CHAPTER 38: MOLECULAR AND CELLULAR MECHANISMS OF MYOCARDIAL ISCHEMIA/REPERFUSION INJURY

Acute myocardial infarction is a leading cause of mortality and morbidity worldwide. It is the result of an acute occlusion of an epicardial coronary artery. The myocardium supplied by the occluded artery suffers severe ischemia. Prolonged ischemia, if there is no collateral circulation, results in irreversible damage to heart muscle, which is replaced by fibrotic noncontracting tissue (myocardial scar). Depending on the size of the territory distal to the occlusion site, the global ventricular function can be significantly impaired, resulting in postinfarction chronic heart failure.¹ Classical studies performed in large animal models demonstrated that blood flow restoration (reperfusion) was able to limit the progression of myocardial injury in a timely manner.^2,3 Reperfusion was shown not only to limit the progression of myocardial death but also to change the pattern of myocardial tissue healing. This concept was rapidly introduced into the clinics, and a long series of successful trials demonstrated that early reperfusion was able to reduce the extent of myocardial injury and, more importantly, to reduce mortality. Since then, timely reperfusion has become the standard treatment for patients suffering an acute myocardial infarction. Extensive preclinical and clinical studies have shown that reperfusion itself induces damage to the formerly ischemic myocardium (Fig. 38–1). Since reperfusion is the prerequisite for myocardial salvage, the damage inflicted on the myocardium during a myocardial infarction is known as ischemia/reperfusion injury (IRI; ie, the result of ischemia- and reperfusion-related damage). Myocardial IRI is a complex phenomenon involving many players, all contributing to the final damage inflicted on the myocardium.¹ Figure 38–2 summarizes the highly complex process of IRI and the multiple mechanisms involved in it. The first critical player is the epicardial artery (represented as 1 in Fig. 38–2). Atherosclerotic plaque rupture with superimposed thrombus results in an abrupt cessation of oxygen and nutrient supply distal to the occlusion site. The opening of the epicardial vessel by mechanical or pharmacologic means, as well as the reduction in thrombus burden by adjuvant antiplatelet/anticoagulant therapies, is only the first step toward the salvage of myocardium. During the reperfusion process (whether it is mechanical by primary angioplasty or pharmacologic by thrombolytics), thrombus material and other plaque debris can be distally embolized, contributing to microvascular obstruction (MVO). Circulating cells contribute to the damage inflicted to the myocardium: activated platelets and leukocytes in the bloodstream not only contribute to the thrombus generation, but also can form plugins that can embolize distally into the microcirculation through resting blood flow across the culprit lesion (a process independent from plaque debris microembolization). The microcirculation is a critical player in the fate of the myocardium during ischemia/reperfusion. Once the epicardial vessel flow is restored, efficient tissue perfusion is dictated mainly by the microcirculation. Plaque debris and platelet/neutrophil aggregates can induce a mechanical obstruction of the microcirculation precluding efficient tissue perfusion despite the opening of the epicardial artery (which is known as the no-reflow phenomenon). The generation of tissue edema following reperfusion can result in external compression of the microcirculation, reducing the perfusion capacity of the capillary network (arrowheads in Fig. 38–2). Finally, the microcirculation can be disintegrated as a result of previous damage, which allows circulating cells to leak into the interstitial space (dashed lines in the capillary of Fig. 38–2). Hemorrhage is especially harmful because of the release of iron, contributing to subsequent inflammatory reactions. Cardiomyocytes that have survived the ischemic phase suffer during the reperfusion period as a result of several intracellular pathways triggered at reperfusion. After the acute phase of the ischemia/reperfusion insult has passed, the significant infiltration of myocardial tissue by inflammatory cells can induce additional damage to the myocardium. In this chapter, we summarize the cellular and molecular mechanisms involved in IRI; however, it is important to keep in mind that it is the combination of these different mechanisms that leads to the progression of myocardial injury at reperfusion and, more importantly, that these mechanisms are interconnected.

Determinants of Infarct Size

After the onset of a myocardial infarction (ie, coronary occlusion), the hypoperfused myocardial region, known as the area at risk (AAR), is at jeopardy of developing irreversible injury. Unrelieved ischemia, in the absence of significant collateral flow, results in death of the entire AAR. Given that today’s mainstay treatment for myocardial infarction is reperfusion, part of the AAR remains free of necrosis: the so-called salvaged myocardium. The seminal studies by Ross and coworkers four decades ago^2,3 first demonstrated that early reestablishment of blood flow (reperfusion) salvages myocardium from infarction. These landmark studies demonstrated that there is a temporal progression of myocardial death during ongoing coronary occlusion: the sooner the reperfusion, the smaller is the proportion of AAR that becomes necrotic. These experimental studies primed the development of reperfusion strategies to limit the size of necrosis in patients suffering an acute myocardial infarction.⁴ Despite being the prerequisite for myocardial salvage, reperfusion per se can cause some damage to the formerly ischemic myocardium, something known as reperfusion injury. Although the contribution of reperfusion injury to final infarct size has been disputed in the past, today it is well accepted that reperfusion can induce additional damage to the myocardium. This view is supported by strong evidence that interventions applied at the end of the ischemic period (ie, coinciding with reperfusion) can reduce infarct size. It was already recognized in the mid-1980s that gentle reperfusion at low pressure resulted in significantly less edema and a smaller infarct size than standard abrupt reperfusion at normal pressure.⁵ This idea was later developed, and a reduction of infarct size by brief episodes of coronary reocclusion/reflow at the time of reperfusion, a strategy called ischemic postconditioning (IPost),^6,7 was demonstrated (discussed later in this chapter). Because these interventions are applied at the end of the ischemic period, they cannot reduce infarct size by reducing ischemic damage and thus must reduce reperfusion-related damage. From these observations, it is clear not only that reperfusion injury contributes to infarct size, but also that all conditioning strategies that protect the myocardium and reduce infarct size act only in conjunction with eventual reperfusion.^8,9

Within a given AAR, both the duration and the severity of coronary blood flow reduction determine the nature and amount of injury.¹ A complete coronary occlusion of less than 20 minutes in duration results in only reversible injury (ie, contractile dysfunction with a slow, but complete, recovery during reperfusion, a phenomenon called myocardial stunning). The underlying mechanisms of the prolonged contractile dysfunction relate to the enhanced formation of reactive oxygen species (ROS) during early reperfusion and impaired excitation-contraction coupling after oxidative modification of the sarcoplasmic reticulum and the contractile proteins.¹⁰ Repeated coronary occlusions of short duration or prolonged moderate reduction in coronary blood flow results in hibernating myocardium, a phenomenon of reduced contractile function with retained viability and, thus, eventual recovery after reperfusion. Hibernating myocardium displays both signs of injury (loss of contractile proteins, small doughnut-like mitochondria, fibrosis) and signs of adaptation (short-term energetic recovery, altered expression of mitochondrial proteins and proteins related to cardioprotection).¹¹ When the reduction in coronary blood flow is severe and lasts longer than 20 to 40 minutes, infarction develops first in the inner subendocardial layers of the core of the AAR and then spreads in a “wavefront” to the outer subepicardial layers and the borders of the AAR over time. The wavefront of infarct development reflects the lateral and transmural distribution of coronary blood flow, which is less in the inner than the outer layers of the myocardium and less in the core than in the borders of the AAR.¹² The evolution of infarction varies with species and depends on the existence and extent of a collateral circulation. Dogs have a well-developed native collateral circulation, and infarction starts after 40 minutes of coronary occlusion and spreads to affect 70% of the AAR after 6 hours.¹² Therefore, in dogs, infarct size is best quantified as a fraction of the AAR and normalized to the residual blood flow.¹³ Pigs have a negligible native collateral circulation, and infarction starts after 15 to 35 minutes of coronary occlusion and affects ≥ 80% of the AAR after 60 to 180 minutes.¹⁴ Primates have few innate collaterals but are relatively resistant to myocardial ischemia; there is no infarction after 40 to 60 minutes of coronary occlusion, and even after 90 minutes of coronary occlusion, infarct size is smaller than in pigs.¹⁵

Fortunately, infarct development in humans is slower than in these large mammals. Even after 4 to 6 hours of coronary occlusion, 30% to 50% of the AAR remains viable and thus salvageable, as one can estimate from magnetic resonance imaging and from the amount of salvage by reperfusion.^16,17,18 Salvageable myocardium remains even after 12 hours from symptom onset.¹⁹ To what extent the resistance of the diseased human heart is attributable to a developed collateral circulation at the time of infarction, as in the native dog heart, or reflects an inherently greater resistance to ischemic injury, such as in the primate heart, or is a result of preceding episodes of brief ischemia/reperfusion is unclear at present. In contrast to prior notions, the hemodynamic situation has little impact on the development of myocardial infarction, and the little impact it has is largely through its impact on coronary blood flow and its spatial distribution; heart rate determines infarct progression to some extent. Variations in coronary blood flow not only determine the nature and extent of myocardial injury, but paradoxically also determine protection from it. Repeated brief episodes of coronary occlusion preceding a prolonged coronary occlusion with reperfusion reduce infarct size (a phenomenon known as ischemic preconditioning [IPC], which is discussed later).²⁰ Likewise, repeated brief coronary occlusion during early reperfusion reduces infarct size, a phenomenon known as IPost⁶ (discussed later).

