Background With moderate reductions in coronary blood flow, adjustments in myocardial metabolism can occur with a normalization of the imbalance between the decreased O2 delivery and tissue O2 demand. This state of “downregulated” metabolism is associated with reduced function and minimal irreversible injury and has been linked to myocardial hibernation. We hypothesized that (1) this process would occur when perfusion was reduced to severely ischemic levels, but only when flow declined at a slow rate rather than abruptly, and (2) this would result in blunted ischemia and reduced myocardial injury for a given period of low blood flow.
Methods and Results The left anterior descending coronary artery in anesthetized open-chest pigs was cannulated and perfused with arterial blood by an extracorporeal perfusion pump. Regional function (percent segment shortening, %SS) was measured with sonomicrometry and a regional coronary vein cannulated for blood gas analysis and lactate measurements. Coronary blood flow (CBF) was reduced to 10% of control either in a step fashion (Fast Ischemia group) or gradually in a linear manner over 70 minutes (Ramp Ischemia group). In all animals, CBF was held for 60 minutes at this 10% level and then followed by 2 hours of reperfusion. In the Ramp Ischemia group, the linear fall in CBF resulted in an initial maintenance of both %SS and myocardial oxygen consumption (MV̇o2) followed by linear decreases in both variables (r=.98 to .99) as flow fell to the 10% level. The relation of MV̇o2 to function was linear (r=.99) over the entire flow range. Although %SS, MV̇o2, CBF, coronary pressure, and hemodynamics during the 10% flow period were not different between groups, the increases in coronary venous lactate and Pco2 and fall in pH were blunted in the Ramp Ischemia group compared with the Fast Ischemia group. With reperfusion, a significant decrease in end-diastolic length was present only in the Fast Ischemia group. Additionally, although the region at risk was not different, infarction was markedly reduced in the Ramp Ischemia group (6.6±1.9%) compared with the Fast Ischemia group (31.4±6.9%).
Conclusions These data are consistent with the hypothesis that the downregulation of myocardial metabolism with gradually decreased flow to severe levels results in reduced myocardial injury for a given period of low flow. We propose that the rate at which blood flow decreases with myocardial underperfusion is a novel determinant of infarct injury. This may have clinical implication in situations in which there is a time-dependent component to the decrease in coronary blood flow in acute ischemic events, ie, thrombus formation at a site of coronary stenosis.
Abrupt occlusion of a coronary artery results in an immediate decrease in myocardial perfusion and subsequent ischemia that can be defined as an acute imbalance between oxygen delivery to the region of the myocardium supplied by that artery relative to its oxygen needs. This is associated with both functional and metabolic consequences. The former is characterized by a rapid fall in contractile motion described as a loss of systolic shortening and wall thickening. This dysfunctional state is fully reversible, provided that reperfusion is instituted before the loss of myocyte viability. Return of function can be rapid, with very short ischemic periods (<2 to 3 minutes), or can require an extended recovery period of hours to days, with longer ischemic events, ie, myocardial “stunning.”1 The metabolic consequences of ischemia include a rapid fall in tissue levels of phosphocreatine (PCr), increases in inorganic phosphate (Pi), decreases in ATP, and a change from aerobic to anaerobic metabolism, with a conversion of net myocardial lactate extraction to production.2 If this period of low flow is sustained for longer than 15 to 20 minutes, there is a progressive loss of myocyte viability, ie, irreversible injury as a function of the duration of the low flow period, collateral flow, risk area, myocardial metabolic demand, and ischemic history, ie, preconditioning. With moderate and sustained decreases in flow, several recent studies have documented the normalization of metabolic indexes of ischemia, including the restoration of PCr, stabilization of ATP levels, and return to net lactate extraction.3 4 5 6 7 These changes are associated with a stable but reduced level of contractile function that has been called “acute perfusion-contraction matching.”8 This state is reversible with restoration of normal flow and is associated with minimal necrosis even with 5 hours of sustained reduced flow.9 These changes have also been linked to the clinical observation of “hibernating myocardium” or “idling myocardium,” coined to describe the chronic decrease in function reversed on revascularization in humans.8 10 11 12
The normalization of indexes of ischemia in the presence of reduced oxygen delivery (supply) implies that a new steady state has been reached and that myocardial oxygen demand has been reduced proportionately. Although the mechanism responsible for this matching has not been clearly defined, the process has been documented by Pantely et al4 and Bristow et al,12 who called it “myocardial downregulation.” The conclusion from these early studies is that the imbalance between supply and demand or the extent of ischemia can be modulated by active reductions in the demand term as supply falls. This implies a tight coupling of tissue respiration and muscle function to oxygen supply and is supported by recent studies in skeletal and cardiac muscles indicating that oxygen availability and tissue oxygenation significantly influence tissue respiration at submaximal V̇o2.13 14 15 In skeletal muscle, reductions in oxygen delivery with hypoxia or ischemia have been shown to result in proportional decreases in function and V̇o2 without significant falls in tissue ATP levels.13 16 This matching of mitochondrial respiration and oxygen utilization to oxygen availability would serve to maintain myocyte viability and ATP levels in the face of reduced perfusion by downregulating function and oxygen demand.
It has recently been shown that there is a time-dependent component to this downregulation process. With abrupt but moderate decreases in myocardial perfusion, there is a rapid fall in PCr, with net lactate production that returns to control conditions, and a stable but reduced ATP level and flow.4 5 However, if blood flow is reduced gradually to the same level, the changes in these metabolic indexes of ischemia are blunted.17 This latter observation suggested to us that the downregulation of myocardial metabolism might also occur with gradual but continuous reductions of blood flow to very low levels normally associated with extensive myocyte loss. We hypothesized that with rapid reductions in coronary blood flow (CBF), such as that occurring with abrupt ligation of a coronary artery, this downregulation process would not have time to develop compared with the situation of a gradual flow reduction. Thus, for a given period of severe low flow, one might expect much less myocardial injury in the latter instance compared with the abrupt decrease in flow. If this is indeed the case, this would suggest that the rate of the decrease in myocardial perfusion with developing ischemia would be a determinant of the extent of infarction injury and might have important implications regarding the time frame in which reperfusion in a clinical setting might potentially salvage ischemic myocardium.
Therefore, the goal of the present study was to answer these questions: (1) Does a gradual reduction in myocardial perfusion result in a matched decrease in contractile function and myocardial oxygen consumption (MV̇o2)? (2) Is this associated with blunted indexes of myocardial ischemia during a period of low flow compared with an abrupt decrease in flow? (3) Is the extent of myocardial infarct injury resulting from a given period of low blood flow reduced when flow falls gradually compared with the abrupt situation? The experimental approach was to reduce CBF in a major coronary artery in either a slow linear fashion or abruptly while measuring regional function, MV̇o2, release of lactate, and the extent of infarction.
The experimental preparation has been described in detail.18 Briefly, domestic farm pigs (18 to 25 kg) were premedicated with ketamine hydrochloride (2.5 mg/kg IM) and anesthetized with sodium thiamylal (15 to 25 mg/kg IV) followed by α-chloralose (100 mg/kg IV) and morphine sulfate (30 mg/h SC). Animals were artificially ventilated at an end-expiratory pressure of 3 to 5 cm H2O. Arterial blood gas and pH levels were monitored (Radiometer ABL-300) and maintained within the physiological range. Animal temperature was monitored (Yellow Springs Instrument) and maintained at 37°C with a circulating-water heating pad. Aortic blood pressure was measured (Statham P23Db) via a catheter placed into the aorta. Left ventricular (LV) pressure, LV end-diastolic pressure, and LV dP/dt were obtained with a catheter-tip manometer (7F, Millar) inserted into the left ventricle via the right carotid artery. A pair of bipolar pacing electrodes was placed on the right atrium to allow cardiac pacing.
