Homage to James B. Herrick: A Contemporary Look at Myocardial Infarction and at Sickle-Cell Heart Disease
2000; Lippincott Williams & Wilkins; Volume: 101; Issue: 15 Linguagem: Inglês
10.1161/01.cir.101.15.1874
ISSN1524-4539
Autores Tópico(s)Blood groups and transfusion
ResumoHomeCirculationVol. 101, No. 15Homage to James B. Herrick: A Contemporary Look at Myocardial Infarction and at Sickle-Cell Heart Disease Free AccessOtherPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessOtherPDF/EPUBHomage to James B. Herrick: A Contemporary Look at Myocardial Infarction and at Sickle-Cell Heart Disease The 32nd Annual Herrick Lecture of the Council on Clinical Cardiology of the American Heart Association Thomas N. James Thomas N. JamesThomas N. James From the Division of Cardiology, Department of Medicine, and Department of Pathology, University of Texas Medical Branch, Galveston. Originally published18 Apr 2000https://doi.org/10.1161/01.CIR.101.15.1874Circulation. 2000;101:1874–1887For a clinical cardiologist, I know of no award more appreciated than selection for the Herrick Lecture. I express my personal gratitude to the Council on Clinical Cardiology for adding me to a long list of distinguished predecessors, among whom I realized to my amazement that every individual, living and dead, is one I have cherished as a personal friend. There are too few advantages to growing old, but realizations such as this are surely to be treasured. Ptolemy, who lived near the beginning of the millennium that we are now about to close, wrote, "As material fortune is associated with the properties of the body, so honor belongs to those of the soul."James B. Herrick has become a revered figure not only in cardiology, but also in hematology. In fact, of his 2 historic contributions in medicine, the first came in 1910,1 when he described the unusual sickle-shaped erythrocytes found in a blood smear from an anemic young black Caribbean dental student in Chicago. His more familiar (to us) landmark article about myocardial infarction came 2 years later, in 1912.2 In it, he reported that myocardial infarction was not an inescapable tocsin of doom, but that it was often survived, sometimes with little lasting damage. Even today, there is too little appreciation that the complex causes of fatal myocardial infarction3 include many factors other than simply the occlusion of a coronary artery.Both his sickle-cell article and the one on myocardial infarction were essentially "case reports," a category today too often greeted by editors with rejection and by readers with disdain. But what case reports Herrick's were! In his sickle-cell article, he imaginatively considered every conceivable cause or explanation for the unusual sickle-shaped erythrocytes he saw and, with astonishing acumen, concluded that the explanation must lie in some abnormal intracellular factor that we now know to be a biochemical alteration of the hemoglobin molecule. The one case of myocardial infarction that he reported was autopsied by Herrick's colleague and friend, Ludwig Hektoen, but to those findings, Herrick added numerous personal experiences with similar but surviving patients from his own clinical practice, combined with a comprehensive review and critique of virtually all of the published literature on the subject. Both articles were written in matchless style and have proven to be of continuing value for nearly 100 years.Few clinical scientists have been remembered for not 1 but 2 signal contributions to medical knowledge, and Herrick's achievement in this regard has captured the imagination of clinical scientists ever since. This seemingly surprising accomplishment in 2 segments of medicine now seldom practiced together must be considered in the context of today's medical specialization. In Herrick's day, cardiology and hematology were not thought of as separate fields, and internal medicine itself was not yet established as a professional discipline. As a general physician, Herrick was fascinated early on with the diagnostic value of careful microscopic study of blood smears taken from his patients, which provided important information about anemia, infection, or leukemia, for example. Similarly, it was only after long cogitation about his own observations on myocardial infarction that he summoned the considerable courage to dispute the long-prevailing concept espoused by Cohnheim and others, ie, that myocardial infarction was always fatal. In retrospect, it is disappointing to read that neither the sickle cell nor the myocardial infarction article attracted the immediate attention they both deserved. This may be the reason why Herrick did not continue investigation into what we now know as sickle-cell anemia. However, the initial disinterest of his peers did not dissuade him from a lifelong and imaginatively productive interest in myocardial infarction.Having been elected a member of the American Society of Hematology in 1960, exactly 50 years after Herrick's sickle-cell