Clin. Cardiol. 14, 708-712 (1991)

The Stunned and Hibernating Myocardium: A Brief Review c. RICHARDCONTI, M.D. Division of Cardiology, Department of Medicine, University of Florida College of Medicine, Gainesville, Florida, USA

Summary: Definitions: Stunned myocardium is viable myocardium salvaged by coronary reperfusion that exhibits prolonged postischemic dysfunction after reperfusion. Hihernuting myocardium is ischemic myocardium supplied by a narrowed coronary artery in which ischemic cells remain viable but contraction is chronically depressed.

Clinical evidence: Stunned myocardium has been identified in the following patient groups: (1) thrombolysis or percutaneous transluminal coronary angiography (PTCA) i n patients with acute evolving infarction; (2) unstable angina; (3) exercise-induced angina; (4) coronary artery spasm; ( 5 ) platelet aggregation or transient thrombosis of a coronary artery; (6) PTCA for chronic myocardial ischemia; and (7) immediately following coronary artery bypass graft (CABG). Evidence of hibernating myocardium (LV dysfunction) is found in the patient with severe coronary artery stenosis, even in asymptomatic patients at rest. Stunned myocardium returns to normal after a prolonged period of time (hours to weeks). Hibernating myocardium returns to normal function rather quickly if the cause is removed. Differentiation: Stunned myocardium can be differentiated from hibernating myocardium by three clinical parameters, namely, LV wall motion, myocardial perfusion, and myocardial metabolism. Stunned myocardium has abnormal wall motion that tends to normalize in response to

inotropes and postextrasystolic potentiation. Perfusion is adequate and metabolism is also adequate. Hibernating myocardium also has abnormal wall motion, which normalizes after nitrates, inotropes, post extrasystolic potentiation (PESP), PTCA, or CABG. Myocardial perfusion is reduced but can be reversed with PTCA or CABG and metabolism is adequate.

Key words: stunned myocardium, hibernating myocardium, left ventricular dysfunction Introduction Most of this review will deal with the stunned myocardium since the animal models and the human representation are more easily studied than the models of hibernating m yocardi um. The word stunned is derived from the Latin extonrrw meaning to thunder, from the old English stunian meiining to crash, and from the old French esfoner meaning to resound. Synonyms include rendering senseless, knocking unconscious, and dazing. Kloner defined the stunned myocardium as a viable myocardium that has been salvaged by coronary reperfusion, yet exhibits prolonged but transient postischemic dysfunction, lasting hours to days.'

Animal Evidence for Stunned Myocardium Address fnr Reprints: C. Richard Conti. M.D. PiiliiiBeach Heafl Association Eminent Scholar (Cardiology) Chief. Division of Cardiology BOX5-277, JHMHC Gninesville, FL 326 10, USA

Received: July 22, 1991 Accepted: July 22, 1991

Most have used the IS minute coronary artery occlusion to investigate stunning. Numerous observations have been made. These include the following: 1. Depression of contractile function for 6 hours after reflow2 2. Prolonged diastolic relaxation time3 3. Depressed levels of adenosine triphosphate (ATP) up to 72 hours4 4. Nonnal myocytes as viewed by light microscopys

C. R. Conti: Stunned and hibernating myocardium

5. Reversible mitochondria1 changes of unknown signiftcance as viewed by electron microscopy’ 6. Stimulation of the ventricle to contract with several inolropes including dopamine, isoproterenol, epinephrine, calcium, hydralazine, and postextrasystolic potentiationb

Mechanisms of Stunned Myocardium There probably is no single mechanism that results in myocardial stunning but several should be considered: 1. Abiiormal energy utilization 2. Prtduction of oxygen-derived free radicals 3. Abnormal calcium flux 4. White cell accumulation in previously ischemic tissue 5 . Microvascular abnormalities 6 . A combination of all of these

Greenfield and Swain’ have shown experimentally in animals after a IS-minute occlusion followed by reperfusion that there is a disruption of energy utilization rather than abnormal energy production. They found a decrease in niyocxrdial creatine kinase (CK) plus a decrease in adenosine diphosphate which is necessary for maximal creatine kinase activity in the animals studied. Przyklenk and Kloner,8 Myers,9 and Grosslo and coworkers have reported that cytotoxic oxygen-derived free radicals (e.g., hydroxide and superoxide anion) are associated with stunning and free radical scavengers attenuate the sluniied myocardium. The hypothesis that stunning is due to altered calcium flux, that is, calcium entry after reperfusion, is supported by the observation that nifedipine given 30 minutes after reperfusion restores myocardial function toward normal. I Conflicting data are reported relating to white cell accumulation i n previously ischemic tissue which results in microvascular plugging after reperfusion. Both of these can decrease blood flow (the low retlow phenomenon) to previously ischemic myocardium and can also be a possible extracellular source of oxygen-derived free radicals.”, Tilmanns et nl. showed that there was altered capillary permeability and red and white cell clumping in the animal model of stunned myocardium suggesting abnormalities at the microvascular level as a possible cause for stunning.”