The heart was exposed via a combination of a sternotomy and left thoracotomy. The distal portion of the left anterior descending coronary artery (LAD) approximately 1.5 to 2.5 cm from the left main bifurcation was cannulated and perfused with arterial blood from the left carotid artery with an extracorporeal perfusion system described below. Typical time between ligation and cannulation was 15 to 20 seconds, with no animals requiring more than 30 seconds to reinstate perfusion. A small catheter was placed into the interventricular vein draining the LAD region to allow for sampling of coronary venous blood for measurement of blood gas levels and pH (ABL-300) as well as total hemoglobin and oxygen content (OSM-3 Hemoximeter, Radiometer). The OSM-3 measures lysed blood hemoglobin absorbance at six wavelengths and incorporates algorithms for automatic correction for nonhuman species, including pig. In repeated trials, we have found this instrument to yield O2 content data comparable to the LEX-O2-Con Instrument (Hospex) but with far superior reliability and reproducibility. Coronary venous blood was also obtained for measurement of total blood lactate with a semiautomatic lactate analyzer (YSI-23L, Yellow Springs Instrument) calibrated at four levels of lactate concentration.
The extracorporeal perfusion system has been described in detail18 and is constructed of short lengths of Tygon and Silastic tubing. This system includes a low-pulsation linear output coronary perfusion pump that can be operated in a controlled-flow or servo-feedback constant pressure mode. One advantage of this system is that it allows the set point to be varied in a user-controlled fashion. In the present experiment, in the controlled-flow mode, the control signal was either decreased in a step fashion or decreased linearly over a defined time interval by use of a voltage divider circuit with the resistance (thus voltage) controlled by a linear potentiometer mounted in an infusion pump (model 975, Harvard Apparatus). The set-point voltage could thus be decreased from a user-defined upper level to a lower level in a linear and reproducible manner. Coronary perfusion pressure was measured via a T-connection close to the cannula end. The extracorporeal circuit also included a small Windkessel (15 mL) and a transit-time flowmeter (Transonics Inc). Blood temperature was measured at the coronary end of the perfusion circuit with a miniature in-line probe (Yellow Springs Instrument) and was maintained at body temperature by circulating water coils on all exposed tubing. Anticoagulation was achieved with sodium heparin (500 U/kg IV) and maintained by hourly supplemental doses of 250 U/kg.
Regional Myocardial Function
Regional myocardial function was measured with a sonomicrometer (Triton Technology) and pairs of lensed sonomicrometer crystals (2 mm) placed into the inner third of the LV wall (6 to 8 mm deep), approximately 1.0 to 1.5 mm apart and perpendicular to the base-apex chord. One set was placed into the externally perfused LAD region and the second set in the normally perfused circumflex (Cx) area in the basal lateral portion of the heart.
Myocardial Infarction Quantification
After termination of the experiment, 150 mg of zinc cadmium fluorescent microspheres (Duke Scientific) in 50 mL Ringer’s solution was rapidly injected into the ascending aorta, the heart was fibrillated with a 60-Hz, 10-V square-wave signal applied directly to the surface of the heart, and the LAD coronary perfusion was stopped. In this fashion, myocardium perfused via the extracorporeal perfusion circuit did not receive microspheres, in contrast to the rest of the heart. The atria, right ventricle, and great vessels were weighed and trimmed from the left ventricle. The LV chamber was then sliced in approximately 10 to 12 pieces from apex to base, with the last slice located just above the cannulation site on the LAD, and the slices were individually weighed. Heart slices were incubated at 37°C in triphenyltetrazolium chloride (TTC) (1%) in PBS, pH 7.0, for 30 to 45 minutes and then placed in 6% formalin overnight. The next day, heart slices were mounted between two sheets of glass, shimmed to a constant width of 4 mm, and visualized on both sides in available and UV light. Three areas were traced: total slice area (TA), microsphere-absent regions (MS−), and areas of infarction (IA) indicated by the absence of the characteristic red TTC stain. Average values for each area were calculated as the mean of the data from the two sides of each slice. For each slice, the masses of the region at risk (RAR) and infarct region (IR) were calculated by the following formulas, in which a mass-per-unit-area factor was derived for each slice (constant thickness) and multiplied by the area of interest: RAR=slice weight/TA×MS− and IR=slice weight/TA×IA. For heart tissue containing no MS− areas, tracing was not necessary, and the entire weight counted as normal tissue. The data from all pieces were summed to yield total LV weight, RAR weight, and IR weight. By this method, both RAR and IR were calculated in units of mass (grams). The RAR was also expressed in the traditional manner as percent of LV. Similarly, IR was also expressed as percent of RAR.
It was our hypothesis that the rate at which myocardial ischemia developed would be an important determinant of the extent of myocardial injury to a given period of low flow. To test this, animals were randomized to two experimental protocols.
(1) In the Fast Ischemia group (n=7), after control measurements, LAD CBF was decreased from normal flow at a coronary pressure of 90 to 100 mm Hg to 10% of that value in a step manner. CBF was held at this level for 60 minutes. At the end of this period, normal coronary perfusion was instituted in a gradual manner over 1 minute, and then coronary pressure was maintained for 2 hours at a level of 75 to 85 mm Hg.
(2) In the Ramp Ischemia group (n=7), CBF was linearly reduced over a period of 70 minutes from normal flow at a coronary pressure of 90 to 100 mm Hg to the same level as achieved in the Fast Ischemia group, that being 10% of normal flow. At the end of this 70-minute period, CBF was held for 60 minutes at this 10% value. Reperfusion was done in exactly the same fashion as in the Fast Ischemia group, also for 2 hours.
After instrumentation, heart rate was paced at 90 to 100 beats per minute with the coronary perfusion pump in the constant-coronary-pressure mode (90 to 100 mm Hg). The CBF in mL/min at this pressure was recorded, and the perfusion pump was then switched to a constant-flow mode at this same flow. Control measurements of hemodynamic, arterial, and coronary venous oxygen content and lactate were taken. CBF was then decreased to 10% of the control value either in a rapid manner (Fast Ischemia group) or gradually over 70 minutes (Ramp Ischemia group). In either case, CBF was held at the 10% level for 60 minutes. At the end of this ischemic period, CBF was returned to the control value over a period of 1 minute, and then the perfusion pump was switched to constant pressure (75 to 85 mm Hg) for the duration of the 2-hour reperfusion period. During this reperfusion period, the cardiac pacing rate was increased to just above intrinsic rate to minimize arrhythmias. Hemodynamic and blood oxygen and lactate content were taken at 5, 10, 15, 20, 30, 45, and 60 minutes of ischemia and 5, 10, 15, 30, 60, 90, and 120 minutes of reperfusion. In the Ramp Ischemia group, measurements were taken every 5 minutes for the duration of the 70-minute ramp protocol. At the end of the reperfusion period, the RAR was determined and the heart tissue processed as described above.