article was published, and a member of our Council on Clinical Cardiology in 1963, 51 years after Herrick's publication on myocardial infarction, I have faithfully retained an active professional and research interest in both hematology and cardiology. It thus appealed to me as a subject for this Herrick Lecture to take a contemporary look at both sickle-cell anemia and myocardial infarction. On examining what I knew and could find about the knowledge of the 2 subjects during Herrick's professional life, it was quickly apparent that there was much more information and active interest in myocardial infarction than in sickle-cell anemia. Even today, there is far more research on myocardial infarction than on sickle-cell anemia, but knowledge about both conditions has grown immensely since early this century. This opportunity to discuss both conditions may be seen as a harmonizing framework resembling Herrick's own contributions. Accordingly, I will discuss myocardial infarction first and at more length than I will sickle-cell heart disease.A Contemporary Look at Myocardial InfarctionIn my half-century of continuing interest in this subject, I have been fascinated with the evolution and successful application of knowledge obtained by an immense cadre of investigators. Two contemporary subjects very intensively studied are the use of serum markers for diagnosis and ischemic preconditioning (with the related conditions of hibernation or stunning). Both subjects have proven to be important in all levels of ischemic heart disease (or, as also lamentably known, "acute coronary syndromes"). It is my growing belief that much more could be understood about both serum marker diagnosis and ischemic preconditioning if greater attention was directed at the role of apoptosis, rather than almost exclusively on necrosis. These 2 very different forms of cell death are found in all biological tissues and clearly play a role in many aspects of normal and abnormal processes in the human heart,4 including myocardial infarction.Apoptosis and Necrosis in Acute Myocardial InfarctionIt is now clear that in experimental5 as well as clinical6 circumstances, both apoptosis and necrosis participate and each has vast and vastly different influence on the evolution of acute myocardial infarction. Although apoptosis has been aggressively studied by gastroenterologists, hematologists, oncologists, immunologists, developmental biologists, and others now for over a quarter of this century, it is only in the past very few years that much attention has been directed at apoptosis by cardiologists, and the term and concept remain vague for most. For too many generations of pathologists, the presence or absence of necrosis formed the ground rules for saying whether myocardial infarction was or was not present. This rigid dogma mystified and misled too many investigators until it was recognized that a second and quite different form of cell death in the myocardium was due to apoptosis, and this recognition affected the diagnosis of human myocardial infarction.789Apoptosis (sometimes referred to as programmed call death) is a genetically determined form of cell death often serving as an essential component of normal postnatal morphogenesis in the human heart,10 but it also can be triggered by noxious stimuli or become uncontrolled after having begun normally, and then it has pathological consequences.4101112131415 Apoptosis is an impressively rapid process, often completed in seconds or minutes when studied with in vitro preparations.1617 Cytological characteristics (again studied especially in vitro) include a lack of swelling or even shrinkage of cell size, blebbing of the plasmalemma, a distinctive clumping and then cleaving of the nucleus, early disappearance of the nucleolus, and generally a morphological preservation of intracellular organelles, such as mitochondria or sarcoplasmic reticulum, but perhaps most importantly, preservation of an intact plasmalemma. By contrast, necrotic cells swell in size, their mitochondria and sarcoplasmic reticulum quickly disintegrate, the nucleus deteriorates nonspecifically, and the plasma membrane typically ruptures to release all intracellular contents into the interstitial space.The noxious local effect of necrotic myocardial debris is further compounded by the harmful effects of neutrophils themselves, thus jeopardizing the viability of other local myocytes that may have escaped the hypoxic effect of coronary occlusion. As an apoptotic cell dies, there is rapid movement of normally intracellular phosphatidylserine molecules to the external surface of the cell, thereby signaling macrophages for phagocytosis.18 If there is a delay in this phagocytosis, the apoptotic cell (especially its nucleus) breaks into numerous individual apoptotic bodies, each still bound by an intact membrane. In most areas of myocardial apoptosis, the phagocytosis remains effective, creating the histological appearance of an orderly process. However, if (for whatever reason) phagocytosis signals fail or the phagocytic capacity simply becomes overwhelmed, the apoptotic cells and apoptotic bodies then break down and become indistinguishable from necrotic ones.At the light microscopic level, one can readily distinguish myocardial apoptosis from necrosis in human cases of infarction (Figures 1 to 3) and, although they may be distinctly separate foci, they are frequently either contiguous or even intermingled. The separate apoptotic foci are sharply demarcated (Figures 1, 2, 4, and 5) and have a relatively homogeneous appearance, with a mixture of both apoptotic and nonapoptotic myocytes, as well as numerous macrophages containing phagocytosed apoptotic bodies (Figures 5B and 6A). In areas where phagocytic capacity has been exceeded, free pools or sheets of apoptotic bodies can be seen (Figure 6B). By contrast, the boundaries of necrotic foci are seldom sharply demarcated and, within such foci, there is a mixture of broken myocytes and numerous neutrophils and lymphocytes (Figure 3). Hemorrhage is often seen within necrotic foci but is rare in apoptosis. Late in necrosis, there is sometimes a loss of all identifiable myocytes and the creation of areas of "liquefaction necrosis," which is never seen in apoptosis.Many of the original criteria for the morphological diagnosis of apoptosis were based on electron microscopic studies of in vitro preparations. With the later development of an immunohistochemical (TUNEL) method for staining broken strands of DNA within nuclei from paraffin-embedded tissue,19 a new opportunity was presented for accurate recognition of apoptotic cells in situ (Figures 1B, 2B, 4B, and 6A). Other diagnostic methods are available for gel electrophoretic examination and analysis by flow cytometry, but these are not the usual components of most human autopsy studies. In addition to the TUNEL method, which has proved especially useful in my own laboratory, several other histochemical methods (some involving fluorescence) are available to stain apoptotic nuclei. There is also a rapidly growing array of special stains to identify either promoters or inhibitors of apoptosis, but there are now so many of these,202122 it is difficult to decide whether the presence or absence of any one or a few of them is functionally significant in a given human case.There is a useful "experiment of nature" for a comparison with human acute myocardial infarction, the multifocal myocardial degeneration found in fatal cases of thrombotic thrombocytopenic purpura (TTP). These numerous foci include not only the working myocardium, but also any component of the cardiac conduction system (where the foci sometimes cause sudden death due to heart block).152324 Every focus of myocardial degeneration in TTP is due to apoptosis (Figures 7 and 8), and I have never seen necrosis in TTP hearts. The histological appearance of apoptotic foci in TTP (Figure 7A) and in acute myocardial infarction (Figures 1A and 2A) is virtually identical. The ubiquity of apoptosis in the focal myocardial degeneration of TTP may be logically attributed to the smaller areas of ischemia produced by the episodic occlusion of capillaries or small arterioles that is typical of TTP and the better opportunity for macrophages to deal with the dead apoptotic myocytes. Acute myocardial infarction causes much larger areas of ischemia, and necrosis is one of the results.Serum Marker DiagnosisApproximately 50 years ago, it was proposed that certain serum enzymes could be measured as a useful reflection of the presence, and even magnitude, of myocardial infarction in human subjects. This measurement would be based on the release of these normally intracellular enzymes during the necrotic death of cardiac myocytes. One of the earliest serum enzymes measured was glutamic oxaloacetic transaminase and, less often, either glutamic pyruvic transaminase or lactic dehydrogenase.25 The subsequent surge of interest initially focused on the first of these, but it then began to wane because glutamic oxaloacetic transaminase was also released by injured skeletal muscle and the liver. The later parade of candidate enzymes, now more correctly termed markers, included myoglobin, creatine kinase and its subforms and, more recently, 2 forms (T and I) of troponin. The most notable common characteristic of all these markers is that they are normally intracellular substances and they are only released in amounts useful for diagnosis when cardiac myocytes break down. Cardiac myocytes typically rupture during necrosis but not apoptosis, although death by either process leaves the myocytes equally dead.In tandem with the growing national and worldwide focus on myocardial infarction and the unsurprising growth of several multibillion-dollar industries dealing with the need for its more accurate, early, and consistent diagnosis, there has been a persistent expressed skepticism for every proposed serum marker diagnostic approach.26272829 One probable