Altered response of reperfused myocardium to repeated coronary occlusion in dogs

Cardiovascular proteomics: past, present, and future

With cardiovascular (CV)-related disorders accounting for the highest mortality rates in the world, affecting the quantity and quality of life of patients and creating an economic burden of prolonged therapeutic intervention, there is great significance in understanding the cellular and molecular alterations that influence the progression of these pathologies. The cellular genotype is regulated by the DNA component, whilst the cellular phenotype is influenced by the protein complement. By improving the understanding of the molecular mechanisms that influence the protein profile, the pathologies that influence the intrinsic functions of the CV system may be detected earlier or managed more efficiently. This is achievable with technologies encompassed by 'proteomics.' Proteomic investigations of CV diseases, including dilated cardiomyopathy (DCM), atherosclerosis, and ischemia/reperfusion (I/R) injury, have identified candidate proteins altered with the pathologic states, complementing past biochemical and physiologic observations. Whilst proteomics is still a relatively new discipline to be applied to the basic scientific investigation of CV diseases, it is emerging as a technique to screen for potential biomarkers in both tissues/cells and biologic fluids (biofluids), as well as to identify the targets of existing therapeutics. By enabling the separation of complex mixtures over numerous dimensions, exploiting the intrinsic properties of proteins, including charge state, molecular mass, and hydrophobicity, in addition to cellular location, the discrete alterations within the cell may be resolved. Proteomics has shown alterations to myofilament proteins including troponin I and myosin light chain, correlating with the reduction in contractility in the myocardium from DCM and I/R. The diverse cell types that coalesce to induce atherosclerotic plaque formation have been investigated both collectively and individually to elucidate the influence of the modifications to single cell types on the developing plaque as a whole. Proteomics has also been used to observe changes to biofluids occurring with these pathologies, a new potential link between basic science and clinical applications. The development of CV proteomics has helped to identify a number of possible protein candidates, and offers the potential to treat and diagnose CV disease more effectively in the future.

Long-term clinical outcomes after rescue angioplasty are not different from those of successful thrombolysis for acute myocardial infarction

AIMS

The long-term value of rescue percutaneous transluminal coronary angioplasty (PTCA) in patients with ST-segment elevation myocardial infarction who received thrombolytic therapy but failed to achieve early recanalization of the artery is still debated. This study aimed to compare long-term outcomes after successful thrombolysis vs. systematic attempted rescue PTCA.

METHODS AND RESULTS

A total of 362 consecutive patients with STEMI hospitalized within 6 h of symptom onset and treated with intravenous thrombolytic therapy were studied. Of these, 345 underwent coronary angiography within 90 min. Sixty per cent of patients achieved TIMI 3 flow and were treated medically; the in-hospital death rate in this group was 4%. Nine per cent of patients had TIMI 2 flow and 31% TIMI 0-1 flow. In this latter group, rescue PTCA was attempted in 85.8% with a hospital death rate of 5.5% (20% with failed vs. 4% with successful rescue PTCA, P=0.03). Eight year actuarial survival without recurrent myocardial infarction was no different in patients who had successful thrombolytic therapy and in patients with attempted rescue PTCA [78 and 95% CI (71-85) vs. 78 and 95% CI (68-87), respectively, hazard ratio: 0.93 (0.52-1.65), P=0.80]. Total mortality, cardiac mortality, and other composite endpoints also did not differ between groups.

CONCLUSION

Routine attempted rescue PTCA 90 min after thrombolytic therapy in patients with persistent occlusion of the infarct-related vessels achieves long-term clinical outcomes which do not differ from those obtained by successful thrombolysis.

