Chapter 15. Myocardial Protection

In the setting of heart surgery, myocardial protection refers to strategies and methodologies used in the operating room to attenuate or prevent perioperative infarction and/or postischemic ventricular dysfunction. This is in contrast to the instance in which a patient presents with an acute myocardial infarction (MI). Here the objective is to reduce infarct size at reperfusion. The underlying pathophysiology in both settings, however, relates to the etiology and consequences of ischemia-reperfusion injury. After surgery the injury manifests by low cardiac output, hypotension, and a need for postoperative inotropic support. The damage may be reversible or irreversible and is differentiated by the presence of electrocardiographic abnormalities, elevations in the levels of specific plasma enzymes or proteins such as creatine kinase and troponin I or T, and/or the presence of regional or global echocardiographic wall motion abnormalities. Depending on the criteria, the incidence of postoperative MI after CABG surgery ranges between 3% and 18%. The incidence of severe ventricular dysfunction, heart failure, and death, despite advances in surgical techniques, ranges between 2 and 15%; the higher mortality rates are associated with high-risk patients with minimal cardiac reserve.

These complications have an enormous impact on both families and society. From an economic standpoint alone, revascularization procedures are costly. In 2004 cardiovascular disease was responsible for more than 1 million hospital stays; it was the most expensive condition treated, with a total cost of more than $44 billion. More than half of the hospitalizations for coronary atherosclerosis were for patients who received percutaneous coronary intervention (PCI) or coronary artery bypass graft (CABG) surgery.1 With respect to CABG surgery alone, the initial hospital cost is approximately $10 billion annually; the complications after CABG consume an additional $2 billion in U.S. health care resources each year.2–5 A reduction in perioperative complications associated with heart surgery could have a significant impact on resource utilization and overall operative costs. Because one cause of morbidity and mortality after heart surgery is ischemia-reperfusion injury, the purpose of this chapter is to review its underlying mechanisms, review the history of myocardial protection, update the reader regarding the current protective techniques, and discuss new strategies under investigation.

The etiologies of perioperative myocardial necrosis and postischemic myocardial dysfunction after cardiac surgery are multifactorial. Myocardial necrosis and associated cardiac biomarkers may arise as a result of ischemia caused by intrinsic coronary disease not amenable to revascularization, anesthetic factors, atrial cannulation, aortic cross-clamping, suturing of heart muscle, plaque and platelet embolism, and graft spasm or thrombosis. Despite the considerable progress made to date, high-risk heart surgery patients including those with unstable angina, poor ventricular function, diabetes, repeat CABG, and advanced age continue to experience postoperative complications such as low cardiac output, perioperative myocardial infarction, and heart failure requiring prolonged intensive care. In many of these patients these complications are caused by ischemia-reperfusion injury and inadequate myocardial protection. Thus there is a compelling unmet need to develop new methods to protect the heart during surgery.

Deleterious Sequelae of Ischemia-Reperfusion

Myocardial ischemia-reperfusion injury can manifest as postischemic “stunning,” which is reversible and/or apoptosis or myocardial infarction, which are irreversible. Myocardial stunning is an injury that lasts for hours to days despite the restoration of normal blood flow. Typically, these patients require some form of temporary inotropic support in the immediate postoperative period in order to maintain an adequate cardiac output. Stunned cardiomyocytes exhibit minimal ultrastructural damage that resolves within hours to days following relief of ischemia. Apoptosis, or programmed cell death, is a pattern of cell death that affects single cells. It is characterized by retention of an intact cell membrane, cell shrinkage, chromatin condensation, and phagocytosis without inflammation.6,7 Apoptotic cell death caused by ischemia-reperfusion contributes significantly to the development of infarction as well as the loss of cells surrounding the infarct area. A large fraction of dying cells may exhibit features of both apoptosis and necrosis, ie, both nuclear condensation and plasma membrane damage. Ultimately, after prolonged ischemia, irreversible cellular injury is characterized by a broad spectrum of reperfusion-related pathologies including membrane destruction, cell swelling, ultra-structural changes of mitochondria, DNA degradation, cytolysis, and the induction of an inflammatory response.8–10

Long-Term Clinical Consequences

Although the consequences of inadequate myocardial protection usually are apparent in the immediate postoperative period, eg, low cardiac output syndrome, the full impact of inadequate protection may not be appreciated for months to years. Klatte and colleagues reported that patients with increased peak creatine kinase-myocardial band (CK-MB) enzyme ratios after CABG surgery exhibited a greater 6-month mortality.11 The 6-month mortality rates were 3.4%, 5.8%, 7.8%, and 20% for patients with peak CK-MB ratios <5, 5-10, 11-20, and >20-fold greater than the upper limits of normal (UNL), respectively. Conversely, the cumulative 6-month survival was inversely related to the peak CK-MB ratio.

In another study, Costa12 and colleagues reported that normal postoperative CK-MB levels were observed in only 38.1% of 496 patients who underwent CABG surgery. When the CK-MB levels were stratified, the incidence of death at 30 days was 0.0%, 0.5%, 5.4%, and 7.0% when the enzyme levels were normal, 1–3, 4–5, and >5-fold higher than ULN, respectively. The one year mortality for these groups was 1.1%, 0.5%, 5.4%, and 10.5%, respectively. The peak postoperative cardiac enzyme level correlated significantly with worse clinical outcome. Thus, although CK-MB elevations are often dismissed as inconsequential in the setting of multivessel CABG surgery, they occur frequently and are associated with significant increases in both repeat myocardial infarction and death beyond the immediate perioperative period. Consistent with this is the evidence by Steuer 13 and associates in which they examined postoperative serum aspartate aminotransferase and CK-MB levels and their relationship to early cardiac-related death and late survival in 4911 patients who underwent CABG consecutively during a 6-year period. They reported that elevated enzyme levels on the first postoperative day greatly increased the risk of early cardiac death and were associated with a 40 to 50% increased risk in late mortality at 7 years.

Finally, in a retrospective analysis Brener et al. showed that CK-MB elevation was common after both percutaneous and surgical revascularization.14 The incidence of CK-MB elevation above the normal range was 90% for CABG surgery and 38% for PCI. The elevations exceeded 10× ULN for 6% of the CABG patients and 5% of the PCI patients. After 3 years average follow-up, the cumulative mortality was 8% for CABG surgery, and 10% for PCI. Thus, even relatively small CK-MB elevations after interventions are associated with increased mortality over time. Similar observations were made when troponin release was used as a biomarker of myocardial injury. Lehrke et al. reported in a series of 204 patients that a serum troponin-T concentration of 0.46 μg/L or more 48 hours after surgery was associated with a 4.9-fold increase in long-term risk of death.15 In short, it appears that biomarkers are frequently increased after CABG surgery, and they are associated with a decrease in short-, medium-, and long-term survival. Although the greater enzyme release may reflect underlying pathology, ie, a heart that is more vulnerable to injury, it is also reasonable to suggest that limiting myocardial necrosis will have a beneficial effect on survival. Therefore, a better understanding of the mechanisms underlying ischemia-reperfusion injury is needed. An appreciation of current techniques and development of new approaches are warranted in order to reduce long-term morbidity and mortality.

Cellular Mediators

The primary mediators of ischemia-reperfusion injury include intracellular Ca2+ overload and oxidative stress induced by reactive oxygen species (ROS) generated at the onset of reperfusion (Fig. 15-1).16 The molecule nitric oxide (NO) also can interact with molecules derived from superoxide (O2-) or peroxides to generate reactive nitrogen species that are capable of inducing injury as well.17 In addition, metabolic alterations occurring during ischemia can contribute directly and indirectly to Ca2+ overload and ROS formation. For example, decreased cytosolic phosphorylation potential, ie, [ATP]/([ADP] × [Pi ]), results in less free energy from ATP hydrolysis than is necessary to drive the energy-dependent pumps (sarcoplasmic reticulum Ca2+-ATPase, the sarcolemmal Ca2+-ATPase) that maintain intracellular calcium homeostasis. 18 With the onset of ischemia and the fall in intracellular pH, there is a concomitant activation of the Na+-H+ exchanger. This results in an accumulation of intracellular Na+. At the onset of reperfusion there is a reversal of the Na+-Ca++ exchanger, which in turn exacerbates intracellular calcium accumulation that leads to injury of the sarcoplasmic reticulum, opening of the mitochondrial permeability transition pore (mPTP), and lethal injury of the myofibrillar contractile elements.19

The metabolic changes that occur during ischemia also reduce the endogenous antioxidant defense systems of cardiac myocytes. The first line of defense against mitochondrial ROS formation and its deleterious effects is the GSH (reduced glutathione)/GSSG (oxidized glutathione) system, which is directly linked to the NADPH:NADP+ ratio via the enzyme glutathione reductase. The depletion of glutathione levels increases ROS formation, oxidative stress, and [Ca2+]i. 20,21 Because NADPH is not formed during ischemia, the normal metabolic mechanism for regenerating the reduced glutathione does not function. Thus, the formation of ROS during reperfusion occurs at a time when the myocyte's endogenous defense mechanisms are depressed. The NADPH:NADP+ ratio is a primary determinant of the redox state of the cell, and there is evidence that redox state plays a key role in determining the bioactivity and redox state of NO.17,22 In addition, there are several reports that in the absence of normal levels of its cofactors, nitric oxide synthase (NOS) itself can generate superoxide anion.23Although systolic intracellular calcium [Ca2+]i may return to normal early in the reperfused stunned myocardium, transient increases in [Ca2+]i can activate Ca2+-dependent isoforms of protein kinase C (PKC), proteases such as calpain, and endonucleases.24 Calpain activation and its subsequent action on contractile proteins have been implicated in the reduction of myofilament Ca2+ sensitivity observed in stunned myocardium.25

Similarly, there is considerable evidence that ROS are involved in mediating myocardial stunning. Various spin-trap agents and chemical probes have demonstrated the rapid release of ROS into the vascular space during reperfusion after brief ischemia in vivo.26 It is also now recognized that mitochondria are a primary source of intracellular ROS in cardiac myocytes.27,28 There is no question that scavengers of ROS and antioxidants, when administered before the onset of ischemia, attenuate myocardial stunning in vitro and in vivo. Moreover, although the evidence is equivocal, efficacy is in some instances maintained when scavengers and antioxidants are administered before or at the onset of reperfusion.16,29 It has been shown that ROS can attack thiol residues of numerous proteins such as the SR Ca2+-ATPase, the ryanodine receptor, and contractile proteins.30 This may explain why myofibrils isolated from in vivo reperfused stunned but not ischemic myocardium exhibit reduced Ca2+ sensitivity.31 More prolonged ischemia, which produces irreversible injury, is associated with more severe intracellular Ca2+ overload and further depletion of endogenous antioxidants, conditions that both contribute to and are exacerbated during reperfusion by the production of ROS. The production of ROS during reperfusion appears to contribute to Ca2+ overload because exposure of normal myocytes to exogenous ROS is associated with increased L-type Ca2+ channel current and increased [Ca2+]i. 21,32,33 Conversely, increases in [Ca2+]i during ischemia-reperfusion may adversely affect mitochondrial function, leading to further ROS production.33,34 Mitochondria can buffer small increases in intracellular Ca2+ via the Ca uniporter, a process that is energetically favorable owing to the [Ca2+] gradient and the mitochondrial membrane potential. During reperfusion, the increase in cytosolic Ca2+ enhances mitochondrial Ca2+ uptake. Because excess cytosolic Ca2+ has been associated with the loss of myocyte viability, mitochondrial Ca2+ buffering is initially cardioprotective.35 However, continued mitochondrial Ca2+ buffering in the face of decreased antioxidant reserves and excess ROS formation sets up a cycle that ultimately may lead to the total collapse of mitochondrial membrane potential and cell death. The synergistic interactions between Ca2+ overload and ROS formation during conditions of decreased antioxidant reserves also may provide an explanation of why ROS scavengers are not very effective at reducing irreversible injury when administered at reperfusion.36

