Myocardial High-Energy Phosphate and Substrate Metabolism in Swine With Moderate Left Ventricular Hypertrophy
Background Although left ventricular hypertrophy (LVH) is frequently associated with impaired coronary vasodilator reserve, it is uncertain whether this leads to myocardial ischemia under physiological conditions. The goal of the present study was to determine whether swine with moderate LVH exhibit metabolic evidence of ischemia when myocardial oxygen requirements are increased.
Methods and Results Myocardial metabolism was evaluated in an open-chest anesthetized preparation at baseline and during dobutamine infusion in 13 adolescent pigs with moderate LVH induced by supravalvular aortic banding and 12 age-matched control pigs. Transmural myocardial blood flow was quantified with radioactive microspheres; the ratio of phosphocreatine to ATP (PCr/ATP) in the anterior LV free wall was measured by 31P–nuclear magnetic resonance; and anterior wall lactate release was quantified from the arterial-coronary venous difference in 14C- or 13C-labeled lactate. In a subset of 5 animals from each group, the metabolic fate of exogenous glucose was determined from the transmyocardial difference in 6-14C-glucose and its metabolites 14C-lactate and 14CO2. Coronary reserve, as assessed by the ratio of blood flow during adenosine infusion to baseline blood flow, was significantly lower in the LVH pigs compared with controls (3.5±0.4 versus 5.5±0.4 mL/g · min, P<.05); however, transmural myocardial blood flow was similar in both groups of pigs, both at baseline and with dobutamine stimulation, probably reflecting the higher coronary perfusion pressure in the LVH pigs. At baseline, PCr/ATP tended to be lower in the LVH pigs (P=.09) but decreased similarly with dobutamine infusion in both groups. Isotopically measured anterior wall lactate release did not differ between the groups at baseline, nor did the increase in lactate release differ during dobutamine stimulation. The uptake of glucose, lactate, and free fatty acids did not differ between the groups in the basal state. However, during dobutamine stimulation, glucose uptake was greater in the LVH group (0.84±0.09 μmol/g · min versus 0.59±0.08 μmol/g · min, P<.05). In a subset of animals, 14C-glucose was used to assess glucose oxidation. These data showed that the LVH animals had a greater rate of glucose oxidation (0.60±0.10 versus 0.28±0.08 μmol/g · min, P<.05) and a greater rate of glucose conversion to lactate (0.20±0.04 versus 0.09±0.02 μmol/g · min, P<.05) compared with the control pigs.
Conclusions These results suggest that despite their reduced coronary vasodilator reserve and the absence of a greater rise in myocardial blood flow to compensate for a substantially higher LV double product, pigs with this model of moderate LVH do not exhibit a greater susceptibility to myocardial ischemia during dobutamine stress. However, LVH pigs exhibit significantly greater use of exogenous glucose during dobutamine stress, as evidenced by increases in both glucose oxidation and anaerobic glycolysis.
In hypertensive patients, left ventricular hypertrophy (LVH) is associated with an increased risk for cardiovascular morbidity and mortality, with the largest proportion of these complications resulting from myocardial ischemia.1 2 Impaired coronary vasodilator reserve, which is often present in animal models and in many hypertensive patients with LVH, is thought to indicate a predisposition to ischemia.1 3 4 5 6 However, because myocardial blood flow in the hypertrophied heart is generally normal in the basal state and rises significantly with increased cardiac workload, impaired coronary reserve need not be equated with myocardial ischemia.3 4 5 6 Evidence of accompanying ischemic metabolic changes or contractile dysfunction is required to confirm the hypothesis that myocardial ischemia occurs under physiological conditions in pressure-overload LVH. Thus far, such evidence is limited and is based primarily on data from dogs with very severe LVH secondary to chronic supravalvular aortic coarctation.7 8 9 10 11 In patients, LVH is usually less severe, and decompensation of LV systolic function is unusual.6 Evidence that ischemia occurs under these conditions is lacking.
Therefore, the present study was undertaken to determine whether swine with moderate LVH, in a range that corresponds more closely to the typical severity of LVH in hypertensive patients (30% to 50%), exhibit physiological evidence of ischemia during pacing-induced tachycardia or dobutamine-induced increases in myocardial oxygen demand. Two complementary techniques were used to detect myocardial ischemia and characterize myocardial substrate metabolism. Anaerobic glycolysis and glucose metabolism were assessed with 14C- or 13C-labeled lactate and glucose, an approach previously shown to be a highly sensitive marker of ischemia during graded reductions in coronary flow.12 This technique does not permit localization of the ischemic region or layer, but it is a very sensitive indicator of ischemia in any transmural layer. Second, the phosphocreatine-ATP ratio (PCr/ATP) was determined by 31P–magnetic resonance spectroscopy (MRS), which is also a sensitive indicator of ischemia in vivo13 14 and affords partial localization to the more vulnerable subendocardial layer.15 16
Pressure-overload LVH was induced by banding of the ascending aorta in 17 pigs. After premedication with ketamine (25 mg/kg IM) and under general anesthesia (halothane, 1% to 3%), 2- to 4-week-old female Yorkshire-Landrace swine underwent a left lateral thoracotomy in the fourth intercostal space. Approximately 3 cm above the aortic valve, the ascending aorta was constricted with a rigid polyethylene band adjusted to create a 20 mm Hg systolic gradient. The 13 pigs that met the prespecified criterion of a ratio of LV to body weight greater than the mean+2 SD of the control group constituted the LVH group. In 6 sham-operated control animals, the same procedure was followed, but the band was not closed. This surgery and the subsequent experiments were conducted in accordance with the guidelines of the American Physiological Society for Animal Use.