Therefore, there are several determinants of myocardial infarct size (Table 38–1). The most widely supported by experimental and clinical studies are (1) the size of the AAR²¹; (2) the duration of myocardial ischemia^12,22; (3) the amount of residual blood flow through collaterals^12,21; (4) the temperature of the tissue during ischemia²³; and (5) the degree of reperfusion injury.¹ In this chapter, we focus on the mechanisms leading to ischemia- and reperfusion-related damage.

Infarct Size Determines Postinfarction Mortality and Morbidity

The final extent of necrosis (infarct size) has been revealed as a major determinant of postinfarction mortality.^24,25 Large infarctions are associated with an irreversible impairment of the contractile function of the heart and concomitant chronic heart failure.⁹ Beyond their impact on left ventricular systolic impairment (quantified as left ventricular ejection fraction [LVEF]), large infarctions are associated with increased risk of sudden death. The widespread implementation of reperfusion strategies has resulted in a significant reduction in the acute mortality associated with myocardial infarction. In fact, in-hospital mortality has decreased from approximately 20% in the late 1980s to approximately 5% today.^26,27 Paradoxically, this massive reduction of infarction-related acute mortality has resulted in an increase in the incidence of chronic heart failure. The obvious reason is that patients with severely depressed cardiac function would not have survived the acute phase of infarction in the past, but with the advent of reperfusion, they now survive the index episode and live with a significantly damaged heart.²⁶ In fact, myocardial infarction is a main contributor to chronic heart failure worldwide. Chronic heart failure treatment is one of the main contributors to healthcare expenditures and, thus, it is important to implement strategies to reduce its incidence. After acknowledging that infarct size is the main determinant of adverse postinfarction outcomes, including heart failure,²⁵ therapies able to reduce infarct size are urgently sought under the hypothesis that smaller infarctions will result in better long-term heart performance and that this will translate into fewer adverse clinical events.^28,29 The identification of therapies able to reduce infarct size, including amelioration of IRI, is a major challenge in the 21st century.

Ischemia/Reperfusion-Induced Cardiomyocyte Death

Cardiomyocyte death is the most important consequence of myocardial IRI and is the main cause of heart failure, arrhythmias, and death in patients with acute myocardial infarction. Cardiomyocyte death secondary to reperfusion injury is mainly dependent on mechanisms taking place within cardiomyocytes themselves. In fact, the most important aspects of IRI in the intact heart can be recapitulated in isolated cardiomyocytes submitted to transient hypoxia/reoxygenation or to similar conditions.^30,31,32 However, as discussed later, other cells can contribute to cardiomyocyte cell death occurring upon myocardial reperfusion. The mechanism of these interactions is complex and incompletely understood, and may take place independently of impairment of microvascular flow. For example, platelet activation and adhesion increase cardiomyocyte death independently of aggregation and of any effects on myocardial flow.^33,34

It is widely admitted that a large fraction of cell death caused by lethal myocardial IRI occurs during the initial minutes of reperfusion, thus requiring early administration of cardioprotective interventions in order to prevent it.³¹ Some studies have suggested the existence of additional delayed IRI during the first hours of reperfusion, but the importance of this phenomenon needs to be established.^35,36 In this context, it is important to notice that noncardiomyocyte cells have a better tolerance to ischemia than cardiomyocytes. This is particularly relevant in the case of endothelial cells, the metabolism of which is largely independent of mitochondrial respiration.³⁷

Although there have been controversies about the contribution of apoptosis to IRI, it is clear now that IRI causes cardiomyocyte death mainly through necrosis.^38,39 It has been described that adult cardiomyocytes do not express executioner caspases 3 or 7 and that their selective genetic ablation does not modify infarct size or postinfarction remodeling in mice submitted to transient coronary occlusions. Recent studies have shown that necrosis may occur in different forms, ranging from sarcolemmal rupture resulting in massive calcium overload to programed forms of necrosis involving precise signal cascades, in particular necroptosis, a form of programmed necrotic cell death dependent on receptor-interacting protein kinase 3.⁴⁰ Reperfused infarcts are mainly composed of contraction band necrosis, a pathologic pattern reflecting sarcolemmal disruption and extreme cell shortening.

Mechanisms of Cardiomyocyte Necrosis During Reperfusion

Studies using magnetic resonance spectroscopy, fluorescence/confocal microscopy, and electrophysiologic methods have allowed us to characterize the changes in ionic concentrations, pH, energy-storing molecules, and transmembrane potentials in cardiomyocytes during ischemia and reperfusion. Severe ischemia causes the immediate arrest of mitochondrial respiration and switch to anaerobic metabolism. This, in combination with very reduced catabolite washout, results in a rapid fall in intracellular pH, reaching 6.4 within a few minutes. Mitochondrial membrane potential progressively dissipates during ischemia, and adenosine triphosphate (ATP) concentration falls and reaches the very low levels, triggering rigor contracture (in approximately 20-40 minutes of severe normothermic ischemia in large animals).⁴¹ Inversion of respiratory complex V (ATP synthase) activity from ATP synthesis to ATP degradation and proton pumping into the intermembrane space delay complete mitochondrial depolarization at the expense of accelerating complete ATP exhaustion.⁴² Very low levels of ATP mark the onset of a progressive increase in cytosolic Ca²⁺, mainly through reverse-mode Na⁺/Ca²⁺ secondary to the Na⁺ overload produced, mainly, by the inactivity of the Na⁺/K⁺-ATPase pump.³⁰

Altered Ca²⁺ Handling and Normalization of Intracellular pH

Reperfusion results in rapid changes in cardiomyocytes that can lead to their survival and progressive functional recovery or to cell death, depending mainly on the situation of the cardiomyocytes at the time of reperfusion. These changes include, predominantly, rapid restoration of energy availability and intracellular pH, additional Ca²⁺ influx and uptake by the sarcoplasmic reticulum (SR), and a marked increase in ROS^43,44 (Fig. 38–3).

Although normalization of ATP concentration caused by reperfusion may take some time, normalization of the free energy of dissociation of ATP is rapid. Mitochondria of ATP-depleted cardiomyocytes may rapidly repolarize and start respiration and ATP synthesis within seconds of restoration of blood flow and oxygen availability.⁴² Normalization of intracellular pH is completed within a few minutes after reperfusion and restoration of catabolite washout, thanks in part to the activation of the sarcolemmal Na⁺/H⁺ exchanger and the Na⁺/CO₃H^– co-transporter.⁴⁵

Altered Ca²⁺ handling is one of the most prominent and relevant aspects of cardiomyocyte reperfusion. Reperfusion causes further Ca²⁺ influx through the Na⁺/Ca²⁺ exchanger in response to Na⁺ overload. During the initial minutes of reperfusion, Na⁺ influx associated with normalization of intracellular pH and passage of Na⁺ from adjacent cells via gap junctions may lead to severe Na⁺ overload in the presence of altered Na⁺ pump function.⁴⁶ Ca²⁺ influx occurring during initial reperfusion adds up to Ca²⁺ overload occurring during ischemia and results in increased cytosolic Ca²⁺ concentration.⁴⁴ An important consequence of increased Ca²⁺ is activation of the Ca²⁺-dependent proteases calpains. Calpains translocate to the sarcolemma during ischemia in response to Ca²⁺ overload, but are activated only during reperfusion when intracellular pH is normalized, as calpains are inhibited by low pH.⁴⁷ Calpain-mediated hydrolysis of fodrin, the cytoskeleton protein responsible or the proper anchorage of the Na⁺ pump to the sarcolemma, has been shown to worsen Na⁺ overload and contribute to cardiomyocyte death, as its pharmacologic or genetic inhibition consistently reduces infarct size.