Novel mechanisms mediating stunned myocardium

Myocardial stunning is defined as the prolonged contractile dysfunction following an ischemic episode that does not result in necrosis, which also occurs in patients with coronary artery disease. There is also evidence to consider myocardial stunning as a fundamental component of hibernating myocardium. Various experimental approaches (from a brief episode to prolonged partial ischemia) and animal models (from rodents to large mammals) have been developed to investigate the pathogenesis of myocardial stunning. Three hypotheses to explain the mechanism, i.e. oxygen radical, Troponin I degradation, and Ca(2+), have been proposed. The first was tested primarily using large mammalian models, whereas the others were tested primarily using rodent models. Recently, the Ca(2+) handling hyothesis has been tested in a large mammalian swine model of myocardial stunning, in which both Ca(2+) and transients and L-type Ca(2+) current density were decreased. Relaxation function and phospholamban phosphorylation are also radically different in large mammalian and rodent models. In addition, troponin I degradation, which was identified as the mechanism of stunning in rodent models, was not found in stunned swine myocardium. Interestingly, the large mammalian model demonstrates that stunning elicits broad changes in gene and protein regulation, some of which have not been observed in the heart previously. The overall genomic adaptation upregulates the expression of survival genes that prevent irreversible damage. Pursuing these new concepts derived from large mammalian models of ischemia/reperfusion will provide more comprehensive mechanistic information underlying myocardial stunning and will serve to devise new therapeutic modalities for patients.

[Ischemic preconditioning. Is it always a beneficial phenomenon?]

INTRODUCTION AND OBJECTIVES

Hearts exposed to reversible ischemia stand a subsequent prolonged episode of coronary artery occlusion (ischemic preconditioning) better. The reduction of infarct size by means of preconditioning has been amply demonstrated, but the relationship between preconditioning and contractile function remains less well defined. In this study we assess the effect of a later ischemia on the regional contractility in a stunned-preconditioned myocardium.

METHODS

We analyze the shortening fraction in the ischemic (dependent on the left anterior descending coronary artery), periischemic and control zone (dependent on the left circumflex coronary artery), using chronic implants of ultrasonic crystals in 17 adult mongrel dogs. In the control series, we quantified the effects of partial (30-60% reduction of coronary flow from the basal) and transitory (15 minutes) ischemic episode in the regional myocardial function in a "virgin" myocardium. In two other series, the myocardium was previously stunned-preconditioned through brief and repeated ischemias. Afterwards, at 5th day (series B) and at 15th day (series C), the dogs were subjected to ischemic episode similar to control ones.

RESULTS

After comparing the results with the control series, we observed that the shortening fraction of the ischemic zone was decreased by 107% (p < 0.01) during partial ischemic episode when it was induced on the 5th day of the stunning-preconditioning (series B).

CONCLUSIONS

In dogs, the brief and repeated episodes of ischemia could condition the contractile function so that a later partial and transitory reduction of coronary flow could induce a severe affectation of contractility expressed as a diskinetic area.

Molecular and cellular mechanisms of myocardial stunning

The past two decades have witnessed an explosive growth of knowledge regarding postischemic myocardial dysfunction or myocardial "stunning." The purpose of this review is to summarize current information regarding the pathophysiology and pathogenesis of this phenomenon. Myocardial stunning should not be regarded as a single entity but rather as a "syndrome" that has been observed in a wide variety of experimental settings, which include the following: 1) stunning after a single, completely reversible episode of regional ischemia in vivo; 2) stunning after multiple, completely reversible episodes of regional ischemia in vivo; 3) stunning after a partly reversible episode of regional ischemia in vivo (subendocardial infarction); 4) stunning after global ischemia in vitro; 5) stunning after global ischemia in vivo; and 6) stunning after exercise-induced ischemia (high-flow ischemia). Whether these settings share a common mechanism is unknown. Although the pathogenesis of myocardial stunning has not been definitively established, the two major hypotheses are that it is caused by the generation of oxygen-derived free radicals (oxyradical hypothesis) and by a transient calcium overload (calcium hypothesis) on reperfusion. The final lesion responsible for the contractile depression appears to be a decreased responsiveness of contractile filaments to calcium. Recent evidence suggests that calcium overload may activate calpains, resulting in selective proteolysis of myofibrils; the time required for resynthesis of damaged proteins would explain in part the delayed recovery of function in stunned myocardium. The oxyradical and calcium hypotheses are not mutually exclusive and are likely to represent different facets of the same pathophysiological cascade. For example, increased free radical formation could cause cellular calcium overload, which would damage the contractile apparatus of the myocytes. Free radical generation could also directly alter contractile filaments in a manner that renders them less responsive to calcium (e.g., oxidation of critical thiol groups). However, it remains unknown whether oxyradicals play a role in all forms of stunning and whether the calcium hypothesis is applicable to stunning in vivo. Nevertheless, it is clear that the lesion responsible for myocardial stunning occurs, at least in part, after reperfusion so that this contractile dysfunction can be viewed, in part, as a form of "reperfusion injury." An important implication of the phenomenon of myocardial stunning is that so-called chronic hibernation may in fact be the result of repetitive episodes of stunning, which have a cumulative effect and cause protracted postischemic dysfunction. A better understanding of myocardial stunning will expand our knowledge of the pathophysiology of myocardial ischemia and provide a rationale for developing new therapeutic strategies designed to prevent postischemic dysfunction in patients.