Exclusion of Animals
It is our experience that acute proximal ligation of the LAD in pigs is associated with high ventricular fibrillation (VF) rates, >40%, probably because of the absence of significant collateral blood flow. In the present experiments, CBF was reduced to 10% of its control value, approximately 0.12 to 0.15 mL · min−1 · g−1, when normalized for the mass of the perfused region. In pilot experiments, we found that this greatly decreased the rate of VF. The rates of animal attrition due to VF and other problems in the present study are as follows. Eighteen animals were anesthetized. One animal was immediately excluded because of pericarditis, leaving 17 that were instrumented. Of these, 9 were randomized to the Fast Ischemia group. Of these 9 animals, 2 developed VF (1 during the ischemic period, 1 during reperfusion) and were excluded, to yield 7 animals completing the protocol. Eight animals of the 17 were randomized to the Ramp Ischemia group. One developed VF during early reperfusion and was excluded, to yield 7 animals in the data set.
At each time point, mean CBF, systolic and diastolic aortic blood pressures, peak LV pressure, LV end-diastolic pressure, LV dP/dt, and segment lengths were read from digitally acquired files (CODAS, Dataq Corp). Data from two or three consecutive cardiac cycles were averaged for each experimental time point. Regional segment lengths, in millimeters, were determined at end diastole and end systole. The end-diastolic length (EDL) was measured after the a wave, at the time of the sharp upsweep in the high-gain LV pressure signal. The end-systolic length (ESL) was defined at the point 20 ms before peak negative dP/dt. Regional myocardial function was calculated as percent segment shortening (%SS) defined as (EDL−ESL/EDL)×100. To allow comparison of changes in EDLs between groups, EDLs of each animal were normalized to a control value of 10 mm defined at the control, preischemia time point. MV̇o2 was calculated as CBF times (arterial O2 content minus coronary venous O2 content) and expressed as mL O2 · min−1 · 100 g wet wt−1. LAD CBF in mL/min was normalized to perfused wet weight of myocardium based on the mass of cardiac tissue that was fluorescent microsphere–negative.
The SEMs given in the figures and text were derived from data sets containing one value from each pig studied in each experimental group (n-1 df). Statistical analyses of the time data were done with a two-level ANOVA (time and treatment) for both the Ramp Ischemia and Fast Ischemia experimental groups. When they were found to be different, linear contrast comparisons were made at each time point, with correction for multiple comparisons. A statistical significance level was set at P=.05. For comparison of group differences with respect to lactate, pH, Pco2, and infarct size, a one-level ANOVA was performed and is indicated in the figures.
In both the Fast Ischemia group and the Ramp Ischemia group, heart rates were similar, approximately 100 to 105 beats per minute at the control point, and remained stable for the duration of the ischemic period (Fig 1⇓). Heart rate was also stable at this level for the duration of the 70-minute ramp period of the Ramp Ischemia group. At the onset of reperfusion, there was an increase in heart rate of approximately 15 to 20 beats per minute in both groups of animals, with no significant differences between the groups. Although there was a gradual decrease in LV systolic pressures over the course of the experimental protocol, there were no significant differences either at the control time point or during ischemia and reperfusion between the Fast Ischemia and Ramp Ischemia groups (Fig 1⇓). Similarly, there were no differences in the calculated heart rate–systolic blood pressure product.
Average data depicting CBF and coronary pressure in the two experimental groups are shown in Fig 2⇓. There were no statistical differences in CBF between groups. Flows at the control time point (1.17±0.15 versus 1.16±0.08 mL · min−1 · g−1) and during the ischemic period (0.11±0.02 versus 0.12±0.02 mL · min−1 · g−1) were virtually identical in the Fast and Ramp Ischemic groups, respectively. As is apparent from Fig 2⇓, CBF in the Ramp group decreased in a linear fashion over the course of the 70-minute ramp period, in contrast to the abrupt decrease in the Fast Ischemia group. Similar to flow, coronary perfusion pressures at control (101.8±7.0 versus 98.3±7.0 mm Hg) and during the ischemic period (15 to 17 mm Hg) were indistinguishable between groups. With reperfusion, there was a substantial hyperemia, with CBF rising to similar levels (2.37±0.31 versus 2.58±0.20 mL · min−1 · g−1) at 5 minutes of reperfusion. At the end of the 120-minute reperfusion, there was a tendency for CBF to remain above preischemia levels in the Fast group (1.62±0.30 mL · min−1 · g−1) compared with the Ramp group (1.25±0.10 mL ·min−1 · g−1) at similar coronary pressures.
Average contractile function in the LAD-perfused and control left Cx regions and normalized EDLs are shown in Fig 3⇓. Under control conditions, there were no significant differences in either %SS (21.8±2.1% versus 26.7±1.2%) or EDL (10.54±0.64 versus 10.98±0.51 mm) in the LAD region in the Fast or Ramp groups, respectively. In the Ramp group, as flow was ramped down, shortening was initially relatively constant from 0 to 20 minutes and then exhibited a gradual decrease from 20 to 70 minutes, falling to a stable level (from −3.2% to −3.9%). This level of paradoxical systolic lengthening during the ischemic period was not different between groups. Similarly, the level of depressed function during the reperfusion period was not significantly different between the Fast and Ramp Ischemia groups. In both groups, contractile function in the Cx area remained stable at or above control levels for the entire experimental protocol.
Within 5 minutes from the onset of the ischemic period, normalized EDL in the LAD region of the Fast Ischemia group increased significantly, by an average of 18.1±2.2% from control (P<.01) (Fig 3⇑, bottom), and remained elevated for the duration of the ischemic period. This occurred coincidently with an increase in LV end-diastolic pressure from a control level of 6.3±0.4 mm Hg to an ischemic value of 9.7 to 10.6 mm Hg. In contrast, in the Ramp Ischemia group, the increase in EDL was gradual and returned to control levels by the end of the ischemic period. This result could not be accounted for by differences in end-diastolic pressure, which actually increased to a greater extent (from 6.9 mm Hg at control to 10.6 to 11.4 mm Hg during ischemia) in the Ramp Ischemia group. The increase in EDL in the Fast group was significantly greater than that observed in the Ramp group at each time point of the ischemic period. On reperfusion, there was a rapid fall in EDL that was significantly greater in the Fast Ischemia group than in the Ramp Ischemia group and was sustained for the duration of the reperfusion period. In contrast, the fall in EDL in the Ramp Ischemia group was transient and reversed to control levels within 30 to 45 minutes. This difference in reperfusion EDL could also not be accounted for by differences in end-diastolic pressure, which actually tended to be greater in the Fast Ischemia group (12 to 15 mm Hg) than in the Ramp Ischemia group (10 to 12 mm Hg).
Average arterial blood oxygen contents were approximately 11.5 to 12.0 vol% in both the Fast and Ramp Ischemia groups and did not vary significantly over the experimental protocol (Fig 4⇓, top). Coronary venous blood oxygen contents at the control time point were also similar between groups, approximately 5.6 to 5.8 vol%. With the onset of the ischemic period in the Fast Ischemia group, myocardial oxygen extraction increased with coronary venous oxygen content, falling to an average of 2 to 2.3 vol% within 5 minutes. In the Ramp Ischemia group, oxygen extraction increased gradually over the ramp flow period, with coronary venous oxygen content falling to a stable level at 55 minutes and remaining at this level (1.7 to 1.9 vol%) for the duration of the ischemic period. This level was not significantly different from the Fast Ischemia group (P>.1). In both groups of animals, reperfusion resulted in an abrupt decrease in oxygen extraction, with coronary venous oxygen contents rising to similar levels of 9.5 to 10.5 vol% at 10 minutes of reperfusion. Oxygen extraction remained reduced in the Fast Ischemia group for the duration of the reperfusion period. In contrast, during reperfusion in the Ramp Ischemia group, oxygen extraction steadily increased toward control levels, with coronary venous O2 contents falling to 7.2±1.4 vol% by the end of the reperfusion, significantly less than that in the Fast Ischemia group (P<.05).