Restoration of ATP availability allows the sarcoplasmic Ca²⁺ pump (sarcoendoplasmic reticulum Ca²⁺ ATPase [SERCA]) to take large amounts of Ca²⁺ into the SR. When the Ca²⁺ capacity of the SR is exceeded, Ca²⁺ is released back into the cytosol through the ryanodine receptor channel (RyR2). This results in Ca²⁺ oscillations that propagate across the cell and may have several adverse consequences, favoring the generation of ventricular arrhythmias, excessive contractile activation, and mitochondrial Ca²⁺ overload and permeability transition.⁴⁸ Excessive contractile activation may cause hypercontracture, resulting in disruption of cardiomyocyte architecture that can cause sarcolemmal rupture and cell death, as demonstrated in studies showing that transient contractile inhibition during the initial minutes of reperfusion may prevent cell death and limit infarct size in different models ranging from isolated cardiomyocytes to intact large animals.⁴⁹

The Mitochondrial Permeability Transition Pore (MPTP)

There is solid evidence demonstrating that mitochondria play a key role in the events determining cardiomyocyte death or survival during reperfusion. Reperfusion results in immediate restoration of respiration, proton gradient across the inner mitochondrial membrane, and ATP synthesis. However, during reperfusion, respiration is associated with increased ROS generation, in part as a result of reoxidation of large amounts of succinate accumulated during ischemia by succinate dehydrogenase and reverse electron transport between complex II and complex I of the respiratory chain.^50,51 However, restoration of mitochondrial membrane potential favors mitochondrial Ca²⁺ uptake through the Ca²⁺ uniporter and Ca²⁺ overload.⁴⁴

Mitochondrial Ca²⁺ overload and increased ROS production are important triggers of mitochondrial permeability transition, an abrupt increase in the permeability of the inner mitochondrial membrane resulting in mitochondrial depolarization and swelling, arrest of respiration, and release of molecules from the mitochondrial matrix into the cytosol.^52,53 Mitochondrial permeability transition is caused by the opening of the mitochondrial permeability transition pore (MPTP), a large conductance channel in the inner mitochondrial membrane allowing passage of large molecules up to 1.5 kDa. MPTP opening was first observed in isolated mitochondria exposed to high Ca²⁺ concentrations, can be triggered by ROS, and is favored by low ATP concentration but is completely inhibited by low pH.⁵⁴

During ischemia, the MPTP does not open because of the inhibiting effect of intracellular acidosis. However, the MPTP may open during the first few minutes of reperfusion when intracellular pH returns to normal values,⁵⁵ and there is strong evidence of the involvement of MPTP opening in cardiomyocyte death secondary to IRI. MPTP has been documented in mitochondrial preparations exposed to Ca²⁺ overload and ROS by monitoring the release of fluorescent molecules previously loaded into mitochondria or mitochondrial swelling, and has been shown to occur in isolated cardiomyocytes submitted to simulated ischemia-reperfusion.⁵⁴ Inhibition of MPTP opening by genetic ablation of cyclophilin D, a molecule that modulates the sensitivity of MPTP to Ca²⁺, has been shown to prevent MPTP opening in isolated mitochondria and cardiomyocytes during simulated reperfusion and to limit infarct size in mice submitted to transient coronary occlusion.⁵⁶ Pharmacologic inhibition of cyclophilin D with cyclosporine, a calcineurin inhibitor used as an immunosuppressant, has also been shown to inhibit MPTP opening in isolated mitochondria and cardiomyocytes and to limit infarct size in preclinical studies, although in a less consistent way.^57,58,59

The understanding of the regulation and function of the MPTP is at present limited by the lack of a precise knowledge of its molecular structure. After different models involving the adenine nucleotide transporter, voltage-dependent anion channels, and other molecules were discarded, more recent studies suggest that the MPTP is formed by the ATP synthase (complex V of the respiratory change), the channel being formed by either the rotor unit or by a conformational change of the dimeric organization of the ATP synthase molecules.⁶⁰ This model, like all previous ones, accounts for the regulatory role of cyclophilin D.

Interaction Between Mitochondria and Sarcoplasmic Reticulum

The mechanism by which MPTP opening may cause necrosis with sarcolemmal rupture within minutes of reperfusion is not well understood. A potential explanation is that it causes hypercontracture. It has been shown that opening of the MPTP in part of the mitochondria may cause hypercontracture in Ca²⁺-overloaded cardiomyocytes by allowing the release of its Ca²⁺ contents while nonpermeabilized mitochondria ensure ATP availability.⁶¹ However, SR-driven Ca²⁺ oscillations may induce MPTP opening.⁴⁸ This is favored by the tight physical connection between the SR and mitochondria mediated by specific anchoring proteins and resulting in the creation of Ca²⁺ micro-domains in which RyR channels and SERCA are in close proximity to the mitochondrial Ca²⁺ uniporter.⁶² These micro-domains allow a preferential communication between SR and mitochondria and may translate SR Ca²⁺ oscillations into mitochondrial Ca²⁺ overload.⁴⁸ Thus, there is an important crosstalk between MPTP opening and Ca²⁺ oscillation-driven hypercontracture as causes of cell death in reperfused cardiomyocytes. The relative importance of these two mechanisms may depend on the conditions, in particular on the duration of the ischemia; the role of MPTP opening appears to be less important after brief periods of ischemia.⁶³

During reperfusion, increased ROS generation inside and outside mitochondria may favor cardiomyocyte cell death by mechanisms other than by directly inducing MPTP opening.⁶⁴ ROS may oxidize tetrahydrobiopterin and cause endothelial nitric oxide synthase dissociation, favoring the generation of toxic peroxynitrite and further oxidative stress and reducing nitric oxide availability and cyclic guanosine monophosphate–protein kinase G signaling. In addition to directly modulating MPTP opening,^65,66 protein kinase G signaling is reduced in reperfused myocardium,⁶⁷ and this favors SR-driven Ca²⁺ oscillations via protein kinase A–mediated effects on phospholamban, a regulator of SERCA.⁶⁸ Figure 38–4 summarizes intracellular changes leading to cell damage/death during reperfusion.

Myocardial IRI injury affects not only the cardiomyocyte compartment, but also the coronary circulation.⁶⁹ Coronary occlusion after atherosclerotic plaque rupture with superimposed thrombosis is the cause and origin of myocardial ischemia, and reperfusion of the occluded coronary artery with restoration of coronary blood flow not only terminates myocardial ischemia but also inflicts additional injury. The spatial and temporal evolution of coronary occlusion and reperfusion determines both the size of the affected myocardial region and the nature of the outcome from ischemia/reperfusion (ie, reversible [stunning] or irreversible [infarction] injury and, vice versa, protection from injury [hibernation, conditioning]).

