INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IN THE HEART, SEVERAL ASPECTS of myocardial bioenergetics, perfusion, and systolic function are known to be transmurally nonuniform and are importantly affected by changes in systemic hemodynamics. These nonuniformities are amplified both during and subsequent to an ischemic insult, especially when a perfusion deficit is induced by a coronary stenosis. In a study (8) of dog myocardium distal to a flow-limiting coronary stenosis, the blood flow deficiency induced changes in myocardial high-energy phosphate (HEP) and P_i levels, which were progressively more severe in the inner layers of the left ventricular (LV) wall. These findings demonstrate that the inner layers of the LV wall are most vulnerable to a coronary stenosis insult. In contrast, in a study (15) of normal hearts at very high work states induced by a high dose of catecholamine stimulation, it was found that HEP changes across the LV wall were uniform with a tendency for most pronounced changes in the outer layers of the LV wall where blood flow was higher. These data suggest that during very high work states the energy expenditure of the subepicardium exceeds that of the subendocardium. This evidence of greater subepicardial (Epi) than subendocardial (Endo) energy expenditure in the catecholamine-stimulated normal heart is at variance with previous observations (13) indicating that under normal perfusion conditions myocardial energy expenditure is greatest in the subendocardium.

The present studies were performed in an open-chest canine model in which wall-thickening fraction, spatially localized ³¹P NMR measurements of HEP and P_i, transmural blood flow, and ischemic region myocardial oxygen consumption (MO₂) were measured during pacing- and dobutamine-induced episodes of tachycardia. In these studies mean myocardial blood flow was maintained constant at control levels during experimental interventions by adjustments of the severity of the imposed coronary artery stenosis. The purpose of the current investigation was to compare the transmural heterogeneity of bioenergetic responses to pacing- and dobutamine-induced tachycardia in myocardium in which flow reserve was virtually eliminated by means of a coronary artery stenosis.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Studies were performed on 24 normal mongrel dogs. All experimental procedures performed were approved by the University of Minnesota Animal Resources Committee. The investigation conformed to the Guide for the Care and Use of Laboratory Animals [DHHS Publication No. (NIH) 85-23, Revised 1985, Office of Science and Health Reports, Bethesda, MD 20892].

Surgical preparation. Twenty-four mongrel dogs weighing 19-25 kg were anesthetized with pentobarbital sodium (30 mg/kg iv), and adequate anesthesia was maintained with a continuing infusion of this agent (4 mg · kg¹ · h¹). The animals were intubated and ventilated with a respirator using room air supplemented with oxygen. A heparin-filled polyvinyl chloride catheter (3.0-mm OD) was inserted into the left femoral artery and advanced into the descending aorta. A left thoracotomy was performed through the fourth intercostal space. The pericardium was opened, and the heart was suspended in a pericardial cradle. Heparin-filled catheters were inserted into the left ventricle through the apical dimple and into the left atrium through the atrial appendage and secured with purse-string sutures. Coronary venous blood sampling was accomplished with a catheter introduced through the right atrial appendage and advanced into the coronary sinus until the tip could be palpated at the origin of the anterior interventricular vein. The proximal left anterior descending coronary artery (LAD) was dissected free, and a hydraulic occluder constructed of polyvinyl chloride tubing (2.7-mm OD) was placed around the artery proximal to the first major arterial branch. A Doppler flow probe was placed on the LAD proximal to the occluder, and a bipolar pacing electrode was sutured onto the right atrial appendage.

NMR measurements. NMR data were acquired at 4.7 tesla using a SISCO (Spectroscopy Imaging System, Fremont, CA) instrument. Spatially localized ³¹P NMR spectroscopy was performed with the rotating-frame experiment using adiabatic plane-rotation pulses for phase modulation (RAPP)-imaging-selected in vivo spectroscopy (ISIS) technique (RAPP/ISIS; Refs. 6, 11, 14). The use of the adiabatic pulses ensured uniform spin rotations within the sensitive volume of the surface coil. Signal origin was restricted to a 18 × 18-mm column perpendicular to the surface coil plane and hence the LV wall. Localization along the column and therefore across the LV wall was achieved using the radio frequency magnetic field magnitude generated by the surface coil (B₁) gradient-based phase encoding. The number of transients accumulated for each phase-encoded step was weighted according to a nine-term Fourier series window, as previously described (6, 14). The phase-encoded data were used to generate a voxel or a "window" that could be shifted arbitrarily by postdata acquisition processing along the phase-encoded direction. Consequently, voxels were generated at different distances or "depths" from the outer LV wall (6, 11, 14). We normally present five voxels centered about 45°, 60°, 90°, 120°, and 135° phase angles as previously described (6, 11, 16). The position of the voxels relative to the coil was set according to the B₁ strength at the coil center, which was experimentally determined in each case by measuring the 90° pulse length with a reference in the center of the coil (6, 11, 14). The 18 × 18-mm column was defined using sech/tanh-modulated, 1.5- to 2-ms-long adiabatic inversion pulses, and 2.5-3.0 G/cm static magnetic field (B₀) gradients. The adiabatic excitation pulse that follows the adiabatic inversion pulse in RAPP-ISIS was based on optimized functions and was typically 1 ms in length (6, 11, 14).