In the banded LVH pigs and sham-operated pigs, experiments were conducted 10 to 14 weeks after surgery, when the pigs reached a weight of 30 to 44 kg. An additional 6 pigs of similar age and weight but without prior surgery were studied with the same protocol.17 This latter group was derived from a larger group included in a previous study concerning metabolic responses to increased work state. No differences were found between these and the sham-operated pigs; therefore, their results were combined to form a 12-pig control group.
The Figure⇓ shows the preparation used in this study, which is similar to that described in greater detail elsewhere.13 14 17 After premedication with ketamine-HCl (20 mg/kg IM), anesthesia was induced and maintained with α-chloralose (100 mg/kg IV followed by 20 to 30 mg/kg · h IV). Pigs were ventilated with 100% oxygen through a tracheotomy by use of a pressure-cycled respirator adjusted to maintain Po2 above 100 mm Hg and a physiological pH. After an initial infusion of 500 mL normal saline, an infusion of 5% dextrose in normal saline was maintained at a rate of 100 mL/h. Both carotid arteries and internal jugular veins were cannulated for pressure monitoring, arterial sampling, and infusions of carbon-labeled substrates and medications. The chest was opened by a midline sternotomy, and the heart was suspended in a pericardial cradle. A catheter was placed in the left atrium for microsphere injection, and bipolar pacing wires were attached to the left atrial appendage. A 24-g Teflon catheter was inserted into the anterior interventricular vein for blood sampling, and a 6F micromanometer catheter (Millar Instruments) was inserted through the left ventricular apex for pressure monitoring. A pair of piezoelectric crystals (Triton Technology) was inserted into the subendocardial layer of the anterior wall approximately 1 cm apart along the presumed axis fiber orientation for measurements of segment shortening. After instrumentation was completed, a 2.5-cm-diameter, two-turn surface coil was loosely sutured to the anterior LV free wall in the region drained by the cannulated vein and immediately adjacent to the ultrasonic crystals. The pig was then placed on a heated water blanket in a plexiglass cradle and transferred to the spectrometer.
After hemodynamic stability was confirmed over a 30-minute period, baseline measurements of hemodynamics, segment shortening, and transmural myocardial blood flow were obtained. Baseline 31P spectroscopy was performed, and simultaneous blood samples were collected from a carotid artery and the anterior interventricular vein for measurements of O2 content; chemical glucose, free fatty acids, and lactate; and concentrations of isotopically labeled substrates and 14CO2. The specific techniques for these measurements are described below. Control measurements were made during atrial pacing at 10 beats above basal heart rate to facilitate gated nuclear magnetic resonance (NMR) spectroscopy at end diastole. Under each experimental condition, measurements were made over a 35-minute period: 10 minutes to reach steady state, 20 minutes for 31P-MRS, and 5 minutes to repeat hemodynamic measurements and obtain blood samples for metabolic analyses.
In five LVH pigs and five control pigs, coronary reserve was assessed during intravenous infusion of adenosine. The infusion was started at 0.25 mg/kg · min and increased by 0.25 mg/kg · min every 2 to 3 minutes. Blood flow measurements were repeated after 5 minutes at 1 mg/kg · min or at the highest infusion tolerated without a decrease in mean arterial pressure below 60 mm Hg.
In six LVH pigs and six control pigs, responses to atrial pacing were examined. The atrial pacing rate was increased at increments of 10 beats per minute (bpm) until AV block occurred. If a ventricular response of at least 150 bpm could not be obtained, pigs were treated with atropine 2 mg IV, and pacing was restarted. After 5 minutes at the highest achieved pacing rate, a complete set of hemodynamic, blood flow, and metabolic measurements was repeated.
The responses to an intravenous infusion of dobutamine were examined in all LVH and control pigs. Dobutamine was started at an infusion rate of 5 μg/kg · min, and the infusion rate was increased at 5-μg/kg · min increments every 10 minutes until the heart rate exceeded 200 bpm or an infusion rate of 15 μg/kg · min was reached. A complete set of hemodynamic, blood flow, and metabolic measurements was obtained after a 10-minute stabilization period at the highest infusion rate. 31P-MRS was performed first, so the isotopic tracer metabolic measurements were obtained after 30 minutes at a constant dobutamine infusion rate.
Hemodynamics and Sonomicrometry
Heart rate, blood pressure, and segment length were monitored continuously and recorded at 100 mm/s on a multichannel recorder (Gould Instruments) before, during, and after 31P-MRS under each experimental condition. LV pressures were recorded immediately before and after each 31P spectrum because radiofrequency noise introduced by the catheter interfered with spectroscopy. End-diastolic segment length was measured at the initial rise of the dP/dt signal, and end-systolic measurements were determined 20 milliseconds before the peak negative dP/dt. Fractional systolic segment shortening was determined as [(end-diastolic length)−(end-systolic length)]/(end-diastolic length). Measurements were averaged from 10 consecutive beats encompassing at least one full respiratory cycle.
Blood Flow Analysis
Myocardial blood flow was measured by the radioactive microsphere technique. Three million 15-μm-diameter radioactive microspheres were injected into the left atrium over a 30-second interval, and a reference sample was simultaneously withdrawn at a rate of 11 mL/min over 4 minutes with a Harvard pump. At the conclusion of the experiment, pigs were killed with a lethal injection of pentobarbital, and the hearts were excised and fixed in 10% formalin for 72 hours. After formalin fixation, myocardium sections of approximately 10 g were cut from the anterior wall immediately below the surface coil and from the posterior wall. Each section was then divided into subendocardial, midwall, and subepicardial layers of equal thicknesses for determination of the transmural distribution of blood flow. Mean transmural blood flow was calculated by averaging of the measurements in the three layers.
