INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

LATE PRECONDITIONING AGAINST myocardial stunning develops within 24 h after an episode of brief ischemia and lasts for at least 60 h after its appearance (20, 21, 23). This was evidenced by the fact that when performing a second brief ischemic period 24 h after the first, the severity of stunning was strongly reduced. In these studies, myocardial stunning was induced by brief periods of myocardial ischemia (coronary artery occlusions, CAOs, followed by reperfusion) (5). Oxygen free radicals are generated during reperfusion and participate in the genesis of stunning (19) and the development of late preconditioning (22). An upregulation of nitric oxide production in the resistance coronary vessels also occurs (11) and nitric oxide triggers and mediates late preconditioning against myocardial stunning (3).

Myocardial stunning, however, can also be induced by exercise-induced ischemia (high-flow ischemia) (24, 25). This experimental procedure is different from that of CAO-induced stunning (1) because during exercise there is an increase in myocardial oxygen consumption (MO₂) associated with limited coronary flow due to the presence of a partial coronary stenosis. During the latter, MO₂ is higher than during CAO at rest, resulting in a myocardial ischemic insult despite a maintained substantial perfusion allowing continued oxygen delivery (10) and the removal of anaerobic metabolism products. Although both exercise-induced ischemia and CAO at rest induce myocardial stunning, the role of oxygen free radicals in its genesis during exercise-induced ischemia is unlikely (8), and it is not known whether late preconditioning develops after exercise-induced ischemia.

Accordingly, the main goal of this study was to investigate this issue, i.e., whether late preconditioning against myocardial stunning occur in a model of exercise-induced ischemia. For this purpose, dogs were chronically instrumented with occluders to temporarily realize a partial coronary stenosis before and during exercise, and exercise-induced ischemia was performed two times, 24 h apart. The second goal of our study was to indirectly determine whether an upregulation of nitric oxide production develops in the coronary vessels during ischemia-induced stunning. Therefore, we also investigated coronary vascular endothelial function.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Surgical preparation. Studies were performed in 21 mongrel dogs (mean weight, 25 ± 1 kg). The animals were anesthetized with pentobarbitone sodium (30 mg/kg iv), intubated, and ventilated. Under sterile surgical conditions, a left thoracotomy was performed through the fifth intercostal space. Fluid-filled Tygon catheters were implanted in the descending thoracic aorta and in the left atrium for measurement of pressure and for fluorescent microsphere injection (Fig. 1). A solid-state pressure transducer (P7A, Konigsberg Instruments, Pasadena, CA) was introduced into the left ventricle (LV). A 10-MHz Doppler flow probe (Crystal Biotech, Hopkinton, MA) and two pneumatic occluders were implanted on the left circumflex coronary artery. Two pairs of ultrasonic crystals were used for measurement of left ventricular wall thickening in the distribution of the left circumflex (ischemic zone) and left anterior descending (nonischemic zone) coronary arteries. One crystal was implanted in the endocardium and the other was sutured to the epicardium. All catheters and wires were exteriorized between the scapulae, and the pneumothorax was evacuated. Cefazolin (1 g iv) and gentamicin (40 mg iv) were administered before and at the end of surgery. The position of all catheters and crystals was confirmed at autopsy. The animal instrumentation and the ensuing experiments were performed in accordance with the official regulations of by the French Ministry of Agriculture.

View larger version (41K): [in this window] [in a new window]   Fig. 1.   Cardiac surgical instrumentation. Phasic waveforms are shown (right). LV, left ventricular; dP/dt, change in pressure over time.

Hemodynamic measurements. Data were recorded on a multichannel recorder (ES 1000, Gould Instruments, Cleveland, OH), digitized on a computer, and analyzed by using the data acquisition software HEM version 1.5 (Notocord Systems, Croissy sur Seine, France). Each measurement was performed over a 2-min period of recording.

