Interregional Differences in the Systolic and Diastolic Response of Nonischemic Myocardium to Remote Coronary Occlusion 

The study conforms to the United Kingdom Animals Act (Scientific Procedures, 1986).

The experimental preparation is described in detail elsewhere. 3,4Briefly, a thoracotomy was performed during halothane anesthesia in 14 beagles of either sex, and the heart was exposed in a pericardial cradle. Instrumentation included a tube mounted on an umbilical tape around the inferior cava vein proximal to the right atrium for caval occlusions; micromanometer-tipped 8-French catheters in the aortic arch and in the left ventricle for measurement of systemic and ventricular pressures; and a Doppler flow probe placed around the proximal ascending aorta. For the measurement of segmental dimensions, pairs of piezoelectric crystals (5 MHz, 2-mm outer diameter) were implanted in the subendocardium approximately 10 mm apart via  small stab wounds. Crystals were implanted in myocardium supplied by the LAD and LCX coronary arteries.

In the animals of group A (LAD ischemia, n = 7) we investigated the effects of myocardial ischemia in an apical area of the anterior wall on remote nonischemic myocardium located in the posterior wall and the basal region of the anterior wall. Moreover, in the basal region, we assessed the response to ischemia in both the short-axis plane and the long-axis plane. A snare was placed around the LAD coronary artery distal to the first diagonal branch, and piezoelectric crystals were implanted in the short-axis plane in the near-apical region of the anterior wall (ischemic area); in the short-axis plane of the posterior wall, midway between the apex and the base; in the short-axis plane of basal myocardium of the anterior wall; and in the long-axis plane of basal myocardium of the anterior wall (fig. 1, left).

In the animals of group B (LCX ischemia, n = 7) we investigated the effects of myocardial ischemia of the posterior wall on remote nonischemic myocardium located in the apicoanterior region of the anterior wall and the basal region of the anterior wall. A snare was placed around the LCX coronary artery distal to the first marginal branch, and piezoelectric crystals were implanted in the short-axis plane of the posterior wall, midway between the apex and the base (ischemic area); in the short-axis plane in the near-apical region of the anterior wall; and in the short-axis plane of basal myocardium of the anterior wall (fig. 1, right).

After instrumentation, halothane was discontinued and fentanyl (100 μg/kg then 2μg · kg⁻¹· min⁻¹) was administered intravenously. Mechanical ventilation (40% oxygen in nitrogen) was adjusted to maintain the end-tidal carbon dioxide concentration at 34–41 mmHg. The preparation was allowed to stabilize for 1 h before the experiment was started. Mid esophageal temperature was continuously measured and maintained at 36–37°C using a servocontrolled heating element incorporated in the operating table.

After completion of baseline recordings, a caval occlusion was performed by sliding down the tube on the tape around the cava vein. A period of 5 min was allowed for hemodynamics to recover after the caval occlusion. The snare around the coronary artery (LAD in group A, LCX in group B) was then abruptly tightened to achieve complete coronary occlusion, and hemodynamic recordings, including caval occlusion, were repeated after 5 min of occlusion.

Hemodynamic signals were recorded during apnea at a sampling frequency of 500 Hz and stored on hard disk. The left ventricular dP/dt was obtained by electronic differentiation and the time constant of isovolumic relaxation (τ) was obtained as previously described. 13Length measures were normalized to an initial end-diastolic length of 10 mm. The following definitions were used to quantify regional wall movement: end-diastolic length (EDL), defined as length at the first upward (positive) deflection of the left ventricular dP/dt signal; length at the end of the isovolumic contraction (IVL), defined as length at the first upslope of the aortic flow signal; end-systolic length (ESL), defined as length at the first return to zero of the aortic flow signal; minimum diastolic length after ESL (L^min); maximum length during systole (L^max).

Total segmental shortening (TS) was defined as EDL − L^min(in millimeters). Shortening during the isovolumic period was defined as EDL − IVL and expressed as a percentage of total segmental shortening. Ejection shortening was defined as IVL − ESL and expressed as a percentage of total segmental shortening. Postsystolic shortening was defined as ESL − L^minand expressed as a percentage of total segmental shortening. Thus, isovolumic shortening, ejection shortening, and postsystolic shortening add up to 100%. Paradoxic systolic bulging was defined as L^max− EDL and expressed as a percentage of EDL. Peak lengthening rates (PLRs; dL/dt^max) were obtained by electronic differentiation. Regional contraction time was defined as the interval between the preceding R wave of the electrocardiogram and L^min.

The regional preload recruitable stroke work (SW) relation was calculated as previously described. 14Briefly, segmental SW was obtained by electronic integration of the area of the segmental pressure-length loop. SW of consecutive beats during a caval occlusion was plotted against the corresponding EDL. Linear least-squares regression analysis was performed to calculate the slope (M^W) and length intercept (L^W) of the SW versus  EDL relation:

For each caval occlusion the regional chamber stiffness constant (k) was obtained by fitting the end-diastolic pressure-length data to equation 2(A is a curve-fitting constant). High correlation coefficients (r > 0.95) were found in all animals.