Analysis of O2 Content and Metabolic Substrates
The O2 content of the arterial and anterior interventricular vein blood samples was determined from the sum of the hemoglobin-bound and dissolved O2 content measured by an oximeter calibrated for peak blood (Radiometer OSM3) and a blood gas analyzer (Radiometer ABL 30), respectively. Anterior wall myocardial oxygen consumption was determined by multiplying the transmurally averaged anterior wall blood flow by the arterial-venous O2 difference and is expressed in micromoles per gram per minute.
The methods for measuring glucose, lactate, and free fatty acid concentration were described previously.18 19 Blood samples for lactate and glucose measurements were mixed immediately with cold 7% perchloric acid and centrifuged. The protein-free filtrate was removed and stored at −4°C for future analysis. Lactate concentrations were determined by an enzymatic spectrophotometric method20 ; glucose concentration was determined by the hexokinase/glucose-6-phosphate dehydrogenase–coupled enzymatic method.21 Blood samples for fatty acid analysis were placed in iced, heparinized glass tubes and centrifuged at 4°C, and the plasma was stored at −4°C. Free fatty acid concentrations were determined with gas chromatography by a modification of the method of Ko and Royer.22 The uptake of these substrates was calculated as the product of the arterial-venous concentration difference times the transmurally averaged anterior wall blood flow. Plasma insulin levels were measured by radioimmunoassay.
Labeled Substrate Analyses
The arterial-coronary venous chemical lactate difference reflects only the net difference between lactate extraction and release. Previous studies demonstrated that there is simultaneous extraction and release of lactate during nonischemic, normoxic conditions in animals and humans.12 18 19 Therefore, in this study, labeled lactate was used to quantify myocardial lactate release. As described previously,12 in eight LVH pigs and seven controls, a priming dose of 20 μCi of L-[1-14C]lactate (specific activity, 55 mCi/mmol) was administered intravenously, followed by an infusion at 25 μCi/h to maintain a stable arterial specific activity of 14C-lactate over the study period. A 30-minute stabilization period was allowed for equilibration of 14C-lactate levels in the arterial circulation and coronary veins.18 From the simultaneously drawn arterial and coronary venous blood samples, lactate was separated from other substrates by ion-exchange chromatography.20 The lactate specific activity was determined by measurement of 14C by scintillation counting and the chemical concentration. A lactate isotopic extraction ratio was determined as follows:
The isotopic extraction was then determined by multiplying the arterial chemical lactate concentration by this isotopic extraction ratio and expressed as micromoles per milliliter. The amount of lactate release in micromoles per milliliter was calculated as the difference between the myocardial lactate extraction determined from the isotope technique and the arterial-coronary sinus chemical concentration difference. Lactate release in micromoles per gram per minute was calculated by multiplying this value by the transmurally averaged blood flow.
In five pigs from each group, dual carbon-labeled substrate experiments were performed. In these pigs, both D-[6-14C] glucose (priming bolus 16 μCi followed by a constant infusion of 10 μCi/h) and L-[U-13C] lactate (priming dose 55 mg followed by a constant infusion of 65 mg/h) were administered. The chemical analysis methods and calculations were described in detail elsewhere.19 In brief, L-[U-13C] lactate content was determined with gas chromatography and mass spectrometry. The amount of lactate release was determined as described above for the 14C-labeled lactate experiments. 14CO2 was collected from blood by a diffusion method. Because glucose was labeled in the C-6 position, oxidation of exogenous glucose resulted in the release of this carbon as 14CO2 in the citric acid cycle. By measurement of the coronary vein (CV)–arterial (A) 14CO2 difference, the amount of glucose being oxidized (in micromoles per milliliter) was calculated as
The amount of glucose oxidized per gram per minute was determined by multiplying this value by the transmurally averaged blood flow.
In the experiments with 13C-lactate and 14C-glucose, the contribution of exogenous glucose to the lactate released or produced was calculated from the observed and theoretical disintegrations per minute of lactate per milliliter of blood in the coronary vein and the specific activity of arterial glucose as
The factor of 2 was used because one molecule of glucose yields two molecules of lactate. The observed disintegrations per minute of lactate per milliliter of blood in the coronary vein was calculated as
The theoretical disintegrations per minute from lactate per milliliter of blood was determined as
31P–Magnetic Resonance Spectroscopy
Measurements were performed with a 1.0-m-bore Philips Gyroscan System operating at 2T (Philips Medical Systems), according to previously described techniques.13 14 15 17 The magnet was shimmed on cardiac water protons to line widths less than 35 Hz. 31P-MRS signals were acquired from the heart with two methods in each animal. First, the pulse length was chosen to provide maximal weighting of the signal from the subendocardium relative to the subepicardium on the basis of computer modeling of the surface coil radiofrequency field and an estimated wall thickness of 10 mm for control animals and 15 mm for LVH animals. Second, further subendocardial weighting was obtained with the Fourier series window localization technique previously described from our laboratory,15 which enhances the contribution of signal from the inner half of the LV wall. For both methods, a repetition rate of 3 seconds was used, and 200 transients were collected for each spectrum. The time required for each MRS acquisition was 20 minutes.