Myocardial Oxygen Consumption
Fast Ischemia Group
At the control time point, average MV̇o2 of the LAD-perfused region in these animals was 7.4±0.5 mL O2 · min−1 · 100 g−1 (Fig 4⇑, bottom). With the onset of the ischemia, MV̇o2 fell to 0.9 to 1.0 mL O2 · min−1 · 100 g−1 and remained at this level for the duration of the ischemic period. The restoration of normal blood flow was associated with an early increase in MV̇o2 toward control levels at 10 minutes, which decreased to a stable but low value of 2.0 to 2.5 mL O2 · min−1 · 100 g−1 over the 120-minute reperfusion period.
Ramp Ischemia Group
At the control time point, the average MV̇o2 in these animals (6.9±0.6 mL O2 · min−1 · 100 g−1) was not different from that in the Fast Ischemia group. Similar to regional function, MV̇o2 remained constant for the first 15 to 20 minutes of the ramp period, followed by a gradual fall in MV̇o2 to a stable level (1.1 to 1.3 mL O2 · min−1 · 100 g−1) for the duration of the 60-minute ischemic period. This value was slightly but not significantly greater than in the Fast Ischemia group. Reperfusion was associated with a transient increase in MV̇o2 at 10 minutes of reperfusion that was significantly greater than in the Fast Ischemia group (P<.05). Also, in contrast to the Fast Ischemia group, there was an improvement in MV̇o2 from 30 to 120 minutes of reperfusion. At the end of the reperfusion period, MV̇o2 had recovered to within 65% of the control value, compared with 31% for the Fast Ischemia group (P<.05).
Marked differences in the changes in coronary venous lactate, pH, and Pco2 were associated with the low-flow period between the two groups of animals (Fig 5⇓). In the case of lactate, arterial and coronary venous levels were similar between the two experimental groups under control conditions, with a small but significant net transcardiac extraction of lactate (Fig 5⇓, top). With the acute reduction in CBF in the Fast Ischemia group, lactate extraction reversed to production within 5 minutes to a peak value of −3.9±0.5 mmol/L. In the Ramp Ischemia group, net lactate extraction was present for the first 20 minutes of the ramp period, with blood flows greater than 0.7 to 0.8 mL · min−1 · g−1. With lower blood flows, net lactate production rose steadily to a peak of −2.2±0.6 mmol/L at the onset of the ischemic period, which was significantly less than that observed in the Fast Ischemia group (P<.05). There were no significant differences in either arterial or coronary venous lactate levels between the groups during the reperfusion period.
As with lactate, there were no differences in either coronary venous blood pH or Pco2 under control conditions between the two groups of animals (Fig 5⇑, middle and bottom). With the reduction in CBF to the same 10% level, the fall in pH and the increase in Pco2 were significantly blunted in the Ramp Ischemia group compared with the Fast Ischemia group (P<.01).
To investigate the relation of blood supply to contractile function, LAD %SS (normalized to control conditions) was plotted against myocardial blood flow for both groups of animals (Fig 6⇓, top). For the Ramp group, it was apparent that function was stable above a flow of approximately 0.8 mL · min−1 · g−1 and decreased gradually at lower flows during the ramp period. A linear regression analysis of the data at flows of 0.8 mL · min−1 · g−1 and below revealed the highly linear nature of this decrease (r=.99). Similarly, there was a clear relation of MV̇o2 to myocardial flow in the Ramp Ischemia group, with an initial flow-independent portion above a level of 0.8 mL · min−1 · g−1, followed by a gradual decrease with lower flows (Fig 6⇓, bottom). This also was highly linear (r=.98). It should be noted that there were no differences in MV̇o2, function, or flow between the Ramp and Fast Ischemia groups either during the control or low-flow conditions.
To investigate the relation of regional MV̇o2 to regional function, the two variables were plotted against each other with data from both experimental groups. As shown in Fig 7⇓ (top), there was a highly linear relation of MV̇o2 to function (r=.99) when data from the Ramp group were analyzed by linear regression. Furthermore, the low-flow data point for the Fast Ischemia group was very close to this relation. These data clearly indicate the tight coupling of MV̇o2 to function under conditions of reduced blood flow.
To compare the dependence of MV̇o2 with oxygen delivery under conditions of reduced perfusion, MV̇o2 was plotted against oxygen delivery calculated as the product of blood flow and arterial oxygen content using blood flow data below 0.8 mL · min−1 · g−1 (Fig 7⇑, bottom). This relation was highly linear, with a value of r=.99. In addition, as shown in Fig 7⇑, bottom, the slope of this relation was strikingly similar to that generated from data from Hogan et al13 obtained from contracting skeletal muscle under conditions of varied levels of hypoxia, illustrating the influence of oxygen delivery on tissue respiration.
There were also marked differences in the degree of infarction between the two experimental groups. The average data for all experiments are shown in Fig 8⇓. The mass of TTC-negative myocardium expressed as absolute weight or as a percent of RAR was considerably less (6.6±1.9%) in the Ramp Ischemia group than in the Fast Ischemia group (31.4±6.9%) (P<.01). The difference in infarction could not be attributed to differences in perfused myocardium. In the Fast Ischemia group, the LAD perfusion mass at risk averaged 40.0±2.1% compared with 36.5±2.6% of the left ventricle in the Ramp Ischemia group. These values were not significantly different (P>.1).
There are three important findings of the present study. (1) The relation of regional function and MV̇o2 to CBF was characterized by a flow-independent region and a linear flow–dependent region. (2) A severe reduction in myocardial perfusion produced in a gradual manner resulted in substantially less myocardial infarction and reduction in segment EDL compared with the condition in which CBF was reduced rapidly. (3) This reduction in myocardial injury was associated with blunted indicators of an imbalance between oxygen delivery and oxygen demand (ischemia), ie, pH, lactate, and Pco2. The results will be discussed in relation to the postulated determinants of infarction.
At present, there are five recognized determinants of irreversible myocardial injury resulting from ischemia: (1) the size of the RAR; (2) the magnitude of the myocardial blood flow reduction and collateral flow; (3) the duration of the ischemic period; (4) global metabolic state, ie, myocardial oxygen demand; and (5) ischemic history, ie, acute preconditioning. Although this is more controversial, inflammatory processes associated with reperfusion, including the activation of the complement cascade and neutrophil accumulation, have also been suggested as contributing to myocardial injury. The results from the present study will be discussed in relation to these recognized determinants of myocardial ischemic injury.
Region at Risk
It is known that the mass of the ischemic myocardium or RAR is a determinant of infarct size.19 20 21 In the present study, the RAR was measured by an anatomic method that delineated the myocardium perfused from the cannulated LAD from the remainder of the heart based on the distribution of fluorescent microspheres. This method is similar to that in common use in other models, ie, rabbits, and yields unambiguous perfusion boundaries. The RAR expressed as either absolute mass or percent of LV weight was not different between the Ramp and Fast Ischemia groups (Fig 8⇑). Thus, a random difference in this variable cannot account for the reduction in infarct size found in the Ramp Ischemia animals.