Broadening the Spectrum of Ischemia-Reperfusion Injury

Historically, myocardial ischemia-reperfusion injury has been characterized as either reversible or irreversible based on staining techniques, enzyme release, and histology. There is now increasing evidence that this injury represents a transition from reversible to irreversible injury and that it occurs as a continuum, not as an all-or-none phenomenon. For example, apoptosis occurs before severe depletion in ATP and loss of membrane integrity but ultimately leads to cell death.37,38 The phenomenon of apoptosis appears to commence during reperfusion with the formation of intracellular ROS and/or intracellular calcium overload39,40 (Fig. 15-2). This process is initiated by translocation of the proapoptotic proteins Bad and Bax from the cytosol to the mitochondrial membrane. Heterodimerization of Bad or Bax with the antiapoptotic Bcl-2 or Bcl-xl can lead to the release of cytochrome c from mitochondria into the cytosol.40–42 Formation of a cytosolic complex consisting of cytochrome c, apoptosis activating factor 1 (APAF-1), and caspase-9 leads to activation of caspase 3 and the cleavage of poly(ADP)-ribose polymerase (PARP) protein. Activation of PARP is the final step in apoptosis, activated in part by DNA fragmentation and the rapid consumption of the remaining adenine nucleotides. As described earlier, the increased intracellular ROS and/or intracellular calcium overload collapse the mitochondrial membrane potential, leading to mPTP opening, which, if not reversed, can result in mitochondrial swelling, rupture of the outer mitochondrial membrane, and release of cytochrome c. However, more commonly, opening of the mPTP leads to necrotic cell death, in which the mitochondrial F0F1 ATP synthase runs in reverse, hydrolyzing ATP in a futile effort to restore mitochondrial inner membrane potential. The depletion of the already-limited cytosolic supplies of ATP compromises the ability to maintain ion homeostasis via the Na+/K+ ATPase, culminating in swelling and rupture of the plasma membrane.

The physiologic relevance of apoptosis during myocardial ischemia-reperfusion has yet to be determined. This is owing to the fact that the majority of reports on apoptosis in this setting have been based on measurements of DNA fragmentation and laddering, the final steps in apoptotic cell death. Once DNA is fragmented, the cell's ability to synthesize new proteins to repair itself is severely compromised, and these cells, even if they survive a first ischemic episode, may die at an accelerated rate during subsequent stress or ischemia. However, studies conducted in other tissues and in isolated cells including cardiomyocytes indicate that the apoptotic program can be detected much earlier. One of the earliest signs of apoptosis is the translocation of phosphatidylserine from the inner face of the plasma membrane to the cell surface, a process that can be detected by annexin V, which has a strong affinity for phosphatidylserine.43 Apoptosis in cardiac myocytes can be demonstrated with fluorescein isothiocyanate (FITC)-conjugated annexin V staining of the plasma membrane much earlier than DNA fragmentation (via the TUNEL assay and DNA laddering).44 There are also reports that this early stage of apoptosis does not irreversibly commit cells to programmed cell death in noncardiac tissue; a significant proportion of myocytes, when subjected to simulated ischemia-reperfusion, exhibit signs of early apoptosis (positive annexin-FITC staining, intact membrane cell death, decreased cell width, and increased mitochondrial [Ca2+]).45,46 In the setting of ischemia-reperfusion injury, it appears that mPTP opening and necrotic cell death contribute more to myocardial injury than apoptosis, and interventions to prevent apoptosis (eg, with caspase inhibitors) have been disappointing.

Thus, it appears that ischemia-reperfusion injury, ie, myocardial stunning, apoptosis, and infarction, manifests in a variety of interrelated ways. Apoptosis may proceed to necrosis when mitochondria are no longer able to withstand the intracellular Ca2+ overload and oxidative stress induced by ROS and when oxidative phosphorylation is unable to keep pace with energy demands. Because of the resulting decrease in the myocardial phosphorylation potential, energy-dependent ion pumps cannot maintain normal ion gradients. This results in cell swelling and, ultimately, loss of membrane integrity. These disturbances can be further exacerbated by the influx of macrophages and leukocytes, complement activation, and endothelial plugging by platelets and neutrophils. If cell death in ischemic reperfused myocardium progresses from apoptosis to necrosis, and if early apoptosis is indeed reversible, then one therapeutic approach for the treatment or prevention of ischemia-reperfusion injury would be to target the early events in apoptosis. Regardless of which stage is being addressed, current cardioprotection strategies are designed to reduce cellular and subcellular ROS formation and oxidative stress, to enhance the heart's endogenous antioxidant defense mechanisms, and to prevent intracellular Ca2+ overload.

In 1883, Ringer described the antagonistic effects of calcium and potassium ions on cardiac contraction. In 1929, Hooker reported the successful resuscitation of dogs with potassium in which ventricular fibrillation was induced by electric shock.47,48 In 1930, Wiggers reported that injections of potassium chloride were capable of abrogating ventricular fibrillation and arrested the heart in diastole. He also demonstrated that revival of the heart was possible using calcium chloride and massage.49 The work by Wiggers led Beck, a thoracic surgeon, to successfully apply defibrillation therapy and save a human life using this method.50 This resulted in a surge in basic and clinical research in the field of fibrillation and defibrillation with the principles applied to cardiac surgery.

With the advent of cardiopulmonary bypass, there was a need to refine techniques to protect the heart and create a quiescent and bloodless field for the unhurried repair of intracardiac defects. Over the next 50 years, a variety of cardioprotective methods and techniques were introduced (Table 15-1). One of the first methodologies evolved from the concept of shielding the heart from perioperative insult using hypothermia. Bigelow and colleagues suggested using hypothermia “as a form of anesthetic” to expand the scope of surgery. The technique “might permit surgeons to operate upon the 'bloodless heart' without recourse to extra corporal pumps, and perhaps allow transplantation of organs.“51

Five years later, Melrose and colleagues reported another way to stop and restart the human heart reliably by injecting potassium citrate into the root of the aorta at both normal and reduced body temperatures.52 Soon thereafter, the clinical application of potassium citrate arrest was adopted by many centers. Interest in using the Melrose technique waned, however, with subsequent reports that potassium citrate arrest was associated with myocardial injury and necrosis. Within a short time, many cardiac surgeons shifted from using potassium- induced arrest to normothermic cardiac ischemia (ie, normothermic heart surgery performed with the aorta occluded while the patient was on cardiopulmonary bypass), intermittent aortic occlusion, or coronary artery perfusion. Experimental and clinical evidence showed, however, that normothermic cardiac ischemia was associated with metabolic acidosis, hypotension, and low cardiac output.53–55

As a consequence, there was a renewed interest in discovering ways to arrest the heart. Bretschneider published the principle of arresting the heart with a low-sodium, calcium-free solution.56 It was Hearse and colleagues, however, who studied the various components of cardioplegic solutions, which led to the development and use of St. Thomas solution.57 The components of this crystalloid solution were based on Ringer's solution with its normal concentrations of sodium and calcium with the addition of potassium chloride (16 mmol/L) and magnesium chloride (16 mmol/L) to arrest the heart instantly. The latter component was shown by Hearse to provide an additional cardioprotective benefit. In 1975, Braimbridge and colleagues introduced this crystalloid solution into clinical practice at St. Thomas Hospital.58

Gay and Ebert showed experimentally that lower concentrations of potassium chloride could achieve the same degree of chemical arrest and myocardial protection afforded by the Melrose solution without the associated myocardial necrosis reported earlier.58,59 Shortly thereafter, Roe and colleagues reported an operative mortality of 5.4% for patients who underwent cardiac surgery with potassium-induced arrest as the primary form of myocardial protection.60 In 1977, Tyers and colleagues demonstrated that potassium cardioplegia provided satisfactory protection in over 100 consecutive cardiac patients.61

By the 1980s, normothermic aortic occlusion had been replaced for the most part by cardioplegia to protect the heart during cardiac surgery. The major controversy at the time was not whether cardioplegic solutions should be used, but what was the ideal composition. The chief variants consisted of (1) the Bretschneider solution, containing primarily sodium, magnesium, and procaine; (2) the St. Thomas solution, consisting of potassium, magnesium, and procaine added to Ringer's solution; and (3) potassium-enriched solutions containing no magnesium or procaine (Table 15-2). Coincident with this controversy, another variant of cardioplegia was introduced, ie, the use of potassium-enriched blood cardioplegia.62,63 Theoretically, blood cardioplegia would be a superior delivery vehicle based on its oxygenating and buffering capacity. Ironically, Melrose and colleagues initially used blood as the vehicle to deliver high concentrations of potassium citrate more than 20 years earlier.

Although hypothermia and potassium infusions remain the cornerstone of myocardial protection today, a variety of cardioprotective techniques and solutions are used that allow patients to undergo heart operations with excellent 30-day outcomes.64,65

Cardioplegic surgery techniques utilize solutions containing a variety of chemical agents that are designed to arrest the heart rapidly in diastole, create a quiescent operating field, and provide reliable protection against ischemia-reperfusion injury. Although contemporary cardiac surgery in low-risk patients is relatively safe, patient characteristics have been changing over the past decade. In addition to coronary heart disease and poor ventricular function, patients are presenting with more comorbidities such as obesity, renal dysfunction, peripheral vascular disease, and emphysema. Despite ongoing improvements in cardioplegic techniques, low cardiac output syndrome (LCOS) frequently occurs after surgery and is a major factor associated with poor outcomes. In the absence of technical complications, the most common cause of postoperative LCOS is inadequate myocardial preservation. For this reason, there is a real need to develop more effective strategies and novel additives to existing cardioplegic solutions. Currently there are two types of cardioplegic solutions that are used: crystalloid cardioplegia and blood cardioplegia. These solutions are administered most frequently under hypothermic conditions.