To minimize the potentially confounding effects of noncontractile components (e.g. , connective tissue), the comparison of stiffness constants between baseline and occlusion was performed over the same range of end-diastolic pressures. Thus, during the ischemic state all data above the end-diastolic pressure of the baseline state were discarded.

Values are the mean ± SD. Means were compared by paired Student t  test. The response of the different nonischemic regions to myocardial ischemia in group A was compared by two-way analysis of variance followed by paired Student t  tests using Bonferroni correction. A P  value less than 0.05 was considered to represent statistical significance.

A 5-min complete LAD occlusion was associated with an increase in end-diastolic pressure and τ whiles decreases in peak positive dP/dt, peak negative dP/dt, and arterial blood pressure occurred (table 1). Note that heart rate remained unaffected. In the ischemic area paradoxic systolic bulging occurred, and the hemodynamically effective shortening during the ejection period was replaced by ineffective postsystolic shortening (table 2). Consequently, no effective SW was performed and the preload recruitable SW (PRSW) relation (baseline: M^W94 ± 24, L^W6.7 ± 1.6) could not be calculated. This severe ischemia-induced systolic dysfunction was associated with an increased stiffness and a decreased PLR.

In the absence of ischemia, chamber stiffness was significantly greater in the long-axis plane of the base than in both the short-axis plane of the base and the posterior wall (table 3). In all three nonischemic areas, remote ischemia was associated with an increase in isovolumic shortening, end-diastolic length, length intercept of the PRSW, and regional chamber stiffness. Stiffening was significantly less pronounced in the posterior wall than at the base (figs. 2 and 3). Moreover, by contrast to the basal area, LAD ischemia was associated with an increase in PLR, an increase in total shortening, and steepening of the slope of the PRSW relation in the posterior wall (fig. 3). This increase in contractility however, resulted only in an increase in isovolumic shortening while ejection shortening decreased in relative terms and remained unchanged in absolute terms (table 3).

The consequences of 5 min of LCX occlusion on global hemodynamics (table 4) and regional function in the ischemic area (table 5) were very similar to that of LAD occlusion described previously.

During baseline, a greater PLR, a lower stiffness, and higher indices of contractility were found at the apex compared to the base (table 6). In both the apical and basal areas of the remote nonischemic anterior wall, LCX coronary occlusion was associated with an increase in chamber stiffness (figs. 3 and 4). Note that this increase in stiffness was significantly more pronounced (P = 0.005) in the basal compared to the apical area (fig. 3). By contrast to the basal area, the apical area exhibited a compensatory increase in contractile function and PLR (table 6, figs. 3 and 5).

To assess the effect of remote ischemia on stiffness in the thickness plane, we reanalyzed data from a previous study from our laboratory. 4In that study, we implanted sonomicrometers at the base in both the short-axis and the thickness plane. Although stiffness constants were derived from the length measures, thickness measures were only used to calculate wall stress. We reanalyzed data from 11 animals that were subjected to a baseline stage and 45 min of severe myocardial ischemia of the anterior wall (LAD supply) during fentanyl anesthesia. In response to remote ischemia, chamber stiffness at the base in the short axis increased from 0.63 ± 0.27 mm⁻¹to 0.81 ± 0.36 mm⁻¹(P < 0.05), whereas, in the thickness plane, chamber stiffness increased from 1.04 ± 0.31 mm⁻¹to 1.43 ± 0.48 mm⁻¹(P < 0.05).

Coronary occlusion results in almost immediate contractile dysfunction of the ischemic area that, in turn, no longer contributes to the mechanical function of the heart as a pump. In these circumstances, the nonischemic myocardium becomes critical for the maintenance of global cardiac function. Not surprisingly, a substantial number of laboratory and clinical studies focused on the contractile response of nonischemic myocardium to remote myocardial ischemia. By contrast, there are only limited data regarding diastolic function in remote nonischemic myocardium.

In a previous investigation, we were able to demonstrate that, during halothane anesthesia (0.8% v/v), myocardial ischemia of the anterior wall is associated with an increased stiffness of remote nonischemic myocardium at the base in dogs with open chests. 1Subsequent work from our laboratory confirmed this finding during isoflurane (1.0% v/v) and fentanyl anesthesia. 3,4In conscious, long-term instrumented dogs, Pagel et al.  2found an increase in stiffness of remote nonischemic myocardium located in an area supplied by the LCX coronary artery during brief (2 min) occlusions of the LAD coronary artery. It is noteworthy that, in all of the aforementioned investigations, the nonischemic area was located at the base, whereas the ischemic area was located in the apical area of the left anterior wall.

