INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

WHEN THE NORMAL HEALTHY HEART is exposed to an increased demand for cardiac power, there is an increase in myocardial oxygen consumption (MO₂) and glucose, lactate, and fatty acid oxidation, (10, 14-16, 19, 24, 26). Frequently, patients with coronary artery disease have sufficient coronary blood flow and MO₂ at rest but will fail to have the normal increase in coronary flow in response to a greater demand for oxygen, such as occurs with physical exercise or adrenergic stimulation (increase in heart rate, afterload, and contractility). Dobutamine stress echocardiography has traditionally been used in clinical settings to identify residual viable myocardium in patients following acute myocardial infarction or those with suspected myocardial hibernation (35, 36) and can also be utilized to diagnose coronary artery disease by increasing demand for oxygen consumption and blood flow in the myocardium. In previously published studies, an infusion of dobutamine to open-chest anesthetized swine resulted in an increase in MO₂, heart rate, and myocardial uptake of fatty acids, lactate, and glucose, but with no evidence of myocardial ischemia (14-16), similar to exercise in healthy humans (10, 19). When the increased energy demand occurs in combination with an inability to increase coronary blood flow, a mismatch occurs between oxygen demand and supply, resulting in stimulation of glycolysis and a switch to lactate production, a condition that is referred to as "demand-induced ischemia" (31).

Little is known about the regulation of substrate metabolism during demand-induced ischemia. Pharmacological agents that partially inhibit fatty acid oxidation (ranolazine, trimetazidine, oxfenicine, and perhexiline) have been shown to reduce the symptoms of demand-induced ischemia in patients with stable angina (3, 7, 21, 40), presumably by reducing inhibition by fatty acid oxidation on pyruvate dehydrogenase, thereby increasing glucose oxidation and decreasing lactate and H⁺ production (20, 23, 40). Because these partial fatty oxidation inhibitors are effective during demand-induced ischemia, one would hypothesize that there is a high rate of fatty acid oxidation under these conditions. A more careful assessment of the metabolic responses during demand-induced ischemia has not been conducted.

The purpose of this study was to examine the contractile and metabolic responses to demand-induced ischemia caused by an inability to increase flow in response to the increased oxygen demand. We hypothesized that demand-induced ischemia would result in no change in free fatty acid uptake or MO₂, but would stimulate nonoxidative glycolysis and lactate production. Studies were performed in an open-chest swine model where flow was restricted and demand was increased with an intravenous infusion of dobutamine.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experiments were performed on seven domestic pigs (mean weight, 40.5 ± 0.9 kg). Studies were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication Number 85-23) and the Institutional Animal Care and Use Committee at Case Western Reserve University.

Surgical preparation. After an overnight fast, animals were sedated with 6 mg/kg Telazol (im), masked down with isoflurane (5%), intubated via a tracheotomy, and ventilated to maintain blood gases in the normal range (PO₂ > 100 mmHg, PCO₂ 35-45 mmHg, and pH 7.35-7.45) (39). Anesthesia was maintained with isoflurane (0.75-1.5%). The heart was exposed via a midline sternotomy with left-side rib resection (30). Heparin was infused to prevent clotting and thrombus formation (200 U/kg bolus, followed by 100 U · kg¹ · h¹ iv), and a 20% triglyceride emulsion was infused (Intralipid 20%, 0.3 ml · kg¹ · h¹ iv) to increase plasma free fatty acids to ~0.6 mM (39). Left anterior descending coronary artery (LAD) blood flow was controlled by an extracorporeal perfusion circuit via a roller pump with blood supplied from the femoral artery, as previously described in detail (27, 30, 39). Arterial blood samples were obtained from a constant flow (10 ml/min) withdrawal loop from the LAD perfusion circuit so that blood sampling would not disturb coronary artery blood flow. A 7-Fr Milar Mikrotip dual-transducer catheter was used to assess left ventricular (LV) pressure. A polyethylene cannula was placed in the anterior interventricular vein and was used to collect venous blood samples from the perfusion zone of the LAD. Studies by Renstrom et al. (33, 34), using a constant infusion of indocyanide green dye, have shown that the anterior interventricular vein in domestic swine receives 91 ± 1% of its blood from the LAD; however, we measured this factor in each animal with a constant infusion of green dye (0.3 mg/min into the LAD perfusion line) under normal conditions and during demand-induced ischemia. Regional segment length was measured in the anterior free wall (LAD bed) using sonomicrometry as previously described (Triton Technologies, San Diego, CA) (16, 27). The crystal pair was positioned at approximately midwall depth. This preparation allowed us to subject the LAD perfusion bed to demand-induced ischemia by decreasing LAD flow by 20% while infusing dobutamine. Regional myocardial contractile dysfunction was assessed during demand-induced ischemia from the decrease in the LV pressure-segment length loop area from normal conditions to demand-induced ischemia conditions in the anterior wall. The right main and circumflex coronary artery blood flows were not restricted, thus their perfusion beds could increase myocardial blood flow in response to the increased demand as previously described (15, 16).