Basic and clinical aspects of myocardial stunning

Although the pathogenesis of myocardial stunning has not been definitively established, the two major hypotheses are that it is caused by the generation of oxygen-derived free radicals on reperfusion and by a loss of sensitivity of contractile filaments to calcium. These hypotheses are not mutually exclusive and are likely to represent different facets of the same pathophysiological cascade. For example, a burst of free radical generation after reperfusion could alter contractile filaments in a manner that renders them less responsive to calcium. Increased free radical formation could also cause cellular calcium overload, which would damage the contractile apparatus of the myocytes. There is now considerable evidence that myocardial stunning occurs clinically in various situations in which the heart is exposed to transient ischemia, such as unstable angina, acute myocardial infarction with early reperfusion, exercise-induced ischemia, cardiac surgery, and cardiac transplantation. Recognition of myocardial stunning is clinically important and may impact patient treatment. Although no ideal diagnostic technique for myocardial stunning has yet been developed, thallium-201 scintigraphy or dobutamine echocardiography are available and can be useful to identify viable myocardium with reversible wall motion abnormalities. An intriguing possibility is that so-called chronic hibernation may in fact be the result of repetitive episodes of stunning, which have a cumulative effect and cause protracted postischemic left ventricular dysfunction. A better understanding of myocardial stunning will expand our knowledge of the pathophysiology of myocardial ischemia and provide a rationale for developing new therapeutic strategies designed to prevent postischemic dysfunction.

Role of adenosine in postischemic dysfunction of coronary innervation

We sought to determine the role of adenosine in the sustained but reversible decrease in cardiac neurotransmission that occurs after brief ischemia. Adult mongrel dogs were anesthetized and instrumented for measurements of heart rate, arterial pressure, and left anterior descending coronary artery (LAD) and left circumflex coronary artery (LCX) flow velocities. Changes in coronary vascular resistance were measured during bilateral stimulation of the stellate ganglia. After beta-adrenergic blockade and bilateral vagotomy, stellate stimulation increased coronary vascular resistance in the LAD and LCX beds 28 +/- 2% and 30 +/- 3%, respectively. After a 15-minute infusion of adenosine into the LAD, the peak increase in LAD resistance was significantly reduced (18 +/- 2%) compared with LCX (34 +/- 5%) and control (P < .05, n = 6) resistance. The LAD response after infusion of the vasodilator papaverine was unchanged (n = 6). Intracoronary infusion of adenosine deaminase (n = 10) but not vehicle (n = 5) into the LAD during a 15-minute LAD occlusion prevented the attenuation in constriction to stellate stimulation. We conclude that adenosine, exogenously infused or endogenously produced, is capable of reducing cardiac neurotransmission.

Can myocardial stunning contribute to myocardial infarction during coronary artery bypass surgery?