Hemodynamic and LV function measurements. Aortic and LV pressures were measured using Spectromed TNF-R pressure transducers positioned at midchest level, and diastolic duration was determined from the LV pressure recordings. Transmural LV wall thickening was measured using the single epicardial microcrystal transducer and the 10-MHz pulsed-Doppler technique (4). All data were recorded on an eight-channel direct-writing recorder (model R14-28, Coulbourne).

Myocardial blood flow measurements. Mean (on-line) LAD blood flow was measured with the Doppler flow probe. Transmural myocardial blood flow was measured using radionuclide-labeled microspheres (8), 15 µm in diameter, suspended in low-molecular-weight dextran. Microspheres labeled with four different radioisotopes (⁵¹Cr, ⁸⁵Sr, ⁹⁵Nb, and ⁴⁶Sc) were agitated in an ultrasonic mixer for 10 min before injection. Microsphere suspension containing 2 × 10⁶ microspheres was injected through the left atrial catheter and flushed with 10 ml of normal saline. A reference sample of arterial blood was drawn from the aortic catheter at a rate of 15 ml/min beginning 5 s before microsphere injection and continuing for 120 s. Radioactivity in the myocardial and blood reference specimens was determined using a gamma spectrometer with a multichannel analyzer (model 5912, Packard Instrument, Downers Grove, IL) at window settings chosen for the combination of radioisotopes used during the study. Activity in each energy window, background activity, and sample weight were entered into a digital computer programmed to correct for overlap between isotopes, for background activity, and to compute the corrected counts per minute per gram of myocardium. Knowing the rate of withdrawal of the reference blood specimen (_r) and the radioactivity of the reference specimen (C_r), we used myocardial radioactivity (C_m) to compute myocardial blood flow (_m) from the equation _m = _r(C_m/C_r). Microsphere blood flow measurements were expressed as milliliters per minute per gram of myocardium.

M

Study protocol group 1. Ventilator rate, volume, and inspired oxygen content were adjusted (on the basis of arterial blood gas and pH measurements) as required to maintain physiological values (n = 11 hearts). Aortic and LV pressures and LV wall-thickening fraction were monitored continuously throughout the study. During each intervention, myocardial contractile function, blood flow, and hemodynamic measurements were collected simultaneously with the acquisition of transmural ³¹P NMR spectra. After control observations, the coronary artery occluder was inflated with a micrometer-driven syringe to reduce mean coronary blood flow (CBF; determined with the Doppler flow probe) to a level associated with a just-detectable decrease in LV wall-thickening fraction. The occluder was subsequently monitored and finely adjusted to maintain CBF at this reduced level until the end of the study. Thus the stenosis was used to maintain a constant level of mean CBF that was close to the basal value and that would be expected to exhaust autoregulation in the subendocardium (2). After we allowed ~10 min to ensure a hemodynamic steady state, atrial pacing was begun at a rate of 200 beats/min, and all measurements were repeated in the presence of the coronary stenosis.

Study protocol group 2. To examine the transmural heterogeneity of bioenergetic and thickening fraction changes in response to dobutamine-induced tachycardia, a second set of experiments was performed (n = 13 hearts). Surgical preparation and measurements were performed as described for group 1. Baseline measurements were obtained, and then infusion of dobutamine (20 µg · kg¹ · min¹ iv) was initiated. After waiting for ~10 min for a new hemodynamic steady state, we inflated the occluder to a degree sufficient to reduce mean LAD blood flow to baseline level. After waiting for 5 min, we repeated all measurements.

Data analysis. HEP and P_i spectral resonances were integrated using SISCO integral software. Hemodynamic data were measured from the strip-chart recordings. Systolic contractile function was obtained from the thickening fraction, which was measured as follows (4). For the whole wall thickening, gating depth (D) was set at 1.0 cm; end-diastolic thickness was measured at the initiation of the upstroke of the LV pressure tracing, whereas end-systolic thickness was measured 20 ms before peak decrease in pressure development over time. The values for 6-10 beats (corresponding to one respiratory cycle) were averaged. Whole wall thickening was defined as the difference between end-systolic and end-diastolic thicknesses divided by the gating depth.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Hemodynamic and LV systolic thickening fraction. Hemodynamic and systolic wall-thickening data are shown in Table 1. In both groups, heart rate increased comparably during pacing and dobutamine (P < 0.01). In group 1, mean aortic pressure and LV systolic pressure (LVSP) decreased significantly during stenosis plus pacing. In group 2, LVSP increased significantly during dobutamine stimulation plus coronary stenosis. LV end-diastolic pressure (LVEDP) rose modestly, but significantly, during pacing in group 1; LVEDP was not affected by dobutamine in group 2. In groups 1 and 2, the rate-pressure product (RPP = heart rate × LVSP) increased significantly during the interventions and increased more in group 2. LV contractile function became dyskinetic in both groups during coronary stenosis.