reason for this recurring skepticism has received too little attention, and that is the fact that both apoptosis and necrosis participate importantly in cell death during every myocardial infarction. Furthermore, the major form of death early in murine experimental myocardial infarction is overwhelmingly apoptotic (it is calculated to be 6 times that due to necrosis).5 Accordingly, the reason for recurring disenchantment with serum marker diagnosis may not be, as popularly believed, that the previously hailed serum marker was not sufficiently specific for myocardial cells, but that no intracellular enzyme is released during apoptotic cell death because the plasmalemma characteristically remains intact until the cell is removed by phagocytosis.Regarding the experimental evidence that the earliest form of death of cardiac myocytes after murine coronary ligation is overwhelmingly due to apoptosis,5 a recent clinical study is especially germane. Schuchert et al30 measured troponin levels as possible early diagnostic markers in patients with acute coronary syndromes who were first encountered and who had blood samples drawn during emergency ambulance contact. They found that these early measurements of troponin levels failed to make the correct diagnosis in 90% of patients who were later determined to have ischemic heart disease. Conversely, the same study found that those few cases with significant troponin elevations early in their clinical course had much poorer prognoses, raising the possibility that necrosis found that early was already preceded by apoptosis. These same caveats may be addressed to the consideration of so-called infarctlets found during surgical or invasive medical procedures31 and the probability that those same circumstances are also when apoptosis may be maximal but undetectable by serum marker diagnosis.The sequential fate of apoptotic cells and apoptotic bodies helps explain 2 other aspects of acute myocardial infarction. (1) The roles of apoptosis and necrosis in every human myocardial infarction need not be functionally independent or unrelated. It is plausible to consider that acute myocardial infarction in human subjects always begins with apoptotic cell death, as it does in experimental murine acute myocardial infarction,5 and that any and all necrosis takes place only when the local phagocytic capacity to remove dead apoptotic cells is exceeded (Figure 6B). This may be explained mathematically as too few macrophages available (for whatever reason) or biochemically as the chemotactic signals from the apoptotic cells failed to be generated or were received by the macrophages but, for some reason, misinterpreted. (2) Although the histological appearance of focal apoptosis in human myocardial infarctions is often remarkably discreet and not mixed with necrosis, frequently the apoptotic and necrotic components adjoin or even intermingle. This is unsurprising, especially if one thinks of these as 2 sequential rather than totally independent forms of cell death. Given the variable intermingling of apoptosis and necrosis in every human myocardial infarction,6 it is then to be expected that some apoptotic cells can be recognized within areas of predominant necrosis. This is sometimes proposed as a criticism of histochemical methods that identify DNA strand breakage as unreliable indicators of apoptosis rather than necrosis, but the criticism loses validity if there is often a mixture of the 2 death processes.Finally, in comparative considerations between data from murine experiments and observations in human subjects with ischemic heart disease, it is important to remember certain time constraints. It is simple to time the events after the ligation of a rat coronary artery, but it is now increasingly appreciated that the human events are far more complex, giving rise to the useful description of a "stuttering onset."32 Some of the obvious sources for this human variability include the waxing and waning of coronary spasm, the clumping and then disintegration of platelet aggregates, the morphology of the obstructing atheroma and especially any hinge-like action from it, and also the extracardiac influences at play, such as hypertension or hypotension, respiratory problems, emotional and reflex neural influences, and a wide assortment of pharmacological agents affecting the heart in different ways. Thus, the precisely timed events in the rat experiments serve as a useful frame of reference relative to human events with myocardial ischemia, but it cannot be surprising that the procession of events in humans is seldom so predictable and is sometimes seemingly contradictory.Ischemic PreconditioningIn 1918, Fred Smith33 heeded the suggestion of his mentor, James B. Herrick, and conducted a series of excellent experiments recording the electrocardiographic effects of ligating a coronary artery in anesthetized dogs. It was the most comprehensive and definitive study of its type at that time, and Smith has received too