The Coronary Circulation as a Determinant of Reperfusion and Reperfusion Injury

Only 2 years after the demonstration by John Ross’s group that reperfusion was the only way to rescue myocardium from impending infarction,^2,3 the dark side of coronary reperfusion became apparent, when Robert Kloner first reported the coronary no-reflow phenomenon.⁷⁰ And again, only a few years later, Keith Reimer and Robert Jennings reported in a series of detailed dog studies that signs of irreversible myocardial injury became particularly manifest during reperfusion¹²; at that time, it was not clear whether the irreversible injury was caused by reperfusion or only became more obvious during reperfusion. The long debate over the existence of lethal reperfusion injury was finally ended by the recognition of the IPost phenomenon when Jakob Vinten-Johansen reported that repeated coronary reocclusion during early reperfusion reduced infarct size in dogs.⁶ The fact that a modified procedure of reperfusion could indeed attenuate irreversible injury once again revived earlier studies⁷ on gentle reperfusion (ie, the reduction of functional and morphologic signs of injury by reperfusion at reduced perfusion pressure⁷¹ or reduced coronary blood flow⁷²). With the unequivocal notion that reperfusion is both mandatory for salvage from impending infarction and causes irreversible injury per se, a complex picture arises in that both ischemia and reperfusion contribute to the ultimate injury, but their individual contribution to ultimate injury depends on the duration and severity of coronary blood flow reduction.⁷³ Although ischemic injury increases with the severity and the duration of coronary blood flow reduction, there is a maximum of reperfusion injury in more moderate ischemic injury.

Manifestations of Myocardial Ischemia/Reperfusion Injury in the Coronary Circulation

The manifestations of myocardial ischemia/ reperfusion in the coronary circulation range from mild and reversible functional impairment to severe and irreversible destruction.

Vascular Permeability and Edema

Edema develops within minutes of acute myocardial ischemia⁷⁴ and is both intracellular and interstitial. Intracellular edema develops in cardiomyocytes and endothelial cells largely as a consequence of the rapidly developing energetic deficit and the reduced function of energy-dependent ion pumps. Interstitial edema develops as a consequence of increased interstitial osmolarity from increased ion and catabolite concentrations and a dysfunction of the endothelial barrier function during myocardial ischemia (as depicted in Fig. 38–2). The endothelial barrier function is made up of endothelial cells, pericytes, and the glycocalyx. Endothelial cytoskeletal derangement and hypercontracture induce gap formation.^75,76,77 Degradation of the glycocalyx also contributes to reduced endothelial barrier function and edema formation⁷⁸ and promotes leukocyte⁷⁹ and platelet adherence.⁸⁰ Upon reperfusion, interstitial edema is greatly enhanced by reactive hyperemia and the rapid washout of osmotically active molecules from the intravascular space. Edema development during reperfusion follows a bimodal pattern,⁸¹ where an initial maximum of water content after 120 minutes is associated with a beginning leukocyte infiltration and a secondary peak after 7 days is associated with enhanced collagen deposition⁸² (Fig. 38–5). Edema during reperfusion has been proposed to reflect the AAR on cardiac magnetic resonance (CMR), but such bimodal pattern questions the use of T2-weighted edema measurement for AAR delineation.⁸³

Vasomotion

The coronary microcirculation distal to a severe coronary stenosis or coronary occlusion has traditionally been considered as maximally dilated after exhaustion of autoregulatory reserve. However, even during myocardial ischemia that limits regional contractile function, a pharmacologically recruitable vasodilator reserve persists⁸⁴ or, conversely, the coronary microcirculation remains responsive to vasoconstrictor mediators, notably α-adrenergic coronary vasoconstriction.^85,86 The release of vasoconstrictor substances such as thromboxane, serotonin,⁸⁷ and endothelin⁸⁸ from the culprit lesion into the microcirculation, in conjunction with the impairment of endothelial function by ischemia/reperfusion per se⁸⁹ or by tumor necrosis factor-α (TNF-α),⁸⁷ can contribute to such enhanced vasoconstrictor responsiveness during myocardial ischemia/reperfusion. With more prolonged ischemia in hibernating myocardium, there is structural remodeling of the microvasculature with hypertrophy of smaller vessels and atrophy of larger vessels, reduced vascular distensibility, and increased vasoconstriction in response to endothelin.⁹

Microembolization

Plaque fissure or rupture occurs spontaneously and is induced traumatically/iatrogenically by percutaneous coronary interventions. Atherosclerotic debris with superimposed thrombotic material is then dislodged and embolized into the coronary microcirculation where it induces patchy microinfarcts with an inflammatory reaction. Such microinfarcts add to the infarct size caused by sustained coronary occlusion with reperfusion.^90,91 Somewhat paradoxically, the increase in TNF-α expression secondary to coronary microembolization can also induce protection and reduce infarct size from subsequent coronary occlusion/reperfusion⁹² such that the actual impact of coronary microembolization on infarct size depends critically on timing and is difficult to predict.

Stasis and Microvascular Obstruction

Myocardial ischemia and reperfusion increase the expression of adhesion molecules such as intercellular adhesion molecules (ICAM), vascular cell adhesion molecules, and selectins on endothelium and circulating cells, and thereby promote the interaction of platelets, leukocytes, and endothelium and the adherence of platelet aggregates and platelet-leukocyte aggregates to the endothelium.⁹³ Such aggregates are either released from the epicardial atherosclerotic culprit lesion and dislodged into the microcirculation or are formed in the coronary microcirculation. These cellular aggregates contribute to the impairment of microvascular perfusion. In pronounced forms of no-reflow, typical erythrocyte aggregates are found that block the capillaries.⁹⁴

Capillary Destruction and Hemorrhage

The most severe form of coronary microvascular injury from myocardial ischemia/reperfusion, as already detailed in the original report of coronary no-reflow by Kloner,⁷⁰ is the massive swelling of capillary endothelial cells with consequent rupture of the vascular wall and leakage of circulating cells into the interstitium (ie, hemorrhage). No-reflow and hemorrhage carry an adverse prognosis.^95,96

Coronary Microvascular Injury: Cause or Consequence of Myocardial Ischemia/Reperfusion?

The available studies indicate a close correlation between infarct size and no-reflow. Still, correlations cannot resolve questions of causality, and the lack of adequate techniques to make serial measurements of infarcted tissue and no-reflow with reasonable spatial resolution is largely responsible for the fact that causality between myocardial infarction and coronary microvascular no-reflow has not been established. With microembolization of atherosclerotic debris, plugging of platelet/leukocyte aggregates, and vasoconstriction in response to soluble mediators, the resulting coronary MVO could be a cause of myocardial infarction. With this rationale, thrombaspiration, protection devices and coronary vasodilators are used to reduce peri-interventional reperfusion injury, but there is no conclusive evidence that these measures indeed reduce infarct size.⁹⁰ Also, recombinant angiopoietin-like protein-4 has been demonstrated to stabilize the endothelial barrier and subsequently reduce infarct size and no-reflow in mice, but it is unclear whether or not this molecule also has cardiomyocyte-protective properties.⁷⁷ Vice versa, there may be primary damage to cardiomyocytes that only subsequently progresses to coronary microvascular damage, as seen in animal models with mechanical occlusion/reperfusion of virgin coronary arteries without a culprit lesion.⁹⁷ Whether cardiomyocyte damage per se is causal for subsequent coronary microvascular damage or both are consequences of the same fundamental pathomechanism (eg, excessive ROS formation) remains unclear.⁹⁸

Most experimental studies on myocardial ischemia/reperfusion are done in healthy and young animals that have a virgin coronary circulation. Such studies accordingly cannot account for the presence of atherosclerosis with impaired endothelial function, exhaustion of autoregulatory mechanisms, and coronary vascular remodeling distal to coronary stenosis,⁹ the development of a significant collateral circulation, and pre-existing myocardial disease (patchy microinfarcts and fibrosis) or adaptation (hibernation). These limitations contribute to the difficulties in translating data from animal studies to the patient with acute myocardial infarction undergoing reperfusion therapy.²⁸