Experimental protocol. After completion of the instrumentation, a continuous infusion of [U-¹⁴C]glucose (0.2 µCi/min) and [9, 10-³H]oleate (0.2 µCi/min) was introduced into the proximal end of the coronary perfusion line at a rate of 0.1 ml/min (see Fig. 1). By infusing tracer directly into the coronary perfusion circuit, the radioactivity dose was greatly reduced, as was the extent of secondary labeling of ¹⁴CO₂ and ³H₂O, which increases the precision of the measurement of oleate and glucose oxidation by the myocardium. After 40 min of tracer infusion, three pairs of arterial and interventricular venous samples were drawn 10 min apart (40, 50, and 60 min after tracer infusion). Sixty minutes after the tracer infusion was initiated, demand-induced ischemia was initiated with an infusion of dobutamine (15 µg · kg¹ · min¹) to increase myocardial oxygen demand and by reducing LAD blood flow by 20% for a period of 20 min. The LAD perfuses approximately one-third of the LV mass, thus during dobutamine infusion the LV mass subjected to demand-induced ischemia was limited to the LAD bed, and the mass perfused by the right main and circumflex coronary arteries presumably experienced the normal dobutamine-induced increase in myocardial blood flow and oxygen consumption (15, 16). Arterial and anterior interventricular venous blood samples were then taken at 63, 66, 70, and 80 min of tracer infusion (3, 6, 10, and 20 min of demand-induced ischemia). Blood samples were analyzed for the concentrations of oxygen, lactate, and glucose in blood and plasma free fatty acids. In addition, samples were analyzed for [¹⁴C]glucose, ¹⁴CO₂, [³H]oleate, and ³H₂O; this information was used to calculate the rates of exogenous glucose and oleate oxidation, as described below. Cardiovascular measurements were recorded immediately before each arterial and venous blood sample collection. Heart rate, left ventricular pressure (LVP), peak positive and negative first derivative of LV pressure (dP/dt), and segment length were continuously recorded using a commercial on-line data acquisition system (Crystal Biotech model CBI8000 with Biopaq software). Small myocardial biopsies (10-20 mg) were taken from the anterior LV free wall with a 14-gauge biopsy needle 5 min before initiating demand-induced ischemia and at 8 and 18 min of demand-induced ischemia. All biopsies were immediately freeze-clamped (3-5 s) on aluminum blocks precooled in liquid nitrogen and stored at 80°C for subsequent analysis. Tissue ATP and lactate were assayed in these samples. After 20 min of demand-induced ischemia, two large (~3 g) punch biopsies were rapidly obtained from the anterior and posterior LV free wall and freeze-clamped in large steel tongs precooled in liquid nitrogen; these samples were assayed for concentrations of tissue glycogen and triglycerides.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Time line depicting the study protocol. Approximately 30 min before the onset of the protocol, heparin and intralipid infusions were started and continued throughout the protocol. Time 0 marked the onset of a 60-min period of normal flow conditions and the start of the tracer infusion. This was followed by a 20-min period of dobutamine infusion (15 µg · kg1 · min1) accompanied by a 20% reduction in flow. Throughout the protocol, arrows mark the times at which blood, indocyanide green dye, and CO2 vial samples, and needle biopsies and terminal punch biopsies were taken.