Regional myocardial blood flow. Regional myocardial blood flow (MBF) was measured by using the fluorescent microspheres technique. Microspheres (3 × 10⁶, 15 ± 1 µm) labeled with fluorescent dye (blue, blue-green, yellow, orange, red, crimson, or far-red) suspended in 0.02% Tween 80 solution (Triton Technology) were sonicated for 20-30 min and vortexed. Arterial blood reference samples were withdrawn at a rate of 7.75 ml/min for a total of 120 s, and microspheres were injected and flushed with saline over a 20-s period via the left atrial catheter. At termination of the study, the animal was given heparin and a lethal dose of pentobarbital sodium. The heart was excised and a dual perfusion was performed. The ascending aorta was cannulated and retrograde perfused (120 mmHg) with 1% Monastral blue dye. The left circumflex coronary artery was cannulated at the site of the occluders and perfused with saline. Afterward, the left ventricle was cut into three to four slices and further divided into endocardium, midmyocardium, and epicardium in the nonischemic and ischemic zones.

1

1

Experimental protocols. All experiments were performed 3 wk after surgery. Before final inclusion in the protocol, each dog performed two training treadmill exercises without stenosis (14 km/h, 13% slope, 10 min) 24 h apart to verify their reproducibility. During these exercises, LV wall thickening increased significantly up to 43 ± 5 from 23 ± 2% and to 39 ± 4 from 21 ± 4% on days 1 and 2, respectively.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Setting of the partial stenosis. Left: the first occluder was deflated and the second occluder was used to perform a 20-s coronary artery occlusion (without stenosis). The release of the occluder was followed by reactive hyperemia. Right: a partial stenosis was induced with the first occluder to abolish the reactive hyperemic response after the 20-s coronary artery occlusion realized with the second occluder (with stenosis).

Statistical analysis. The data reported are means ± SE. Statistical analysis was performed by using StatView and SuperANOVA (Abacus Concepts, Berkeley, CA). Comparisons between the two treadmill exercises and recovery periods were performed by using a two-way ANOVA for repeated measures, followed, if necessary, by contrast analysis corrected for repetition (Bonferroni method). The coronary reactivity on day 1 and day 2 was analyzed by using a two-way ANOVA for repeated measures followed by a paired Student's t-test. A P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Hemodynamics. Hemodynamic data are summarized in Table 1. None of the parameters was affected by stenosis compared with baseline values. During exercise, all parameters increased significantly. Whereas LV end-diastolic pressure increased by 3 ± 3 mmHg during a test exercise without stenosis, it rose significantly to 26 ± 2 mmHg during exercise in the presence of stenosis (P < 0.05). On day 1, heart rate was 120 ± 4 beats/min at rest, increased significantly to 221 ± 10 beats/min during exercise, and remained statistically elevated during the first 30 min after exercise. Mean arterial pressure, LV pressure, LV end-diastolic pressure, LV dP/dt, and rate-pressure product recovered within 5 min after exercise completion. On day 2, the values of all hemodynamic parameters investigated at baseline, during exercise, and during recovery did not differ significantly from those measured on day 1.

LV regional MBFs. MBFs measured at baseline, at rest under stenosis, during exercise, and during recovery on day 1 are summarized in Table 2. At baseline, MBFs were not significantly different in the nonischemic and ischemic zones. Coronary artery stenosis did not significantly affect MBFs. During exercise, transmural MBF in the nonischemic zone increased markedly; in contrast, in the ischemic zone, transmural MBF remained constant. At rest, there was a transmural gradient of myocardial perfusion favoring the subendocardium in both nonischemic and ischemic zones, with an endocardium/epicardium ratio of 1.40 ± 0.04 and 1.42 ± 0.11, respectively. During exercise, however, this perfusion gradient fell substantially to 0.74 ± 0.06 (P < 0.05) in the ischemic zone, whereas it remained similar in the nonischemic zone (1.37 ± 0.03). After exercise completion, MBFs returned to their respective resting values within 1 h. In the ischemic zone, the release of the occluder did not significantly change MBFs. On day 2, MBF in both ischemic and nonischemic zones were not significantly different from day 1 (Table 2). The ratios of ischemic to nonischemic blood flows calculated during exercise revealed no significant differences between days 1 and 2 (transmural, 0.49 ± 0.10 vs. 0.46 ± 0.09, respectively).