As depicted in Figure 38–2, circulating cells are critical players in IRI. During ischemia, there is a systemic proinflammatory milieu resulting in the activation of white blood cells (WBC) and platelets, among other cells. Within the WBC, neutrophils play a critical role in the acute phase of myocardial infarction. As described earlier in this chapter, circulating neutrophils and platelets aggregate and form plugins that, upon reperfusion, are distally embolized and can be physically stuck in the microcirculation, contributing to MVO.^{99,100,101,102} The presence of MVO contributes to maintain tissue ischemia despite the opening of the epicardial vessel, something known as the no-reflow phenomenon (see above). Besides the role of neutrophil/platelet plugins in MVO, these cells infiltrate the postischemic myocardial tissue and contribute to the reperfusion-related damage. The presence of neutrophils infiltrating the reperfused myocardium has been traditionally believed to represent a pathophysiologic response to injury¹⁰³; however, this response also contributes to the damage triggered by reperfusion. In fact, influx of neutrophils in the postischemic myocardium is detected long before there is any need for scavenging of any debris. Neutrophils are typically found in close connection with platelets, and the interaction between both cell types is known to occur in many acute stress conditions, not just in IRI. Prolonged ischemia not only affects cardiomyocytes, but endothelial cells are also affected, resulting in a dysfunctional endothelium. Upon reperfusion, activated neutrophils bind to the activated endothelium as a first step to infiltrate the myocardial tissue. The migration of neutrophils is initiated by tethering and rolling on inflamed vessels, a process mediated by endothelial selectins.¹⁰⁴ Activation of integrins then allows firm adhesion, after which leukocytes actively crawl on the endothelium before they extravasate into the reperfused myocardium.¹⁰⁵ A distinct feature of leukocytes recruited to inflamed vessels is the rapid shift from a symmetric morphology into a polarized form, in which intracellular proteins and receptors rapidly segregate.¹⁰⁶ The relevance of neutrophil/platelet interaction in the infiltration of the former into the postischemic tissue has been recently unraveled: neutrophils recruited to injured vessels extend a domain into the lumen and scan for the presence of activated platelets. Only when productive neutrophil-platelet interactions occur do neutrophils organize additional receptors needed for intravascular migration.¹⁰⁷ The contribution of neutrophil infiltration to reperfusion-related damage is established from preclinical studies. The depletion of neutrophils (even by employing leukocyte filters) has been shown to reduce infarct size and preserve coronary endothelial function in several models.^{108,109,110,111} The use of antineutrophil serum, resulting in complete removal of neutrophils from the circulation without affecting other WBC, has been demonstrated to reduce infarct size, clearly demonstrating the relevant role of neutrophils in the damage associated with IRI.¹⁰⁸ Once the role of neutrophils in damage associated with IRI was determined, more sophisticated approaches targeting different neutrophil adhesion receptors were tested. The use of antibodies against Mac-1,^112,113 CD18,¹¹⁴ ICAM-1,^115,116 and P-selectin^117,118 resulted in significant reduction of infarct size in different, but not all,¹¹⁹ preclinical studies. Unfortunately, this solid experimental evidence has not been translated into a clinical benefit. The administration of a monoclonal antibody against CD18 in myocardial infarction patients undergoing thrombolytic therapy¹²⁰ or primary angioplasty¹²¹ did not reduce infarct size in two independent, randomized, placebo-controlled clinical trials. Other indirect approaches targeting postinfarction inflammation, such as the use of the anti-C5 complement antibody pexelizumab, did not reduce infarct size or improve outcomes in randomized clinical trials.^122,123 These neutral effects in the clinical arena have reduced the interest in developing strategies to inhibit neutrophil infiltration into the postischemic myocardium as a therapy to reduce infarct size. However, the clear preclinical evidence¹²⁴ and the potential confounders affecting clinical results leave a door open for this strategy. It might be the case that the targets chosen for the translation into clinical studies (blocking neutrophil-endothelium adhesion) were not the ideal ones. In this regard, it has been recently proposed that the mechanism by which the β₁-selective adrenergic receptor antagonist metoprolol reduces infarct size^18,125 involves the interaction between neutrophils and platelets during IRI.¹²⁶

More than 50 years ago, Jennings et al¹²⁷ first suggested that reperfusion may actually hasten the necrotic process of myocytes irreversibly injured during ischemia, with the concept that reperfusion could induce the death of reversibly injured cardiomyocytes.^8,9 Currently, despite inducing additional damage to the myocardium,¹²⁸ reperfusion is the mainstay treatment for patients with acute ST-segment elevation myocardial infarction (STEMI). Therefore, any therapy able to reduce the size of infarction must do so in conjunction with reperfusion (either mechanically [primary angioplasty] or pharmacologically [thrombolytics]). Despite significant data demonstrating that reperfusion-induced injury significantly contributes to final infarct size (reviewed earlier in this chapter), at present, there is still no effective therapy in routine clinical use for reducing lethal myocardial reperfusion injury in reperfused STEMI or cardiac surgery patients.¹²⁹ Therefore, reperfusion injury is still a neglected therapeutic target.^1,31 In the following sections, we review the strategies potentially available to attenuate IRI and for protecting the heart against acute myocardial infarction and myocardial reperfusion injury.

Ischemic Conditioning

The possibility of “preparing” the myocardium for a subsequent ischemic insult is termed conditioning in the context of this chapter. The execution of brief episodes of ischemia and reperfusion is known as ischemic conditioning if these brief episodes of ischemia occur before the index episode (eg, brief coronary occlusion/reperfusion before prolonged coronary occlusion–myocardial infarction) or after the index episode (eg, brief episodes of coronary occlusion/reperfusion at the end of a prolonged coronary occlusion–myocardial infarction). This conditioning can be local (ie, the brief episodes of ischemia occur at the same organ) or remote (ie, the brief episodes of ischemia occur at a different distant organ) (Fig. 38–6).¹³⁰ The existence of “natural” local IPC has been known for several decades; for example, patients suffering preinfarction anginal episodes (representing brief episodes of spontaneous occlusion/reperfusion [eg, thrombosis-spontaneous lysis] of the coronary artery preceding STEMI) have smaller infarctions and better prognoses than patients without preinfarction angina (ie, straight coronary occlusion).¹³¹ In the following sections, we describe different forms of ischemic conditioning as well as their clinical applications.

Ischemic Preconditioning

For patients suffering a STEMI or undergoing cardiac surgery (a paradigmatic “model” of planned myocardial IRI), the phenomenon known as ischemic conditioning provides an endogenous strategy that is capable of protecting the heart from the detrimental effects of acute IRI and that has the therapeutic potential to improve clinical outcomes in patients with ischemic heart disease.

This remarkable cardioprotective strategy is the term given to a number of related endogenous cardioprotective procedures, which are all based on rendering the heart tolerant to acute IRI by “conditioning” it with one or more brief cycles of ischemia and reperfusion. Discovered by Murry, Jennings, and Reimer in 1986,^20,132 it remains the most consistent and powerful form of cardioprotection at our disposal to date. What makes it so important is that it can be seen in all species examined, including man, and can be readily applied to other organs and tissues (reviewed in Yellon and Downey¹³³ and Bulluck and Hausenloy¹³⁴). Other than direct reperfusion of a blocked coronary artery, IPC remains the most powerful intervention for reducing myocardial infarction size in the ischemic heart.

IPC has been shown to induce two distinct windows of cardioprotection. The first, usually referred to as classic preconditioning, occurs immediately after the IPC stimulus and lasts 2 to 3 hours followed by a disappearing of the effect. This is followed by a second window of protection or delayed effect appearing 12 to 24 hours later and lasting 48 to 72 hours.^135,136 Despite intensive investigation, the actual mechanisms that mediate the cardioprotective effect induced by IPC are not fully understood. Importantly a large number of signaling pathways have been identified, which are briefly mentioned below (for more comprehensive review, see Yellon and Downey¹³³ and Heusch¹³⁷). In brief, the sublethal cycles of short bursts of ischemia and reperfusion, which make up the IPC stimulus, produce a number of endogenous biological factors (ie, autacoids) from the myocyte, including adenosine, bradykinin, endothelin, acetylcholine, and opioids, which can bind to their respective receptors on the plasma membrane. This will direct the appropriate cardioprotective communication pathway that will convey the protective signal to the mitochondria. It is believed that, within the cell, a signaling ROS is produced that activates protein kinases, such as Akt, Erk1/2, tyrosine kinase, and protein kinase C, that provide the “memory” for the cardioprotective effect such that the effect lasts up to 2 to 3 hours (for classical or acute IPC); the role of protein kinase C in such memory is somewhat controversial.¹³⁸ In the second window of protection or delayed conditioning, these protein kinases activate transcription factors (such as STAT1/3, NFκB, AP-1, Nrf2, and HIF-1α), which mediate the synthesis of distal mediators (such as inducible nitric oxide synthase, heat shock protein, and cyclooxygenase-2), which then induce the cardioprotective effect 12 to 24 hours following the IPC stimulus.¹³⁹ In terms of preventing myocardial IRI, IPC has been shown to recruit prosurvival signaling pathways at the onset of reperfusion, including the reperfusion injury salvage kinase (RISK) pathway, first identified by Yellon’s group in 2002 and comprising Akt and Erk1/2,^140,141 and the survivor activator factor enhancement (SAFE) pathway, comprising TNF-α and JAK-STAT3 and identified by Lecour’s group in 2008.^141,142,143