Analytic methods. Arterial and venous pH, PCO₂, and PO₂ were determined on a blood gas analyzer (NOVA Profile Stat 3, NOVA Biomedical; Waltham, MA), and hemoglobin concentration and saturation were determined on a hemoximeter (Avoximeter; San Antonio, TX). Blood samples for glucose, lactate, and [¹⁴C]glucose were deproteinized in ice-cold 1 M perchloric acid (1:2 vol/vol) and analyzed for glucose and lactate using enzymatic spectrophotometric assays on a 96-well plate reader as previously described (13, 45). Blood samples for [¹⁴C]glucose and [¹⁴C]lactate measurements were neutralized with K₂CO₃, and the neutral eluate was run through ion-exchange resin columns (Bio-Rad AG 50W-X8 Resin and Bio-Rad AG1-X8 Formate Resin) to separate [¹⁴C]glucose as previously described (10, 43). Total glucose concentration and ¹⁴C activity were then measured in the eluate to calculate [¹⁴C]glucose specific activity. Plasma [³H]oleate concentration was measured by extracting the fatty acids from 0.5 ml of plasma in 3 ml of heptane-isopropanol (3:7) and counting the organic phase as previously described (44). ³H₂O concentration was measured by distilling 0.5 ml of plasma in custom-made, modified Hickman stills (Kontes Glass, Custom Shop). Blood ¹⁴CO₂ concentration was measured by expelling ¹⁴CO₂ with the addition of concentrated lactic acid and trapping it in hyamine hydroxide as previously described (9, 45). Plasma free fatty acids were measured using a commercially available enzymatic spectrophotometric kit (Wako Chemicals; Richmond, VA).

Calculations. The net uptake (µmol/min) for glucose and free fatty acids were calculated as the product of the arterial and coronary venous substrate concentration difference and myocardial blood flow. The rate of exogenous glucose and fatty acid oxidation (µmol/min) were calculated as the product of myocardial blood flow (ml/min) and the release of either ¹⁴CO₂ or ³H₂O (dpm/ml) into the coronary vein, divided by the arterial specific radioactivity of glucose or free fatty acids (dpm/µmol) (43). The interventricular venous concentrations of ¹⁴CO₂ and ³H₂O were corrected for dilution of blood derived from coronary arteries other than the LAD by multiplying the measured values by the concentration of green dye in venous plasma divided by the concentration in arterial plasma (33, 34). Myocardial blood flow (ml/min) was measured from the calibrated pump flow of the coronary perfusion line. MO₂ was calculated as the product of the arterial and venous oxygen concentration difference times the myocardial blood flow.

Statistical analysis. All hemodynamic parameters; rates of free fatty acid, glucose, and lactate uptakes; rates of free fatty acid and glucose oxidation; concentrations of tissue ATP, lactate, triglyceride, and glycogen; and regional anterior wall work and power index were compared between normal conditions and demand-induced ischemia using paired t-tests. Lactate production and MO₂ over time were analyzed using a one-way repeated measure analysis of variance. Arterial and venous concentrations and differences were compared using a two-way, repeated-measure analysis of variance, using a Student-Newman-Keuls test for post hoc comparisons. All significance tests were performed at the 0.05 level of significance. All values are reported as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Hemodynamics. Hemodynamic variables are listed in Table 1. Demand-induced ischemia resulted in significant increases in heart rate (48 ± 8%), LV peak positive (85 ± 9%) and negative dP/dt (29 ± 9%), and rate pressure product (44 ± 7%). There was no increase in MO₂ in the LAD perfusion bed during the period of demand-induced ischemia (Table 1 and Fig. 2).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Net myocardial lactate production (bottom) and oxygen consumption (MO2) (top) plotted as a function of time during normal conditions and demand-induced ischemia. * P < 0.05, significantly different from normal conditions.

Substrate metabolism. The arterial concentrations of lactate and glucose did not change significantly during the experiment (Fig. 3). Arterial concentration of free fatty acid was unchanged until the final 20 min of demand-induced ischemia when arterial free fatty acid was slightly higher than all other time points; this was matched by a similar rise in venous free fatty acid, resulting in no change in arteriovenous difference for free fatty acid. However, the arteriovenous difference increased for glucose and the uptake of lactate converted to production (Fig. 3).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Arterial and venous concentrations for lactate, glucose, and free fatty acids (FFA) plotted as a function of time during normal conditions and demand-induced ischemia.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Rates of FFA uptake measured with [3H]oleate and exogenous fatty acid oxidation measured from the production of 3H2O during normal conditions and demand-induced ischemia. * P < 0.05, significantly different from normal conditions.

Regional left ventricular wall work and power. Sonomicrometry was used to assess anterior wall work and power index as calculated from the segment length-LVP loop area. Figure 5 presents an example of segment length-LVP loops generated during both the normal conditions and during demand-induced ischemia from a representative animal. Anterior wall work index was significantly decreased by 38 ± 10% (169 ± 35 vs. 109 ± 28 mmHg · mm; P < 0.05) during demand-induced ischemia (Fig. 6). Anterior wall power index was unchanged during this period (299 ± 47 vs. 309 ± 86 mmHg · mm · s¹). The estimated rate of total ATP production (11.5 ± 1.2 and 12.1 ± 1.3 µmol/s) was constant; however, there was a switch to lactate production during demand-induced ischemia, which contributed 1.2 ± 0.3 µmol ATP/s, which accounted for 10 ± 2% of the total energy expenditure during demand-induced ischemia. Because there were no changes in either the anterior wall power index or the estimated total ATP production, the anterior wall energy efficiency index was unchanged during the demand-induced ischemia period (Table 1).