The summed free induction decays were processed by an exponential multiplication of 15 Hz, a convolution difference of 200 Hz, and phasing with zero and first-order phase correction. The peak areas of the PCr and the α- and β-phosphates of ATP were determined by Lorentzian deconvolution with Phillips Medical Systems software, and the ratio of PCr and ATP peak areas (PCr/ATP) was determined by dividing the PCr resonance area by the mean of the α- and β-ATP areas. The effect of partial saturation was examined in three animals by determining the PCr/ATP ratio and spectra obtained by 3- and 15-second repetition rates. The fully relaxed PCr/ATP ratio was found to be 1.43 times greater than the partially saturated ratio in both one-pulse and Fourier series window acquisitions. The saturation factor was not different under basal versus high work state. However, because saturation factors were not measured in every animal, the data are presented without correction for partial saturation. In addition, no correction was made for the contribution of NAD+ and NADH to the α-ATP resonance because the sum would not be expected to change.
The significance of differences between measurements obtained at baseline and with each intervention in the two groups was examined with repeated-measures ANOVA.23 When a significant change (P<.05) was identified by the F test, the Newman-Keuls multiple comparison procedure was used to evaluate the significance of differences between individual groups and conditions. In addition, the changes in certain variables between baseline and dobutamine or atrial pacing are of interest as potential indicators of metabolic or functional alterations related to LVH. Paired Student’s t tests were used to identify significant intergroup differences in the magnitude of change in these variables between baseline and stress conditions. All data are reported as mean±SEM. A probability value <.05 was used as the threshold for statistical significance. Values of .05<P<.10 are indicated for informational purposes.
Degree of LVH
Table 1⇓ shows the values for LV weight and body weight and the ratio of LV to body weight. Of the 17 banded pigs, 13 met the predetermined criterion for LVH of a ratio of LV to body weight greater than the mean+2 SD of the controls (≥2.80 g/kg), and these 13 animals constitute the experimental group. In these animals, the ratio was 38% greater than in the control group (3.31±0.06 versus 2.40±0.06 g/kg), and there was no overlap between animals in the two groups. Similarly, mean LV wall thickness measured post mortem was 42% greater in the LVH animals (17.3±0.6 versus 12.3±0.3 mm).
Hemodynamics and LV Function
Table 2⇓ gives the hemodynamic measurements at baseline and during dobutamine infusion. There were no significant differences between control and LVH animals in the basal state, except for the higher LV systolic pressure in the banded LVH group.
Dobutamine was infused at a mean rate of 14 μg/kg · min in the control pigs and 13.5 μg/kg · min in the LVH animals. This resulted in a significant increase in heart rate and LV systolic pressure in both groups. Because of the aortic constriction, LV systolic pressure rose more and to much higher levels with dobutamine in the LVH group (262±15 versus 164±9 mm Hg), whereas distal to the aortic band systolic pressure was lower in the LVH group during dobutamine. The magnitude of the changes between baseline and dobutamine in the two groups was different only for LV end-diastolic pressure and segment shortening fraction. With dobutamine, LV end-diastolic pressure declined by 3 mm Hg in the controls, while it rose by 5 mm Hg in the LVH animals (within-group changes were not significant, but P<.01 for intergroup differences). Dobutamine increased the shortening fraction by 0.02 in the control group but decreased the shortening fraction by 0.03 in the LVH pigs (intergroup differences, P<.05). However, the much higher LV systolic pressure in the LVH group confounds the interpretation of changes in systolic shortening.
Blood Flow and O2 Consumption
Table 3⇓ gives the LV anterior wall blood flow and O2 consumption measurements. The LV posterior wall blood flow measurements were similar to the anterior wall measurements in each layer under each experimental condition, indicating that the NMR surface coil and interventricular vein catheter did not affect myocardial perfusion, but these measurements are not shown because no metabolic and functional data were obtained from this region. Myocardial blood flow and its transmural distribution did not differ significantly between the LVH and control pigs, either in the basal state or after adenosine. However, adenosine produced a smaller proportional increase in transmurally averaged blood flow in the LVH pigs (3.5±0.4-fold from baseline, from 1.15±0.12 to 4.04±0.42 mL/g · min) than in the controls (5.5±0.4-fold from baseline, from 0.86±0.12 to 4.69±0.61 mL/g · min, P<.01). Because pressure in the proximal aorta was not measured in the LVH pigs, coronary resistance could not be calculated in that group. However, the much higher LV systolic pressure in the LVH pigs indicates that their mean coronary perfusion pressure was probably higher, suggesting a higher minimal coronary resistance in the LVH pigs. The subendocardial to subepicardial blood flow ratio declined with adenosine in both groups (1.18±0.11 to 0.64±0.05 and 1.10±0.09 to 0.53±0.12 in the control and LVH pigs, respectively), but the intergroup differences were not significant.
During dobutamine infusion, blood flow increased significantly in all layers in both groups, albeit substantially less than with adenosine. Again, there was an insignificant trend toward higher values in the LVH pigs. With dobutamine, the subendocardial to subepicardial blood flow ratio declined modestly in both groups, but these changes were also not significant. Anterior wall O2 consumption rose significantly and to a similar extent in both groups with dobutamine. This change reflected primarily the rise in blood flow.
Table 4⇓ provides the 31P-MRS measurements during the two interventions. At baseline, although there were no significant differences between the two groups, the PCr/ATP ratio measured by both techniques tended to be lower in the LVH pigs (1.32±0.04 versus 1.45±0.07, P=.06, by the one-pulse acquisition; 1.26±0.06 versus 1.39±0.06, P=.09, by the Fourier series window technique). The PCr/ATP ratio declined significantly during dobutamine stimulation in both groups of animals, but the magnitude of change in PCr/ATP with dobutamine was similar.