Myocardial stunning commonly occurs after cardiopulmonary bypass (CPB). Myocardial stunning can be cardioprotective under some conditions, but in other situations may actually contribute to myocardial infarction (MI). Vascular endothelial stunning may be one of the mechanisms by which myocardial stunning can cause MI. It has been found that 15 minutes of reversible ischemia is enough to cause elevations in vascular resistance and impairment of vasodilator responsiveness. However, no correlation between contractile dysfunction and microvascular stunning has been observed. Transduction defects (increased oxygen extraction and consumption despite normal regional oxygen blood flow and delivery) may be another mechanism by which stunning predisposes to MI, indicating uncoupling of substrate utilization from energy production. In addition, abnormalities in wall motion, oxygen free radical toxicity, hypotension, use of inotropic agents (leading to increased oxygen consumption, high heart rates, and arrhythmias) increase the risk of cellular necrosis. Following CPB, flow limitations due to diffuse atherosclerosis in some areas may result in poor contractility, and newly grafted areas have a high probability of becoming ischemic and stunned. These areas are likely to contribute to MI.

Effect of brief myocardial ischemia on sympathetic coronary vasoconstriction

The purpose of the present study was to determine whether sympathetic coronary vasoconstrictor responses are altered after brief ischemia and reperfusion. Adult mongrel dogs were anesthetized and instrumented for measurements of heart rate, arterial pressure, left ventricular pressure, left ventricular dP/dt, anterior myocardial wall thickening, and left circumflex coronary artery (LCX) and left anterior descending coronary artery (LAD) blood flow velocities. Changes in coronary vascular resistance were recorded during intravenous bolus doses of norepinephrine and bilateral electrical stimulation of the stellate ganglia. After beta-adrenergic blockade and bilateral vagotomy, electrical stimulation of the stellate ganglia increased coronary vascular resistance in the LAD and LCX beds by 38 +/- 5% and 39 +/- 5%, respectively. After a 15-minute LAD occlusion, repeat electrical stimulation produced increases in coronary resistance of 16 +/- 3% and 45 +/- 8%, respectively (p less than 0.05 for the LAD before versus after the occlusion). The peak increase in coronary vascular resistance to two doses of norepinephrine was unchanged. After a shorter period of myocardial ischemia (7 minutes), similar increase in coronary resistance to stellate stimulation were observed before (27 +/- 4%) and after (26 +/- 6%) myocardial ischemia. The mechanism of this impaired sympathetic coronary vasoconstriction was further tested by examining the responses to bretylium and tyramine. Brief ischemia did not alter the coronary constrictor responses to either bretylium or tyramine, suggesting that mechanisms governing prejunctional release of norepinephrine are intact in the postischemic coronary arterial bed.(ABSTRACT TRUNCATED AT 250 WORDS)

Clinical and experimental aspects of myocardial stunning

Although the mechanisms involved in stunning remain incompletely defined, it appears that intracellular calcium overload, sarcoplasmic reticulum dysfunction, and the generation of OFR are important components of post-ischemic myocyte dysfunction. It is likely that a variety of mechanisms, some possibly remaining to be elucidated, are operative in the pathogenesis of stunning, and that the contribution of a particular process may be influenced by the model and the method of inducing ischemia. Myocardial stunning has been shown to be prevalent in patients with diverse cardiac diseases. Small clinical trials have suggested that electrocardiography, echocardiography, and radionuclide imaging techniques may be useful in identifying patients with stunned myocardium. In patients with depressed cardiac performance due to stunning, therapy with inotropic agents may recruit the viable but injured myocardium to contract and improve cardiac output in the short term. An important issue that will be addressed over the next decade is whether aggressive therapy aimed at reducing myocardial stunning in stable patients should be attempted. Some authorities have suggested that stunning may represent an adaptive response to limit reperfusion injury, and that interfering with this response may not be beneficial in the long term. Further investigation into the cellular and molecular basis of ischemic injury should provide insight into these and other important aspects of myocardial stunning. Methods of attenuating postischemic ventricular dysfunction that appear convincing in the research laboratory may not translate to clinical benefit when applied to humans.