Myocardial Blood Flow/Collateral Flow
Residual flow or collateral blood flow is a strong determinant of myocardial ischemic injury.20 21 22 23 Although it is possible that unequal blood flow is responsible for the difference in infarction observed in the present study, this is unlikely. In our model, there would be two potential sources of blood delivery to the region supplied by the cannulated LAD: residual flow through the LAD artery and collateral flow. We used a precision coronary perfusion pump that made it possible to lower blood flow to the same level in both groups of animals in a reproducible fashion. Perfusion blood flow was not different either during control or at any time during the 10% low-flow period (Figs 2⇑ and 6⇑). Although collateral flow was not measured, it is also unlikely that this was a confounding factor, for the following reasons. First, it is commonly accepted, based on numerous studies, that innate collateral blood flow in swine is very low in contrast to dogs24 25 26 and is approximately 0.01 mL · min−1 · g−124 25 and inversely related to bed size. Bed size in absolute mass or percent of LV in the present study was the same between groups (Fig 8⇑). Second, the similarity of coronary perfusion pressures at equivalent blood flows during the ischemic period in the two groups of animals (Fig 2⇑, bottom) suggests that nutritive myocardial perfusion from collateral sources, if present, was small. Third, even if collateral blood flow were present, its magnitude would have to be quite large to result in the magnitude of the difference in infarct size reported here. We estimate that for the 75% reduction in infarct size for the 60-minute period of ischemia in the Ramp group, collateral blood flow would have to be even greater than that found in dogs (0.2 to 0.3 mL · min−1 · g−1). This is unlikely to be present in these normal pigs in the absence of a chronic stenosis. Thus, for the reasons given, it is unlikely that the difference in infarction was due to differences in myocardial blood flow.
Duration of Ischemic Period
The duration of the period of blood flow deficit is clearly a determinant of myocardial injury. With restoration of blood flow within 15 to 20 minutes, even in the absence of collateral flow, the extent of infarction is minimal.27 28 29 The development of irreversible injury beyond this time is highly dependent on the species investigated and the innate collateral circulation present. In dogs, with an average transmural collateral flow of approximately 0.2 to 0.3 mL · min−1 · g−1,21 22 fully developed infarction occurs within 3 to 6 hours.29 30 In collateral-deficient species such as pig and rabbit, infarction develops more rapidly, reaching a plateau within 60 to 90 minutes.28 29 In the present experiments, average myocardial blood flow during the low-flow period was intermediate between these two extremes, approximately 0.12 mL · min−1 · g−1, and by design was maintained for the same period of time in both experimental groups (Fig 2⇑). However, it should be noted that depending on the definition of myocardial ischemia, the period during which inadequate flow was present could actually be interpreted as being considerably longer in the Ramp Ischemia group. For example, if one considers the fall in contractile function as an index of ischemia, then it could be argued that ischemic conditions were already present at approximately 20 minutes into the ramp period when contractile function fell below control (Fig 3⇑). Similarly, if one considers the transition of net myocardial consumption of lactate to production as an index of ischemia, then in the Ramp Ischemia experiments, ischemia was present for the latter half of the ramp period (Fig 5⇑). In any case, it is clear that a difference in the duration of the ischemic period is not likely to be responsible for the reduced injury result and actually might have been expected to result in enhanced injury in the Ramp group.
Although it has been suggested that MV̇o2 is an important independent determinant of infarction injury,31 32 33 this concept is not universally accepted. On one hand, it seems intuitive that since ischemia is a condition defined in relation to metabolic state, a reduction in energy needs would decrease the magnitude of the supply/demand imbalance for a given flow reduction and thus delay high-energy phosphate loss and the development of irreversible injury. In support of this position, there have been numerous reports that a reduction in myocardial metabolism by ventricular unloading reduces infarct size after acute coronary ligation in dogs.34 35 Furthermore, the depletion of high-energy phosphate levels occurs considerably more slowly with total ischemia in vitro than with severe ischemia in vivo36 and has been attributed to the reduced metabolism of isolated tissue devoid of electrical and mechanical activity. Yet, there is only a poor correlation of hemodynamic determinants of MV̇o2 with infarction,20 and the reduction of myocardial metabolism with β-blockers has not been clearly shown to be protective.37 Nonetheless, in a careful study in which the independent effects of both MV̇o2 and collateral flow were evaluated, it was shown that a low MV̇o2 before and within the first few minutes of coronary occlusion (90 minutes), but not later in the ischemic period, resulted in less infarction compared with the situation with a high MV̇o2.38 Those data suggest that the MV̇o2 at the onset of ischemia may be of greater importance than the MV̇o2 during the ischemic period. This point may have a bearing on the interpretation of the present study. Although MV̇o2 and flow during the ischemic period were not different between the Ramp and Fast Ischemia groups, the myocardium was never subjected to a large mismatch between MV̇o2 and delivery in the Ramp group, since MV̇o2 fell as a function of flow (Fig 7⇑, bottom). In contrast, in the Fast group, reduced delivery occurred rapidly when MV̇o2 was initially at its normally high value; thus, the mismatch was large early in ischemia.
Several investigators,12 39 40 41 including ourselves,42 have advanced the concept that the myocardium may possess and use mechanisms that enable it to downregulate its regional energy expenditure in the presence of reduced blood supply and that this may be beneficial in terms of myocyte viability. Although other investigators have shown that this process may occur with moderate reductions in CBF,3 4 5 6 7 the present study is the first to suggest that this may have beneficial consequences in the presence of severe blood flow reductions. However, a simple downregulation of energy consumption does not appear to be a completely satisfactory explanation for the difference in infarct size between groups. The problem is that both the supply (CBF×arterial oxygen content) and the demand terms (MV̇o2) as well as contractile function were similar in both groups of animals (Figs 2 through 4⇑⇑⇑) during the ischemic period. The data do suggest that how the myocardium got to that low-flow/low-metabolism state may be important. As discussed above, the magnitude of the imbalance between supply and demand early in the low-flow period may be a critical factor.
Evidence is accumulating that ischemic history influences the extent of myocardial infarction resulting from a period of low blood flow. This phenomenon, called preconditioning, has been defined as the reduction in infarct size occurring in response to a sustained period of ischemia preceded by short periods of ischemia and reperfusion.43 44 There are several similarities and important differences between the reduction in infarct size reported in the present study and that resulting from preconditioning. (1) The magnitude of protective effect of gradual ischemia (75% reduction in infarct size) is comparable to that found in preconditioning models. (2) The reduced lactate formation and fall in pH during the sustained ischemic period observed in the present study (Fig 5⇑) are similar to those described in preconditioned animals.45 46 47 (3) Both phenomena indicate a reduced imbalance between myocardial supply versus demand (ischemia) during the sustained low-flow period. These similarities support the possibility that what we report here may be “preconditioning without reperfusion.” However, since the protective effect in the present study occurred in the absence of prior reperfusion, with blood flow always decreasing during the ramp period, the present phenomenon does not fall under the strict definition of preconditioning. Also, we can state that the processes known to be stimulated during reperfusion, including leukocyte accumulation, oxygen-derived free radical release, hyperemic flow/enhanced release of endothelium-derived relaxing factor, PCr overshoot, and transient calcium entry, which have been suggested to influence myocyte viability, are not likely to have contributed to the present result. It should be noted that some literature indicates that reperfusion is required48 and that the enhanced oxygen free radical release occurring with occlusion/reperfusion events participates in preconditioning.49 Nonetheless, it is attractive to speculate that the mechanism responsible for the reduction of injury with preconditioning is in fact identical to the one that is involved here, with a gradual reduction in CBF. Further investigation will be needed to test this possibility.