Clinical Evidence for Stunned Myocardium Nunicrous observations have been made following thronibolysis or percutaneous transluminal coronary angiogrophy (PTCA) in patients with evolving acute myocardial infarction. Prolonged recovery of myocardial function has also been noted in patients with unstable angina. folltwing exercise-induced angina, following PTCA for chronic myocardial ischemia, and following coronary artery bypass graft surgery. In this latter example, a stun-

709

ned myocardium is implied since patients often require long-term inotropic support after cardiopulmonary bypass. Anderson,14 Stack,15 and RedutoI6 and their colleagues have shown clearly that following thrombolysis for acute myocardial infarction there is gradual continued improvement of ventricular function weeks after successful reperfusion. This gradual improvement was not seen in patients whose infarct-related artery was not reperfused. In the unstable angina patients, Nixon,I7 Satler,I8 and colleagues have shown convincing cardiac echo evidence of temporary ventricular dysfunction during angina-free periods. They have also shown that isoproterenol improves contraction of these “stunned myocardium” analogous to results in animals. Robertson and colleaguesI9 have also shown cardiac echo evidence of regional wall motion abnormalities 30 minutes after exercise, particularly in patients with multivessel disease. Camici et a/.,*” using positron emission tomography (PET) studies, have also shown prolonged alteration in metabolism using fluorodeoxyglucose (i.e., decreased myocardial uptake of glucose). Further clinical evidence for a stunned myocardium comes from Wijns and colleagues,21who evaluated diastolic abnormalities of the ventricle and showed marked decrease in compliance during balloon occlusion and reduced compliance up to 15 minutes after repetitive PTCA.

The Hibernating Myocardium Hibernate derives from the Latin hibernus meaning wintry. To hibernate is to spend a period, either seasonal or diurnal in a state of deep sleep, especially one marked by a distinct lowering of metabolism and body temperature. A second definition is “a state of death-like sleep.” Rahimtoola?? defined hibernating myocardium as ischemic myocardium (that is supplied by narrowed coronary artery) in which cells remain viable but contraction is chronically depressed. The depressed contractile function reduces oxygen demand, which presumably protects these myocytes. Keller and Cannon?’ reported a model of hibernating myocardium using isolated perfused rat heart by using graded reductions in coronary artery pressure. Table I summarizes their observations. They demonstrated significant reductions in myocardial oxygen consumption and contractile performance that returned to control when coronary artery pressures were returned to baseline. With modest decrease in coronary perfusion pressure these changes were associated with a decrease in myocardial creatine phosphate but no lactate production or change in myocardial pH or adenosine triphosphate (ATP). With further reduction in coronary perfusion pressure, creatine phosphate decreased further, myocardial lactate production increased, myocardial pH decreased, and ATP decreased. These observations suggest that traditional

Clin. Cardiol. Vol. 14, September 1991

710

TArjt,F,

I

Hibernating myocardiumf4

Modest I CPP Moderute

ICPP

MVO:,

Contractile function

Myocardial creatine phosphate

Lactate production

Myocardial PH

Myocardial ATP

1

1

1

0

0

0

1

I

11

T

I

1

“Isolated perfused rat heart model. Ahhrc.viurions: ATP = adenosine triphosphate; CPP = coronary perfursion pressure; MVO;?=myocardial oxygen consumption. Based on data from Ref. 23.

markers of myocardial ischemia need not be present in the “hibemating” myocardium but oxygen delivery may be an important determinant of myocardial function in this model of decreased coronary perfusion pressure.

Clinical Evidence for Hibernating Myocardium Several observations have been made suggesting the presence of a hibernating myocardium in humans:24.25

I . Reversal of chronic ventricular asynergy by nitrates, postextrasystolic potentiation, inotropes, exercise, and revascularization. 2. There is some correlation with the severity of coronary artery disease and the presence or absence of coronary artery collaterals. 3. There is some correlation with quantitated (histologic) myocardial loss (e.g., if‘ nitrates are given and there is only 10% cell death there will be reversible function of the ventricle. In contrast, if 50% of the cells are dead there will be little reversible function. 4. Reversal of a thallium perfusion defect in akinetic myocardium following coronary artery bypass surgery.

Summary of Difference Between Stunned and Hibernating Myocardium Table I1 outlines some differences between stunned, hibernating, and infarcted myocardium. The stunned myocardium occurs after ischemia and after reperfusion. Hibernating myocardium occurs during ischemia. The clinical causes of stunned myocardium include episodes of severe ischemia due to either exercise, PTCA occlusion, coronary artery spasm, platelet aggregation, or transient thrombosis of a coronary artery. In contrast, the clinical cause of hibernating myocardium is severe coronary artery stenosis even under resting conditions. Stunned myocardium returns to normal after a prolonged period of time (hours to weeks). In contrast, hibernating myocardium has an indefinite delay, but returns to normal function rather quickly if the cause is removed. A clinical example of the stunned myocardium is the reperfused evolving myocardial infarction. The clinical example of the hibernating myocardium is LV dysfunction prior to CABG or PTCA compared with normalized function following CABG or PTCA. As summarized in Table 111, one can clinically differentiate stunned from hibernating myocardium using three parameters (i.e., wall motion, perfusion, and metabolism).