Crystalloid Cardioplegia

One of the first cardioplegic solutions used to protect the heart during surgery consisted of a cold crystalloid formulation. It was developed on the premise that its administration would protect the heart by reducing its metabolism and aid the surgeon by providing a bloodless field. Over the years a number of crystalloid cardioplegic solutions have been developed that contain different ingredients. The rationale for these ingredients has been based on the need to: (1) induce a rapid cardiac arrest using potassium or magnesium; (2) reduce energy demand and conserve ATP reserves; (3) maintain intracellular ionic and metabolic homeostasis; (4) lower myocardial oxygen consumption; (5) enhance anaerobic and aerobic energy production utilizing glucose and amino acids; (6) stabilize the pH using bicarbonate, phosphate, or histidine buffers; (7) stabilize cellular membranes by using steroids, oxygen free radical scavengers such as glutathione, calcium antagonists, and/or procaine; (8) avoid intracellular calcium overload by providing a hypocalcemic environment and adding magnesium; and (9) prevent cellular edema by the addition of colloids such as mannitol to maintain normal oncotic pressures.

There are basically two types of crystalloid cardioplegic solutions: the intracellular type and the extracellular type.82 Both types can be used as preservation solutions for organ transplantation. The intracellular types are characterized by absent or low concentrations of sodium and calcium. The extracellular types contain relatively higher concentrations of sodium, calcium, and magnesium. Both types avoid concentrations of potassium >40 mmol/L (typical range is 10 to 40 mmol/L), contain bicarbonate for buffering, and are osmotically balanced. Examples of various crystalloid cardioplegic solutions are shown in Table 15-2.

Operative Procedure

Although the degree of core cooling varies from center to center, patients undergoing cardiac surgery are placed on cardiopulmonary bypass (CPB) and often cooled to between 33 and 28°C. To initiate immediate chemical arrest, the solution is infused after cross-clamping the aorta through a cardioplegic catheter inserted into the aorta proximal to the cross-clamp. The catheter may or may not be accompanied by a separate vent cannula. The cold hyperkalemic crystalloid solution then is infused (antegrade) at a volume that generally does not exceed 1000 mL. One or more infusions of 300 to 500 mL of the cardioplegic solution may be administered if there is evidence of electrical heart activity resumption or if a prolonged ischemic time is anticipated. If myocardial revascularization is being performed, the aortic cross-clamp can be removed after completing the distal anastomoses, and the heart reperfused while the proximal anastomoses are completed using a partial occlusion clamp. Alternatively, the proximal grafts can be performed after the distal grafts have been completed with the cross-clamp still in place (the single-clamp technique). Another approach is to perform the proximal aortic grafts first and then to cross-clamp the aorta and infuse the cardioplegic solution. When valve repair or replacement is being performed, the crystalloid cardioplegia can be administered directly into the coronary arteries via cannulation of the coronary ostia. Crystalloid cardioplegia also can be administered retrograde via a coronary sinus catheter with or without a self-inflating silicone cuff.

Outcomes

Although there is a concern that crystalloid cardioplegia lacks blood components and therefore has a limited oxygen-carrying capacity, this has not been shown to be clinically relevant. Likewise, although there is preclinical evidence that hyperkalemic crystalloid cardioplegia may damage coronary vascular endothelium and impair the ability of endothelial cells to replicate and produce endothelial-derived factors, this has not been shown to be of clinical significance.83,84 In fact, there are numerous clinical studies that demonstrate that crystalloid cardioplegia is as effective as blood cardioplegia, particularly in centers where it is the primary form of protection.85,86

Cold Blood Cardioplegia

Cold blood cardioplegia, widely employed throughout the world, is the cardioplegic technique used most often in the United States. The rationale for using blood as a vehicle for hypothermic potassium-induced cardiac arrest is that it: (1) provides an oxygenated environment and a method for intermittent reoxygenation of the heart during arrest; (2) limits hemodilution when large volumes of cardioplegic solution are used; (3) affords an excellent buffering capacity and osmotic properties; (4) allows for electrolyte composition and pH that are physiologic; (5) offers a number of endogenous antioxidants and free-radical scavengers; and (6) is less complex than other solutions to prepare.

Although there are a variety of formulations, it is usually prepared by combining autologous blood obtained from the extracorporeal circuit while the patient is on cardiopulmonary bypass with a crystalloid cardioplegic solution that consists of citrate-phosphate-dextrose (CPD), tris-hydroxymethyl-aminomethane (tham), or bicarbonate (buffers), and potassium chloride. The CPD is used to lower the ionic calcium; the buffer is used to maintain an alkaline pH at approximately 7.8. The final concentration of potassium used to arrest the heart is approximately 20 to 25 mEq/L. After the initial induction dose for rapid arrest, subsequent administrations may be intermittent or continuous and utilize a concentration of 8 to 10 mEq/L (the low concentration maintenance dose)87,88 (Table 15-2).

Before administering blood cardioplegia, the temperature of the solution usually is lowered with a heat-exchanging coil to between 12 and 4°C. The ratio of blood to crystalloid varies among centers, with the most common ratios being 8:1, 4:1, and 2:1. This, in turn, affects the final hematocrit of the blood cardioplegia infused. For example, if the hematocrit of the autologous blood obtained from the extracorporeal circuit is 30, these ratios would result in a blood cardioplegia with a hematocrit of approximately 27, 24, and 20, respectively.

The use of undiluted blood cardioplegia, or “miniplegia” (using a minimum amount of crystalloid additives), also has been reported to be effective. Petrucci et al. studied the use of all blood miniplegia in a clinically relevant swine preparation and compared miniplegia with crystalloid cardioplegia. They concluded that the use of all blood miniplegia was effective or superior in the acutely ischemic heart.89 Velez and colleagues tested the hypothesis that an all-blood cardioplegia in an acute ischemia-reperfusion canine preparation (66:1 blood:crystalloid ratio) would provide superior protection compared with a 4:1 blood cardioplegia delivered in a continuous retrograde fashion.90 They found very little difference between the animal groups with respect to infarct size or postischemic recovery of function. This is consistent with the findings by Rousou and colleagues years earlier that it is the level of hypothermia that is important in blood cardioplegia, not necessarily the hematocrit.91

With respect to efficacy, there are numerous preclinical studies and nonrandomized and randomized clinical trials that demonstrate cold blood cardioplegia is an effective way to provide excellent myocardial protection. Many of these same studies have suggested that cold blood cardioplegia is superior to cold crystalloid cardioplegia. It is important to note that other investigators have shown crystalloid cardioplegia to be as cardioprotective and cost-effective, if not more so. Unfortunately, even the most contemporary clinical trials comparing the efficacy of blood cardioplegia with crystalloid cardioplegia have involved single-center studies, enrolled a limited number of patients, focused on a specific subset of patients, and/or omitted details regarding clinical management. In 2006, Guru et al. reported the results of a meta-analysis of 34 clinical trials that compared blood cardioplegia to crystalloid cardioplegia. A lower incidence of low cardiac output syndrome and CK-MB release was observed in patients administered blood cardioplegia. No difference however was observed in the incidence of perioperative myocardial infarction or mortality.92 Jacob et al. analyzed clinical data from 15 randomized trials. Although eight of the trials reported statistically significant outcomes favoring blood cardioplegia, and five showed statistically significant differences in enzyme release, the bulk of the evidence favoring one over the other was inconclusive.86

Warm Blood Cardioplegia

The concept of using warm (normothermic) blood cardioplegia as a cardioprotective strategy dates back to the 1980s. In 1982 Rosenkranz and colleagues reported that warm induction with normothermic blood cardioplegia, with multidose cold blood cardioplegia maintenance of arrest, resulted in better recovery of function in canines than a similar protocol using cold blood induction.93 In 1986, Teoh and colleagues provided experimental evidence that a terminal infusion of warm blood cardioplegia before removing the cross-clamp (a “hot shot“) accelerated myocardial metabolic recovery.94 This was followed by a report in 1991 by Lichtenstein and colleagues that normothermic blood cardioplegia in humans is an effective cardio-protective approach.76 They compared the results of 121 consecutive patients undergoing CABG surgery who received antegrade normothermic blood cardioplegia operations to an historical group of 133 patients who received antegrade hypothermic blood cardioplegia. The operative mortality in the warm cardioplegic group was 0.9% compared with 2.2% for the historical controls.

Despite these encouraging reports, there are concerns that for any given patient, it is difficult to determine how long a warm heart can tolerate an ischemic insult if the infusion is interrupted or flow rates are reduced owing to an obscured surgical field or a maldistribution of the cardioplegic solution. Another concern is the report by Martin and colleagues that warm cardioplegia is associated with an increased incidence of neurologic deficits.43 In a prospective, randomized study conducted in more than 1000 patients, the efficacies of continuous warm blood cardioplegia with systemic normothermia (≥35°C) were compared with intermittent cold oxygenated crystalloid cardioplegia and moderate systemic hypothermia (≤28°C). Although operative mortalities were similar (1.0 versus 1.6%, respectively), the incidence of permanent neurologic deficits was threefold greater in the warm blood group (3.1 versus 1.0%). In this study, it appears that warm blood cardioplegia offered no distinct advantage over cold crystalloid cardioplegia, and it may be suboptimal if its delivery is interrupted for any reason. Subsequent studies have indicated however that appropriately designed protocols using intermittent antegrade warm blood cardioplegia provide clinically acceptable myocardial protection.95–97

Tepid Blood Cardioplegia

Both cold blood (4 to 10°C) and warm blood cardioplegic solutions (37°C) have temperature-related advantages and disadvantages. As a consequence, a number of studies were performed in the 1990s to determine the optimal temperature. Hayashida and colleagues were one of the first groups to study specifically the efficacy of tepid blood (29°C) cardioplegia.98 In this study, 72 patients undergoing CABG were randomized to receive cold (8°C) antegrade or retrograde, tepid (29°C) antegrade or retrograde, or warm (37°C) ante-grade or retrograde blood cardioplegia. Although protection was adequate for all three, the tepid antegrade cardioplegia was the most effective in reducing anaerobic lactate acid release during the arrest period. These authors reported similar findings when the tepid solution was delivered continuously retrograde and intermittently antegrade.99 In contrast, Baretti et al. reported in anesthetized open-chest swine placed on cardiopulmonary bypass that tepid normokalemic continuous antegrade blood cardioplegia was associated with an increased incidence of intermittent fibrillation.100

Subsequent to this report, Mallidi et al. analyzed the effects of cold blood and warm or tepid blood cardioplegia on early and late outcomes after CABG surgery. Warm blood cardioplegia was used in 4532 patients; cold blood cardioplegia was used in 1532 patients. Actuarial survival at five years was 91.1% in the warm blood cardioplegia group, and 89.9% in the cold blood cardioplegia group (p = 0.09). They concluded that warm or tepid blood cardioplegia may be associated with better early and late event-free survivals. For the most part, however, although tepid blood cardioplegia may be safe and effective, the majority of studies have been single-center studies conducted in relatively small cohorts of patients. Whether tepid cardioplegia confers superior protection over other current methodologies remains to be determined.101

Methods of Delivery

In addition to a variety of formulations and temperatures, there are also many different ways of administering the cardioplegic solutions (Fig. 15-3). As one might expect with so many options, the optimal delivery method also remains controversial. The various methods include intermittent antegrade, antegrade via the graft, continuous antegrade, continuous retrograde, intermittent retrograde, antegrade followed by retrograde, and simultaneous antegrade and retrograde infusions. Although all methods generally are good, comparisons are difficult because there are numerous confounding factors such as (1) the composition of the solution, (2) the temperature of the solution, (3) the duration of the infusion, (4) the infusion pressure, (5) the type and complexity of the operation, (6) the need for surgical exposure, and (7) the expected versus actual cross-clamp time.