LV regional myocardial function. As illustrated in Fig. 3, partial coronary artery stenosis did not significantly change LV regional myocardial function in the ischemic zone. During exercise, LV wall thickening increased significantly by 71 ± 11 from 23 ± 2% in the nonischemic zone, whereas it decreased dramatically in the ischemic zone (84 ± 5 from 27 ± 3%) (Table 3). LV wall thickening returned to the corresponding baseline value within 10 min after exercise completion in the nonischemic zone. In contrast, ischemic LV wall thickening remained significantly depressed 1 h after the end of exercise, whereas MBF returned to baseline, indicating myocardial stunning. Ten hours later, myocardial function in the ischemic zone had recovered 85 ± 2% of its baseline value (23 ± 3 and 27 ± 3%, respectively), and statistically completely recovered at 24 h (25 ± 3%). Comparison of day 1 with day 2 revealed no significant differences in myocardial function in either ischemic or nonischemic zones. Time courses for recovery in ischemic LV wall thickening on days 1 and 2 were superimposable (Fig. 3). In addition, the total deficits in systolic wall thickening were not significantly different between days 1 and 2 (442 ± 73 vs. 423 ± 86 arbitrary units, respectively).

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Plots of wall thickening changes, expressed as percent reduction from baseline (B), during stenosis at rest (St), during exercise (Ex), and on day 1 during the recovery period (open symbols) and on day 2 (solid symbols) in the nonischemic zone (top) and in the ischemic zone (bottom) (n = 9 dogs).

Coronary artery responses to acetylcholine and nitroglycerin and during reactive hyperemia. As summarized in Table 4, administration of both acetylcholine and nitroglycerin decreased mean arterial pressure and increased heart rate similarly on days 1 and 2. As illustrated in Fig. 4, decreases in coronary vascular resistance were similar on days 1 and 2 in response to acetylcholine and nitroglycerin. Also, peak blood flow velocity values after the release of the 20-s CAO were not significantly different on days 1 and 2 (Fig. 4). Flow repayment-to-flow debt ratio did not significantly differ between day 2 (3.2 ± 0.3) and day 1 (3.6 ± 0.2).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Decreases in coronary vascular resistance (%) in response to nitroglycerin (3-30 µg/kg iv) and acetylcholine (0.3-3 µg/kg iv), and peak increases in coronary blood flow velocity during reactive hyperemia on day 1 (open circles, open bar) and on day 2 (solid circles, hatched bar) at baseline.

Additional studies. In the subgroup of four dogs in which a more severe stenosis was realized, MBFs were transmurally reduced to 0.52 ± 0.17 from 1.20 ± 0.10 ml · min¹ · g¹ and in the subendocardium layer to 0.34 ± 0.15 from 1.50 ± 0.24 ml · min¹ · g¹ (n = 4). As illustrated in Fig. 5, LV wall thickening was abolished during exercise (105 ± 8%) but returned completely to its baseline value at 24 h. Comparison of day 1 and day 2 revealed no significant difference in myocardial function either in nonischemic or in ischemic zones (Table 5). In addition, total deficits in systolic wall thickening were not significantly different between days 1 and 2 (536 ± 114 vs. 486 ± 103 arbitrary units, respectively).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Plots of wall thickening changes expressed as percent reduction from B, St, Ex, and during the recovery period on day 1 (open circles) and day 2 (solid circles) in the ischemic zone. Myocardial blood flows were severely reduced during exercise, and the stenosis was released just after the end of exercise on day 1 (n = 4 dogs). Although LV wall thickening was abolished during exercise, no differences were observed between day 1 and day 2.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Plots of wall thickening changes expressed as percent reduction from B, during coronary artery occlusion (CAO), and during the recovery period on day 1 (open circles) and day 2 (solid circles) in the ischemic zone (n = 4 dogs). The time course for recovery in wall thickening was significantly reduced on day 2 compared with day 1, demonstrating late preconditioning in a model of occlusion-reperfusion. *P < 0.05 significantly different from day 1.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates that in contrast with what is observed in occlusion-reperfusion models, no late preconditioning against myocardial stunning develops in a model of exercise-induced ischemia in dogs. Comparison of wall thickening recovery between days 1 and 2 shows that the first episode of ischemia (exercise performed under partial coronary stenosis) does not protect against myocardial stunning 24 h later when a similar ischemic episode is repeated.