Although the end effectors of cardioprotection in classical and delayed IPC remain unclear, it has been suggested that preservation of mitochondrial function with less calcium overload, attenuated ROS production, and MPTP inhibition all contribute to the protective effect.^133,144,145 It is hoped that functional genomics of myocardial tissue may provide further insights into the potential mechanisms underlying IPC strategies.¹⁴⁶

Clinical Application of Ischemic Preconditioning

The first clinical study to apply an IPC stimulus in patients was by Yellon’s group in 1993 in the setting of coronary artery bypass grafting (CABG) surgery.¹⁴⁷ They found that intermittent clamping and declamping of the aorta of patients undergoing CABG surgery preserved myocardial ATP levels in a similar manner to that seen by Murry and colleagues in their seminal preconditioning paper. Since then, a number of small studies¹⁴⁸ have confirmed the cardioprotective effect of IPC, as measured by serum cardiac enzymes, in terms of reducing the extent of perimyocardial injury in patients undergoing CABG surgery; a recent meta-analysis of 22 trials of 933 patients found that IPC reduced ventricular arrhythmias, decreased inotrope requirements, and shortened the intensive care unit stay.¹⁴⁹ Despite these potential beneficial effects, the need to intervene on the heart directly and the inherent risk of thromboembolization arising from clamping an atherosclerotic aorta have prevented IPC from being adopted in this clinical setting.¹⁵⁰

Interestingly it has been suggested that the phenomenon of IPC can be observed in a number of clinical scenarios in which the heart is able to protect itself with brief episodes of ischemia (eg, in the setting of “warm-up angina,” which refers to the occurrence in which a patient with stable ischemic heart disease is able to increase his or her exercise tolerance following an episode of angina followed by a period of rest).¹⁵¹ In addition, antecedent angina prior to an acute myocardial infarction, or preinfarct angina, has also been suggested to account for the cardioprotective effects of smaller myocardial infarction size and improved clinical outcomes.¹³¹

Ischemic Postconditioning

The major disadvantage of IPC as a cardioprotective strategy is the requirement to intervene prior to the index ischemic event, which in the case of an acute myocardial infarction is not possible. However, in 2003, Zhao et al⁶ made the exciting discovery that the heart could be protected against acute myocardial infarction by interrupting myocardial reperfusion with several short-lived episodes of myocardial ischemia; they termed this phenomenon IPost.¹⁵² These authors found that applying three cycles of 30-second left anterior descending coronary artery reocclusion and reflow within 1 minute of myocardial reperfusion could reduce myocardial infarction size by 44% in the canine heart.⁶ In addition, IPost was found to confer a myriad of protective effects with preserved endothelial function and reduced levels of myocardial edema, oxidative stress, and polymorphonuclear neutrophil accumulation, findings that were consistent with there being less myocardial reperfusion injury in postconditioned hearts.⁶

Interestingly, the concept of modifying reperfusion as a strategy to limit myocardial infarction size had already been introduced in the 1980s with the strategies of gentle⁵ and gradual¹⁵³ reperfusion.⁷ Indeed, in 1996, Na et al¹⁵⁴ had already coined the term postconditioning when they described the phenomenon by which intermittent reperfusion, induced by ventricular premature beats, prevented reperfusion-induced ventricular fibrillation in ischemic feline hearts. However, the concept of IPost has captured the imagination and restored efforts to target myocardial reperfusion injury as a therapeutic strategy for reducing myocardial infarction size. Importantly, this phenomenon has been shown to reduce myocardial infarction size in rodents, rabbits, pigs, and other species, including man, although the cardioprotective effect of IPost does not appear to be as robust as IPC.^137,155

IPost appears to share many of the same signaling mechanisms recruited at the time of reperfusion by IPC,^134,137 which include the activation of cell-surface receptors on the cardiomyocyte (such as adenosine, bradykinin, opioids) and the recruitment of prosurvival signaling pathways (such as the RISK, SAFE, and cyclic guanosine monophosphate pathways)¹⁵⁶ that converge on mitochondria and mediate cardioprotection by preserving mitochondrial function (less calcium overload, attenuated oxidative stress, and inhibited MPTP opening).

Ischemic Postconditioning in ST-Segment Elevation Myocardial Infarction

The ability to apply the therapeutic intervention at the onset of reperfusion in STEMI patients has greatly facilitated the translation of IPost into the clinical setting,^73,157 where it was shown that following direct stenting, IPost, performed within 1 minute of reflow by four episodes of 1-minute inflation and 1-minute deflation of an angioplasty balloon positioned upstream of the stent, reduced enzymatic myocardial infarction size (total creatine kinase) by 36% and improved myocardial perfusion (assessed by myocardial blush grade).¹⁵⁷ This study provided the first evidence that confirmed the existence of myocardial reperfusion injury in man,¹⁵⁸ as it clearly demonstrated that intervening at the onset of reperfusion was able to reduce myocardial infarction size.

A number of small studies have confirmed the myocardial infarction–limiting effects of IPost using serum troponin release,¹⁵⁹ myocardial single-photon emission computed tomography,^160,161 and cardiac magnetic resonance imaging,¹⁶² with apparent long-term benefits on cardiac function.¹⁶¹ However, a number of recent studies have also failed to show any beneficial effects with IPost,¹⁶³ and there are even some studies reporting possible detrimental effects with IPost.^164,165 Although recent meta-analyses have confirmed the myocardial infarction–limiting effects of IPost in STEMI patients,^166,167,168 the largest clinical study of more than 700 STEMI patients failed to find any benefits with IPost in terms of ST-segment resolution, peak creatine kinase myocardial band levels, myocardial blush grade, or major adverse cardiovascular event at 30 days.¹⁶⁹ The reasons for these discrepancies are not fully understood but may have a variety of explanations.^28,170,171

Remote Ischemic Conditioning

The major disadvantage of IPC and IPost is that they both require an intervention to be applied to the heart directly; therefore, with the discovery of the phenomenon of remote preconditioning by Przyklenk and colleagues,¹⁷² the idea that conditioning may indeed have a practical role to play in protecting the heart from IRI has been realized. Remote ischemic conditioning (RIC) is the phenomenon whereby the application of one or more brief cycles of nonlethal ischemia and reperfusion to an organ or tissue protects the heart against a lethal episode of acute IRI.

The actual mechanism through which an episode of brief ischemia and reperfusion in an organ or tissue away from the heart exerts protection against a subsequent sustained insult of acute myocardial IRI injury is currently unclear.^173,174,175 It has been suggested that the underlying mechanistic pathways and signaling cascades activated within the protected organ may be similar to those recruited in the setting of IPC and IPost (Fig. 38–7).^137,173 What remains uncertain is the mechanistic pathway responsible for conveying the cardioprotective signal from the remote preconditioned organ or tissue to the heart. The current paradigm suggests a neurohormonal pathway is central to the protective effect underlying RIC. In this regard, it has been shown that a substance or humoral factor generated by the preconditioning ischemia is transported to the heart to elicit protection. It has been shown that blood taken from a rabbit that has been preconditioned reduced myocardial infarction size when transfused into a naïve rabbit,¹⁷⁶ suggesting the transfer of one or more humoral cardioprotective factors. Furthermore hexamethonium (a ganglion blocker),¹⁷⁷ resection of the neural innervation of the limb,^178,179 genetic inhibition of preganglionic vagal neurons in the brainstem,¹⁸⁰ and resection of the vagal nerve supply to the heart¹⁸¹ have all been shown to abrogate the myocardial infarction–limiting effects of limb RIC, suggesting the requirement for an intact neural pathway for RIC cardioprotection. The exact details of the neural pathway have not been completely elucidated. The stimulation of the neural pathway in the RIC-treated organ or tissue appears to be a result of the local production of autacoids such as adenosine^182,183 and bradykinin.¹⁸⁴ Furthermore, proteomic analysis of plasma harvested from RIC-treated animals has failed to identify the cardioprotective factor(s), although the evidence suggests that the factor(s) is thermolabile and hydrophobic and is in the range of 3 to 8.5 kDa in size.^{182,183,184,185,186} Recent studies have provided experimental evidence implicating calcitonin gene-related peptide,¹⁸⁷ stromal derived factor-1α,¹⁸⁸ nitrite,¹⁸⁹ and microRNA-144¹⁹⁰ as possible mediators of RIC cardioprotection, although conclusive evidence for their role as the bloodborne transferrable cardioprotective factors in RIC is lacking.