Substrate Concentrations and Myocardial Substrate Uptake and Use
Table 5⇓ shows the measurements of lactate, glucose, and free fatty acids in arterial and anterior coronary venous blood and the uptake of these substrates calculated from the chemical arterial-venous difference and the anterior wall blood flow. There were no significant differences in the arterial concentrations of any of the substrates, either at baseline or with dobutamine. Baseline plasma insulin concentrations were similar in the LVH and control groups (10±2 versus 8±2 μU/mL, respectively). During dobutamine infusion, insulin levels rose significantly in both groups but to comparable values (25±5 versus 22±4 μU/mL, respectively).
Table 5⇑ also shows the values of isotopically measured anterior wall lactate release. At baseline there was net extraction of lactate in both groups associated with similar small amounts of isotopically measured lactate release. With dobutamine, lactate release increased in both groups. The greater mean lactate release during dobutamine in the LVH group was the result of one pig that exhibited a marked increase, but the intergroup differences did not approach significance (P=.19).
The major intergroup difference was the greater glucose use with dobutamine in the LVH pigs. Glucose uptake increased significantly only in the LVH group, and both the magnitude of the increase in glucose uptake (0.52±0.13 versus 0.21±0.08 μmol/g · min, P<.05) and total glucose uptake during dobutamine (0.84±0.09 versus 0.59±0.08 μmol/g · min, P<.05) were significantly greater in the LVH group. In contrast to the differences in glucose use, increases in lactate and free fatty acid uptake during dobutamine infusion were similar in both groups.
Table 6⇓ shows the measurements of glucose uptake and the metabolic fate of glucose in the five pigs from each group studied with 14C-glucose. There were no significant differences in the baseline measurements of glucose uptake, although a higher proportion of the extracted glucose was oxidized (66±4% versus 48±5%, P<.05). During dobutamine, anterior wall glucose uptake tended to be higher in the LVH pigs (1.19±0.21 versus 0.71±0.15 μmol/g · min, P=.07), and glucose oxidation was significantly greater (0.60±0.10 versus 0.28±0.08 μmol/g · min, P<.05). In addition to the greater rate of glucose oxidation in the LVH pigs, more exogenous glucose was released as lactate (0.20±0.04 versus 0.09±0.02 μmol/g · min, P<.05) in the LVH pigs, indicating higher anaerobic and aerobic glycolytic flux. The amount of lactate release and the proportion of lactate derived from exogenous glucose did not differ significantly between the groups of LVH pigs and controls studied with 14C-glucose; however, the small number of animals and the variability of these measurements limit the power of this study to detect such differences.
Changes During Atrial Pacing
Table 7⇓ shows the measurements in the six animals from each group that underwent atrial pacing. Although heart rate increased comparably in both groups, this increase was accompanied by an increase in myocardial blood flow and anterior wall O2 consumption only in the control animals. In contrast, blood flow and O2 consumption actually declined in the LVH animals. This paradoxical response most likely reflects a decrease in LV volume and wall stress, which is suggested by the decline in end-diastolic segment length. The absence of any increase in O2 consumption in the LVH pigs and the modest increase in the control group account for the absence of any significant changes in PCr/ATP or lactate release in either group with atrial pacing.
This study was designed to determine whether pigs with a moderate degree of LVH manifest metabolic evidence of ischemia either in the baseline state or during stress. Somewhat surprisingly, despite the presence of significant impairment of coronary reserve in the LVH animals, compared with controls they did not manifest greater changes in either PCr/ATP or isotopically measured lactate release in response to a nearly threefold increment in myocardial oxygen requirements. However, the LVH animals did exhibit a significantly greater increment in glucose uptake during dobutamine stress than control pigs that was associated with a greater increase in both glucose oxidation and glucose metabolism to lactate.
Failure to Detect Increased Vulnerability to Myocardial Ischemia in Moderate LVH
The primary observation in the present study was the similar changes in both PCr/ATP and lactate release during dobutamine infusion in the LVH and control pigs, arguing against a greater vulnerability to myocardial ischemia in pigs with moderate LVH under these conditions. This lack of metabolic evidence for greater ischemia in LVH is particularly noteworthy because the LVH animals achieved a significantly higher LV rate-pressure product and exhibited some evidence of systolic and diastolic LV dysfunction relative to the control group (a decrease in segment shortening and a rise in end-diastolic pressure). The greater wall thickness in the LVH group most likely normalized wall stress in the LVH animals despite their higher LV systolic pressure.24 This, in turn, may explain the comparable values of MV̇o2 per gram in the two groups despite the substantially higher double product in the LVH pigs. It should be noted, however, that although MV̇o2 per gram was similar in the two groups, the MV̇o2 per heart was much higher in the LVH pigs, as expected.
It should be noted that lactate release increased and PCr/ATP fell modestly in both groups. The findings in the control group are consistent with our previous report of metabolic responses to increased work state induced by dobutamine infusion or dobutamine plus aortic constriction.17 Because the metabolic changes in this earlier study were associated with a decrease in the subendocardial to subepicardial blood flow ratio, as was also the case in the present study, one interpretation is that they may reflect “demand” ischemia of the subendocardium in normal pigs under these experimental conditions. Importantly, in the present study, despite the large increment in MV̇o2 with dobutamine in the LVH pigs, they did not exhibit greater changes in either metabolic index than the control group. Of note, dobutamine stimulation in the LVH pigs increased MV̇o2 per gram to a value (13.1±1.6 μmol/g · min) intermediate between that obtained in normal pigs with dobutamine stimulation and the combination of dobutamine stimulation plus aortic constriction in the previous study (11.5±1.1 and 14.9±1.2 μmol/g · min, respectively); this was associated with changes in the endocardial to epicardial ratio, PCr/ATP, and lactate release in the LVH pigs, which were also intermediate in magnitude to those observed with these increases in work state in normal pigs. This correspondence provides further evidence that the metabolic changes in the present study are more closely related to work state than to LVH.