Antegrade Cardioplegia

The most widely used method of delivering cardioplegic solutions entails the injection of the solution into the aortic root while the ascending aorta is cross-clamped, the antegrade method. After the initiation of cardiopulmonary bypass (CPB), chemical arrest can be achieved quickly with the infusion of the solution through a catheter inserted into the aorta proximal to the cross-clamp (close to the aortic root). The catheter may or may not be accompanied by a separate vent cannula. The induction dose usually has a higher concentration of potassium than subsequent maintenance doses, and is infused at a rate of 250 to 300 mL per minute to ensure aortic valve closure. The induction rate of infusion ranges between 10 mL/kg and 15 mL/kg body weight. The typical aortic root perfusion pressure is maintained between 60 mm Hg and 80 mm Hg. Adjustments in rates of administration are made to accommodate patients with hypertrophied ventricles. Intermittent maintenance doses of 300 to 500 mL of low-dose cardioplegia are administered every 15 to 20 minutes or earlier if there is evidence of resumption of electrical activity. During CABG surgery, in order to minimize cold ischemia and aortic cross-clamp time, the heart can be reperfused earlier by removal of the aortic cross-clamp as soon as the distal anastomoses are completed; the proximal anastomoses can be performed using a partial occlusion clamp. Alternatively, the proximal grafts can be performed after the distal grafts have been completed with the cross-clamp still in place (the single-clamp technique).

The administration of antegrade cardioplegia depends on a competent aortic valve and hence is liable to be ineffective and relatively contraindicated in patients with significant aortic valve incompetence. In such cases, antegrade cardioplegia can be administered directly into the coronary arteries via cannulation of the individual coronary ostium. This technique is used in cases requiring the opening of the ascending aorta, such as in aortic valve replacement, ascending aortic aneurysm repair, and aortic root replacement procedures. Coronary ostial infusion may be necessary for induction in patients with severe aortic incompetence in the presence of coronary sinus pacing leads. Potential complications using this approach include coronary dissection, unintentional selective delivery into left anterior descending or left circumflex artery because of a short left main coronary artery and the late occurrence of coronary ostial stenosis.102

Retrograde Cardioplegia

Retrograde cardioplegia entails placement of a coronary sinus catheter with or without a self-inflating silicone cuff and infusion of the cardioplegic solution either for induction or for maintenance of cardioplegia. This approach originated with a concept developed by Pratt in 1898, who suggested that oxygenated blood could be supplied to the ischemic heart via the coronary venous system.103 Sixty years later, Lillehei and colleagues used retrograde coronary sinus perfusion to protect the heart during aortic valve surgery.67 Today, it is an accepted method for delivering a cardioplegic solution and is used frequently as an adjunct to antegrade cardioplegia. Placement of the catheter is often facilitated by the use of catheters that are precurved and the simultaneous use of TEE imaging to guide placement. Care must be taken to avoid rupture of the coronary sinus, an uncommon but real and potentially fatal complication. This can be prevented by infusing the solution at a rate that ensures that the coronary sinus profusion pressure does not exceed 45 to 50 mm Hg. The retrograde approach can be used for continuous or intermittent delivery of blood or crystalloid cardioplegia as well. In situations where the native coronary arteries are severely stenotic or totally occluded, the antegrade approach may result in an uneven distributed in the myocardium. In this situation retrograde infusion or infusion via the vein grafts as they are completed can be complementary. Although most routine on-pump cardiac procedures can be done with good outcomes when only the antegrade technique is used, patients with poor ventricular function, high-risk patients requiring long aortic cross clamp times, and those with occlusive coronary disease may benefit from a dual approach, ie, both antegrade and retrograde techniques are used.

Another advantage to the retrograde technique is that it may be effective in reducing the risk of embolization from saphenous vein grafts that could occur during antegrade perfusion for reoperative coronary artery bypass surgery. The retrograde approach also has the theoretical advantage of ensuring a more homogeneous distribution of the cardioplegic solution to regions of the heart that are poorly collateralized.

Despite these advantages, retrograde cardioplegia is not without its limitations. Numerous experimental and clinical studies have shown that cardioplegia administered via the coronary sinus can result in a poor distribution of the solution to the right ventricle. This may be related to the variable venous anatomy of the heart and drainage via the Thebesian veins. The anterior region of the right ventricle is poorly drained by the coronary sinus, and it is not uncommon for the heart to have a number of coronary sinus anomalies. Thus, the retrograde approach may result in the heterogeneous distribution of cardioplegia. Nevertheless, it is felt that coronary sinus retroperfusion facilitates right ventricular hypothermia as the effluent travels through conductance vessels. The hypothermia may counteract the potentially decreased cardioplegic nutritive supply because of the shunting of blood via Thebesian veins. Because hypothermia is an important element of cardioprotection afforded the right ventricle by retrograde cardioplegia, it is unlikely that warm or tepid retrograde cardioplegia would confer similar protection to this region of the heart.

To address these concerns, a technique for simultaneously delivering cardioplegia both antegrade and retrograde is available. The feasibility and safety of this approach was first reported in 1984.104 Subsequently, Cohen and coworkers used sonicated albumen and transesophageal echocardiography (TEE) intraoperatively to assess the effects of delivering a cardioplegic solution antegrade and retrograde simultaneously.105 They reported that perfusion of the left and right anterior ventricles was most consistently achieved using the simultaneous approach. They noted, however, that antegrade infusion was associated with superior perfusion of the left ventricle when compared with retrograde delivery alone and that right ventricular perfusion was inconsistent with both antegrade and retrograde delivery. Thus, it is still unclear when the simultaneous use of both methods is most appropriate.

Continuous Cardioplegia versus Intermittent Cardioplegia

Cardioplegic surgery with aortic cross clamping may necessitate interruption of coronary flow for variable lengths of time. Logically, continuous cold blood cardioplegia would seem to be the better technique to use to address this problem. The major advantage of using intermittent cardioplegia is the ability to achieve and sustain a dry, quiescent operative field. Although a continuous infusion, especially if it is oxygenated, has the theoretical advantage of minimizing ischemia, from a practical standpoint, it is unlikely that this can be achieved reliably. Whether one uses continuous retrograde infusion via the coronary sinus, or handheld coronary ostial infusion in the open aortic root, with few exceptions, both necessitate interruption of the surgical steps while the infusion is delivered.

Many attempts have been made to sort out the clinical superiority of continuous infusion. One such study compared the efficacy of continuous versus intermittent cold blood cardioplegia administered in a retrograde fashion. Seventy patients undergoing CABG surgery for three-vessel coronary artery disease were prospectively randomized to receive retrograde cold blood cardioplegia either intermittently or in a continuous fashion. Ventricular performance was assessed by left and right ventricular stroke work index, and cardiac output. Degree of myocardial injury was assessed by measuring release of the biomarkers lactate and hypoxanthine. The investigators found continuous delivery of retrograde cold blood cardioplegia to be superior in preserving ventricular performance and minimizing myocardial injury.106 A limitation of the study was the relatively small number of patients studied.

In conclusion, there is still controversy regarding superiority of blood versus crystalloid cardioplegia, the ideal temperature, and best method of delivery. A recent survey of practice patterns in the United Kingdom, in 2004, showed that 56% of all cardioplegia on-pump surgery was performed with cold blood cardioplegia, whereas warm blood cardioplegia was used in 14% of cases. In the same survey 14% of surgeons used crystalloid cardioplegia, 21% used retrograde cardioplegia, and 16% did not use any cardioplegia, preferring to use cross-clamp fibrillation only. Based on these observations and experiences in the United States, it appears that most surgeons favor cold intermittent blood cardioplegia. Nevertheless, a wide spectrum of variation exists among heart centers; there is no international consensus regarding the ideal cardioplegic solution or its use.107 Awareness of the options allows the surgeon to utilize the best approach that meets the individual needs of the patient.

Intermittent cross-clamping with fibrillation (ICCF) and systemic hypothermia with intermittent elective fibrillatory arrest are the most common forms of noncardioplegic heart surgery used today. The objective is to provide a relatively quiescent surgical field without having to arrest the heart with a cardioplegic solution.

Intermittent Aortic Cross-Clamping with Fibrillation

This technique is the earliest method developed to protect the heart during surgery and is still used at many centers. The patient is placed on cardiopulmonary bypass, the ascending aorta is cannulated, and generally a dual-stage single venous cannula is used. Often the patient is cooled to between 30 and 32°C. This technique allows the surgeon to operate in a relatively quiet field during ventricular fibrillation. In the setting of CABG surgery, the aortic cross-clamp is removed intermittently after the completion of each graft. The duration of fibrillation is determined by how long it takes to perform the distal anastomosis. After completion of the revascularization, the heart is defibrillated and the proximal anastomoses is performed on the beating heart using an aortic partial occlusion clamp. This technique is particularly useful in patients with cold agglutinin disease, an autoimmune phenomenon in which antibody directly agglutinates human red cells at low temperatures. In these patients, open-heart operations with hypothermia carry the risk of red cell agglutination and may result in hemolysis, myocardial infarction, renal insufficiency and cerebral damage.

As a result of increasing pressures to reduce cost and yet maintain acceptable levels of cardioprotection, there is an interest in using this approach. There are a number of reports that indicate that satisfactory protection can be conferred using this technique. In 1992, Bonchek et al. reported a large clinical series in which the advantages and safety of this technique were meticulously analyzed.108 In this study, the authors reviewed the outcomes of the first 3000 patients at their institution who underwent primary CABG using the ICCF technique. Preoperative risk factors, eg, age, gender, left ventricular dysfunction, preoperative use of the intra-aortic balloon pump (IABP), the urgency of operation, and operative deaths were analyzed. In this series, 29% of the patients were older than 70 years of age, 27% were females, 9.7% had an ejection fraction of less than 0.30, 13% had an MI less than 1 week preoperatively, and 31% had preinfarction angina in the hospital. Only 26% underwent purely elective operations. Using the noncardioplegic cardioprotective technique, the authors reported an elective operative mortality rate of 0.5%, an urgent mortality rate of 1.7%, and an emergency rate of 2.3%. Postoperatively, inotropic support was needed in only 6.6% of the patients, and only 1% required insertion of the IABP. It is important to note, however, that this was a retrospective, single-center institutional experience. The findings would have been enhanced if the analysis had included a similarly matched group of patients at the same institution in which cardioplegic arrest had been employed. Nevertheless, the findings do suggest that noncardioplegic strategies can provide satisfactory myocardial protection even in high-risk patients.