Furthermore, the generation of this cardioprotective factor has been shown to be dependent on an intact neural pathway to the RIC-treated limb,¹⁹¹ with neural stimulation of the limb using direct nerve stimulation,¹⁹² electroacupuncture,¹⁹³ topical capsaicin,¹⁹² or transcutaneous electrode stimulation¹⁹⁴ generating a bloodborne transferrable cardioprotective factor and reducing myocardial infarction size in animal models. Finally, Jensen et al¹⁹⁵ have confirmed the need for an intact neural pathway to the limb by showing that the bloodborne transferrable cardioprotective factor was not produced by limb RIC in diabetic patients with a sensory neuropathy of the limb. In the myocardium, RIC activates both the RISK and the SAFE pathways, and their activation is species dependent¹⁹⁶; in humans, RIC activates STAT5 in the myocardium.¹⁹⁷

Studies are required to tease out the exact interplay between the neuronal and hormonal pathways underlying RIC and to identify the bloodborne cardioprotective factor(s) that mediate RIC cardioprotection.

Clinical Application of Remote Ischemic Conditioning

The discovery by Kharbanda et al¹⁹⁸ that cardioprotection could be achieved noninvasively by simply inflating and deflating a blood pressure cuff placed on the upper arm or thigh has facilitated a simple means of translating this phenomenon into the clinical setting. With the rush to clinical trial, the RIC stimulus to the limb itself has not been properly characterized and appears to be taken from experimental animal models most often using 5 to 15 minutes. However, the most effective limb RIC stimulus is still not known. Importantly, however, RIC can be delivered at different time points with respect to the ischemia/reperfusion insult, including 24 hours prior to the index ischemia (delayed remote IPC), immediately prior to the index ischemia (remote IPC), after the onset of index ischemia but prior to reperfusion (remote ischemic perconditioning),¹⁹⁹ at the onset of reperfusion (remote IPost),^200,201 and even 15 to 30 minutes into reperfusion (delayed remote IPost),²⁰² allowing its application in a wide variety of clinical settings of acute IRI.

Remote Ischemic Conditioning in Cardiac Surgery

RIC using limb ischemia was first investigated by Guanydin et al in 2000²⁰³ in a small eight-patient study, although perioperative myocardial injury was not assessed. In 2006, Cheung et al²⁰⁴ were the first to demonstrate a cardioprotective effect with limb RIC in children undergoing cardiac surgery for congenital heart disease. They used four 5-minute cycles of lower limb ischemia and reperfusion induced by inflating and deflating a blood pressure cuff on the thigh and demonstrated reduced perioperative myocardial injury (less troponin I), reduced inotrope requirements, and decreased airway pressure. Yellon’s group went on to report beneficial effects in terms of a 43% reduction in perioperative myocardial injury (72-hour area under the curve troponin T)²⁰⁵ in adults undergoing CABG surgery.

Although recent meta-analyses appeared to confirm the cardioprotective effect of limb RIC in terms of reducing perioperative myocardial injury, the fundamental question was whether such a supposed benefit could be seen in terms of clinical outcome.^{206,207,208,209} Most important in this regard are the recently reported large prospective multicenter randomized clinical trials (adequately powered to detect major adverse cardiovascular events) that confirm that limb RIC has no beneficial effects on major clinical outcomes in patients undergoing cardiac surgery.^210,211,212

There have been many proposed reasons why RIC has not been observed in the setting of CABG, and this is the subject of much debate.^{213,214,215,216}

Remote Ischemic Conditioning in Planned Percutaneous Coronary Intervention

The setting of planned percutaneous coronary intervention (PCI) has also been used to examine the potential cardioprotective effects of RIC. It is known that injury occurs in about 30% of stable patients and up to 80% of unstable patients undergoing urgent PCI. The injury is measured by the release of serum cardiac enzymes. However, the etiology of this form of injury is not a result of acute IRI per se, but is mainly caused by acute ischemic injury (arising from distal branch occlusions and coronary embolization), which are complications that are more frequent following multivessel and complex PCI.²¹⁷ In 2006, Iliodromitis et al²¹⁸ were the first to investigate limb RIC in this clinical setting and found that in 41 patients RIC using bilateral upper arm cuff inflations/deflations actually exacerbated myocardial injury. In 2010, in a much larger study, Hoole et al²¹⁹ found that this intervention reduced the magnitude of PCI-related myocardial injury. The reasons for this discrepancy are not clear, but a recent meta-analyses found benefit in patients undergoing elective coronary intervention.²²⁰

Remote Ischemic Conditioning in ST-Segment Elevation Myocardial Infarction

Compared to the above settings of CABG and PCI, the clinical setting of STEMI provides the most promising setting to investigate any cardioprotective therapy that targets myocardial IRI. To date, a number of proof-of-concept studies have reported cardioprotective effects with limb RIC in STEMI patients treated by primary angioplasty. This form of treatment is effective when given by paramedics in the ambulance,²²¹ as well as on arrival at the hospital prior to primary angioplasty^222,223 and at the onset of reperfusion at the time of primary angioplasty.²²⁴ Importantly, as with the setting of CABG mentioned earlier, outcome studies will inform us as to whether RIC can indeed find a place in cardiovascular medicine, and this is now being jointly investigated in the CONDI2 and ERIC-PPCI trials (ClinicalTrials.gov Identifier: NCT01857414). The aim of these studies is to demonstrate whether RIC can reduce the rates of cardiac death and hospitalization for heart failure at 12 months.

Pharmacologic Cardioprotection

Finding appropriate pharmacologic therapy to protect against IRI has been ongoing for many years. It was with the improved understanding of the pathophysiology of ischemia/reperfusion, as well as the advent of the mechanisms associated with preconditioning, that the identification of a number of molecular targets amenable to pharmacologic manipulation became apparent.^{132,133,136,139,140,141,218}

The history of pharmacologic cardioprotection has been hugely disappointing, with antioxidants, magnesium, calcium channel blockers, anti-inflammatory agents, erythropoietin, and atorvastatin all failing to reduce myocardial infarction size and/or improve clinical outcomes.²²⁵ A number of pharmacologic cardioprotective strategies, such as adenosine²²⁶ and glucose-insulin-potassium therapy,²²⁷ have had mixed results, with cardioprotective efficacy depending on study design. A more targeted pharmacologic approach has also failed to limit myocardial infarction size or improve clinical outcomes in recent clinical studies that investigated the cardioprotective effects of therapeutic hypothermia,^228,229 mitochondria-targeted agents (eg, MTO-131),²³⁰ cyclosporine A (CsA), and modulation of nitric oxide signaling using nitrite or inhaled nitric oxide.^231,232 Although CsA was shown to reduce myocardial infarction size in an initial proof-of-concept clinical study,²³³ it failed to improve clinical outcomes in a subsequent large multicenter clinical trial (the CIRCUS trial),²³⁴ illustrating the challenge of translating cardioprotection into clinical benefit. The reason for the failure of CsA to reduce myocardial infarction size and improve clinical outcomes in STEMI patients is not completely understood. A number of factors to consider include the inconclusive preclinical data, with some experimental studies failing to show a cardioprotective effect with CsA administered at reperfusion; the limited clinical data with only one positive clinical study with CsA in STEMI; the use of the CicloMulsion formulation of CsA; and the potential failure of CsA to reach its molecular target in time.^235,236