Although the metabolic changes with stress did not differ between groups, there was a trend toward a lower baseline PCr/ATP ratio in the LVH pigs compared with the controls that approached statistical significance (P=.09). A similar finding was reported in dogs with more severe LVH with 31P-MRS25 and is also consistent with some26 27 28 29 30 but not all29 31 32 33 measurements in hypertrophied animals and humans. The mechanism for such a reduction in baseline PCr/ATP is unclear, but because biochemical analyses on tissue samples have shown a reduction in myocardial ATP content, it is likely that both PCr and ATP concentrations are lower.25 One possibility is that a lower PCr/ATP ratio may be the result of chronic or intermittent ischemia. However, the available evidence from the present study and from previous work by others does not support this explanation.25 Despite the limitation of coronary vasodilator reserve, baseline myocardial blood flow was not reduced in the LVH pigs, and its transmural distribution was preserved. Although impaired O2 diffusion in the absence of blood flow abnormalities could also cause myocardial ischemia, the lack of evidence of increased anaerobic glycolysis at rest or with increased O2 requirements weighs against ischemia arising from this mechanism. Zhang et al25 found that increasing blood flow with the coronary vasodilator adenosine had no effect on the depressed PCr/ATP ratio of dogs with LVH and concluded that a lower basal PCr/ATP did not represent ischemia.
Relation of Present Study to Earlier Reports
The present results differ from several earlier studies that used blood flow or contractility measurements as markers of ischemia. In dogs with severe LVH, the increases in subendocardial blood flow during atrial pacing, catecholamine infusion, coronary vasodilation, and exercise were often less than those seen in other myocardial layers, implying impairment of subendocardial reserve.3 4 5 6 In subsets of these animals that exhibited evidence of heart failure, contractile responses during these stresses were also abnormal, especially in the subendocardium.9 10 11 However, even in these studies, the absolute blood flow to the subendocardium was usually preserved, and the functional measurements must be interpreted with caution because loading conditions in the LVH dogs differed from those in the control dogs, as they did in the present study. Relatively few studies have sought metabolic evidence of myocardial ischemia in LVH in vivo. Using atrial pacing in dogs, Bache et al7 found evidence of net chemical myocardial lactate production in some but not all animals at the highest pacing rate. More recently, Zhang et al25 studied dogs with severe LVH induced by supravalvular banding. These workers found a decline in subendocardial PCr/ATP in LVH animals at a pacing rate of 240 that was associated with a blunted rise in subendocardial blood flow and a significant reduction in the subendocardial to subepicardial blood flow ratio. Although it is likely that these metabolic changes reflected myocardial ischemia in dogs with LVH, it should be noted that these animals had significantly higher LV rate-pressure products, higher LV end-diastolic pressure, and most likely higher LV wall stress than the control dogs, in contrast to the response to atrial pacing in the current study. In addition, the canine LVH model used by Zhang et al25 and in other studies caused a much greater degree of LVH than was present in the current study. However, the degree of LVH in the present study is more characteristic of that encountered in patients with essential hypertension, whereas the 75% to 100% increases in LV mass in the canine models exceed those seen in all but the most severely hypertrophied patients.1 Another potentially relevant difference is the duration of LVH. In the present study, animals were studied approximately 3 months after aortic banding. In the canine studies, measurements were typically performed 1 year or more after banding. A longer duration of pressure overload may have caused not only a more severe degree of LVH but also more severe myocardial fibrosis and greater changes in LV metabolism. Finally, higher levels of stress than those achieved in the present study may still have provoked ischemia.
The response of the coronary circulation and potentially myocardial metabolism may vary between models of LVH. Banding of the ascending aorta is used widely because it can induce significant hypertrophy without impairing coronary perfusion.3 4 5 7 8 9 10 11 25 In this regard, these results should not be extrapolated to valvular aortic stenosis, where coronary perfusion pressure may be impaired. This model also differs from chronic hypertension because in the current model, diastolic pressure in the proximal aorta does not rise in proportion to systolic pressure. As a result of this latter difference, coronary reserve may be reduced to a greater extent with aortic banding than with hypertension. Thus, it is not likely that systemic hypertension producing an equivalent degree of LVH would be more predisposed to ischemia.
Interspecies differences may also play a role in the differing metabolic findings in the present study. Our previous work demonstrated that the myocardial metabolic response to an increased work state in pigs differs significantly from that in dogs.17 34 35 Plasma free fatty acid concentrations and myocardial free fatty acid uptake are lower under basal conditions in pigs than in dogs.36 Furthermore, important differences are present in the coronary circulation of pigs and dogs.37
It is also possible that the techniques used in the current study were not sensitive enough to detect localized areas of ischemia. The spectroscopic techniques used afford only partial subendocardial localization.15 Furthermore, because of possible changes in the size or geometry of the left ventricle during dobutamine infusion, it is possible that spectra may not have been acquired from the same myocardial region throughout the protocol. However, the isotopic technique for measuring myocardial lactate release has proved to be very sensitive, both in animal models and in patients,12 18 and the consistency of the findings with the two approaches adds confidence to the MRS results. Finally, higher levels of stress than those achieved in this study might have provoked ischemia.
Increased Glycolysis in LVH Pigs
There were significant differences in the metabolism of exogenous glucose in the LVH pigs. At baseline, the proportion of glucose oxidized, but not the total amount of glucose uptake, was higher in the LVH pigs. More importantly, during dobutamine stress, both exogenous glucose uptake and oxidation were significantly greater in LVH pigs, as was glucose metabolism to lactate. Although fatty acid uptake rose in both groups with dobutamine, the rate of free fatty acid oxidation was not measured. Therefore, it is not possible to determine whether the increase in glucose use represents a change in substrate preference in the LVH pigs, but this possibility raises several intriguing issues.