In 2002, Raco and colleagues reported their experience in 800 consecutive CABG operations performed by a single surgeon using ICCF in both elective and nonelective settings. The patients were divided into three cohorts: (1) elective, (2) urgent, and (3) emergent. The mean age, number of distal grafts, and mortality among the three groups were comparable. The mortality rate in the elective, urgent, and emergent groups were 0.6%, 3.1%, and 5.6%, respectively, and consistent with outcomes associated with cardioplegic surgery. Because this report reflects the experience of a single surgeon, there is the concern that the technique may not be generally applicable. Regardless, the findings do support the notion that ICCF is a safe technique for both elective and nonelective patients undergoing CABG surgery.109–111 In 2003, Bonchek et al. reported their experience in 8300 patients who underwent CABG surgery without cardioplegia. The unadjusted mortality rates in elective, urgent, and emergent patients was 0.9%, 1.5%, and 4.0%, respectively. The overall mortality was 1.7%, a rate considerably lower than the 3.27% predicted based on the Society of Thoracic Surgeons National Database model.112 This experience represented the work of five surgeons, three of whom had not been trained in noncardioplegic surgery; it provides additional evidence that ICCF is an effective form of cardioprotection. Evidence that the cardioprotective effects of ischemic preconditioning (IPC) may contribute to the efficacy of ICCF in the human heart and have resulted in a resurgence of interest in using this method of protection, especially in the United Kingdom. Results in animal studies indicate that the protective effect of ICCF is blocked by protein kinase C inhibitors and KATP channel antagonists, both of which have been implicated in the signaling pathways involved in IPC.113 Regardless of the mechanism, there are numerous reports that indicate that noncardioplegic strategies, such as ICCF, can provide satisfactory myocardial protection even in high-risk patients.114

Systemic Hypothermia and Elective Fibrillatory Arrest

Elective fibrillatory arrest is another safe approach to protect the heart during noncardioplegic heart surgery. The use of systemic hypothermia (26 to 30°C) and maintenance of systemic perfusion pressure between 80 and 100 mm Hg are key elements. This approach is particularly applicable in the setting of the severely calcified “porcelain aorta,” where clamping the aorta may be associated with increased risk of stroke and aortic dissection. Under these circumstances the distal anastomoses are performed locally, occluding the coronary artery using vascular clamps or sutures. The proximal anastomoses can be completed during short periods of hypothermic circulatory arrest. Alternatively, the proximal anastomoses can be based entirely on an in situ internal thoracic artery. Using this approach one can avoid manipulation of the aorta altogether.

Akins, in 1984, reported a low incidence of perioperative MI and a low hospital mortality rate in 500 consecutive patients using this technique.74 In 1987, Akins and Carroll assessed that late results of hypothermic fibrillatory arrest in 1000 consecutive patients undergoing nonemergent CABG surgery using hypothermic fibrillatory arrest. They concluded that the technique is effective for myocardial preservation and yields excellent event-free survival. Potential disadvantages with this approach include: (1) the surgical field may be obscured as a result of existing collateral circulation; (2) ventricular fibrillation may be associated with increased muscular tone and thus compromise the surgeon's ability to position the heart for optimal exposure; (3) aortic valvular regurgitation may be exacerbated; and (4) it is generally not applicable for intracardiac procedures. An advantage with the technique is that it can be used when aortic occlusion or cardioplegic arrest is not desirable, eg, in the setting of a calcified ascending aorta.

This approach can also be used in patients undergoing mitral valve surgery.115 Imanaka et al., in 2003, published a retrospective observational study in which mitral valve surgery was performed in 27 patients with ischemic mitral regurgitation using perfused ventricular fibrillation. Concomitant procedures included CABG in 23 patients and the Dor procedure in 5 patients. Surgery was performed using moderate hypothermia (∼28°C) and fibrillatory arrest; flow rates on bypass were maintained at 2.4 L/min/m2 and the perfusion pressure was 70 mm Hg. Among these select patients, the mortality rate was 3.7%. The authors concluded that extended periods of ventricular fibrillation during hypothermic surgery can be well tolerated without excessive morbidity and mortality and is appropriate in patients in whom aortic cross-clamping is unsuitable or the duration of cross-clamping is expected to be long.116

There are a number of physiologic processes and pharmacologic agents known to confer protection against ischemia-reperfusion injury in the experimental setting. Although the current evidence in support of the efficacy of these novel approaches is largely limited to preclinical studies, several trials are underway to determine the clinical relevance of these new approaches. The purpose of this section is to review emerging cardioprotective strategies that hold the most promise.

Physiologic Processes

Ischemic Preconditioning

“Precondition with ischemia” is an endogenous adaptive phenomenon whereby the heart becomes more tolerant to a period of prolonged ischemia if first exposed to brief episodes of coronary artery occlusion. This adaptation to ischemia was first described by Murry and colleagues and is referred to as classic,first window, or early phase ischemic preconditioning (IPC).75 IPC is associated with a reduction in infarct size, apoptosis, and reperfusion-associated arrhythmias.117 IPC has been demonstrated in every animal species studied and appears to persist as long as 1 to 2 hours after the ischemic preconditioning stimulus.118,119 It becomes ineffective when the sustained ischemic insult exceeds 3 hours. This suggests that the protection is conferred only when prolonged ischemia is followed by timely reperfusion.

Subsequent studies have revealed that this endogenous defense mechanism can manifest itself in multiple ways. After the acute phase of preconditioning disappears, a second phase of protection appears 24 hours later and is sustained for up to 72 hours. This has been referred to as the second window of protection, late-phase preconditioning, or delayed preconditioning. Unlike classic IPC, which protects only against infarction, the late phase protects against both infarction and myocardial stunning.120,121

Cellular Mechanisms of IPC

These reports of adaptation to ischemia have resulted in major investigative efforts to elucidate the intracellular mechanism(s) that underlie the heart's endogenous defenses against ischemia-reperfusion injury. The assumption is that a better understanding of these mechanism(s) could lead to the development of potent new therapeutic modalities that are more effective in treating or preventing the deleterious consequences of ischemia-reperfusion injury. One of the earliest hypotheses was that stimulation of cardiomyocyte adenosine A1 and/or A3 receptors was the primary mediator of acute ischemic preconditioning.119,122 Subsequent studies have revealed that in addition to adenosine, there are multiple guanine nucleotide-binding (G) protein–coupled receptors (GPCRs) that, once activated, can mimic the infarct-reducing effect of ischemic preconditioning, eg, bradykinin, endothelin, and α1 -adrenergic, muscarinic, angiotensin II, and delta-opioid receptors (Fig. 15-4). The infusion of exogenous agents that mimic ischemic preconditioning is referred to as pharmacologic preconditioning. Exactly which of these receptors is the most important in mediating endogenous preconditioning is unknown because there appear to be species differences and redundant signaling pathways. Regardless, it is now thought that these triggers of IPC result in alterations in certain enzymes, such as tyrosine kinases, isoforms of PKC, and mitogen-activated protein kinases (p38 and extracellular signal regulated kinase [ERK]) that, in turn, confer protection against irreversible injury before the onset of prolonged ischemia.

Interestingly, IPC-induced cardioprotection appears to require repopulation of receptors and activation (or, in some instances, reactivation) of “prosurvival” kinases upon relief of sustained ischemia. In this regard, Hausenloy and Yellon introduced the term Reperfusion Injury Salvage Kinase (RISK) pathway to represent the PI3K–Akt and ERK 1/2 pro-survival kinases activated at the time of reperfusion and proposed that manipulation and upregulation of the RISK pathway may represent another approach to myocardial protection.123

Although the identity of the end effector(s) of IPC remain speculative, significant evidence has accumulated indicating that the cardiomyocyte mitochondria are key targets of conditioning-induced protection (Fig. 15-4).124,125 Specifically, inhibition of mPTP and the opening of the mitochondrial KATP (mKATP) channel have been implicated as the effectors of IPC.126,127 Under normal conditions the mitochondrial inner membrane is impermeable to most metabolites and ions and the mPTP is closed. Although the molecular structure of the pore has yet to be determined, it is characterized by the formation of a large conductance megachannel that is regulated by cyclophilin D in the matrix. Although early investigations implicated the voltage-dependent anion channel in the outer mitochondrial membrane and the adenine nucleotide translocator (ANT) in the inner membrane in addition to cyclophilin D, genetic studies have refuted that model. Mouse models in which all ANT isoforms were deleted still exhibited mPTP opening; this was also the case for deletion of VDAC isoforms. However, deletion of cyclophilin resulted in hearts that were much more resistant to I/R injury, and further studies revealed that although the threshold for mPTP opening was greatly increased, it was still possible to trigger pore opening. It was concluded that cyclophilin D plays an important regulatory role in mPTP opening, but the molecular composition remains uncertain. Under conditions of stress, the mPTP may open, resulting in depolarization of the inner mitochondrial membrane and an influx of water and ions into the matrix because of its high oncotic pressure. Matrix swelling expands the highly folded inner membrane, but ultimately ruptures the outer membrane, resulting in release of cytochrome c and other pro-apoptotic factors. Even in the absence of the outer membrane rupture, loss of mitochondrial membrane potential results in ATP hydrolysis by the F0F1 ATP synthase in an effort to restore membrane potential. This futile cycling accelerates energy depletion.

An ATP-sensitive potassium channel in the mitochondrial inner membrane (mKATP) has been implicated on the basis of pharmacologic effects of diazoxide and pinacidil (channel openers) and 5-hydroxydecanoate and glibenclamide (channel closers). Many pharmacologic studies have demonstrated a protective role for the putative mKATP, although its molecular composition remains unknown. Garg and Hu have proposed that PKC activation enhances the import of plasma membrane KATP channels into mitochondria. This was based on their observation that in COS-7 cells, Kir6.2 protein (a subunit of KATP channels) and channel activity increased in mitochondria after PMA treatment, and this increase was inhibited by the selective PKC inhibitor chelerythrine. Pharmacologically triggered opening of the mKATP channel has been shown to reduce calcium overload, mitochondrial free-radical production, and swelling and to preserve ATP levels after ischemia/reperfusion.128

Although early-phase preconditioning shares many of the same signaling mechanisms with late-phase preconditioning, the most obvious difference between the two is the apparent requirement for protein synthesis in the latter. Late-phase IPC has been shown to be associated with the upregulation of various proteins, including, but not limited to, heat-shock proteins, inducible NOS (iNOS), cyclooxygenase 2, heme oxygenase, and manganese superoxide dismutase.129,130 There are, however, conflicting reports on what specific proteins are upregulated during late-phase preconditioning, which may be because of species differences as well as stimulus-specific responses.