The conclusion from the foregoing discussion is that the mechanism responsible for the protective effect observed in the present study cannot be strictly accounted for by alterations in those variables recognized by most investigators to influence the extent of myocardial infarction. At present, it is not clear whether mechanisms of inflammation occurring during reperfusion, including activation of the complement cascade and leukocyte sequestration, which have been reported by some to influence infarct size, are involved in the phenomenon we describe. However, the large magnitude of the protective effect observed here would argue that these are unlikely to be major contributors, as has been shown to be the case with preconditioning.
One result of the present study was that myocardial blood flow was a strong determinant of regional contractile function. In the Ramp Ischemia group, a proportional decrease in blood flow below 0.8 mL · min−1 · g−1 resulted in a linear decrease in LAD segment shortening (Fig 6⇑). This result is consistent with published data. Gallagher and coworkers50 observed a nearly linear relation between reductions in LV subendocardial or transmural blood flow and regional systolic thickening in dogs. In contrast, an exponential relation between subendocardial blood flow and subendocardial segment function was described by Vatner51 in conscious dogs. It should be noted that below the knee, the relation was quite linear in this latter study. Also, as in the present study, Vatner found a narrow flow-independent region at basal flow. It should be noted that the presence of this flow-independent portion may simply be due to a small degree of coronary vasodilation and relative overperfusion in our two studies. Regardless, the nearly linear decrease in contractile function with reduced myocardial perfusion has been called “perfusion-contraction matching”8 and has been suggested to be involved in myocardial hibernation. Our data are consistent with this interpretation.
At any given work level, ATP turnover occurs at a rate necessary to conduct the work required. Since it is known that myocyte ATP content remains nearly constant over a wide range of work intensities,52 the rate of ATP hydrolysis must be tightly coupled to the metabolic processes that rephosphorylate ADP. This tight coupling of ATP turnover to production is exemplified by the strong correlation between contractile work intensity (ATPase activity) and tissue respiration as measured by oxygen uptake (V̇o2) under conditions of unrestricted blood flow. Under conditions of restricted oxygen supply occurring with reduced myocardial perfusion, a tight relation of contractile function and MV̇o2 continues to exist. This is shown in Fig 7⇑, top, in the present study. The nature of this relation is unclear. Since the oxygen extraction ratio is normally high in the heart, a severe reduction of oxygen supply must be accompanied by a fall in MV̇o2, once the upper limit of oxygen extraction is reached. Here, in the Ramp group, this limit is approximately 80% to 85% extraction and occurred at a blood flow of approximately 0.3 mL · min−1 · g−1 at 55 to 60 minutes of the ramp period (Figs 2⇑ and 4⇑). This maximal value of extraction is remarkably similar to that described for hypoxic skeletal muscle.13 As discussed above, the fall in contractile function with reduced perfusion has been shown in this and other studies. Importantly, however, both MV̇o2 and contractile function began to fall at a blood flow of approximately 0.8 mL · min−1 · g−1 at 20 minutes of the ramp period. This is well before the maximal extraction limit was reached at a level of 0.3 mL · min−1 · g−1. The explanation for the tight linear relation between MV̇o2 and function below the plateau level with reduced flow is of interest.
Since contractile function (tension development) is an important determinant of MV̇o2, it is possible that the decrease in MV̇o2 is secondary to the fall in function. However, this cannot fully account for the magnitude of the decrease in MV̇o2 with reduced function. It has been shown that the slope of the relation between segment shortening and MV̇o2 when function is decreased to dys- kinesis with intracoronary lidocaine is reasonably linear but, surprisingly, very shallow, with an MV̇o2 at akinesis of approximately 67% of control.53 This is in contrast to the relation we describe, which is considerably steeper, with an MV̇o2 of approximately 2 mL · min−1 · 100 g−1 at akinesis (Fig 7⇑). This level is close to that described for the basal metabolic state of nonworking muscle (2 to 2.5 mL · min−1 · 100 g−1). Together, these data indicate that the reduction in function is only one component responsible for the decrease in MV̇o2 at lower flows. The nature of this other component is not clear but may relate to the question of what limits myocardial oxidative phosphorylation in the presence of reduced oxygen delivery. As recently reviewed by Balaban,54 possible candidates include (1) concentrations of ATP, ADP, and Pi; (2) the phosphorylation potential described by [ATP][ADP][Pi]; (3) the mitochondrial and cytoplasmic redox potential described by [NADH][H+]/[NAD]; (4) PCr; and (5) molecular oxygen. Although the present study does not clearly distinguish between these possibilities, the relation of MV̇o2 to oxygen delivery is similar to that described for an in vivo skeletal muscle preparation (Fig 7⇑, bottom). It was observed in that model that O2 delivery modulated tissue respiration similarly whether achieved by hypoxia or by ischemia in the absence of marked decreases in tissue ATP levels.13 16 In addition, our finding has remarkable similarities to a study by Marshall55 showing that a tight linear relation existed between MV̇o2 and flow and also between MV̇o2 and developed pressure in the isolated perfused rabbit heart. The linear relation between MV̇o2 and oxygen delivery or blood flow described in both skeletal and cardiac muscle under conditions of reduced delivery can be called “perfusion-metabolism matching” as a counterpart to the phenomenon of “perfusion-contraction matching.”
Recent data suggest that perfusion-metabolism matching may be important for the limitation of ischemic injury under conditions of reduced CBF. In the case of moderate, sustained reductions of blood flow, it is now apparent that the myocardium makes a metabolic adjustment characterized by a restoration of PCr levels, normalization of lactate production, reduced but stable ATP levels, and depressed but stable contractile dysfunction.4 5 6 7 It has also been shown that this state can be maintained with depressed but stable ATP and contractile function without evidence of infarction for 5 hours3 9 up to 2 weeks.56 This phenomenon has been suggested to be a condition of downregulated metabolism and ATP consumption to match the reduction in perfusion.12 A recent study by Arai17 demonstrated the time dependence of this process and showed that slow reductions in myocardial perfusion to moderate levels result in blunted indexes of ischemia, including lactate release and fall in ATP and PCr levels, compared with an abrupt decrease. Our results demonstrating a reduction in lactate formation and fall in coronary venous pH and a reduced rise in coronary venous Pco2 in the Ramp Ischemia group appear to be consistent with the findings of Arai. However, it remains possible that the reduced lactate formation observed here may be due to a depletion of tissue glycogen occurring during the slow reduction in blood flow compared with the Fast Ischemia group. Additional experiments will be necessary to evaluate this possibility.