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Representative left ventricular pressure (LVP)-segment length loops generated during normal conditions and demand-induced ischemia for one animal. Anterior wall work index decreased by 55%, heart rate increased from 117 to 185 beats/min, and anterior wall power index decreased by 19% during demand-induced ischemia relative to the normal conditions in this particular animal.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Anterior wall work and power index in the anterior LV free wall. Anterior wall work index was calculated as the area contained within the LV pressure-segment length loops during normal conditions and demand-induced ischemia. Anterior wall power index was calculated as the rate of work production. P < 0.05, significantly different from normal conditions.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present investigation is the first to show that demand-induced ischemia resulting from dobutamine infusion with flow restriction activates nonoxidative glycolysis, but does not change MO₂, anterior wall power, or the myocardial energy efficiency index. Whereas there was no change in the rate of tracer-measured free fatty acid uptake, exogenous free fatty acid oxidation was decreased. Thus demand-induced ischemia stimulated glycolysis and lactate production, but did not effect fatty acid uptake despite a paradoxical fall in exogenous fatty acid oxidation.

Previous studies in patients with coronary artery disease demonstrated stimulation of lactate production when the heart was stressed by atrial pacing (31, 42). In the present investigation there was a dramatic switch to lactate production when dobutamine infusion was accompanied by a 20% reduction in LAD flow. Similarly, Parker et al. (31) demonstrated a significant increase in lactate production during angina induced by atrial pacing in patients with coronary artery disease. The switch to net lactate production corresponded with the onset of chest pain and the depression of the S-T segment during progressive atrial pacing. Another group of patients who experience exertional angina but have normal coronary arteries have been found to have limited coronary flow reserve due to a dysfunction of small coronary arteries (5). These patients have been reported to have a limited ability to increase coronary blood flow (6) and abnormal lactate metabolism in response to atrial pacing (4). Wisneski et al. (43) reported similar increases in lactate production during atrial pacing both in patients who experienced mild chest pain as well as in patients who did not experience clinical symptoms of ischemia. Thus the present investigation demonstrates that there is increased net glycogen breakdown, glucose uptake, and tissue lactate accumulation and a fall in ATP content during demand-induced ischemia, illustrating the activation of nonoxidative glycolysis under these conditions.

The primary source of glycolytic substrate for lactate production during demand-induced ischemia was endogenous glycogen stores (~84%), with a lesser contribution from exogenous glucose (16%). The total net lactate production in the LAD bed during demand-induced ischemia was measured, and the approximate lactate production was calculated from the conversion of exogenous glucose to lactate and the glycogen breakdown that was converted to lactate. The measured lactate production during the demand-induced ischemia period was 1,407 µmoles, and the mean expected lactate production from exogenous glucose and glycogen breakdown was 1,602 µmoles. These calculations were based on the following assumptions; 2 lactate molecules were produced for every glucose and/or glycosyl unit broken down, and the mean LAD bed weight was 42.0 g. The predobutamine glycogen concentration was assumed to be 31.1 µmol/g wet wt in the LAD bed [i.e., the same value as in the CFX bed at the end of the protocol, as previously found in nonischemic unstressed myocardium (38), thus the total glycogen breakdown during demand-induced ischemia was 1,789 µmoles (31.1 µmol/g  9.8 µmol/g × 42 g × 2 lactates/glycosyl unit)]. From the assumption that ~25% of glycolytic flux was subsequently oxidized and 75% was converted to lactate, 1,342 µmoles of lactate were produced from glycogen during the 20 min of demand-induced ischemia. The lactate that was produced from exogenous glucose [260 µmoles (13 µmols/min × 20 min)] was then added, resulting in a mean expected total of 1,602 µmoles of lactate production during demand-induced ischemia. This value was in the same range as the integral of the net lactate production, which was calculated as 1,113 µmoles released from the myocardium (venous-arterial lactate difference times blood flow; Fig. 2) and 294 µmoles of lactate that accumulated in the tissue [(10.9 µmol/g  4.1 µmol/g) × 42 g], for a total of 1,407 µmoles. Thus there was good agreement between the estimated lactate production from glycogen breakdown and glucose conversion to lactate and the total of lactate production.