little credit for that work. In addition to anatomical and electrocardiographic correlations, Smith wrote that he was particularly impressed with the special ischemia produced in the posterior papillary muscle of the left ventricle when the left circumflex artery was ligated. Later examination of this same phenomenon by Jennings and colleagues34 has contributed enormously to our current knowledge of the metabolic, biochemical, and anatomic dimensions of acute myocardial infarction.In a landmark study by Murry et al,35 one that should still be the "gold standard" to which virtually any subsequent study of preconditioning must be compared, there were four 5-minute periods of ischemia produced by temporarily occluding the proximal portion of the canine left circumflex coronary artery, each separated by 5 minutes of reperfusion. The surprising discovery was that a later infarct produced (by prolonged occlusion of the left circumflex artery) after such "preconditioning" was only one-quarter the size of an infarct produced without preconditioning. Their provocative conclusion, which quickly caught the attention of others, was their stated suggestion that in humans, multiple episodes of angina may delay cell death and thus allow the salvage of jeopardized myocardium by reperfusion.Regrettably, neither Smith33 nor Jennings et al34 described certain relevant anatomic features of the normal canine left circumflex coronary artery and the subsequent ischemia produced by its occlusion. In the canine heart, the left circumflex artery is the primary blood supply to the atrioventricular (AV) node and, combined with the terminal branches of the large single septal branch of the canine left anterior descending artery, provides the blood supply to the canine His bundle and proximal bundle branches.3637 As a consequence, some (possibly major) alteration of AV conduction is to be anticipated when the dog's left circumflex artery is ligated. This is in considerable contrast to human coronary anatomy, where the predominant blood supply to the AV node and His bundle is from the right coronary artery.3638 Given the very large number of studies done and still being done with ligation of the canine left circumflex artery, I am surprised not to have read of heart block as a complication. Furthermore, this exact same area of myocardial ischemia generates the Bezold-Jarisch reflex,39 which also contributes to various degrees of heart block. One must suspect that the several fatal experiments necessarily discarded in the original study35 and, often, subsequently by others may have been due to lethal electrical instability unamenable to attempted restoration of cardiac rhythm.Another peculiarity of the canine left circumflex artery is the 1 major source of collateral circulation left after it is occluded. When the canine sinus node artery, located >95% from the distal portion of the right coronary artery,36 is cannulated and ligated for experimental selective perfusion of the sinus node,40 it has such copious collateral blood flow from atrial branches of the left circumflex artery that the experiments so conducted can be continued for many hours without any apparent impairment of sinus node function. In fact, in ≈15% of such canine experiments, the retrograde collateral blood flow into the occluded sinus node artery produces a pulsatile pressure equal to central aortic pressure.41 It is apparent that substantial collateral circulation exists in the canine heart connecting the left circumflex and right coronary arteries and that when the left circumflex artery is proximally occluded, variable but often copious collateral flow will come to it by this transatrial route.42 Why this does not more often prevent or at least modulate intense ischemia in the dog's left ventricular posterior papillary muscle (consistently found by both Smith33 and Jennings et al34 ) is puzzling. However, in studies such as those characteristic of experimental ischemic preconditioning by intermittent occlusion and release of the left circumflex coronary artery, transatrial anastomotic flow could be a source of error.A more important problem arises from a contemporary lack of due consideration of the role of apoptosis and its participation in ischemic preconditioning. This is not to say that apoptosis has not been considered at all in this subject, for it has,43444546 but it has been considered primarily in experimental preparations that are both far removed from the classic canine experimental model35 and that often use arguably valid simulations of ischemia for in vitro experiments. What has not been done, to my knowledge, is to examine the role of apoptosis in the results of either repeated intermittent occlusions (or reperfusions) or sustained occlusion of the anesthetized dog's left circumflex artery. As was apparent from my discussion of apoptosis in relation to serum marker diagnosis, the necessary future consideration of apoptosis poses especially relevant questions in the