Myocardial blood flow. The regional myocardial blood flow measurements are shown in Table 2. In group 1, stenosis plus pacing tended to increase Epi and decrease Endo blood flow, which resulted in a decrease of the Endo-to-Epi blood flow ratio (Endo/Epi) (P < 0.01). In group 2 application of the coronary stenosis and dobutamine infusion resulted in an increase of blood flow in the Epi layer and decrease of Endo flow, resulting in a marked decrease of the Endo/Epi (P < 0.01). In both groups, during tachycardia induced by rapid pacing or dobutamine, myocardial blood flow was lower in all myocardial layers in the ischemic region compared with the nonischemic region (Table 2) as was the Endo/Epi (P < 0.01).

MO₂. MO₂ in the ischemic region tended to increase during tachycardia stresses but did not achieve statistical significance (Table 3).

Myocardial HEP and P_i levels. A transmural set of myocardial spectra from a group 1 experiment is shown in Fig. 1, and a set from a group 2 experiment is shown in Fig. 2. Blood flows, RPP, and MO₂ data are given in the figure legends. In these transmural spectra every other voxel is virtually nonoverlapping; however, adjacent voxels partially overlap. The spectrum at the bottom of the transmural stack (voxel 1) is always located near the coil and hence the outer myocardial wall; therefore, it arises from the subepicardium and accordingly it is labeled as "Epi." The location of this voxel is easily identified because it includes a resonance (not shown) at ~15 ppm arising from a phosphonate compound contained in a small chamber placed in the center of the surface coil plane on the outer LV wall (11, 14, 16). Spectra from the midwall (voxel 3) and the subendocardium (voxel 5) are labeled as "Mid" and "Endo," respectively, and have been rigorously tested previously (11, 14, 16). Prominent CrP and ATP resonances are observed in all voxels as expected (Fig. 1A). Under baseline conditions, P_i was too low to be detected in any myocardial layer at the signal-to-noise ratio of the spatially localized spectra, which was consistent with previous findings in the in vivo canine heart under baseline conditions (11, 14, 16). As illustrated in Fig. 1B, during stenosis plus pacing the decrease of CrP and increase of P_i peaks are more severe in the inner layers of the LV wall. This change was concordant with the ischemic regional blood flow pattern (see Fig. 1). A set of spectra obtained from a heart in group 2 is illustrated in Fig. 2. During stenosis plus dobutamine a significant increase of P_i and decrease of CrP were observed across the LV wall with a tendency toward more pronounced changes in the outer layers of the LV wall where the regional blood flow was higher (Tables 2 and 4).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Transmurally differentiated 31P NMR spectra obtained under basal conditions (A) and during coronary stenosis and pacing at 200 beats/min (B) from a heart in group 1 showing marked decrease of creatine phosphate (CrP) and increase of inorganic phosphate (Pi) levels in the inner layers of the LV wall during the period of ischemia. Each transmural data set consists of a stack of 5 spectra corresponding to voxels centered on phase angles of 45°, 60°, 90°, 120°, and 135°. The 135° voxel (corresponds to subepicardium) and 45° voxel (corresponds to subendocardium) are the outermost and innermost voxels respective to the surface coil. Endo, subendocardial voxel; Mid, midwall voxel; Epi, subepicardial voxel. These three voxels are nonoverlapping. ATP, ATP, and ATP refer to the resonances of the three ATP phosphates. Under basal conditions myocardial blood flow (MBF) from Epi, Mid, and Endo layers was 1.08, 1.11, and 1.21, respectively, and changed to 1.29, 0.98, and 0.71 in response to the coronary stenosis and rapid pacing, respectively. Redistribution of MBF away from subendocardium was accompanied by a significant decrease of CrP and increase of Pi in the inner layers (*) of the left ventricular (LV) wall during the period of ischemia (B).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Transmurally differentiated 31P NMR spectra obtained from a heart in group 2 under basal conditions (A) and during coronary stenosis and dobutamine infusion (B) showing marked decrease of CrP and increase of Pi levels across the LV wall. Under the basal conditions the MBF from Epi, Mid, and Endo layers was 0.98, 1.06, and 1.12, respectively, and changed to 1.91, 1.73, and 1.04 in response to coronary stenosis and dobutamine stimulation, respectively. A significant decrease of CrP and an increase of Pi (B) did not correspond to the pattern of myocardial hypoperfusion.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present data indicate that, in a constant coronary flow model of myocardial underperfusion, tachycardia stresses resulting from either pacing or dobutamine are associated with a common pattern of transmural blood flow maldistribution; i.e., Epi blood flow exceeds that present in the subendocardium. However, the HEP and P_i responses to the two interventions are dissimilar. During pacing the alterations in HEP and P_i are largely confined to the inner myocardial layers and are therefore concordant with the transmural blood flow gradient. In contrast, during dobutamine infusion the HEP and P_i abnormalities are transmurally severe and tend to be more pronounced in the subepicardium, a pattern that is discordant with the transmural blood flow gradient.