Although increased glycolytic metabolism occurs with ischemia, the concurrent increase in glucose oxidation suggests an alternative explanation for enhanced glucose metabolism in LVH. Both increased glycolytic enzyme activity and enhanced glucose uptake have previously been observed in animal models of LVH.38 39 40 Indeed, an increase in glycolytic capacity accompanies chronic pressure overload, even in the absence of LVH.41 A reduction in fatty acid oxidation has also been noted in homogenates of hypertrophic hearts,42 but the activity of enzymes of β-oxidation has been variably affected.40 41 A preference for glucose over fatty acids in hypertensive rats has also been demonstrated by autoradiographic techniques.43 However, none of these findings provide conclusive evidence for altered substrate use in vivo. The present study extends these observations to a large mammalian model and provides direct evidence of increased glucose oxidation and uptake. The mechanism for this apparent preference for glucose remains to be determined. One potential explanation, hyperinsulinemia, does not appear to be operating because there were no significant differences in insulin concentrations between the groups. Differences in the number or accessibility of glucose transporters could also be responsible, but these have not yet been studied in hypertrophy models.
A difference in substrate preference could provide a partial explanation of the lower PCr/ATP ratio in LVH. A decrease in PCr/ATP has been found when glucose is the sole substrate in perfused hearts.44 The relevance of this finding in vivo, where a mixture of substrates is available, is uncertain, and the differences in glucose use observed under these experimental conditions were relatively minor. However, this laboratory has recently demonstrated that during isoproterenol stress, PCr/ATP in the right ventricle is significantly lower when free fatty acid oxidation is inhibited and glucose becomes the predominant substrate.45
Thus, the present study suggests that pigs with moderate LVH and reduced coronary reserve do not exhibit greater metabolic evidence of ischemia under conditions of moderate increases in workload than control animals. However, it is quite possible that even a moderate degree of LVH could increase the predisposition to ischemia if coronary stenoses were present because of diminished downstream perfusion pressure. Certainly, the majority of coronary events in hypertensive patients occur in individuals with accompanying atherosclerotic coronary artery disease, and such synergism may very well be operative.
It is also possible that ischemia may occur as LVH becomes more severe or more long-standing. In the canine model, this may be a factor in the evolution of cardiomyopathy and heart failure observed in some but not all animals. Further studies are necessary to elucidate whether ischemia is an important factor in the evolution of heart failure in patients and experimental models with pressure-overload LVH.
Finally, the present study indicates that significant differences in myocardial metabolism occur with a clinically relevant degree of LVH. Even after 3 months, pigs with moderate LVH exhibit reduced basal PCr/ATP and increased use of exogenous glucose. The mechanism of these changes and their physiological significance remain to be determined.
This work was supported in part by NIH grants HL-28547, HL-02155, HL-07544, and HL-49944 and the Department of Veterans Affairs Medical Research Service. We would like to acknowledge the contributions of Sean Steinman, Maria Mayr, and Gail Cassifer, without whose technical assistance this work would not have been possible.
- Received August 9, 1994.
- Revision received October 13, 1994.
- Accepted October 31, 1994.
- Copyright © 1995 by American Heart Association
Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Prognostic implications of echocardiographically determined left ventricular mass in the Framingham Heart Study. N Engl J Med. 1990;332:1561-1566.
Bache RJ, Vrobel JR, Ring WS, Emery RW, Anderson RW. Regional myocardial blood flow during exercise in dogs with chronic left ventricular hypertrophy. Circ Res. 1981;49:76-87.
Bache RJ. Effects of hypertrophy on the coronary circulation. Prog Cardiovasc Dis. 1988;31:412-431.
Massie BM. Hypertensive left ventricular hypertrophy: pathophysiology and response to therapy. In: Parmley WW, Chatterjee K, eds. Cardiology. Philadelphia, Pa: Lippincott; 1992;2; chap 25.
Bache RJ, Arentzen CE, Simon AB, Vrobel JR. Abnormalities in myocardial perfusion during tachycardia in dogs with left ventricular hypertrophy. Circulation. 1984;69:409-417.
Fujii AM, Vatner SF, Serur J, Als A, Mirsky I. Mechanical and inotropic reserve in conscious dogs with left ventricular hypertrophy. Am J Physiol. 1986;251:H815-H823.
Fujii AM, Gelpi RJ, Mirsky I, Vatner SF. Systolic and diastolic dysfunction during atrial pacing in conscious dogs with left ventricular hypertrophy. Circ Res. 1988;62:462-470.
Hittinger L, Shannon RP, Kohin S, Lader AS, Manders WT, Patrick TA, Kelly P, Vatner SF. Isoproterenol-induced alterations in myocardial blood flow, systolic and diastolic function in conscious dogs with heart failure. Circulation. 1989;80:658-668.
Hittinger L, Shannon RP, Kohin S, Manders WT, Kelly P, Vatner SF. Exercise-induced subendocardial dysfunction in dogs with left ventricular hypertrophy. Circ Res. 1990;66:329-343.
Guth BD, Wisneski JA, Neese RA, White FC, Heusch G, Mazer CD, Gertz EW. Myocardial lactate release during ischemia in swine: relation to regional blood flow. Circulation. 1990;81: 1948-1958.
Schaefer S, Camacho SA, Gober J, Obregon RG, DeGroot MA, Botvinick EH, Massie BM, Weiner MW. Response of myocardial metabolites to graded regional ischemia: 31P NMR spectroscopy of porcine myocardium in vivo. Circ Res. 1989;64:968-976.