Downing and Chen57 recently reported that severe global ischemia for 2 hours in isolated perfused neonatal pig hearts was associated with increased O2 extraction, reduced MV̇o2 and cardiac function, and normal PCr and glycogen levels, with relatively well-preserved ATP levels. They concluded that when myocardial O2 supply is limited, myocardial function diminishes and serves to preserve critical energy stores and prevent irreversible injury. Our results in vivo are similar in many respects. However, direct comparison of the results from our in vivo study with those obtained in isolated neonatal hearts with global ischemia is problematic. First, it has been shown that high-energy phosphate depletion is reduced with global ischemia in vitro compared with regional ischemia in vivo36 and attributed to the electrical quiescence and reduced mechanical function found with global ischemia. It is likely that the dramatically reduced heart rates (decrease from 180 to between 30 and 60 beats per minute) observed by Downing and Chen during global ischemia may have influenced the result. It is also possible that ischemia occurring in that preparation with relatively lengthy time from excision to perfusion (up to 10 minutes) may have introduced some degree of myocardial preconditioning. In contrast, in our model, both heart rates and ventricular loads were constant between the experimental groups. Thus, these strong determinants of myocardial oxygen demand are not likely to have played a role. Nonetheless, the data from both of these studies yield the same conclusion: that with limited oxygen supply, myocardial contractile function and oxygen consumption fall proportionately and are associated with reduced myocyte injury.
Quantification of the mass of infarction in the present study was done with the histochemical method of tissue incubation with TTC. It has been reported that in some instances, the ability of TTC to discriminate between viable and dead tissue may be influenced by some pharmacological agents58 and is dependent on reperfusion time. Thus, the possibility exists that the reduced infarct size present in the Ramp Ischemia group was simply an artifact of the technique and if recovery and longer perfusion times were used in combination with histological assessment, this difference would not be present. Unfortunately, this cannot be easily done in this model. However, most studies find that a TTC or para-nitroblue tetrazolium estimate of infarct size after 1 to 2 hours of reperfusion is accurate in representing the ultimate infarct size as determined days later by histology.59
The reduction in infarct size in the Ramp Ischemia group occurred in the absence of an improvement in systolic segment shortening during the 120-minute reperfusion period compared with the Fast Ischemia group (Fig 3⇑). We interpret this as being due to myocardial stunning and speculate that function would recover substantially in the Ramp group over the course of several days to weeks. We would not expect this to occur in the Fast Ischemia group, in which significant infarction is present. This speculation is based on experimental data from preconditioning models in which infarct size is reduced but severe stunning remains.60 In addition, reports from both clinical61 and laboratory studies9 indicate that several days to weeks is required for a gradual return of wall function after reperfusion of hibernating myocardium.
The experimental model of myocardial infarction currently used usually involves acute ligation of a coronary artery within seconds, from a level of normal coronary artery flows and perfusion pressures to zero flow (excluding collaterals). It should be considered that this may only mimic a clinical situation in the individual with unstable angina, in whom a major coronary artery would be rapidly occluded from a basal metabolic and flow state. However, comprehensive data exist that thrombus formation and occlusion of a coronary artery at a site of plaque rupture and the progression of coronary stenosis to occlusion may occur over several minutes.62 In this situation, coronary occlusion is the end result of plaque rupture, formation of an intraplaque thrombus, extension to an intraluminal or mural thrombus, and complete or subtotal vessel occlusion. Thus, one might expect that, as the thrombus develops and occupies more of the cross-sectional diameter of the vessel, CBF would fall, not over seconds, but at some rate dependent on the geometry of the thrombus within the vessel and its rate of growth. The data from the present study indicate that the rate at which CBF falls is a determinant of the extent of infarction for a period of severe low flow. Thus, our results may have some implications regarding the rate of infarction development in humans with acute coronary occlusion. It should be noted that 70 minutes was arbitrarily chosen for the ramp period. It remains to be determined whether significant reduction in infarction would also occur with faster ramps. It seems possible that a delay of infarct development due to the phenomenon we describe here, as well as preconditioning, may increase the window of time available for myocardial salvage with reperfusion in humans. Extrapolating from the present study, an individual in whom the reduction in CBF to low levels occurred over a period of an hour would be expected to benefit from reperfusion at late times compared with the individual in whom sudden occlusion of a coronary artery occurred.
In summary, the present study demonstrates that the extent of myocardial infarction is dramatically less for a given period of reduced flow when blood flow falls gradually. The slow reduction in blood flow is associated with a concomitant fall in regional contractile function and MV̇o2 and reduced lactate release. These data suggest that the rate at which myocardial ischemia develops may be an important determinant of the extent of myocardial infarction resulting from a period of low blood flow. Although the data are consistent with models of hibernation and preconditioning, the mechanism responsible for this effect warrants further investigation.
This work was supported by funds provided by the Cigarette and Tobacco Surtax Fund of the State of California through the Tobacco-Related Disease Research Program of the University of California, grant RT400, to Dr Ito. The author would like to acknowledge Barry Peters, David Stein, Maureen Lee, and Paul Tang for their expert technical assistance in the completion of these studies.
- Received August 2, 1994.
- Revision received November 9, 1994.
- Accepted November 13, 1994.
- Copyright © 1995 by American Heart Association
Braunwald E, Kloner RA. The stunned myocardium: prolonged post-ischemic ventricular dysfunction. Circulation. 1982;66:1146-1149.
Jennings RB, Murry CE, Steenbergen C, Reimer KA. Development of cell injury in sustained acute ischemia. Circulation. 1990;82(suppl II):II-2-II-12.
Pantely GA, Malone SA, Rhen WS, Anselone CG, Arai A, Bristow J, Bristow JD. Regeneration of myocardial phosphocreatine in pigs despite continued moderate ischemia. Circ Res. 1990;67:1481-1493.
Arai AE, Pantely GA, Anselone CG, Bristow J, Bristow JD. Active downregulation of myocardial energy requirements during prolonged moderate ischemia in swine. Circ Res. 1991;69:1458-1469.
Schaefer S, Schwartz GG, Wisneski JA, Trocha SD, Christoph I, Steinman SK, Garcia J, Massie BM, Weiner MW. Response of high-energy phosphates and lactate release during prolonged regional ischemia in vivo. Circulation. 1992;85:342-349.
Schulz R, Rose J, Martin C, Brodde O, Heusch G. Development of short-term myocardial hibernation: its limitation by the severity of ischemia and inotropic stimulation. Circ Res. 1993;88:684-695.
Ross J Jr. Myocardial perfusion-contraction matching: implications for coronary heart disease and hibernation. Circulation. 1991;83:1076-1083.
Matsuzaki M, Gallagher KP, Kemper WS, White FC, Ross J Jr. Sustained regional dysfunction produced by prolonged coronary stenosis: gradual recovery after reperfusion. Circulation. 1983;68:170-182.
Rahimtoola SH. A perspective on the three large multicenter randomized clinical trials of coronary bypass surgery for chronic stable angina. Circulation. 1985;72(suppl V):V-123-V-135.
Bristow JD, Arai A, Anselone CG, Pantely GA. Response to myocardial ischemia as a regulated process. Circulation. 1991;84:2580-2587.
Hogan MC, Arthur PG, Bebout DE, Hochachka PW, Wagner PD. Role of O2 in regulating tissue respiration in dog muscle working in situ. J Appl Physiol. 1992;73:728-736.
Rumsey WL, Schlosser C, Nuutinen EM, Wilson DF. Cellular energetics and the oxygen dependence of respiration in cardiac myocytes isolated from adult rat. J Biol Chem. 1990;265:15392-15399.
Hogan MC, Nioka S, Brechue WF, Chance B. A 31P-NMR study of tissue respiration in working dog muscle during reduced O2 delivery conditions. J Appl Physiol. 1992;73:1662-1670.