Clinical Relevance

There is considerable circumstantial evidence that ischemic preconditioning occurs in the human. Investigators have reported that patients experiencing angina before an MI have a better in-hospital prognosis and a reduced incidence of cardiogenic shock, fewer and less severe episodes of congestive heart failure, and smaller infarcts as assessed by cardiac enzyme release.131 Moreover, follow-up studies suggest that patients who have had angina before an infarct have better long-term survival rates.132–134 There are also a myriad of reports that patients who undergo percutaneous coronary interventions (PCI) have an enhanced tolerance to ischemia after the first balloon inflation, provided that the first balloon inflation exceeds 60 to 90 seconds.118 Chest pain severity, regional wall motion abnormalities, ST-segment elevation, QT dispersion, lactate production, and CKMB release all have been reported to be attenuated in this setting as well.135,136

In patients undergoing PCI, a preconditioning-like effect has been mimicked by the administration of a variety of pharmacologic agents that are known to induce preconditioning in animal studies. For example, the administration of adenosine before PCI has been reported to attenuate myocardial ischemic indices during the first balloon inflation.137 Administration of other agents, such as bradykinin and nicorandil (a KATP channel opener), also have been reported to produce similar effects.138,139 Conversely, the administration of aminophylline (a nonselective adenosine receptor antagonist), glibenclamide (a KATP channel blocker), or naloxone (an opioid receptor blocker) reportedly abolishes the effects of ischemic preconditioning during PCI.140,141 Additional studies provide evidence of delayed pharmacologic preconditioning in the clinical setting. Leesar and colleagues reported that a 4-hour intravenous infusion of nitroglycerin (an NO donor) 24 hours before PCI decreased ST-segment changes and reduced chest pain during the first balloon occlusion compared with patients treated with saline vehicle.142 An earlier report by this same group indicated that delayed preconditioning with nitroglycerin decreased exercise-induced ST-segment changes and improved exercise tolerance. Thus, there are observational studies that support the hypothesis that myocardial protection conferred by ischemic preconditioning and its possible mediators in animal studies is translatable to humans. It is important to note, however, that classic or early ischemic preconditioning observed in animals is associated with a reduction in infarct size, but not protection against stunning, and that many of the clinical studies are either retrospective in nature or have used surrogate markers of injury as end points.

With respect to a role for IPC during cardiac surgery, numerous small trials have been conducted.143 One of the first studies was conducted by Yellon and colleagues in patients undergoing CABG surgery.144 Patients were subjected to a protocol that involved two cycles of 3 minutes of global ischemia. The aorta was cross-clamped intermittently and the heart was paced at 90 beats per minute to induce ischemia. This was followed by 2 minutes of reperfusion before a 10-minute period of global ischemia and ventricular fibrillation. Myocardial biopsies were obtained during the 10-minute period of global ischemia, and ATP tissue content was measured. The results showed that the ATP levels in the biopsies obtained from patients subjected to the preconditioning-like protocol were higher. However, because ATP content is not a marker of necrosis, a follow-up study was performed, and troponin T serum levels were measured. In this study, the investigators reported that troponin release was attenuated in the patients subjected to the preconditioning protocol. In 2002, Teoh and colleagues reported that IPC conferred myocardial protection beyond that provided by intermittent cross-clamp fibrillation in patients undergoing CABG.79 Other investigators have reported similar findings.

Thus, a number of cardiac surgical studies suggest IPC may be effective in the setting of aortic cross-clamping and administration of cardioplegia. It is important to note, however, that to date the total number of patients studied has been relatively small, and the outcomes have been limited to surrogate markers of myocardial necrosis, viz., CK-MB levels and troponin release, and not clinical endpoints. This explains in part why IPC has not been adopted as an adjuvant approach among the myocardial protection techniques employed to date. A more promising strategy may be to develop a better understanding of the intracellular events and effectors that confer protection and then design the appropriate pharmacologic agent to mimic the phenomenon.

Postconditioning

The phenomenon of ischemic postconditioning (PostCond) was first reported by Zhao et al. in the canine model.145 The term refers to rapid intermittent interruptions of blood flow in the early phase of reperfusion, ie, relief of ischemia in a stuttered or staccato manner. Although the cellular mechanisms underlying PostCond are poorly defined, they appear to involve many of the same signal transduction pathways that are involved in IPC, including cell surface receptor signaling, pro-survival kinases, the mPTP, and the mKATP channel. Although the duration and frequency of reperfusion may be variable, for the most part, the cycles that induce PostCond are measured in seconds in smaller species, and slightly longer in larger animals and humans, justifying the name stuttering reperfusion. The reduction in infarct size appears to be comparable with that observed with IPC. Preclinical studies conducted in multiple models and species (including dog, rat, rabbit, mouse, and pig) have demonstrated a reduction in infarct size that ranges from 20 to 70%. The restoration of blood flow in a stuttering manner during early reperfusion is of major interest to clinicians because it holds particular promise for patients presenting with an acute MI. In the surgical setting, PostCond could be applied in the operating room after release of the aortic cross-clamp.

Evidence that PostCond exists in humans was first reported in patients undergoing PCI. Patients receiving brief balloon inflations/deflations in the initial minutes of reperfusion during PCI exhibited smaller ST-segment changes and lower levels of total creatine kinase release compared with patients that were not subjected to stuttering reperfusion. More recently, Darling et al. conducted a retrospective chart review in patients undergoing emergent cardiac catheterization for ST-segment elevation MI (STEMI).146 The hypothesis was that outcome would be better in patients undergoing multiple balloon inflations after primary angioplasty. Patients were divided into two cohorts: those who, at the discretion of the interventional cardiologist, received one to three balloon inflations, and those in whom four or more inflations were applied. In this retrospective analysis, peak CK release was less in patients requiring ≥four inflations. In a separate study by Lønborg et al., the cardioprotective effects of PostCond in patients treated with PCI was evaluated using MRI.147 These investigators reported that mechanical PostCond appeared to be independent of the size of myocardium at risk. The findings were consistent with the concept that stuttering reperfusion confers cardioprotection during percutaneous interventions.

In the context of heart surgery, Luo reported a beneficial effect of surgical PostCond in 24 patients undergoing repair for Tetralogy of Fallot (TOF) at the time of aortic declamping. The postconditioning protocol consisted of aortic reclamping for 30 seconds and declamping for 30 seconds. The process was repeated twice. The intervention was reported to reduce perioperative troponin T and CK-MB release and decreased the need for inotropic support after surgery.148 A similar finding was reported by the same investigator in a study of adult patients undergoing valve surgery and children undergoing corrective surgery using cardioplegia.

Thus, there is evidence that a PostCond protocol may be of benefit to patients undergoing heart surgery. Although PostCond may offer more promise than IPC in terms of clinical application, it is important to note that both are invasive in nature. Ultimately, the elucidation of mechanisms underlying PostCond may hold the most promise for development of new therapeutic approaches to cardioprotection.

Remote Ischemic Preconditioning

Remote ischemic preconditioning is a phenomenon whereby brief ischemia of one organ or tissue confers protection on a distant naive organ or tissue against a sustained ischemia-reperfusion injury. Remote ischemic preconditioning (RIPC) was first described by Przyklenk and colleagues in 1993.149 In the original study, the investigators questioned whether IPC protected only the heart cells exposed to brief coronary artery occlusions or was it possible for repetitive or stuttering occlusions in a remote naive vascular bed to reduce infarct size in the area subjected to prolonged ischemia. They used a canine preparation in which a branch of the circumflex coronary artery was subjected to four episodes of 5-minute occlusion and reperfusion; this was followed by 1 hour occlusion of the left anterior descending (LAD) coronary artery. After 4.5 hours of reflow, infarct size in the distribution of the LAD was measured. A marked reduction in infarct size was observed. Since then, numerous other investigators have confirmed these findings, and the phenomenon has been observed in various species and with different organs. Brief occlusions of the renal and mesenteric arteries and brief restriction of blood flow to the skeletal muscle of the lower limb have been shown to reduce myocardial infarct size by up to 65%.150 As a consequence, RIPC is now also referred to as interorgan preconditioning.

Not surprisingly, multiple mechanisms have been implicated to play a role in triggering and mediating remote preconditioning, including both humoral factors (such as adenosine, bradykinin, and calcitonin gene-related peptide), and neuronal factors, followed by activation of one or more kinases (including p38MAPK, ERK1/2 and JNK). To date the mechanism(s) underlying RIPC remain largely speculative. As with many other cardioprotective interventions, lack of a clear understanding of the molecular basis of the phenomenon has not dampened enthusiasm to apply the approach clinically as a possible new approach to cardioprotection in humans.

One of the first studies involving patients was conducted in 17 children undergoing congenital heart surgery with cardiopulmonary bypass.151 Brief intermittent lower limb ischemia was associated with attenuated troponin release and a reduction in the need for postoperative inotropic support. Other investigators have reported similar findings in patients undergoing both adult heart surgery and resection of abdominal aortic aneurysms. For example, in one study involving 23 adult patients undergoing on-pump CABG surgery in which cold blood cardioplegia was used, RIPC was associated with a 42% reduction in total troponin T release. RIPC was induced by three 5-minute cycles of right forearm ischemia by inflating a blood pressure cuff on the upper arm to 200 mm Hg with an intervening 5-minute reperfusion period. The control group had a deflated cuff placed on the upper arm for 30 minutes. The outcome was similar to another study by Hausenloy et al., in which 57 patients undergoing elective CABG surgery were randomly assigned to RIPC or control groups. During the CABG operation, either intermittent cross-clamping or cardioplegia was used. RIPC was associated with a reduction in perioperative troponin T release by 43%.152 Whether this mode of cardioprotection will become a standard of care will depend on whether or not RIPC can be shown to have a salutary effect on more robust clinical endpoints such as perioperative MI, stroke, and death.

Autophagy

Autophagy is the process whereby a double-membrane structure called the autophagosome sequesters cytoplasmic components such as ubiquitinated protein aggregates or organelles including mitochondria, peroxisomes, and endoplasmic reticulum. It is involved in degradation of long-lived proteins and the removal of excess or damaged organelles. The outer autophagosomal membrane fuses with a lysosomal membrane to deliver its contents into an autophagolysosome where the cargo is degraded by lysosomal hydrolases and the resulting macromolecules recycled153 (Fig. 15-5).