In the absence of myocardial ischemia, the base is stiffer than the apex. This may relate to structural differences in myocytes, geometric arrangement of muscle fibers, or tethering of the base to the mitral valve apparatus. All these factors may limit the ability of the base to stretch passively and may cause further stiffening in the settings of altered ventricular loading conditions brought about by myocardial ischemia. The current study shows that stiffening of the nonischemic myocardium occurred regardless of the localization of the ischemic and the nonischemic area. In addition, at the base, stiffening was not restricted to the short-axis plane but could be observed in the long-axis plane also. Moreover, reanalyzing data from our previous study revealed that stiffening at the base occurs also in the thickness plane. Therefore, stiffening of nonischemic myocardium in response to remote ischemia appears to occur independently of the localization of the ischemic area, the localization of the nonischemic area, and the plane used to assess regional function.

What are the potential implications of our findings? Pagel et al.  2observed that 1.1 and 1.6 minimum alveolar concentration desflurane and 1.6 minimum alveolar concentration isoflurane attenuated the increase in stiffness of nonischemic myocardium, but also attenuated the ischemia-induced increase in left ventricular end-diastolic pressure. Because we observed that an increase in stiffness occurs in all parts of the nonischemic area, we suggest that stiffening of remote nonischemic myocardium is a prerequisite of the ischemia-induced increase in left ventricular end-diastolic pressure.

The majority of investigations that assessed the systolic response of nonischemic response found hyperkinesia (i.e. , an enhanced contractile function). 2,11,12,15,16However, without apparent reason, several studies, including studies from our laboratory, were unable to show hyperkinesia in remote nonischemic myocardium. 1,3,4,17–21Including the current investigation, we are aware of only three studies 2,4that assessed contractility of remote nonischemic myocardium using the PRSW, an index of contractility independent of loading conditions. In these studies, changes in the slope of the PRSW paralleled changes in shortening fraction. Thus, load dependence is an unlikely explanation for the absence of hyperkinesia in remote nonischemic myocardium in some studies, whereas it is present in most.

In nonischemic areas exhibiting an increase in contractility, we did not observe an increase in hemodynamically effective ejection shortening in nonischemic myocardium. Instead, isovolumic shortening increased, leading to the paradoxic outward movement (bulging) of the ischemic area. Noma et al.  15demonstrated that, at low levels of preload, hyperkinesia in nonischemic myocardium is predominantly caused by an increase in isovolumic shortening, whereas, at higher levels of preload, hyperkinesia reflects an increase in ejection shortening. A simple explanation for this observation may be that a high level of preload stretches the ischemic area already at end-diastole, so additional stretching forces, imposed by isovolumic contraction of nonischemic myocardium, can have only a limited effect. In keeping with this hypothesis is the observation that the extent of systolic bulge is inversely related to preload. 22,23 

Divergent findings were reported concerning PLRs in remote nonischemic myocardium: although some investigators reported an increase in PLR, 2,20others observed an unchanged or even reduced PLR. 3,4,20,24,25In the current study, an increased PLR occurred in nonischemic segments that exhibited an increased contractility. By contrast, an unchanged PLR occurred in regions with unchanged contractility. This pattern is consistent with findings of in vitro  and in vivo  studies that have shown a close correlation of the extent of systolic shortening and the PLR. 26,27 

In the absence of myocardial ischemia, apical and basal myocardium differ in contractility, PLR, and stiffness. 7–10The current investigation shows that during conditions of remote ischemia, contractility increases predominantly in the already more-active apical myocardium, whereas stiffness increases predominantly in the less-compliant basal region. Thus, ischemia of the posterior wall increases preexisting regional differences in mechanical function in the anterior wall. In the absence of ischemia, mechanical function did not differ between myocardium from the posterior wall and from the base. However, ischemia of the anterior wall led to interregional differences in contractility, peak lengthening, and regional chamber stiffness. Therefore, it appears that LAD and LCX coronary artery occlusion both induce mechanical heterogeneity, not only between ischemic and nonischemic segments, but also within nonischemic segments.

We studied anesthetized and animals with open chests, but our findings concord with previous studies performed in conscious animals, particularly concerning the degree of stiffening of nonischemic myocardium. 2Previous work showed the load dependency of the response of nonischemic myocardium to remote ischemia. 15,22In the current study, we did not systematically alter loading conditions during the ischemic state; therefore, we cannot exclude the possibility that regional differences in nonischemic mechanics may be modified by alterations in loading conditions.

In conclusion, the current study shows that coronary occlusion is associated with an increased stiffness in remote nonischemic myocardium that occurs regardless of the localization of the ischemic and nonischemic areas. Moreover, regional differences may exist in both the diastolic and the systolic response to remote ischemia.