Arai AE. Hypokinesis during gradual onset of mild ischemia occurs without decrease in phosphorylation potential: a model of acute myocardial hibernation. Circulation. 1992;86(suppl I):I-281. Abstract.
Ito BR, Roth DR, Engler RL. Thromboxane A2 and peptidoleukotrienes contribute to the myocardial ischemia and contractile dysfunction in response to intracoronary infusion of complement C5a. Circ Res. 1990;66:596-607.
Schaper W. Residual perfusion of the acutely ischemic heart muscle. In: Schaper W, ed. The Pathophysiology of Myocardial Perfusion. New York, NY: Elsevier/North Holland Biomedical Press; 1979:345-378.
Reimer KA, Jennings RB, Cobb FR. Animal models for protecting ischemic myocardium: results of the NHLBI cooperative study: comparison of unconscious and conscious dog models. Circ Res. 1985;56:651-665.
Jugdutt BI, Hutchins GM, Bulkley BH, Becker LC. Myocardial infarction in the conscious dog: three dimensional mapping of infarct, collateral flow and region at risk. Circulation. 1979;60:1141-1150.
Schaper W. The Collateral Circulation of the Heart. Amsterdam: North Holland Publishing Co; 1971.
Harken AH, Simson MB, Hasilgrove J, Wetstein W, Harden WR, Barlow CH. Early ischemia after complete coronary occlusion in the rabbit, dog, pig, and monkey. Am J Physiol. 1981;241:H202-210.
Jennings RB, Schaper J, Hill MJ, Steenbergen C, Reimer KA. Effect of reperfusion late in the phase of reversible ischemic injury. Circ Res. 1985;56:262-278.
Hornefer PJ, Healy B, Gott VL, Gardner TJ. The rapid evolution of a myocardial infarction in an end-artery coronary preparation. Circulation. 1987;76(suppl V):V-39-V-42.
Schaper W. Experimental infarcts and the microcirculation. In: Hearse DJ, Yellon DM, eds. Therapeutic Approaches to Myocardial Infarct Size Limitation. New York, NY: Raven Press; 1984:79-90.
Reimer KA, Lowe JE, Rasmussen MM, Jennings RB. The wavefront phenomenon of ischemic cell death, I: myocardial infarct size vs. duration of coronary occlusion in dogs. Circulation. 1977;56:786-794.
Maroko PR, Kjekshus JK, Sobel BE, Watanabe T, Covell JW, Ross J Jr, Braunwald E. Factors influencing infarct size following experimental coronary artery occlusions. Circulation. 1971;43:67-82.
Pennock JL, Pae WE, Pierce WS, Waldhausen JA. Reduction of myocardial infarct size: comparison between left atrial and left ventricular bypass. Circulation. 1979;59:275-279.
Axlerod HI, Galloway AC, Murphy MS, Laschinger JC, Grossi EA, Baumann FG, Colvin SB, Hunter CE, Glassman E, Spencer FC. A comparison of methods for limiting myocardial infarct expansion during acute reperfusion: primary role of unloading. Circulation. 1987;76(suppl V):V-28-V-32.
Jennings RB, Reimer KA, Hill ML, Mayer SE. Total ischemia in dog hearts, in vitro, 1: comparison of high energy phosphate production, utilization, and depletion, and of adenine nucleotide catabolism in total ischemia in vitro vs. severe ischemia in vivo. Circ Res. 1981;49:892-900.
Hearse DJ, Yellon DM, Downey JM. Can beta blockers limit myocardial infarct size? Eur Heart J. 1986;7:925-930.
Downing SE, Chen V. Myocardial hibernation in the ischemic neonatal heart. Circ Res. 1990;66:763-772.
Fedele FA, Gewirtz H, Capone RJ, Sharaf B, Most AS. Metabolic response to prolonged reduction of myocardial blood flow distal to a severe coronary artery stenosis. Circulation. 1988;78:729-735.
Rahimtoola SH. The hibernating myocardium. Angiology. 1989;40:209-221.
Schaper W, Ito BR. The energetics of “stunned myocardium.” In: de Jong JW, ed. Myocardial Energy Metabolism. Dordrecht, The Netherlands: Martinus Nijhoff Publishers; 1988:203-213.
Lawson CS, Downey JM. Preconditioning: state of the art myocardial protection. Cardiovasc Res. 1993;27:542-550.
Murry CE, Jennings RB, Reimer KA. New insights into potential mechanisms of ischemic preconditioning. Circulation. 1991;84:442-445. Editorial.
Murry CE, Richard VJ, Reimer KA, Jennings RB. Ischemic preconditioning slows energy metabolism and delays ultrastructural damage during a sustained ischemic episode. Circ Res. 1990;66:913-931.
Asimakis GK, Inners-McBride K, Medellin G, Conti VR. Ischemic preconditioning attenuates acidosis and postischemic dysfunction in isolated rat heart. Am J Physiol. 1992;263:H887-H894.
Kida M, Fujiwara H, Ishida M, Kawai C, Ohura M, Miura I, Yabuuchi Y. Ischemic preconditioning preserves creatine phosphate and intracellular pH. Circulation. 1991;84:2495-2503.
Ovize M, Przyklenk K, Kloner RA. Partial coronary stenosis is sufficient and complete reperfusion is mandatory for preconditioning the canine heart. Circ Res. 1992;71:1165-1173.
Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: is the protective effect mediated by free-radical-induced myocardial stunning? Circulation. 1988;78(suppl II):II-77. Abstract.
Gallagher KP, Kumada T, Koziol JA, McKown MD, Kemper WS, Ross J. Significance of regional wall thickening abnormalities relative to transmural myocardial perfusion in anesthetized dogs. Circulation. 1980;62:1266-1273.
Vatner SF. Correlation between acute reductions in myocardial blood flow and function in conscious dogs. Circ Res. 1980;47:201-207.
Katz LA, Swain JA, Portman MA, Balaban RS. Relation between phosphate metabolites and oxygen consumption of heart in vivo. Am J Physiol. 1989;256:H265-H274.
Gayheart PA, Vinten-Johansen J, Johnston WE, Hester TO, Cordell RA. Oxygen requirements of the dyskinetic myocardial segment. Am J Physiol. 1989;257:H1184-H1191.
Balaban RS. Regulation of oxidative phosphorylation in the mammalian cell. Am J Physiol. 1990;258:C377-C389.
Marshall RC. Correlation of contractile dysfunction with oxidative energy production and tissue high energy phosphate stores during partial coronary flow disruption in rabbit heart. J Clin Invest. 1988;82:86-95.
Bolukoglu H, Liedtke J, Nellis SH, Eggleston AM, Subramanian R, Renstrom B. An animal model of chronic coronary stenosis resulting in hibernating myocardium. Am J Physiol. 1992;263:H20-H29.
Downing SE, Chen V. Acute hibernation and reperfusion of the ischemic heart. Circulation. 1992;85:699-707.
Miyamae M, Fujiwara H, Kida M, Yokota R, Tanaka M, Katsuragawa M, Hasegawa K, Ohura M, Koga K, Yabuuchi Y. Preconditioning improves energy metabolism during reperfusion but does not attenuate myocardial stunning in porcine hearts. Circulation. 1993;88:223-234.
Fuster V, Badimon JJ, Badimon L. Clinical-pathological correlations of coronary disease progression and regression. Circulation. 1992;86(suppl III):III-1-III-11.