TABLE II Some differences between stunned, hibernating, and infarcted myocardium Condition of myocardium

Time of occurrence

Stunned

After ischemia and reperfusion

Hi beriiatiiig

During ischemia

1n farcted

After ischemia

Clinical cause Discrete episodes of severe ischemia e.g., exercise, spasm, platelet aggregation, transient thrombosis Patent but severe coronary artery stenosis Coronary artery occlusion

Return of myocardial function

Clinical example

Prolonged delay

Reperfused evolving MI

Indefinite delay

LV function prior to CABG or PTCA Failure of reperfusion

No recovery

S. P. Karas et al.: Restenosis after coronary angioplasty

l'rapadil, an inhibitor of PDGF-induced cellular proliferation, has been shown to reduce intimal thickness following balloon injury in the atherosclerotic rabbit.ImInsulin-like growth factor-I (IGF-I) acts synergistically with PDGF to promote smooth muscle cell replication in vitrO,.VJ.111 Production and release of IGF- 1 by smooth muscle cells is regulated by growth hormone. Angiopeptin is a synthetic analog of somatostatin which blocks the release of growlh hormone by the pituitary. Pretreatment of rats and rabbits with this compound was found to greatly inhibit vascular smooth muscle proliferation following denudation of the carotid arteries.'". Its effects on human restenosis are currently under investigation at Georgetown University. Hypertrophy of the vascular muscle layers seen in hypertension may be mediated by the local production of angiotensin 11.102 Local angiotensin systems may also be involved in the myoproliferative response to vascular injury. Studies have clearly shown a reduction in intimal thickness following carotid injury in rats treated with the angiotensin converting enzyme inhibitor ~ilazapril.~" A multicenter clinicill study in patients following PTCA is currently in progress at European and U.S. enters, including our own.

All potential pharmacologic treatments for restenosis are limited by the difficulty in maintaining a high concentration of the drug at the desired site of activity for an extended period of time without systemic toxicity. Methods hilvl! been devised to deliver agents directly to injured vascular segments. Perforated balloon catheters have been uscd to inject high concentrations of heparin into the arterial wall without significant vascular tra~ ma. " '~ Genetically modified endothelial cells have been successfully implanted onto vascular grafts"Mand onto the denuded luminal surface of porcine iliac arteries.'O5 Recently, recombinant genes were successfully transferred directly into endothelial cells and myocytes by introducing a retroviral vector to the vessel wall through a catheter.lo5 Using this technology, recombinant genes could be delivered to the arterial wall during PTCA and induce prolonged It~calproduction of antithrombotic or antiproliferative suhstances in order to inhibit restenosis.Io6Stents are being developed which may store and slowly release high concentrations of inhibitory agents locally over an extended pcriod of time. A recent study investigated the effect of heparin bonding on the development of neointimal proliferation after placement of a tantalum stent in porcine carotid arteries. New Devices

The development of new techniques in the treatment of coronary artery disease has offered new hope that the luminal diameter could be restored without inducing myointimal proliferation.

797

Since events during angioplasty appear to influence the later development of restenosis, much of the research on new devices has been focused on improving the luminal diameter and geometry at the time of PTCA. The use of coronary artery stents may be one way to optimize final lesion morphology. Initial animal studies showed minimal restenosis associated with placement of an intracoronary stent.Ios Studies in humans have shown restenosis rates of about 20% following placement of a Palmaz-Schatz stent. l1(] One needs to be aware, however, that reports of restenosis rates following stent placement vary considerably depending on the definition of restenosis used. For example, in the same group of patients studied 2 4 months after placement of a Palmaz-Schatz. stent, Ellis found that 20% had restenosis as defined as a >50% stenosis at the site of the stent.l"'When restenosis was defined as a decrease in luminal diameter of >0.72 mm, the rate of restenosis was 50%. Although relatively fewer patients undergoing stent placement developed significant lesions, half of them showed significant intimal proliferation. In a recently published study of the self-expanding Wallstent, Sermys et al. reported a restenosis rate of 33%, based on a change of ~ 0 . 7 2mm in minimal luminal diameter.'ll The restenosis rate was 13% when defined as a stenosis of >50%. These data suggest that perhaps the benefit of stent placement may be due to stretching of the vessel and thus optimizing luminal diameter, rather than by limiting the degree of intimal proliferation. Our studies in swineBoand those of others in humans'". have confirmed this suspicion. Results from our swine model indicate that the luminal diameter was significantly greater following stent placement compared with the diameter after balloon dilatation. Furthermore, the long-term loss in luminal diameter with stents was comparable to that seen following the use of other modalities. Using atherectomy devices, atherosclerotic plaque can be shaved from the luminal surface and subsequently removed from the body.'13 Cenain features of atherectomy led to expectations that this technique might be associated with less intimal reaction and restenosis than balloon angioplasty. Atherectomy usually results in a smooth, widely patent vessel, often with a residual stenosis of