overall concept of ischemic preconditioning, which is now one of the most actively investigated subjects in cardiology. This is driven to some extent by the enormously valuable prospect of finding some biochemical mediator of ischemic preconditioning that could benefit human subjects with coronary disease, not to mention the predictable benefit to the financial assets of the commercial enterprise providing it. There is the vexing possibility that no such substance exists.It is my expectation that it will be unwelcome news to say that alternating ischemia and reperfusion may not actually reduce the volume of myocardium later lost with sustained or long-duration occlusion of the canine left circumflex artery, but that it may reduce only the volume of recognizable (histologically or biochemically) necrosis. Perhaps each of the previous 5-minute periods of occlusion and/or the intervening similar periods of reperfusion had themselves neatly eliminated segments of myocardium by apoptosis. It is already known that brief periods of various forms of injury (ie, hypoxia, irradiation, thermal injury) in a variety of tissues typically produce apoptosis, whereas the same type of injury in the same experimental preparation, presented more intensely or for longer periods of time, leads entirely to necrosis.4748 Similar experiments have been conducted in the kidneys with precisely the same alternating ischemia and reperfusion protocols as in the heart, and they produced apoptosis.49In a recent splendid review exploring the paradox of ischemic preconditioning,50 there is an amazing litany describing work in rats, rabbits, pigs, and dogs and in cell culture, tissue culture, and dispersed cells that examines the biochemical nature of the preconditioning stimulus, as well as possible surrogate alternatives to infarction (eg, arrhythmogenesis, contractile changes, rate of ATP depletion, vascular consequences, and platelet aggregation). This enormous amount of research has still led to frequent conclusions that are in consistent disagreement and controversy and to pleas for further study because "definitive proof is lacking" or the question "remains unresolved." Are we simply learning more and more about less and less?A different but also comprehensive review of ischemic preconditioning51 suggested that the most direct evidence that the human myocardium can be preconditioned comes from studies performed in isolated, cultured human cardiomyocytes subjected to simulated ischemia and reperfusion.52 However, closer examination of that work5354 reveals that the human cardiomyocytes were obtained from biopsies of the right ventricular outflow tract during surgery to repair cases of tetralogy of Fallot, circumstances that could be interpreted as hardly normal myocardium. Furthermore, the following methodological details are especially relevant. The necessary dispersal of biopsied myocytes to be cultured introduces inescapable distortion of these myocytes, including damage to intercellular junctions with various degrees of healing, the necessary elimination of the normal collagenous framework within which normal myocytes function, and the required use of collagenase and trypsin to produce this separation. The culture medium contained bovine serum to which had been added penicillin, streptomycin, and betamercaptoethanol. Ischemia was simulated by placing cells in an acrylic chamber flushed with 100% nitrogen, followed by exposure to low-volume anoxic perfusion (hypoxia, not ischemia), and then 100% nitrogen to reduce the Pco2, with pH adjusted to 7.4 and osmolality corrected with sodium hydroxide and hydrochloric acid. Many of these actions are standard components of certain cell-culture methods, but they are hardly reassuring about the comparability of these in vitro circumstances to either viable or infarcting myocardium. The experimental simulation of canine myocardial ischemia is far more difficult than some may want to think.Two observations on the entire concept of ischemic preconditioning as a putatively beneficial exercise raise some concern about the validity of the original deductions, although one must admire the cautious choice of the term preconditioning in the original work,35 rather than a more implicit or even explicit claim of benefit. (1) Despite the immense amount of research into the subject of ischemic preconditioning by investigators of great talent from many different venues, no one has yet found the elusive responsible substance that might mediate preconditioning. (2) In a recent comprehensive and very informative review50 by investigators especially knowledgeable on the subject, the title emphasized the paradox inherent in ischemic preconditioning. There has been, despite frequent claims of widespread confirmation of the original work by a myriad of investigative teams, little or no examination of the possibility that the penultimate infarction after the preconditioned dog's sustained occlusion of its left circumflex ar
Referência(s)