One of the first studies to suggest autophagy is an adaptive process responsive to stress in the heart was the report by Decker et al. in which they described an association between the formation of autophagosomes and an increase in degenerating mitochondria in rabbit hearts exposed to hypoxia and reperfusion. Reperfusion restored contractility and injured myocytes underwent a cellular repair process that involved a marked increase in lysosomal autophagy. These investigators concluded that this process is important in the efforts to repair cardiac cells during and after hypoxia.154

Autophagy has now been reported to be upregulated in isolated cells subjected to simulated ischemia and reperfusion and rodent models of ex vivo and in vivo ischemia reperfusion injury. Shimomura showed that upregulation of autophagy in HL-1 myocytes protected against cell death induced by simulated ischemia reperfusion (sI/R) whereas inhibition of autophagy enhanced cell death.155 Dosenko et al. was one of the first to report a more direct linkage between autophagy and cytoprotection and observed that autophagy may have a protective effect during anoxia-reoxygenation.156 Subsequently, upregulation of autophagy induced by sI/R in HL-1 myocytes was reported to be protective, whereas inhibition of autophagy enhanced cell death. Hamacher-Brady and colleagues subjected HL-1 cells to sI/R as an in vitro model of ischemia reperfusion. Using three-dimensional high-resolution fluorescence imaging, they analyzed the autophagic response to sI/R. They observed that autophagy is an important underlying protective response against sI/R injury.157 Matsui et al. found that glucose deprivation increased the number of autophagosomes in neonatal cardiac myocytes, and that inhibiting autophagy with 3-methyladenine (3-MA) enhanced cell death induced by glucose deprivation.158 Yan et al. reported that cardiac myocytes with enhanced autophagy were negative for apoptosis, whereas apoptotic cells were negative for autophagy in a porcine model of chronic myocardial ischemia and hibernating myocardium.159

Gurusamy et al. investigated the role of autophagy during ischemia-reperfusion injury and reported that increased BAG-1 expression in the heart correlated with the onset of protection in an in vivo model of myocardial stunning.160 Collectively, these studies suggest that upregulation of autophagy promotes survival during stress such as ischemia-reperfusion.

There is now direct evidence that autophagy plays an important role in mediating ischemic and pharmacologic preconditioning. Yitzhaki et al. investigated the effect of the adenosine preconditioning agent 2-chloro-N(6)- cyclopentyladenosine (CCPA) on autophagy and cell survival after sI/R and GFP-LC3 infected HL-1 cells and neonatal rat cardiomyocytes. Autophagy was induced within 10 minutes of adenosine treatment and increased autophagy was evident when examined 24 hours later. Inhibition of autophagy resulted in significant loss of both immediate and delayed cytoprotection against sI/R as measured by release of lactate dehydrogenase. To assess autophagy in vivo, transgenic mice expressing the red fluorescent autophagy marker mCherry-LC3 were treated with CCPA. Treated hearts revealed a large increase in the number of autophagosomes. Subsequent ex vivo and in vivo studies examining the role of autophagy in pharmacologic and ischemic preconditioning have provided strong evidence that autophagy is required for the protective effect. Based on these findings, it appears that autophagy may serve as an important mediator of protection of the preconditioning agent CCPA.161 A more detailed understanding of the role of autophagy in myocardial protection may lead to a new therapeutic approach to the management and treatment of ischemia-reperfusion injury.

Pharmacologic Agents

Considerable progress has been made towards limiting myocardial ischemia-reperfusion injury in the setting of global ischemia associated with surgical procedures. Despite considerable efforts to translate promising experimental results into clinical therapies, interventions to date have yielded few successes with the exceptions of adenosine and glucose-insulin-potassium. These disappointments led the National Heart Lung and Blood Institute to convene a working group to discuss the reasons for the failure to translate potential therapies for protecting the heart from ischemia and reperfusion. The working group concluded that cardioprotection in the setting of acute myocardial infarction, cardiac surgery, and cardiac arrest requires more relevant animal models corresponding to the human conditions of atherosclerosis, hypercholesterolemia, hypertension, diabetes, and advanced age, all of which are increasingly recognized to interfere with cardioprotective strategies.162 At the same time, a number of recent studies provided a sense of optimism regarding the encouraging clinical data for the following cardioprotective interventions.

Adenosine

There is considerable experimental evidence that activation of various adenosine receptor subtypes results in cardioprotection similar to that induced by IPC. Preischemic administration of the nucleoside adenosine retards the rate of ischemia-induced ATP depletion, prolongs the time to onset of ischemic contracture, attenuates myocardial stunning, enhances postischemic myocardial energetics, and reduces infarct size.163

There are at least four distinct adenosine receptor subtypes, which are designated A1, A2a, A2b, and A3. They couple to a variety of guanine nucleotide-binding (G) proteins (Go, Giα2, Giα3, Gq, and Gs) depending on the receptor subtype and tissue studied. There is definitive evidence that two, and possibly three, of these receptors are expressed in the adult human heart. Recent preclinical reports suggest that an adenosine A2b receptor agonist confers cardioprotection when administered before the onset of ischemia (preconditioning) and at reperfusion (postconditioning).164

With respect to cardiac surgery, there have been a limited number of clinical trials performed. Fremes and colleagues reported the results of an open-label, nonrandomized CABG surgery study in which adenosine administration was combined with antegrade warm blood cardioplegia. The adenosine concentrations studied were 15, 20, and 25 mol/L. These investigators reported that adenosine could be added safely as a supplement to cardioplegic solutions, but it had no effect on myocardial function at the doses studied.165

Cohen and colleagues observed a similar lack of efficacy in a phase II double-blind, placebo-controlled trial performed in patients undergoing CABG surgery. Patients were treated with placebo (saline) or warm blood cardioplegia supplemented with 15, 50, or 100 μM adenosine. These investigators found that adenosine had no effect on survival, the incidence of MI or the incidence of low cardiac output syndrome. A limitation of this study was the use of low concentrations of adenosine in the setting of warm blood cardioplegia. The nucleoside is metabolized rapidly to inosine and hypoxanthine, and the half-life in blood is measured in seconds, thus limiting its potential effect.166

Mentzer and colleagues reported a beneficial effect in an open-label, single-center study in which the safety, tolerance, and efficacy of high doses of adenosine were studied when added to cold blood cardioplegia in CABG surgery patients. Sixty one patients were randomized to receive standard cold blood cardioplegia or cold blood cardioplegia containing one of five adenosine doses ranging between 100 μM and 2 mM. Invasive and noninvasive studies of myocardial function were obtained sequentially after bypass. Parameters included the recording of inotropic utilization rates for the postoperative treatment of low cardiac output. Blood samples were collected for the measurement of nucleoside levels. High-dose adenosine treatment was associated with a 249-fold increase in the plasma adenosine concentration and was associated with a reduction in postbypass inotropic drug utilization, improved regional wall motion, and global function measured by transthoracic echocardiography.167

Subsequently, Mentzer and colleagues examined the effects of high-dose adenosine treatment in 253 patients randomized to one of three treatment arms. This was a double-blind, placebo-controlled multicenter trial using cold blood cardioplegia; adenosine was administered in three different concentrations and rates. Invasive and noninvasive measurements of ventricular performance were obtained before, during, and after surgery. The results of this study revealed a trend toward a decrease in high-dose inotropic agent utilization rates and a lower incidence of perioperative MI. A posthoc composite outcome analysis revealed that patients who received the high-dose adenosine were less likely to experience high-dose dopamine and epinephrine use, insertion of an intra aortic balloon pump, MI, or death.168

In summary, there is preclinical and clinical evidence that adenosine is a cardioprotective agent. Its clinical use is limited because large doses are associated with marked hypotension in patients not on cardiopulmonary bypass. The identification of a selective adenosine receptor subtype, eg, an adenosine A2b agonist that confers protection without systemic hypotension could lead to the development of a drug that is effective in preventing perioperative MI and stunning similar to that observed with late-phase preconditioning.

Acadesine

This agent is a member of a class of drugs referred to as adenosine regulating agents. It is a purine nucleoside analog that purportedly raises adenosine tissue levels selectively during ischemic conditions.169 The mechanism by which this agent regulates adenosine levels during ATP catabolism has not been fully established. Early preclinical studies have indicated that acadesine treatment: (1) improves left ventricular wall motion after intermittent ischemia, (2) attenuates frequency of ventricular arrhythmias, and (3) attenuates myocardial stunning and preserves myocardial function after cardiac arrest and cold cardioplegia. The observations led to five large-scale clinical trials in CABG surgery patients in the 1990s. The findings, however, were inconclusive due in part to the fact they were powered to detect only effect sizes of 50% or more.

As a consequence, Mangano combined the data from all five trials and performed a meta-analysis approach. The entire clinical experience of acadesine in more than 4000 CABG patients were analyzed to determine the effects of this agent on prespecified perioperative outcomes of MI, stroke, and cardiac death. The results of the meta-analysis indicated that acadesine given intravenously before and during surgery along with a fixed concentration in cardioplegia solution was effective in reducing perioperative MI, cardiac death, and combined adverse cardiovascular outcomes.170

Subsequently, Mangano et al. examined the two-year all-cause mortality after perioperative MI in a follow-up of the Acadesine 1024 Trial. Although the primary outcome was negative, the findings at two years among patients experiencing postreperfusion MI indicated a significant four-fold reduction in mortality (15/54 patients versus 3/46 patients).171 Based on these findings, a large scale Phase III study was initiated in 2010 (RED CABG). This study was stopped after 30% enrollment, however, due to futility. Whether the apparent lack of efficacy was due to clinical trial design, the characteristics of this class of drugs, or properties unique to agent itself, is unknown. It is important to note, however, that the salutary effects of this agent may not be apparent immediately after the operation.

Sodium-Hydrogen Exchanger Inhibitors

The sodium-hydrogen exchangers (NHEs) are a family of membrane proteins with nine isoforms that are involved in the transport of hydrogen ions in exchange for sodium ions. NHE-1 is the isoform that is expressed in the heart and may play a minor role in the normal excitation-contraction coupling process; however, it has been implicated in the etiology of arrhythmias, stunning, apoptosis, necrosis associated with acute myocardial ischemia-reperfusion injury, postinfarction ventricular remodeling, and heart failure.

The driving forces for Na+/H+ exchange are the relative transmembrane N+ and H+ gradients. The activity of the exchanger is regulated by the interaction of the H+ with a sensor site on the exchanger protein and can be additionally modulated by phosphorylation.172 During ischemia, cytoplasmic pH falls as low as 6.6 because of increased production of H+ from anaerobic glycolysis, but upon reperfusion, NHE-1 is activated to restore intracellular pH by exchange of intracellular protons for extracellular sodium. The resulting accumulation of intracellular Na+ is further exacerbated by diminished activity of Na+/K+ ATPase as a result of ischemia and lowered ATP availability. The increased intracellular Na+ competes for sites on the Na+/Ca++ exchanger and can actually drive it to run in reverse, resulting in cytosolic calcium overload. As noted earlier, calcium overload has numerous adverse consequences including activation of calcium-dependent proteases and phospholipases, gap junction dysfunction, and triggering of the mPTP culminating in membrane rupture and cell death (see Fig. 15-1).