Preconditioning causes improved wall motion as well as smaller infarcts after transient coronary occlusion in rabbits

BACKGROUND

A brief coronary occlusion before a more prolonged occlusion results in less myocardial infarction than the longer occlusion alone. However, the effects of this preconditioning on recovery of systolic function after coronary occlusion have not been determined.

METHODS AND RESULTS

Ultrasonic crystals implanted in rabbit myocardium measured segment length in the distribution of a branch of the left coronary artery that was fitted with a snare occluder. Rabbits were randomly allocated to either nonpreconditioned or preconditioned groups. Rabbits in the latter group underwent preconditioning with a 5-minute coronary occlusion followed by 10 minutes of reperfusion. Then the coronary artery was occluded for 20 minutes in all rabbits, after which it was allowed to reperfuse for 90 minutes. The hearts were then excised, and infarct size was measured by staining with triphenyltetrazolium chloride. During coronary occlusion, all hearts except one demonstrated either akinesis or paradoxical bulging. Five minutes after release of the 20-minute occlusion, active shortening had returned in the preconditioned rabbits and averaged 27.9 +/- 16.6% of baseline shortening. At the same time, paradoxical lengthening persisted in nonpreconditioned rabbits (-15.5 +/- 19.8% of baseline). By the end of the 90-minute reperfusion period, segment shortening averaged 40.1 +/- 8.4% of baseline in preconditioned rabbits and only 6.2 +/- 12.0% in nonpreconditioned rabbits (p less than 0.05). Infarct size as a percentage of risk area was significantly smaller in preconditioned rabbits as well (3.0 +/- 1.6% versus 28.8 +/- 7.0%, p less than 0.002) and likely accounted for the improved shortening.

CONCLUSIONS

We conclude that a brief coronary occlusion before a more prolonged occlusion results in not only reduced infarct size but also significantly better recovery of systolic function.

Rapid expression of heat shock protein in the rabbit after brief cardiac ischemia

The effect of brief myocardial ischemia on the expression of heat shock protein (HSP 70) was examined in an in vivo rabbit model of myocardial ischemia using Northern blotting. Functional studies were carried out in the open-chested anesthetized rabbit. The large marginal branch of the left circumflex was occluded four times for 5 min. Using piezoelectric crystals implanted midwall in the ischemic zone, end-diastolic length, end-systolic length, and percent segmental shortening were assessed. Expression of HSP 70 was measured by Northern blotting. A single 5-min coronary occlusion doubled the expression of HSP 70 whereas four cycles of 5 min of ischemia/5 min of reperfusion resulted in a threefold increase in HSP 70 mRNA (P less than 0.001). Measurements with the piezoelectric crystals showed mild myocardial dysfunction concomitant with the increase in HSP 70. This increase in HSP 70 mRNA after repetitive brief ischemia was transient, occurring as early as 1 h and returning to baseline by 24 h after ischemia. Western blot analysis with a monoclonal antibody to HSP 70 was used to compare sham and postischemic myocardial HSP 70 levels. Changes in the amount of HSP 70 were evident as early as 2 h and were even more striking at 24 h.

Mechanism of myocardial "stunning"