Schaefer S, Schwartz GG, Gober J, Wong A, Camacho SA, Massie BM, Weiner MW. Relationship between myocardial metabolites and contractile abnormalities during graded regional ischemia: 31P NMR spectroscopy of porcine myocardium in vivo. J Clin Invest. 1990;85:706-713.
Path G, Robitaille R-M, Merkle H, Tristani M, Zhang J, Garwood M, From AHL, Bache RJ, Ugurbil K. Correlation between transmural high energy phosphate levels and myocardial blood flow in the presence of graded coronary stenosis. Circ Res. 1990;67: 660-673.
Massie BM, Schwartz GG, Garcia J, Wisneski JA, Weiner MW, Owens T. Myocardial metabolism during increased workstates in the porcine left ventricle in vivo. Circ Res. 1994;74:64-73.
Gertz EW, Wisneski JA, Neese RA, Bristow JD, Searle GL, Hanlon JT. Myocardial lactate metabolism: evidence for lactate release during net chemical extraction in man. Circulation. 1981;63:1273-1279.
Wisneski JA, Gertz EW, Neese RA, Gruenke LD, Morris DL, Craig JC. Metabolic fate of extracted glucose in normal human myocardium. J Clin Invest. 1985;76:1819-1827.
Fleischer WR. Enzymatic methods for lactic and pyruvic acid. In: MacDonald RP, ed. Standard Methods in Clinical Chemistry. New York, NY: Academic Press, Inc; 1970;6:24-59.
Barthelmai W, Czok R. Exzymatische bestimmungen der glucose in blut, liquor and harn. Klin Wachenschr. 1962;40:585-589.
Ko H, Royer ME. A gas-liquid chromatographic assay for plasma free fatty acids. J Chromatogr Sci. 1974;88:253-263.
Zar JH. Biostatistical Analysis. Englewood Cliffs, NJ: Prentice-Hall, Inc; 1974:151-155.
Zhang J, Merkle H, Hendrich K, Garwood M, From AHL, Ugurbil K, Bache RJ. Bioenergic abnormalities associated with severe left ventricular hypertrophy. J Clin Invest. 1993;92:993-1003.
Fox AC, Wikler NS, Reed GE. High energy phosphate compounds in the myocardium during experimental congestive heart failure: purine and pyrimidine nucleotide, creatine, and creatine phosphate in normal and failing hearts. J Clin Invest. 1965;44:202-218.
Pool PE, Spann JF Jr, Buccino RA, Sonnenblick EH, Braunwald E. Myocardial high energy phosphate stores in cardiac hypertrophy and heart failure. Circ Res. 1967;21:365-373.
Cooper J, Satava RM, Harrison CE, Coleman HN. Mechanisms for abnormal energetics of pressure-induced hypertrophy in cat myocardium. Circ Res. 1973;133:214-223.
Scheuer J. Metabolic factors in myocardial failure. Circulation. 1973;87(suppl VII):VII-54-VII-57.
Swain JL, Serbina RL, Peyton RB, Jones RN, Wechsler AS, Holmes EW. Derangements in myocardial purine and pyrimidine nucleotide metabolism in patients with coronary artery disease and left ventricular hypertrophy. Proc Natl Acad Sci U S A. 1982;79: 655-659.
Bittl JA, Ingwall JS. Intracellular high-energy phosphate transfer in normal and hypertrophied myocardium. Circulation. 1987;75(suppl I):I-96-I-101.
Wexler LF, Lorrell BH, Monomura S-I, Weinberg EO, Ingwall JS, Apstein CS. Enhanced sensitivity to hypoxia-induced diastolic dysfunction in pressure-overload left ventricular hypertrophy in the rat: role of high energy phosphate depletion. Circ Res. 1988;62: 766-775.
Conway MA, Radda G. Detection of low phosphocreatine to ATP ratio in failing hypertrophied human myocardium by 31P magnetic resonance spectroscopy. Lancet. 1991;388:973-976.
Balaban RS, Kantor HL, Katz LA, Briggs RW. Relation between work and phosphate metabolites in the in vivo paced mammalian heart. Science. 1986;232:1121-1123.
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.
Barrett EJ, Schwartz RG, Francis CK, Zaret BL. Regulation by insulin of myocardial glucose and fatty acid metabolism in the conscious dog. J Clin Invest. 1984;74:1073-1079.
Weaver ME, Pantely GA, Bristow JD, Ladley HD. A quantitative study of the anatomy and distribution of coronary arteries in swine in comparison with other animals and man. Cardiovasc Res. 1986;20:907-917.
Bishop SP, Altschuld RA. Increased glycolytic metabolism in cardiac hypertrophy and congestive failure. Am J Physiol. 1970;218:153-159.
Smith SH, Kramer MF, Reis I, Bishop SP, Ingwall JS. Regional changes in creatine kinase and myocyte size in hypertensive and non-hypertensive cardiac hypertrophy. Circ Res. 1990;67:1334-1344.
Taegtmeyer H, Overturf ML. Effects of moderate hypertension on cardiac function and metabolism in the rabbit. Hypertension. 1988;11:416-426.
Wittels B, Spann JF Jr. Defective lipid metabolism in the failing heart. J Clin Invest. 1968;47:1787-1794.
Yonekura Y, Brill AB, Som P, et al. Regional myocardial substrate uptake in hypertensive rats: a quantitative autoradiographic measurement. Science. 1985;227:1494-1496.
Schwartz GG, Greyson C, Wisneski JA, Garcia J. Inhibition of fatty acid metabolism alters myocardial high-energy phosphates in vivo. Am J Physiol. 1994;267:H224-H231.