The EXPEDITION trial was initiated to address the safety and efficacy of NHE-1 inhibition by cariporide in the prevention of death or MI in patients undergoing CABG surgery. High-risk CABG patients (n = 5,770) were randomized to receive either intravenous cariporide or placebo. The composite endpoint was assessed at 5 days, and patients were followed for up to 6 months. The results revealed an 18.3% relative risk reduction (RRR) in the incidence of death or MI at 5 days (p = 0 .0002). The RRR for death or MI at day 30 and month 6 was 16.1% (p = 0.0009) and 15.7% (p = 0.0006), respectively. When analyzed separately, the RRR in the incidence of MI alone at 5 days was 23.8% (p = 0.000005) and at month 6 was 25.6% (p = 0.000001). The mortality rate, however, increased at 5 days from 1.5% in the placebo group to 2.2% in the group with cariporide. This was associated with an overall increase in the incidence of cerebrovascular events. Thus, although cariporide treatment was effective in reducing the incidence of nonfatal MI, its efficacy was associated with toxicity, and the overall assessment of benefits and risks associated with cariporide indicated that the imbalances in the safety profile outweighed the reduction in the observed MI rate. Thus, it is unlikely that cariporide will be used clinically. The importance of the study, however, is that myocardial necrosis after CABG is higher than previously appreciated and it suggested NHE-1 inhibitors represent a new class of drugs that hold promise for reduction of myocardial infarction associated with ischemia-reperfusion injury.81

Glucose-Insulin-Potassium

There are numerous studies that suggest glucose-insulin-potassium (GIK) infusions are effective in reducing perioperative MIs, postischemic myocardial dysfunction, and atrial fibrillation in patients undergoing heart surgery.173 The rationale for this form of treatment is based on the concept that insulin stimulates potassium reuptake through stimulation of the Na+,K+-ATPase while it stimulates glucose uptake for glycolytic energy production. High glucose, insulin, and an increased glycolytic flux also increase pyruvate generation and in turn preserve the citric acid cycle. Additionally, glycolytic ATP protects membranes, drives uptake of Ca2+ by the sarcoplasmic reticulum, and improves sodium homeostasis of ischemic myocardium.

Despite a strong rationale for its application in surgery, the efficacy of GIK in heart surgery remains controversial. This is due in part to the mixed results with the use of GIK to treat patients with acute myocardial infarction. In 1997, Fath-Ordoubadi and Beatt showed a benefit of treatment in a meta-analysis of patients who received GIK reperfusion therapy for acute myocardial infarction.174 Van der Horst reported a benefit of GIK in 940 randomized STEMI patients treated by PCI. It was only in the subgroup of 156 patients without heart failure, however, that a significant reduction in mortality was seen.175 In a follow-up report, Timmer et al. were unable to demonstrate a therapeutic benefit in patients without signs of heart failure. Thus, a role for GIK in the treatment of acute myocardial infarction (STEMI) is yet to be determined.176

With respect to cardiac surgery, Lazar et al. conducted a study to determine whether tight perioperative glycemic control in diabetic CABG patients with a modified GIK solution would optimize myocardial metabolism and improve perioperative outcomes. Patients (n = 141) were randomized to receive GIK or no GIK. The preoperative patient profiles were similar in age, sex, ejection fraction, urgency of surgery, or the type of diabetes. Although the 30-day survival was comparable in both groups, the GIK-treated patients had significantly higher cardiac indices and less need for inotropic support. There was also a lower incidence of atrial fibrillation. Follow-up data two years after surgery were available in 60 of 70 (83.3% GIK) and 60 of 69 (86.9% GIK) patients. Survival was better in the patients who received GIK; the investigators attributed this to long-lasting benefits from perioperative GIK treatment.177

In another single-center study, Quinn et al. reported their results with a prospective, randomized, double-blind, placebo-controlled trial in 280 nondiabetic CABG surgery patients. They found that GIK treatment was associated with fewer episodes of low cardiac output, less inotropic support postoperatively, and a reduction in serum cardiac troponin I levels. The authors concluded that GIK is an effective, inexpensive, and safe adjunctive cardioprotective technique.80

In contrast, S. Bruemmer-Smith et al. reported that GIK infusion during surgery failed to reduce myocardial cellular damage as measured by cTnI levels six hours after cardiopulmonary bypass. It was associated with increased hyperglycemia. In this study, only 42 patients were enrolled although it was a randomized, prospective, double-blinded study in patients undergoing elective CABG surgery. The patients received either GIK or placebo administered via a central line. The groups were well-matched for aged and number of bypassed vessels.178 In another study, Lell et al. reported on 46 patients undergoing elective off-pump CABG surgery; the patients received either normal saline or a GIK infusion for 12 hours. These investigators were unable to demonstrate a beneficial effect on cardiac performance using the clinical measurements of cardiac index and inotropic requirements. They did report the finding of a persistent hyperglycemia, however, despite the use of supplemental insulin.179 Finally, Barcellos et al. reported their experience with 24 patients with type-2 diabetes mellitus who underwent CABG surgery. Patients were administered GIK or subcutaneous insulin from the onset of anesthesia until 12 hours postoperatively. The use of GIK neither improved the cardiac index nor reduced the use inotropic agents albeit it did improve glucose control.180

Thus, the evidence supporting the efficacy of GIK in the setting of heart surgery is still controversial. Although a meta-analysis of randomized studies using GIK suggests that it may improve postoperative recovery of contractile function and reduce atrial arrhythmias, the individual studies have involved small numbers of patients and, therefore, insufficiently powered to detect efficacy. Until a large randomized, multicenter clinical trial has been performed, the use of GIK as a form of myocardial protection will remain controversial.

In an effort to minimize complications associated with cardiopulmonary bypass, such as stroke and a systemic inflammatory response, the number of coronary artery bypass graft operations performed without cardiopulmonary bypass (off-pump: OPCAB). The assumption is that avoidance of aortic cross-clamping will be associated with a lower incidence of cerebrovascular complications and perioperative MI, lower rates of renal and respiratory failure, less postoperative bleeding, less pain, and shorter hospitalization. There is a concern, however, that OPCAB may be associated with incomplete revascularization, a higher incidence of perioperative myocardial infarction, and a reduction in long-term graft patency.181–183 Although the relative benefits of OPCAB versus on-pump CABG surgery remain to be determined, as many as 20% of CABG patients undergo the procedure off-pump. Thus it is important to appreciate the principles and methods of myocardial protection as it relates to beating heart surgery.

The acceptance of OPCAB is due in part to the development and refinement of a myriad of surgical aids that allow for stabilization and local immobilization of the heart during grafting. Techniques include the temporary occlusion of the coronary artery or shunting of blood during sustained coronary artery occlusion. Because temporary occlusion of an already diseased artery may aggravate ongoing ischemia, pharmacologic agents and nonpharmacologic maneuvers are used to protect the heart during displacement required to access lateral and inferior wall vessels. Many of these interventions are designed to reduce myocardial oxygen demand at a time when supply is limited.

Ischemia during temporary occlusion of a coronary artery during OPCAB may last from 6 to 25 minutes based on the surgeon's experience, quality and size of the vessel, and adequacy of the exposure. Most patients with preexisting severe coronary heart disease have experienced self-limiting episodes of ischemia during daily life and may have imparted a certain degree of tolerance to the surgically induced ischemia. This tolerance has been documented using ECGs, transesophageal echocardiography, and continuous SVO2 monitoring. In order to better understand the differences between OPCAB and on-pump surgery, Chowdhury et al. investigated the release pattern of various cardiac biomarkers in 50 patients undergoing cardioplegic and noncardioplegic coronary artery bypass surgery. The various markers included cardiac troponin I, heart-type fatty acid-binding protein, CK-MB, high-sensitivity C-reactive protein and myoglobin. Measurements were obtained at baseline and then at hourly intervals up to 72 hours after completion of the last anastomosis in the noncardioplegic OPCAB group, and after release of aortic cross clamp in the cardioplegic group. They observed a greater release of cardiac troponin I, high-sensitivity C-reactive protein, and heart-type fatty acid-binding protein in the cardioplegic group. They concluded that OPCAB surgery is associated with less injury.184 It is important to note, however, that efficacy was measured using surrogate markers of protection and the actual incidence of perioperative death and MI were similar in both groups of patients. Finally, whether or not OPCAB surgery is associated with less injury, patients undergoing this form or revascularization are still at risk of ischemic injury.

One approach to minimizing the risk of injury is to reduce myocardial oxygen demand. Pharmacologic beta-blockade is frequently instituted to reduce inotropy and achieve negative chronotropy using ultra-short-acting beta blockers such as esmolol and labetalol. Another approach is to optimize the systemic mean blood pressures while reducing afterload. Calcium channel blockers, such as diltiazem, have been used to afford an effective reduction in blood pressure while minimizing a depression in myocardial contractility that may occur with beta blockers. Patients who become hypertensive during the operation may benefit from intravenous nitrates, which allows for coronary vasodilation and increased blood flow via collaterals. Gentle core cooling, by allowing the body temperature to drift to 35–36°C and deepening the level of anesthesia are concurrent measures that can also be employed.

Ischemic preconditioning is another approach that has been used to impart tolerance to ischemia-reperfusion injury. Although its efficacy has yet to be established in the context of meaningful clinical outcomes, it may be a reasonable approach given the findings by Shroyer et al. These investigators analyzed both short-term and long-term outcomes in 2203 patients randomized to undergo either on-pump or off-pump surgery. The primary endpoints included mortality and complications such as reoperation, need for new mechanical support, cardiac arrest, coma, stroke or renal failure before discharge or within 30 days of surgery. The long-term outcome consisted of the composite endpoint of death from any cause, repeat revascularization procedure, or a nonfatal myocardial infarction within one year. Secondary endpoints included graft patency and neurologic outcomes. Although there were no differences at 30 days, they observed that OPCAB patients had a higher incidence of death from cardiac causes within one year, tended to have lower graft patency and had fewer arteries bypassed. The incidence of stroke or resource utilization was similar for both on-pump and off-pump procedures.185 Although OPCAB can be performed safely, the advantages over on-pump procedures remain to be elucidated. Moreover, the survival findings suggest that this approach to myocardial revascularization does not obviate the need to develop new methods to prevent complications associated with ischemia-reperfusion injury.

Although considerable progress has been made in the field of cardioprotection, the ideal solution, technique, or delivery method has yet to be discovered. This is due, in part, to the complex nature of ischemia-reperfusion injury and our limited understanding of the endogenous mechanisms that exist. Likewise, effective cardioprotection in the operating room can no longer be evaluated in the context of 30-day mortality, as we now know that the injurious consequences of inadequate myocardial protection may not be manifest for months to years after the heart operation.

Mrs. Kristyn Hagood provided research support and was an active participant in the preparation of this manuscript.