It is important to note that the high rate of nonoxidative glycolysis during demand-induced ischemia accounts for only 10% of the total energy expenditure, but it clearly reflects a dramatic disruption to normal myocardial metabolism as seen in the marked lactate efflux and accumulation in the tissue. It has been suggested that lactate and H⁺ accumulation and efflux during ischemia have negative effects on the ability of cardiac muscle to maintain Ca²⁺ homeostasis and to use the energy released from the breakdown of ATP to perform contractile work (8, 29, 37). Drugs that decrease the rate of nonoxidative glycolysis by partial inhibition of fatty acid oxidation, such as trimetazidine (20) or ranolazine (28), improve the symptoms of patients with exercise-induced angina, possibly by increasing pyruvate oxidation in the mitochondria, thus reducing lactate accumulation and efflux. This mechanism remains to be demonstrated under clinically relevant in vivo conditions.

We observed that the amount of anterior wall work for each heartbeat was decreased by 38% from control conditions to demand-induced ischemia (Fig. 6, left). Because the heart rate increased by 47% during demand-induced ischemia, anterior wall power (taken as anterior wall work × heart rate) remained constant (Fig. 6, right). Moreover, the estimated rate of total ATP production was constant, thus because the anterior wall power index and the estimated total ATP production were constant, the anterior wall energy efficiency index was unaffected by demand-induced ischemia (Table 1). To our knowledge, this is the first demonstration that under conditions of high nonoxidative glycolysis, ATP depletion, and lactate accumulation, the myocardium does not become energetically less efficient.

During demand-induced ischemia, tracer-measured free fatty acid uptake was unchanged, but there was a fall in exogenous free fatty acid oxidation. These results suggest that labeled free fatty acids that are taken up during demand-induced ischemia are primarily stored as intracellular triglycerides. This concept is supported by studies in isolated myocytes that found an 81% increase in incorporation of [¹⁴C]palmitate into intracardiac triglycerides following exposure to 5 µM epinephrine (41). It is likely that there is also an increase in the breakdown of intracardiac triglyceride stores during demand-induced ischemia. The early study by Kreisberg (22) showed that when the isolated rat heart was perfused with [¹⁴C]oleate, epinephrine resulted in a 20% decrease in oleate oxidation and a doubling in the rate of oleate storage as triglyceride (from 9% to 22% of the oleate uptake). In addition, the rate of glycerol release doubled, reflecting greater intracellular triglyceride breakdown. These findings suggest that under normal unstimulated conditions, there is a slow turnover of the intracellular triglyceride pool; however, when the heart is stimulated with a -adrenergic agonist, there is an increase in the rate of triglyceride synthesis and breakdown (12, 25). Recent work from Goodwin and Taegtmeyer (11, 12) in the isolated buffer-perfused rat heart showed that stimulation of MO₂ and contractile function with 1 µM epinephrine stimulates both lipolysis and triglyceride synthesis, and thus greater turnover of the intracellular triglyceride pool. Clearly, further work is necessary before the effects of demand-induced ischemia on the metabolism of intracardiac triglycerides are understood. However, we speculate that both endogenous triglycerides and newly synthesized triglycerides from exogenous fatty acids could contribute to total energy production during demand-induced ischemia, and thus our measure of exogenous fatty acid oxidation underestimates the true contribution of fatty acids during demand-induced ischemia.

Study limitations. The present investigation did not assess the well-described transmural gradient in myocardial blood flow that occurs with a reduction in coronary flow at rest (17, 38) or in response to demand-induced ischemia (1, 2). Reductions in coronary artery flow by ~60% results in a large decrease in the subendocardial-to-subepicardial flow ratio, and greater glucose extraction (38), lactate production (13), and glycogen breakdown (18, 38) in the subendocardium relative to the subepicardium. The studies by Bache and co-workers (1, 2) demonstrate a transmural gradient in flow during demand-induced ischemia in exercising dogs with coronary flow restriction, thus suggesting that in our swine model there are gradients in both flow and nonoxidative metabolism during demand-induced ischemia. Future studies with microsphere-measured myocardial flow and transmural assessment of metabolism need to be performed to address this issue. An additional limitation in this study is that serial measurements of myocardial triglyceride and glycogen content were not performed. Glycogen breakdown likely occurred at a higher rate during the initial minute of demand-induced ischemia, as noted previously in dogs following coronary ligation (18). It is also important to understand the time course of the changes in intracardiac triglyceride content.