Among the numerous mechanisms proposed for myocardial stunning, three appear to be more plausible: 1) generation of oxygen radicals, 2) calcium overload, and 3) excitation-contraction uncoupling. First, the evidence for a pathogenetic role of oxygen-derived free radicals in myocardial stunning is overwhelming. In the setting of a single 15-minute coronary occlusion, mitigation of stunning by antioxidants has been reproducibly observed by several independent laboratories. Similar protection has been recently demonstrated in the conscious animal, that is, in the most physiological experimental preparation available. Furthermore, generation of free radicals in the stunned myocardium has been directly demonstrated by spin trapping techniques, and attenuation of free radical generation has been repeatedly shown to result in attenuation of contractile dysfunction. Numerous observations suggest that oxyradicals also contribute to stunning in other settings: after global ischemia in vitro, after global ischemia during cardioplegic arrest in vivo, and after multiple brief episodes of regional ischemia in vivo. Compelling evidence indicates that the critical free radical damage occurs in the initial moments of reflow, so that myocardial stunning can be viewed as a sublethal form of oxyradical-mediated "reperfusion injury." Second, there is also considerable evidence that a transient calcium overload during early reperfusion contributes to postischemic dysfunction in vitro; however, the importance of this mechanism in vivo remains to be defined. Third, inadequate release of calcium by the sarcoplasmic reticulum, with consequent excitation-contraction uncoupling, may occur after multiple brief episodes of regional ischemia, but its role in other forms of postischemic dysfunction has not been explored. It is probable that multiple mechanisms contribute to the pathogenesis of myocardial stunning. The three hypotheses outlined above are not mutually exclusive and in fact may represent different steps of the same pathophysiological cascade. Thus, generation of oxyradicals may cause sarcoplasmic reticulum dysfunction, and both of these processes may lead to calcium overload, which in turn could exacerbate the damage initiated by oxygen species. The concepts discussed in this review should provide not only a conceptual framework for further investigation of the pathophysiology of reversible ischemia-reperfusion injury but also a rationale for developing clinically applicable interventions designed to prevent postischemic ventricular dysfunction.

Myocardial stunning in dogs: preconditioning effect and influence of coronary collateral flow

In open-chest dogs the left anterior descending (LAD) coronary artery diagonal branch was encircled with a pneumatic occluder. Pairs of ultrasonic crystals were inserted into LAD myocardium and remote normal muscle. The coronary artery was occluded for 5 minutes, followed by 10 minutes of reperfusion. This occlusion-reperfusion cycle was repeated 12 times, and after a final 90-minute reperfusion period the hearts were removed and stained with triphenyltetrazolium chloride. No heart had evidence of necrosis. Baseline shortening normalized for end-diastolic length averaged 10.4 +/- 1.0% in the LAD area and 7.4 +/- 0.8% in the remote normal myocardium. When analyzed as a percentage of baseline, segment shortening in the normal myocardium was not significantly altered by LAD occlusion and reperfusion. In contrast, during occlusions the LAD myocardium paradoxically lengthened. With the initial reperfusion, shortening was significantly depressed to 28.6 +/- 8.6% of baseline. Although with subsequent reperfusions the return of function progressively decreased, the rate of deterioration was markedly attenuated after the first occlusion. By the end of the protocol many LAD segments lengthened paradoxically even after reperfusion, but in five hearts in which active contraction was preserved there was no significant change in regional function after the third cycle, suggesting a protective or preconditioning effect of earlier ischemia. There was a moderately good correlation between collateral blood flow and the degree of dysfunction following the initial 10-minute reperfusion (r = -0.73). This correlation deteriorated during subsequent reperfusion periods, implying that collateral blood flow can be a predictor of the extent of myocardial stunning only after the initial one or two reperfusion cycles. Thereafter other as yet unidentified factors make baseline collateral flow unimportant.

Effect of repetitive brief episodes of cardiac ischemia on 31P magnetic resonance spectroscopy in the cat

