mitral insufficiency (24). Delayed valve closure in our study was also accompanied by annular area and septal-lateral diameter dilatation, perhaps partially due to a decrease in presystolic mitral annular reduction. Although delayed valve closure in early systole was associated with mitral regurgitation, mitral regurgitation was holosystolic and changes in valvular and subvalvular geometry in mid and late systole were also associated with and may have contributed to the genesis of mitral insufficiency. Whether the decrease in presystolic annular reduction contributed significantly to mitral regurgitation is speculative, but prior experiments suggest that a weak and/or mistimed atrial systole may affect valvular competence (3,4, 13, 25). Conversely, evidence for nonregurgitant mitral valve closure without proper atrial contraction also exists (28,31), implying that atrial contraction may not always be necessary for competent valve closure under all physiological conditions.
It is noteworthy that during pLCX ischemia both atrial and annular dilatation were predominantly in the septal-lateral dimension. Interestingly, such coupling was not observed between the respective LV and annular dimension. These findings strongly suggest that atrial rather than ventricular dilatation exerts the predominant effect on mitral annular dimensions in late diastole during acute ischemia. Under normal conditions, annular septal-lateral diameter reduction before systole acts to “preposition” the leaflets for closure before the rise in LV pressure in early systole. Decreased reduction in this dimension, as seen during pLCX occlusion, would delay leaflet coaptation simply by increasing the distance the leaflets must travel to coapt. In an ovine model of tachycardia-induced cardiomyopathy, a 25% increase in end-diastolic septal-lateral diameter, similar to the increase observed in the current study, was proposed to be the underlying mechanism of mitral regurgitation (26). Furthermore, atrial systole gives rise to late diastolic ventricular vortices that form below the mitral leaflets (6, 12) and may aid in valve closure (10). Reduced atrial systole could negatively influence these normal flow patterns. Thus dysfunctional atrial systole and annular reduction in late diastole could contribute to mitral insufficiency seen during acute posterolateral ischemia.
The above experimental findings suggest that atrial contraction is closely coupled to presystolic mitral annular area reduction. Decreased atrial contractile function, as observed during acute posterolateral ischemia, was associated with diminished presystolic mitral annular reduction and a larger mitral annular size at end diastole, which may in turn affect valvular competence.
The myocardial marker method requires suturing miniature tantalum markers to the structures of interest, and the collective mass of the markers may alter normal valve mechanics; however, the markers used in this study were small (4–8 mg) and biologically inert and were shown in our previous studies not to effect leaflet edge opening velocities derived via echocardiography. Therefore, we do not expect that the implanted markers affect the normal dynamics of the mitral leaflets, annulus, left atrium, or LV. Conversely, radiopaque marker technology offers highly accurate data with resolution of <0.1 mm (18). The sequence of coronary artery occlusions was not randomized because of practical considerations as mentioned above, and therefore the cumulative effect of ischemia during the last coronary occlusion (pLCX) cannot be excluded completely. However, the animals were allowed to return to baseline hemodynamics after each ischemic insult before subsequent coronary occlusion. As the baseline hemodynamics did not differ between the three interventions, a cumulative effect of ischemia with each coronary artery occlusion seems less likely. Other limitations of this study are inherent in the use of any animal model. Some anatomic differences exist between sheep and human mitral annulus (30), and therefore the atrial contribution to the structure and dynamics of the mitral annulus in sheep may not be the same as in humans, although considerable variability in the amount of atrial muscle fibers within the human annulus has been reported (1). However, because significant presystolic annular contraction has been reported in both sheep (8) and humans (19), interspecies differences may actually be unimportant. Furthermore, several reliable models of human cardiac physiology and pathophysiology have been established with the ovine model (15, 16, 20).
We appreciate the superb technical assistance provided by Mary K. Zasio, Carol W. Mead, and Maggie Brophy.
This work was supported in part by National Heart, Lung, and Blood Institute (NHLBI) Grant HL-29589. T. A. Timek, P. Dagum, F. Tibayan, and D. T. Lai are Carl and Leah McConnell Cardiovascular Surgical Research Fellows. T. A. Timek is a recipient of The Thoracic Surgery Foundation Research Fellowship Award. T. A. Timek and P. Dagum were also supported by NHLBI Individual Research Service Awards HL-10452–01 and HL-09569, respectively. D. T. Lai was supported by a fellowship from the American Heart Association, Western States Affiliate.
Address for reprint requests and other correspondence: N. B. Ingels, Jr., Laboratory of Cardiovascular Physiology and Biophysics, Research Institute, Palo Alto Medical Foundation, 795 El Camino Real (AMES Bldg.), Palo Alto, CA 94301 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
June 20, 2002;10.1152/ajpheart.00149.2002
- Copyright © 2002 the American Physiological Society