Transmural blood flow distribution during tachycardia in flow-limited myocardium. Cardiac contraction generates extravascular compressive forces that are the greatest in the deeper myocardial layers and normally impede perfusion markedly in the subendocardium during systole. Even during diastole, a transmural pressure gradient exists, which equals intrathoracic pressure at the epicardial surface and matches or exceeds cavity pressure in the subendocardium. Under normal conditions, autoregulation of the resistance vessels compensates for the inhibitory extravascular forces in the subendocardium. Unlike the situation in the subendocardium, in the subepicardium some flow normally continues throughout systole because extravascular pressure is lower than the perfusion pressure. Thus, whereas the duration of diastole and diastolic coronary pressure are major determinants of Endo perfusion, Epi flow can also be influenced by the difference between tissue and coronary pressure during systole. In the present study, the stenosis was slowly tightened until a barely perceptible decrease of systolic wall thickening was observed. This degree of stenosis results in exhaustion of the autoregulatory reserve in the resistance vessels of the subendocardium, whereas the Epi vessels still have the capacity for further vasodilation. Consequently, when the duration of diastole available for perfusion of the subendocardium was shortened by the tachycardia produced by pacing or dobutamine, the Endo vessels could not undergo further vasodilation and blood flow fell. In contrast, the Epi vessels were able to undergo additional vasodilation in response to the increased metabolic demands produced by pacing or dobutamine, so that Epi flow increased (Table 2).

Transmural distribution of metabolite levels and contractile work. In the group 1 studies the stenosis alone was associated with slight reductions of Endo blood flow and systolic wall thickening (data not shown), suggesting the presence of slight metabolic stress. During pacing Epi blood flow was increased somewhat relative to baseline values and was identical to that in a nonischemic reference region. Therefore, compared with the normal zone, oxygen delivery to the subepicardium was normal, and the increased rate of energy utilization associated with the elevated heart rate remained in balance with energy-generating capacity; hence, P_i/CrP, CrP, and ATP were unaffected. However, in the subendocardium, blood flow fell during pacing relative to baseline values and was significantly lower than that in the nonischemic region. The decrease in Endo blood flow during pacing likely occurred because the stenosis had resulted in exhaustion of autoregulatory reserve in the subendocardium. As a result, the decrease in duration of diastole during pacing-induced tachycardia could not be countered by further vasodilation, resulting in a decrease in Endo blood flow. In the subendocardium, energy supply was perfusion limited compared with the normal zone, and demand exceeded energy-generating capacity with the result that the levels of CrP/ATP and the P_i/CrP were affected. These changes are consistent with the presence of ischemia (8, 15). Thus during rapid pacing the metabolic ischemia pattern across the LV wall was concordant with the blood flow distribution. Although a similar transmural redistribution of blood flow occurred during dobutamine-induced tachycardia, loss of phosphocreatine characteristic of ischemia was seen in the subepicardium (Table 4). This likely occurred because the inotropic effect of dobutamine caused a greater increase in energy demand than rapid pacing. What is interesting, however, is that the ischemic changes in the subendocardium during dobutamine were not greater than those observed during pacing. This suggests that after the subendocardium reached the level of energy supply-demand imbalance, it did not respond to the inotropic stimulation (and associated increase in energy demands) produced by dobutamine.

Limitations. MO₂ measurements in a regional ischemic model have some limitations. Unless contamination of anterior coronary venous blood by efflux from a nonischemic region is rigorously excluded, the calculated ischemic zone MO₂ may be inaccurate. Although it is likely that our coronary venous samples represent mainly ischemic zone efflux, a small but unknown degree of contamination cannot be excluded. However, even if some degree of contamination were present, the conclusions drawn from the data would not be significantly modified. For example, if the extraction fraction of contaminating blood was less than that in ischemic zone blood, then actual ischemic zone MO₂ would be underestimated, not overestimated. Hence, the reported MO₂ measurements indicate that there was significant residual oxidative phosphorylation in the ischemic zone.