Angina is characterized by brief periods of ischemia followed by reperfusion; the cumulative effect of these episodes on energetics of the myocardium has not been fully elucidated. This study used an in vivo feline model for the assessment of high-energy phosphate compounds during brief sequential periods of ischemia and reperfusion. Nine adult, open-chest, anesthetized cats were prepared with a reversible occluder around the proximal left anterior descending artery and a 1.2-cm-inside diameter coil sutured on the myocardial surface in the distribution of the left anterior descending coronary artery. Levels of PCr, Pi, and ATP (beta-phosphate signal) were measured by 31P MRS in a GE CSI 2-T NMR spectrometer/imager. Measurements were obtained during a control period and during three successive occlusion-deocclusion periods of roughly 12 and 20 min' duration, respectively. The last deocclusion period was observed for 60 min. Electron microscopy was performed in two animals. PCr declined (P less than 0.01) rapidly following each occlusion to 51 +/- 5.2% (occlusion 1), 53 +/- 5.8% (occlusion 2), and 48 +/- 5.7% (occlusion 3) of the control value by 6 min. Pi rose (P less than 0.01) with the three sequential occlusions to 253 +/- 46, 288 +/- 57, and 277 +/- 46%, respectively. PCr and Pi returned to baseline promptly with reperfusion, while ATP showed a gradual decline throughout the experiment, decreasing to 77 +/- 7.2% of control at the end of the last reperfusion (P less than 0.05). Although PCr returned to baseline during reperfusion, ATP did not, suggesting a reduction in the nucleotide pool. These findings indicate that the repeated episodes of ischemia, which are insufficient to produce necrosis, can have an effect on myocardial high-energy phosphate metabolism as evidenced by mild depletion of ATP.

Unstable angina

Unstable angina can manifest as an array of symptom complexes. In some patients, medical therapy will stabilize the episodes of angina, and only predismissal exercise testing or angiography (or both) will be necessary. At the other end of the spectrum are patients with rest angina or multiple episodes of silent ischemia who are refractory to medical therapy and experience undetected microinfarction. Most of these patients require immediate catheterization and subsequent intervention with intra-aortic balloon pulsation, percutaneous transluminal coronary angioplasty, or coronary artery bypass grafting. An entire spectrum of manifestations exists between these two extremes. One challenge during the 1990s will be better stratification of patients with unstable angina so that safe, efficient, cost-effective treatment strategies can be appropriately applied to all patients.

Sensitization of reperfused myocardium to subsequent coronary flow reductions. An extension of the concept of myocardial stunning

The purpose of the present study was to evaluate the response of briefly ischemic and reperfused myocardium to subsequent moderate reductions of coronary arterial flow. In mongrel dogs, a carotid to left anterior descending coronary shunt was constricted to produce moderate coronary flow reductions (50-60% of control) and to thereby reduce regional systolic thickening (measured by echocardiography or sonomicrometry). First, we demonstrated an abnormal response of reperfused myocardium to subsequent flow reductions. We performed two episodes of coronary shunt stenosis, with an intervening 5-minute complete coronary shunt occlusion followed by 30 minutes of reperfusion. In a control group, the same two shunt stenoses were done, but no intervening shunt occlusion was performed. In the control dogs, repeated coronary shunt stenosis that produced equivalent perfusion reductions also produced equivalent declines in regional wall thickening. In contrast, in the intervention group (animals undergoing the intervening occlusion-reperfusion sequence between two shunt stenoses), the second coronary shunt stenosis produced an exaggerated decline in regional systolic thickening, even though the decline in myocardial perfusion was similar to the first stenosis. Second, we sought to demonstrate the mechanism of the exaggerated decline of the reperfused myocardium to subsequent moderate flow reductions. Again, two groups of animals were studied. Each group underwent two episodes of coronary shunt stenosis with an intervening sequence of 5 minutes of complete shunt occlusion and 30 minutes of reperfusion. In addition, one of the groups received an infusion of the oxygen free radical scavengers superoxide dismutase and catalase during the occlusion-reperfusion sequence. In the superoxide dismutase and catalase-treated animals, the decline in regional systolic function during the postreperfusion shunt stenosis was similar to the preocclusion stenosis. Thus, oxygen free radical scavengers blocked the exaggerated contraction decline in response to the postreperfusion flow reduction. We conclude that briefly ischemic and reperfused myocardium displays an exaggerated response to subsequent coronary arterial flow reductions and that this response is a subtle manifestation of postischemic ventricular dyskinesis, or "stunning." The mechanism is probably oxygen free radical toxicity.