Coronary artery stenosis was set up at rest before onset of the treadmill exercise without inducing any significant change in mean CBFV and without altering wall thickening in the ischemic zone. Measurement of MBFs confirmed basal perfusion was not significantly changed at that time. However, coronary stenosis created a severe regional imbalance between myocardial metabolic demand and supply, resulting in myocardial ischemia (10, 25). In addition, during exercise, tachycardia and elevated end-diastolic pressure were responsible for changes in MBF distribution (5) resulting in a fall in the endocardium to an epicardium flow ratio (10, 24). Return to a normal balance between myocardial oxygen supply and demand was progressive, and all hemodynamic parameters returned to their corresponding baseline values within 30 min after exercise completion. MBFs measured before and after stenosis relief (1 h after the end of exercise) were not significantly different. However, despite a normal myocardial perfusion, regional myocardial function in the ischemic zone remained depressed for a long time, indicating a true myocardial stunning. Wall motion recovered only after 24 h. On the next day, when the same protocol was repeated, myocardial perfusion in both ischemic and nonischemic zones was not significantly different from day 1. In addition, hemodynamic parameters and wall thickening in the nonischemic zone measured at rest, during exercise, and during the recovery period were similar on days 1 and 2, suggesting that a similar degree of metabolic myocardial demand and of myocardial stunning was achieved.

Ischemic preconditioning effectively limits necrosis (14, 18) but fails to attenuate myocardial stunning (13, 17). It develops within minutes after the first ischemic period and disappears within 2 to 4 h (15). However, a late phase of preconditioning against myocardial stunning develops 24-48 h after the first ischemia (3, 20, 21). In these in vivo studies, myocardial stunning was induced by brief periods of myocardial ischemia (CAO at rest followed by reperfusion) which, when repeated 24 h later, results in reduced stunning severity. The lack of delayed protection in our experimental model contrasts with these previous reports and with our present CAO-reperfusion experiments confirming the existence of late preconditioning against myocardial stunning in the experimental setting of occlusion-reperfusion. It could be hypothesized that the degree of ischemia achieved in our study was insufficient to reach a myocardial preconditioning threshold (12). However, this is unlikely, because in preliminary experiments (see METHODS) using the same protocol, a 20-min exercise induced myocardial stunning lasting more than 24 h, and a 30-min exercise induced severe arrhythmias and patchy necrosis, suggesting that the ischemic insult of the 10-min exercise model was already severe enough. Nevertheless, to further test this hypothesis, experiments using a 10-min exercise protocol were performed in the presence of a more severe stenosis resulting in greater flow deficit during exercise. LV wall thickening was abolished during exercise, but no late preconditioning was observed on day 2. Finally, although unlikely, we cannot completely rule out the possibility that the training sessions performed by our dogs may have induced a late preconditioned state that somehow could have obfuscated the presence of late preconditioning against stunning.

Although both exercise-induced ischemia and CAO at rest followed by reperfusion induce myocardial stunning, the pathophysiological mechanisms might be different (2, 4). Oxygen consumption is higher and a substantial residual perfusion persists during exercise-induced ischemia, which allows continued oxygen delivery (6) and removal of anaerobic metabolism products. Furthermore, reactive oxygen species contribute to the pathogenesis of myocardial stunning consecutive to CAO and subsequent reperfusion (19) and participate in the late preconditioning (21). In contrast, free radicals are unlikely to play a fundamental role in high-flow ischemia since Homans et al. (8) reported that exercise-induced stunning was not alleviated by superoxide dismutase and catalase. It is important to emphasize that our model is characterized by the lack of any "reperfusion phenomenon" and although stenosis was relieved just after the end of exercise in four additional dogs, the presence of a reactive hyperemic response did not induce the development of late preconditioning. Interestingly, an upregulation of coronary vascular nitric oxide production occurs after a brief period of ischemia in coronary resistance vessels in conscious dogs (11); and nitric oxide has been demonstrated to trigger and mediate late preconditioning against myocardial stunning in models of CAO followed by reperfusion (3). Contrasting with these reports, we were unable to detect any significant difference in the CBFV responses to acetylcholine or nitroglycerin, during reactive hyperemia, or during any late preconditioning.

In conclusion, in contrast with experimental models of CAO followed by reperfusion at rest, exercise-induced ischemia does not induce any late preconditioning against myocardial stunning. This finding suggests that the mechanisms responsible for the development of late preconditioning are different from those inducing the regional contractile dysfunction. It is tempting to speculate that late preconditioning against myocardial stunning requires a "zero flow" CAO and/or subsequent reperfusion to upregulate the nitric oxide pathway.