Metoprolol

On a more optimistic note, there are a number of pharmacologic strategies that have shown promise. An interesting example of this is metoprolol. Early β-blocker therapy in reperfused STEMI patients is controversial and had largely been investigated in the pre-reperfusion era. However, recently, Ibanez and colleagues demonstrated that intravenous (IV) metoprolol administered prior to reperfusion in a porcine model reduced myocardial infarction size.²³⁷ They followed this by undertaking a clinical trial (METOCARD-CNIC trial) demonstrating that IV metoprolol administered in the ambulance prior to primary angioplasty reduced myocardial infarction size and improved clinical outcomes (as a secondary end point) in anterior STEMI patients presenting early (< 6 hours from symptom onset).^18,125 More importantly, in the METOCARD-CNIC trial, not only did patients receiving pre-reperfusion IV metoprolol have CMR-evaluated smaller infarctions^18,238 and better long-term LVEF,¹²⁵ but also the incidence of left ventricular severe systolic dysfunction was significantly reduced.¹²⁵ Recently, the results of the EARLY BAMI trial have been reported. This trial recruited 600 STEMI patients (any location) presenting within 12 hours from symptom onset. Patients were randomized to IV metoprolol (10 mg) or placebo.²³⁹ The primary end point was myocardial infarction size as assessed by CMR 1 month after infarction. The trial was neutral, and infarct size was not smaller in patients allocated to IV metoprolol. There were no signs of adverse effects in patients receiving IV metoprolol, and the incidence of ventricular fibrillation was significantly lower in metoprolol-treated patients. These data support the safety of this strategy in Killip I-II STEMI patients.

There are important differences between the METOCARD-CNIC and EARLY BAMI trials. Dose and timing of IV metoprolol administration were different between trials. In contrast to the METOCARD-CNIC trial, in the EARLY BAMI trial, patients received only one 5-mg dose at recruitment, and per protocol, the second dose was given in the catheter lab immediately before PCI. In fact, the first dose of metoprolol did not have any effect on blood pressure or heart rate, suggesting an underdosing effect. In this regard, a recent subanalysis from the METOCARD-CNIC trial demonstrated that the longer the “onboard” metoprolol time at the time of reperfusion, the higher the infarct-reduction effect.²⁴⁰ In fact, patients receiving IV metoprolol close to reperfusion had a mild protective effect, whereas those with a long time between the metoprolol 15-mg bolus and reperfusion had the largest reduction in infarct size and greatest improvement in long-term LVEF. These differences in dose and timing of metoprolol administration might explain the different conclusions from both trials. Given the clear safety profile and the low cost of this therapy (< $2.00), it is worth continuing the clinical research and performing a definite large end-point–powered trial. In the near future, the MOVE ON! trial (Ibanez and Fuster, principal investigators) will be initiated, and more than 1200 anterior STEMI patients will be recruited and randomized to IV metoprolol (15 mg immediately after diagnosis is made in the out-of-hospital setting) or placebo. The primary end point will be the composite of cardiovascular death, heart failure, implantable cardioverter-defibrillator insertion, or severe left ventricular dysfunction.

P2Y₁₂ Inhibition

As we know, optimal medical management of acute coronary syndromes at the time of initial presentation consists of many facets that include analgesia, supportive therapies where required (eg, inotropes, mechanical intra-aortic balloon pump), and dual antiplatelet therapy consisting of both the cyclooxygenase inhibitor aspirin and an adenosine diphosphate P2Y₁₂ receptor antagonist (clopidogrel, prasugrel, or ticagrelor). These latter antiplatelet interventions are orally loaded at the time of presentation in an attempt to reduce platelet aggregation and optimize postinterventional outcomes and coronary flow in the culprit vessel(s). Of great interest is the unanticipated significant mortality benefit of the P2Y₁₂ inhibitor ticagrelor in the PLATO study.²⁴¹ However, the benefits of this drug may not be fully explained by a pure antiplatelet effect. In this regard, ticagrelor has been shown to increase the levels of extracellular adenosine,²⁴² a mediator known to exert a wide range of benefits, including vasodilation, inhibition of platelet aggregation, and leukocyte adherence to the vessel wall. In addition, it has recently been shown that cangrelor is significantly cardioprotective in mice,²⁴³ rats,²⁴⁴ rabbits,¹⁷⁰ and primates.²⁴⁵ Interestingly, the protection conferred by cangrelor is dependent upon the presence of blood, with no evidence of protection ex vivo in crystalloid-perfused Langendorff heart.^170,244 Downey’s group has shown that this protection is mediated through pathways typically recruited by ischemic conditioning, suggesting that P2Y₁₂ inhibition, via a blood component, leads to conditioning-like protection.^170,244 Therefore, IV P2Y₁₂ inhibition may thus have the dual advantage of optimizing both platelet inhibition and offering cardioprotection. Consequently, although the protective potential of these agents in the clinical arena has yet to be fully realized, given the strength of preclinical data and plausible mechanism, there is clear potential for harnessing and optimizing P2Y₁₂ inhibition at the time of reperfusion with the exciting prospect for further improving clinical outcomes.

Other Promising Agents

Another pharmacologic approach that has shown promise is that of exenatide, a synthetic version of exendin-4 (a peptide isolated from the saliva of the Gila lizard). Exenatide is an analogue of glucagon-like peptide-1 (GLP-1), a hormone that lowers blood glucose by stimulating insulin secretion.²⁴⁶ Interestingly, GLP-1 and its analogues, such as exenatide, have been reported in experimental animal studies to reduce myocardial infarction size through the activation of prosurvival intracellular signaling pathways when administered prior to reperfusion.^247,248 Clinical studies have found that administering exenatide prior to primary angioplasty increased salvage index in STEMI patients,^{248,249,250,251} with most benefit observed in patients presenting with shorter ischemic times (< 132 minutes).²⁵⁰ Whether this effect of exenatide translates to improved clinical outcomes in reperfused STEMI patients is not known and remains to be determined in a large randomized clinical trial.

Atrial natriuretic peptide given prior to myocardial reperfusion has also been shown to reduce myocardial infarction size in experimental animal studies. This is believed to occur via the activation of known prosurvival signaling pathways.²⁵² Kitakaze et al²⁵³ translated this therapeutic approach in a large clinical study comprising 569 STEMI patients, in which administering carperitide (an atrial natriuretic peptide analogue) at the time of primary PCI was associated with a modest reduction in myocardial infarction size. Further studies are required to confirm these findings and investigate whether this therapeutic approach can improve clinical outcomes in reperfused STEMI patients.

Combination Reperfusion Therapy: A Novel Therapeutic Strategy

As detailed earlier, most attempts to reduce myocardial infarction size in STEMI patients have relied on using a single agent to target one single component of myocardial reperfusion injury. As detailed in Figure 38–2, myocardial IRI is the result of several mechanisms, and thus targeting on individual phenomenon will unlikely reduce infarct size. The possibility of targeting several mechanisms simultaneously (either with one agent targeting different layers or by several agents administered simultaneously) is attractive, although not widely undertaken. The use of combination reperfusion therapy is still in its infancy, but because of the complex pathophysiology associated with IRI, this could be considered an important option. Interestingly, the RISK and SAFE signaling pathways have been shown to interact in remote limb perconditioning in combination with local IPost.²⁵⁴ Alburquerque-Béjar et al²⁵⁵ recently combined RIC with glucose-insulin-potassium and exenatide in a porcine acute myocardial infarction model and demonstrated an additive benefit in terms of myocardial infarction size reduction. The Combination Therapy in Myocardial Infarction (COMBAT-MI) study (ClinicalTrials.gov identifier: NCT02404376) is set to investigate the potential benefits of using RIC with exenatide on myocardial infarction size reduction in STEMI patients treated by primary PCI. Although an initial small clinical study failed to show an additive cardioprotective effect with RIC and IPost administered in combination in reperfused STEMI patients,²⁵⁶ a recently published large study found increased myocardial salvage in patients given RIC and IPost in combination when compared with patients given IPost alone.²⁵⁷